1 INFO-VAX	Wed, 13 Apr 2005	Volume 2005 : Issue 205       Contents: Re: (history of) BACKUP/ALIAS  Re: (history of) BACKUP/ALIAS  Re: (history of) BACKUP/ALIAS  Re: (history of) BACKUP/ALIAS  Re: (history of) BACKUP/ALIAS  Re: (history of) BACKUP/ALIAS 5 Re: newbie needs help with copying file with only FID 5 Re: newbie needs help with copying file with only FID $ Re: Operating System Security Survey$ Re: Operating System Security Survey$ Re: Operating System Security Survey$ Re: Operating System Security Survey Re: OS Security Poll Re: OS Security Poll Re: OS Security Poll Re: OS Security Poll Re: OS Security Poll Re: Race for dual core 8086s Re: Race for dual core 8086s2 ridiculous GBLPAGES Autogen values under VMS 7.2-26 Re: ridiculous GBLPAGES Autogen values under VMS 7.2-2 Re: RRD43 and CD-RW P Silly Question:  Any easy workarounds of the limitation of <1Gb Boot disks on eaP Re: Silly Question:  Any easy workarounds of the limitation of <1Gb Boot disks o" Re: UAF search for last login date" Re: UAF search for last login date' Unix - VMS compatible encryption method + Re: Unix - VMS compatible encryption method + Re: Unix - VMS compatible encryption method + Re: Unix - VMS compatible encryption method + Re: Unix - VMS compatible encryption method + Re: Unix - VMS compatible encryption method  Re: VMS 8.2 or 7.3-2  F ----------------------------------------------------------------------  + Date: Tue, 12 Apr 2005 17:47:16 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)& Subject: Re: (history of) BACKUP/ALIAS$ Message-ID: <d3h1j3$40q$1@online.de>  C In article <1113267502.831703.137260@l41g2000cwc.googlegroups.com>, ' "AEF" <spamsink2001@yahoo.com> writes:     >  > JF Mezei wrote: 3 > > Phillip Helbig---remove CLOTHES to reply wrote: F > > > tried out all the combinations.  But the HELP says that /NOALIAS > willH > > > IGNORE alias file entries---I don't want them ignored, I want themI > > > copied, not as files, but as aliases.  I can't see why anyone doing  > an* > > > image copy would want anything else. >  > H > What happens is that the data for alias files are ignored. The aliases > ARE copied as aliases.  A OK.  But the combination of /[NO]ALIAS with /IMAGE should not be   allowed.  An IMAGE is an IMAGE.   G > > However, onsider that the alias is really just an extra record in a  > > directory file.  > I > Correct. Using /IMAGE/ALIAS to build a save set will save an extra copy H > of the file (in the save set) per alias entry for that file. This just > wastes time.  I With /NOSAVESET, though, apparently it doesn't make a difference, as far   as I can tell.  F > OK. Even better: Explicitly state that skipping the alias files with- > /NOALIAS will not prevent a proper restore.   G Also, if it is allowed, explain if the behaviour with /IMAGE depens on  9 whether /SAVESET is specified (implicitly or explicitly).    ------------------------------  + Date: Tue, 12 Apr 2005 17:55:41 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)& Subject: Re: (history of) BACKUP/ALIAS$ Message-ID: <d3h22t$40q$2@online.de>  H In article <1113269015.042171.66050@f14g2000cwb.googlegroups.com>, "AEF"! <spamsink2001@yahoo.com> writes:    F > Specifying the /IMAGE qualifier without also specifying /NOALIAS canF > result in incomplete disk or file restoration operations. Therefore,G > Compaq strongly recommends that you specify /NOALIAS with /IMAGE when * > performing image mode backup operations.  @ How's that?  It seems to me that the opposite would be true, if H anything; /ALIAS means to copy stuff which otherwise wouldn't be copied.  F > If you do not specify /NOALIAS, the /ALIAS qualifier is activated by
 > default. > F > [*** OK, at this point I am still confused. Why is it recommended toF > NOT use the default? However, the remainder below makes sense to me. > ***] > F > When you create a save set using /IMAGE and /ALIAS (explicitly or byI > default) in OpenVMS Versions 6.2 and 7.0, BACKUP saves only one copy of G > a file: either the alias file entry or the primary file entry. If the E > primary file entry is not saved in the save set, subsequent restore E > operations for this save set would restore the file using its alias I > entry, causing the file header of the created file to contain the wrong  > file name.  " OK.  What happens with /NOSAVESET?  A I think I understand what happens with /SAVESET.  The wording is  1 confusing, but makes sense once it is understood.   ? I find it really bizarre though that for something so basic as        $  BACK/IMAGE DISK1: DISK2:    H I would need to specify a non-default qualifier (/NOALIAS) to get it to  work.   A Am I correct in assuming that the behaviour might have once been  I non-intuitive, but changed, so that on 7.1 or later (certainly on 7.3 or  F later), BACK/IMAGE DISK1: DISK2: /NOSAVESET will work correctly, even ( though /ALIAS is apparently the default?   ------------------------------    Date: 12 Apr 2005 11:50:43 -0700$ From: "AEF" <spamsink2001@yahoo.com>& Subject: Re: (history of) BACKUP/ALIASC Message-ID: <1113331843.115154.193970@z14g2000cwz.googlegroups.com>   / Phillip Helbig---remove CLOTHES to reply wrote: D > In article <1113269015.042171.66050@f14g2000cwb.googlegroups.com>, "AEF" " > <spamsink2001@yahoo.com> writes: > D > > Specifying the /IMAGE qualifier without also specifying /NOALIAS can = > > result in incomplete disk or file restoration operations. 
 Therefore,D > > Compaq strongly recommends that you specify /NOALIAS with /IMAGE when, > > performing image mode backup operations. > A > How's that?  It seems to me that the opposite would be true, if B > anything; /ALIAS means to copy stuff which otherwise wouldn't be copied.     ) Compaq is gone. Don't worry about it! :-)   ' Really, I can't help you with this one.     E > > If you do not specify /NOALIAS, the /ALIAS qualifier is activated  by > > default. > > E > > [*** OK, at this point I am still confused. Why is it recommended  toD > > NOT use the default? However, the remainder below makes sense to me.  > > ***] > > E > > When you create a save set using /IMAGE and /ALIAS (explicitly or  byC > > default) in OpenVMS Versions 6.2 and 7.0, BACKUP saves only one  copy of E > > a file: either the alias file entry or the primary file entry. If  the G > > primary file entry is not saved in the save set, subsequent restore G > > operations for this save set would restore the file using its alias E > > entry, causing the file header of the created file to contain the  wrong  > > file name. > $ > OK.  What happens with /NOSAVESET?    9 /ALIAS is not relevant to non-save_set  image operations.     B > I think I understand what happens with /SAVESET.  The wording is3 > confusing, but makes sense once it is understood.  > @ > I find it really bizarre though that for something so basic as >   >    $  BACK/IMAGE DISK1: DISK2: > F > I would need to specify a non-default qualifier (/NOALIAS) to get it to > work.  > B > Am I correct in assuming that the behaviour might have once beenG > non-intuitive, but changed, so that on 7.1 or later (certainly on 7.3  orG > later), BACK/IMAGE DISK1: DISK2: /NOSAVESET will work correctly, even   * > though /ALIAS is apparently the default?    A What versions of VMS are you running? Alpha or VAX? Or Integrity?   3 On your save sets: what versions were they made on?     E I'd always use /NOALIAS with /IMAGE when copying files from disk to a F save set. I'd not even bother with /alias or /noalias otherwise exceptG when restoring from those bad save sets from VMS 6.2 and 7.0 and older. F For disk-to-disk image non-saveset operations I can't imagine that anyG use of /alias or /noalias would ever make any difference except perhaps E in performance in which case I'd go with /noalias if that weren't the  default.  A Why don't you try some of these things and tell us what you find?    ------------------------------  + Date: Tue, 12 Apr 2005 19:07:39 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)& Subject: Re: (history of) BACKUP/ALIAS$ Message-ID: <d3h69q$bkl$1@online.de>  C In article <1113331843.115154.193970@z14g2000cwz.googlegroups.com>, ' "AEF" <spamsink2001@yahoo.com> writes:    ; > /ALIAS is not relevant to non-save_set  image operations.   G Is that stated anywhere in HELP or in the documentation?  If it is not  H relevant, then it shouldn't be possible to specify the qualifier at all.  C > What versions of VMS are you running? Alpha or VAX? Or Integrity?   H My "production" stuff is 7.3 VAX and 7.3-1 ALPHA (hopefully upgraded to G 7.3-2 this weekend).  I'm testing some old hardware etc and thus found  I myself making a disk-to-disk non-saveset image copy of a VAX system disk  G with VMS 7.1, which is what sparked my question.  I don't need to read  I old savesets.  I might be doing image backups and restore with /SAVESET,  : either to disk or to tape, but only with VMS 7.3 or newer.  G > I'd always use /NOALIAS with /IMAGE when copying files from disk to a H > save set. I'd not even bother with /alias or /noalias otherwise exceptI > when restoring from those bad save sets from VMS 6.2 and 7.0 and older. H > For disk-to-disk image non-saveset operations I can't imagine that anyI > use of /alias or /noalias would ever make any difference except perhaps G > in performance in which case I'd go with /noalias if that weren't the 
 > default.  * That seems to be what I think is the case.  C > Why don't you try some of these things and tell us what you find?   C Lack of time!  This is actually just peripheral to what I'm really  G concerned with, sorting out old hardware etc.  I usually first want to  G see if a machine works at all, so I boot it.  However, I don't want to  I use the newest VMS, since its firmware might not be up-to-date etc.  For  I example, if I have an old version of VMS on a small disk, I want to boot  B that if I can.  A newer VMS might not fit on that disk, and if it H doesn't boot, I don't know right away what the problem is.  This is all  hardware from 1988--1996.    ------------------------------    Date: 12 Apr 2005 20:14:37 -0700$ From: "AEF" <spamsink2001@yahoo.com>& Subject: Re: (history of) BACKUP/ALIASB Message-ID: <1113362077.945738.53440@o13g2000cwo.googlegroups.com>  / Phillip Helbig---remove CLOTHES to reply wrote: E > In article <1113331843.115154.193970@z14g2000cwz.googlegroups.com>, ( > "AEF" <spamsink2001@yahoo.com> writes: > = > > /ALIAS is not relevant to non-save_set  image operations.  > D > Is that stated anywhere in HELP or in the documentation?  If it is not E > relevant, then it shouldn't be possible to specify the qualifier at  all.    , Don't worry about it. Just don't specify it.  C Look at DIRECTORY: You can specify DIRECTORY/DATE/TOTAL and it will C work, though no dates will appear and it will take much longer than  DIRECTORY/TOTAL.    E > > What versions of VMS are you running? Alpha or VAX? Or Integrity?  > F > My "production" stuff is 7.3 VAX and 7.3-1 ALPHA (hopefully upgraded toB > 7.3-2 this weekend).  I'm testing some old hardware etc and thus found E > myself making a disk-to-disk non-saveset image copy of a VAX system  diskC > with VMS 7.1, which is what sparked my question.  I don't need to  read@ > old savesets.  I might be doing image backups and restore with	 /SAVESET, < > either to disk or to tape, but only with VMS 7.3 or newer. > G > > I'd always use /NOALIAS with /IMAGE when copying files from disk to  a C > > save set. I'd not even bother with /alias or /noalias otherwise  exceptD > > when restoring from those bad save sets from VMS 6.2 and 7.0 and older.F > > For disk-to-disk image non-saveset operations I can't imagine that any C > > use of /alias or /noalias would ever make any difference except  perhaps E > > in performance in which case I'd go with /noalias if that weren't  the  > > default. > , > That seems to be what I think is the case.  > Actually, upon further thought, perhaps /NOALIAS is useful forG incremental backups, but might not be needed if you include /FAST. When G I get time I'll check it out, but I can only do so for VMS 6.1 and 6.2. E One can use /EXCLUDE=[SYS*.SYSCOMMON...] to avoid redundant copies of F common files for incremental saves of system disks, but maybe /NOALIASB would accomplish the same thing? I'll try it when I get some time.  , OK. Here's what I recommend for other cases:  3 For image backup save operations to a save set, use F BACKUP/IMAGE/NOALIAS. If a tape is the target, add /VERIFY, of course!4 Doing incremental saves later, use /RECORD now. Etc.  F For image backup restore operations from a save set, use BACKUP/IMAGE.  F For image backups from disk to disk (no save set) use BACKUP/IMAGE. If? you are worried that this doesn't work, check your results with ( DIRECTORYZZZ/GRAND [*...] on both disks.  C For selective restores from image save sets (using /SELECT), do not E specify /alias -- do not specify /noalias. Check your results. If you E think you are missing files, repeat the operation with /ALIAS (should E only be needed for "older" save sets which you say you don't have, so  don't worry about it!)  2 And my final word of advice: Don't worry about it!    E > > Why don't you try some of these things and tell us what you find?  > D > Lack of time!  This is actually just peripheral to what I'm reallyE > concerned with, sorting out old hardware etc.  I usually first want  toE > see if a machine works at all, so I boot it.  However, I don't want  toE > use the newest VMS, since its firmware might not be up-to-date etc.  For E > example, if I have an old version of VMS on a small disk, I want to  bootC > that if I can.  A newer VMS might not fit on that disk, and if it E > doesn't boot, I don't know right away what the problem is.  This is  all  > hardware from 1988--1996.     C Well, I certainly don't have the time or the resources to try these 2 things for you and I've told you what I *do* know.   ------------------------------  + Date: Wed, 13 Apr 2005 05:20:35 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)& Subject: Re: (history of) BACKUP/ALIAS$ Message-ID: <d3ia72$l0r$1@online.de>  H In article <1113362077.945738.53440@o13g2000cwo.googlegroups.com>, "AEF"! <spamsink2001@yahoo.com> writes:    5 > For image backup save operations to a save set, use H > BACKUP/IMAGE/NOALIAS. If a tape is the target, add /VERIFY, of course!   I use /VERIFY on disks as well!   H > For image backups from disk to disk (no save set) use BACKUP/IMAGE. IfA > you are worried that this doesn't work, check your results with * > DIRECTORYZZZ/GRAND [*...] on both disks.  E I'm pretty sure this would give the same results whatever the /ALIAS  E mechanism used for the restore, as long as all files are there, i.e.  , whether they are just aliases or real files.  E > Well, I certainly don't have the time or the resources to try these 4 > things for you and I've told you what I *do* know.   Thanks!    ------------------------------  # Date: Tue, 12 Apr 2005 17:54:13 GMT ( From: "Hein" <hein.nomail@hp.nomail.com>> Subject: Re: newbie needs help with copying file with only FID2 Message-ID: <9XT6e.3618$hB1.3244@news.cpqcorp.net>  8 "derek pietro" <djpietro@earthlink.net> wrote in message< news:1113327181.630595.87910@z14g2000cwz.googlegroups.com...H > After a bit of work using Heins code chunk to get me started (Thanks !  F > I presume I need to use block IO to ensure the EOF gets written into  F You could indeed. Per $WRITE documentation: "For sequential files, RMSK maintains a logical end of file to correspond to the last block and highest   byte written within the block. "  J However, I would consider using UNDEFINED. RMS will maintain EOF for that,H and it will do buffering (MBC, MBF, RAH, WBH, SQO options). Or, and thisD gets ugly, call the undocumented, unsupported but used by the LinkerF SYS$MODIFY before the last record, and switch to variable or stream or	 whatever. F Or you could QIO the RECATTR. There are plenty examples around, but it remains tricky.    Or... zero out the last bytes?  H > My attempts at block IO on output cause faults ... outrab.rab$l_rop |=
 RAB$V_BIO;  % Ha! Classic C programmer error. RTFM! 3 V_xxx is a bit Vield so use as outrab.rab$v_bio = 1 A M_xxx is a Mask, so use as you did: outrab.rab$l_rop |= RAB$M_BIO    Hein.    ------------------------------    Date: 12 Apr 2005 11:09:29 -0700- From: "derek pietro" <djpietro@earthlink.net> > Subject: Re: newbie needs help with copying file with only FIDC Message-ID: <1113329369.563223.223100@o13g2000cwo.googlegroups.com>   D thanks again - I still suffer the problem of having a dest file that3 has its EOF in the block following the end of data. F The source file somehow has EOF inside its last data record (that is -- inside a block that actually has data in it.)   G So I guess I'm still doing something wrong - but with block set it runs  and doesnt crash.   
 Ideas anyone?    ------------------------------   Date: 12 Apr 2005 18:50:47 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)- Subject: Re: Operating System Security Survey , Message-ID: <3c2jk6F6kvechU2@individual.net>  3 In article <WCV7eWew7S20@eisner.encompasserve.org>, 0 	Kilgallen@SpamCop.net (Larry Kilgallen) writes:K > Sue pointed this out on her non-disclosure mailing list, but since I went I > to the site and saw it with my own eyes, I feel free to point out that:  > # > 	http://www.securitypipeline.com/  >  > offers a survey: > * > 	Which operating system is more secure?  > 
 > 	Windows > 	Linux
 > 	Mainframes  > 	Macs  > 	VMS > 	BSD9 > 	The skill of the administrator matters more than which  > 	platform you use. > I > While administrator skill matters somewhat, you can't make a silk purse  > out of a sow's ear.  > . > It would be a shame if Windows won the vote.  A See my other posting on this subject.  Under the right conditions , there is no reason why Windows couldn't win.   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: Tue, 12 Apr 2005 16:32:57 -0400 ' From: Dave Froble <davef@tsoft-inc.com> - Subject: Re: Operating System Security Survey 0 Message-ID: <115obh8or98rn59@corp.supernews.com>   Bill Gunshannon wrote:5 > In article <WCV7eWew7S20@eisner.encompasserve.org>, 2 > 	Kilgallen@SpamCop.net (Larry Kilgallen) writes: > K >>Sue pointed this out on her non-disclosure mailing list, but since I went I >>to the site and saw it with my own eyes, I feel free to point out that:  >># >>	http://www.securitypipeline.com/  >> >>offers a survey: >>* >>	Which operating system is more secure?  >>
 >>	Windows >>	Linux
 >>	Mainframes  >>	Macs  >>	VMS >>	BSD9 >>	The skill of the administrator matters more than which  >>	platform you use. >>I >>While administrator skill matters somewhat, you can't make a silk purse  >>out of a sow's ear.  >>. >>It would be a shame if Windows won the vote. >  > C > See my other posting on this subject.  Under the right conditions . > there is no reason why Windows couldn't win. >  > bill >   5 Not a chance if we keep packing that ballot box.  :-)    ------------------------------   Date: 12 Apr 2005 23:32:29 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)- Subject: Re: Operating System Security Survey , Message-ID: <3c344dF6kfj47U2@individual.net>  0 In article <115obh8or98rn59@corp.supernews.com>,* 	Dave Froble <davef@tsoft-inc.com> writes: > Bill Gunshannon wrote:6 >> In article <WCV7eWew7S20@eisner.encompasserve.org>,3 >> 	Kilgallen@SpamCop.net (Larry Kilgallen) writes:  >>  L >>>Sue pointed this out on her non-disclosure mailing list, but since I wentJ >>>to the site and saw it with my own eyes, I feel free to point out that: >>> $ >>>	http://www.securitypipeline.com/ >>>  >>>offers a survey:  >>> + >>>	Which operating system is more secure?   >>>  >>>	Windows 	 >>>	Linux  >>>	Mainframes >>>	Macs >>>	VMS  >>>	BSD : >>>	The skill of the administrator matters more than which >>>	platform you use.  >>> J >>>While administrator skill matters somewhat, you can't make a silk purse >>>out of a sow's ear. >>> / >>>It would be a shame if Windows won the vote.  >>   >>  D >> See my other posting on this subject.  Under the right conditions/ >> there is no reason why Windows couldn't win.  >>   >> bill  >>   > 7 > Not a chance if we keep packing that ballot box.  :-)   A And still people here questioned my statement that it was another  useless poll!!!!   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: Tue, 12 Apr 2005 21:16:57 -0400 ' From: Dave Froble <davef@tsoft-inc.com> - Subject: Re: Operating System Security Survey 0 Message-ID: <115os5ninb94l24@corp.supernews.com>   Bill Gunshannon wrote:2 > In article <115obh8or98rn59@corp.supernews.com>,, > 	Dave Froble <davef@tsoft-inc.com> writes: >  >>Bill Gunshannon wrote: >>6 >>>In article <WCV7eWew7S20@eisner.encompasserve.org>,3 >>>	Kilgallen@SpamCop.net (Larry Kilgallen) writes:  >>>  >>> M >>>>Sue pointed this out on her non-disclosure mailing list, but since I went K >>>>to the site and saw it with my own eyes, I feel free to point out that:  >>>>% >>>>	http://www.securitypipeline.com/  >>>> >>>>offers a survey: >>>>, >>>>	Which operating system is more secure?  >>>> >>>>	Windows
 >>>>	Linux >>>>	Mainframes 	 >>>>	Macs  >>>>	VMS >>>>	BSD; >>>>	The skill of the administrator matters more than which  >>>>	platform you use. >>>>K >>>>While administrator skill matters somewhat, you can't make a silk purse  >>>>out of a sow's ear.  >>>>0 >>>>It would be a shame if Windows won the vote. >>>  >>> D >>>See my other posting on this subject.  Under the right conditions/ >>>there is no reason why Windows couldn't win.  >>>  >>>bill  >>>  >>7 >>Not a chance if we keep packing that ballot box.  :-)  >  > C > And still people here questioned my statement that it was another  > useless poll!!!! >  > bill > F Well, yeah, you're 100% right.  It's an OPINION poll, and everyone is I going to vote for their favorite OS.  No, not opinion, it's a popularity  6 poll.  Just like for president.  Any opinions on that?   Dave   ------------------------------   Date: 12 Apr 2005 18:49:25 GMT( From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: OS Security Poll , Message-ID: <3c2jhkF6kvechU1@individual.net>  T In article <DA4AD590CAF06845B671C398333A89C6094ABA01@ohms.electric.ci.austin.tx.us>,2 	"Stuart, Ed" <Ed.Stuart@austinenergy.com> writes:8 > Vote your opinion at: http://www.securitypipeline.com/% > <http://www.securitypipeline.com/>   >   E Yet another meaningless poll.  System Security has absolutely nothing F to do with "opinion".  After all, I have a box running Windows NT thatF is more secure than anything else.  It is not connected to the outsideE world by any fashion, iit is turned off, it is locked inside a Mosler F Safe and even I don't remember the combination.  Let's see some script kiddie break into that one. :-)    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: Tue, 12 Apr 2005 23:06:33 +0200 & From: Paul Sture <paul.sture@decus.ch> Subject: Re: OS Security Poll , Message-ID: <3c2riiF6glkv3U1@individual.net>   Bill Gunshannon wrote:  V > In article <DA4AD590CAF06845B671C398333A89C6094ABA01@ohms.electric.ci.austin.tx.us>,4 > 	"Stuart, Ed" <Ed.Stuart@austinenergy.com> writes: > 8 >>Vote your opinion at: http://www.securitypipeline.com/% >><http://www.securitypipeline.com/>   >> >  > G > Yet another meaningless poll.  System Security has absolutely nothing H > to do with "opinion".  After all, I have a box running Windows NT thatH > is more secure than anything else.  It is not connected to the outsideG > world by any fashion, iit is turned off, it is locked inside a Mosler H > Safe and even I don't remember the combination.  Let's see some script! > kiddie break into that one. :-)  >   7 Oops, if it's turned off, it isn't running anything :-)    ------------------------------    Date: 12 Apr 2005 16:27:11 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)  Subject: Re: OS Security Poll 3 Message-ID: <GetVCt3EvhL5@eisner.encompasserve.org>   W In article <3c2jhkF6kvechU1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes: V > In article <DA4AD590CAF06845B671C398333A89C6094ABA01@ohms.electric.ci.austin.tx.us>,4 > 	"Stuart, Ed" <Ed.Stuart@austinenergy.com> writes:9 >> Vote your opinion at: http://www.securitypipeline.com/ & >> <http://www.securitypipeline.com/>  >>   > G > Yet another meaningless poll.  System Security has absolutely nothing H > to do with "opinion".  After all, I have a box running Windows NT thatH > is more secure than anything else.  It is not connected to the outsideG > world by any fashion, iit is turned off, it is locked inside a Mosler H > Safe and even I don't remember the combination.  Let's see some script! > kiddie break into that one. :-)   D It seems, Bill, that you did not actually look at the survey, as you@ would have seen the last option matches your opinion quite well.   ------------------------------   Date: 12 Apr 2005 23:30:56 GMT( From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: OS Security Poll , Message-ID: <3c341fF6kfj47U1@individual.net>  3 In article <GetVCt3EvhL5@eisner.encompasserve.org>, 0 	Kilgallen@SpamCop.net (Larry Kilgallen) writes:Y > In article <3c2jhkF6kvechU1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes: W >> In article <DA4AD590CAF06845B671C398333A89C6094ABA01@ohms.electric.ci.austin.tx.us>, 5 >> 	"Stuart, Ed" <Ed.Stuart@austinenergy.com> writes: : >>> Vote your opinion at: http://www.securitypipeline.com/' >>> <http://www.securitypipeline.com/>   >>>  >>  H >> Yet another meaningless poll.  System Security has absolutely nothingI >> to do with "opinion".  After all, I have a box running Windows NT that I >> is more secure than anything else.  It is not connected to the outside H >> world by any fashion, iit is turned off, it is locked inside a MoslerI >> Safe and even I don't remember the combination.  Let's see some script " >> kiddie break into that one. :-) > F > It seems, Bill, that you did not actually look at the survey, as youB > would have seen the last option matches your opinion quite well.  3 That option does  not make the poll any more valid.    Take this poll:   0              Have you stopped beating your wife?              ___   Yes              ___   No      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: Tue, 12 Apr 2005 21:26:17 -0400 ' From: Dave Froble <davef@tsoft-inc.com>  Subject: Re: OS Security Poll 0 Message-ID: <115osn97nk4t43e@corp.supernews.com>   Bill Gunshannon wrote:5 > In article <GetVCt3EvhL5@eisner.encompasserve.org>, 2 > 	Kilgallen@SpamCop.net (Larry Kilgallen) writes: > Y >>In article <3c2jhkF6kvechU1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:  >>W >>>In article <DA4AD590CAF06845B671C398333A89C6094ABA01@ohms.electric.ci.austin.tx.us>, 5 >>>	"Stuart, Ed" <Ed.Stuart@austinenergy.com> writes:  >>> : >>>>Vote your opinion at: http://www.securitypipeline.com/' >>>><http://www.securitypipeline.com/>   >>>> >>> H >>>Yet another meaningless poll.  System Security has absolutely nothingI >>>to do with "opinion".  After all, I have a box running Windows NT that I >>>is more secure than anything else.  It is not connected to the outside H >>>world by any fashion, iit is turned off, it is locked inside a MoslerI >>>Safe and even I don't remember the combination.  Let's see some script " >>>kiddie break into that one. :-) >>F >>It seems, Bill, that you did not actually look at the survey, as youB >>would have seen the last option matches your opinion quite well. >  > 5 > That option does  not make the poll any more valid.  >  > Take this poll:  > 2 >              Have you stopped beating your wife? >              ___   Yes >              ___   No  >  >  > bill >   3 False delima!  Logic 101.  (But I failed spelling.)   = The answer is of course no.  Can't stop if you never started.   E Good job folks, VMS is now in first place, ahead of the knowledgable  E administrator, and far ahead of anything else.  Too bad HP marketing  ( won't mention this 'unbiased' poll.  :-)   ------------------------------  # Date: Tue, 12 Apr 2005 17:42:42 GMT % From: "Bob Lail" <Robert.Lail@hp.com> % Subject: Re: Race for dual core 8086s 2 Message-ID: <mMT6e.3616$IJ1.1171@news.cpqcorp.net>  ; "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message  < news:1113265709.5054c9add17bffdd710f2f6a4bd3f22e@teranews...G > AMD announced it would release dual core 64 bit 8086s. Now, Intel has I > leaked news that it may release its dual core 8086 before that, leaving  > IA64 behind in the dust. > J > Another example of how competition in the 8086 space will force Intel toG > bring its 64 bit 8086 up and that it won't be able to keep the market J > spaces for IA64 and 8086 separate for very long. Already IA64 was pushedF > aside for workstation and low/mid range systems, leaving it only for > "big iron" systems.    JF  J I think you are reading that announcement incorrectly. Intel is releasing J their X86 desktop dual core chip early, only in a bid to be first with an M x86 dual core, not an EMT64 Xeon that would compete with Opteron. That's not  J due till next year. Montecito (dual core Itanium II) is still on schedule K for late this summer.  This is about being first, but then IBM (Power), HP  M (PA-RISC) an Sun (UltraSPARC)already have dual core chips out there and have  J for a while, so it's really only smoke and mirror marketing to the masses.  	 \Bob Lail      --  
 Robert G Lail  Pre-Sales Solution Consultant  Corporate Accounts Hewlett-Packard Company 
 Merrimack, NH  TEL: 603.424.6272  CEL: 603.315.0556    ------------------------------  % Date: Tue, 12 Apr 2005 17:20:38 -0400 ( From: Bill Todd <billtodd@metrocast.net>% Subject: Re: Race for dual core 8086s = Message-ID: <GYGdnS1oaoE1psHfRVn-iA@metrocastcablevision.com>    Bob Lail wrote:    ...   7   Montecito (dual core Itanium II) is still on schedule  > for late this summer.   I I think not:  the last dates I've heard (fairly consistently for a while  D now) were to ship sometime in Q4 but with no real volume until 2006.   - bill   ------------------------------    Date: 12 Apr 2005 13:44:00 -0700! From: chessmaster1010@hotmail.com ; Subject: ridiculous GBLPAGES Autogen values under VMS 7.2-2 C Message-ID: <1113338640.689275.276760@l41g2000cwc.googlegroups.com>   A When I have upgraded an Alpha from VMS 7.2-1 to VMS 7.2-2 AUTOGEN D insists on setting GBLPAGES to ridiculously high values, higher than the amount of system memory.  A The following is from AGEN$PARAMS.REPORT on an AS4100 with system ! memory of 2GB (4194304 pagelets).    GBLPAGES parameter information: 4 - AUTOGEN parameter calculation has been overridden.E The calculated value was 50397698. The value 50452146 will be used in + accordance with the following requirements: % GBLPAGES has been increased by 54448. ! GBLPAGES minimum value is 150000.   G If I run AUTOGEN later with feedback it doesn't reduce it; it increases  it more:   GBLPAGES parameter information: C Feedback information. Old value was 50452146, New value is 50577907  Maximum used GBLPAGES: 205216 # Global buffer requirements: 3145728 1 Pagelets reserved for memory resident sections: 0   B Does it hurt anything (i.e. waste memory) to leave GBLPAGES set soA high? Should I manually set it to a reasonable number after every  AUTOGEN?   ------------------------------  # Date: Tue, 12 Apr 2005 23:37:19 GMT # From: Beach Runner <bob@nospam.com>3? Subject: Re: ridiculous GBLPAGES Autogen values under VMS 7.2-2o: Message-ID: <PYY6e.40141$vd.30260@tornado.tampabay.rr.com>  " chessmaster1010@hotmail.com wrote:  C > When I have upgraded an Alpha from VMS 7.2-1 to VMS 7.2-2 AUTOGEN/F > insists on setting GBLPAGES to ridiculously high values, higher than > the amount of system memory. > C > The following is from AGEN$PARAMS.REPORT on an AS4100 with systemm# > memory of 2GB (4194304 pagelets).m > ! > GBLPAGES parameter information:S6 > - AUTOGEN parameter calculation has been overridden.G > The calculated value was 50397698. The value 50452146 will be used in - > accordance with the following requirements:e' > GBLPAGES has been increased by 54448.s# > GBLPAGES minimum value is 150000.A > I > If I run AUTOGEN later with feedback it doesn't reduce it; it increasese
 > it more: > ! > GBLPAGES parameter information:sE > Feedback information. Old value was 50452146, New value is 50577907< > Maximum used GBLPAGES: 205216f% > Global buffer requirements: 3145728l3 > Pagelets reserved for memory resident sections: 0o > D > Does it hurt anything (i.e. waste memory) to leave GBLPAGES set soC > high? Should I manually set it to a reasonable number after everye
 > AUTOGEN? > D It's not ridiculous.  It only has pointers to the page and won't be . actually allocated unless the system needs it.   ------------------------------    Date: 12 Apr 2005 11:44:19 -0700 From: jordan@ccs4vms.com Subject: Re: RRD43 and CD-RWB Message-ID: <1113331459.728813.52300@f14g2000cwb.googlegroups.com>  F I used an RRD43 on my desktop VS3100-30 until a few months ago when anG RD45 became available.  I also use a Yamaha 4416 CDR/CDRW drive on thata+ box, and a Yamaha F1 CDRW on my home Alpha.z  G The RRD43 could read every disk I burned on the 4416, and all the diskseG burned on the F1 at lower speed.  It had some trouble with disks burned E on the F1 at max media speed (16X).  That is, until I ran out of Sony @ and TDK 16X media.  That media had a blue tint on the data side.  C Current media from both vendors is silver on the bottom.  The RRD43tE could not read any of these disks burned on either burner.  The RRD45R? can read them just fine.  Other sources have indicated that thetG 'cheaper' and more common dye types provide the silver color (and don't,D last as long for archival use), and that the blue (especially deeper2 blue) and gold dyes are older/better/more durable.  G Even an old RRD42 was able to read the old blue Sony disks burned at 1Xe( or 2X.  No go at all on the newer media.   Rich   ------------------------------  % Date: Wed, 13 Apr 2005 13:23:35 +1000sG From: "Christine Ricketts/Andrew Stewart" <christiner@uxnxixtxe.com.au>nY Subject: Silly Question:  Any easy workarounds of the limitation of <1Gb Boot disks on eal, Message-ID: <425c9115_1@news.iprimus.com.au>   Greetings Oh VMS Wizards,   9 I am refurbishing two MicroVAX 3100s (model DV-31BT1-A01)-5 for a mate who lives far from the bright city lights.C9 One to use and another for a complete set of spare parts.-  9 RZ23 (100MB) disks were being swapped for RZ28s (2Gb) butI: I couldn't boot from the latter.  From memory, at power up3 they print KA42?? V1.4 then count down from F to 0.E: When I typed in BOOT DKAnnn, they echoed back -DKAnnn then> printed out an error message.  Sorry, forgot to write it down.  > I dragged out later MicroVAX 3100s (a model 80 and a model 40)) which happily booted from the same RZ28s.N< From the depths of my memory came the information that early6 MicroVAXen boot ROMs could not handle >1Gb boot disks.  9 I can get around this problem fairly easily by scroungingh9 500MB SCSI disks, either 3.5 inch internally or 5.25 inchN: RZ5x's externally in SZ12s, but the evil thought struck me$ that perhaps there were others ways.  3 1/ Were there updated Boot ROMs for these machines?l  @ 2/ From "Version 4.4 VAX/VMS Internals and Data Structures" 1988> p669 "VMB can prompt for the name of any standalone program toD be loaded into memory" and "VMB can load the contents of a bootstrap block from the system disk".  C Is there an easy way to boot from an RZ23 with a minimum VMS systemtA then immediately reboot to an RZ28?  The RZ23 wouldn't be a wastef: as it could have a swap and/or page file(s), or user data.  / Apologies in advance if these are stupid ideas.t   -- Regards, Andy.  ( NB.  Remove eXcess if replying by email.  3 Blessed Are The Cracked, For They Let In The Light.E Attributed to Spike Milligan.,   ------------------------------  % Date: Tue, 12 Apr 2005 23:22:47 -0500 / From: Chris Scheers <chris@applied-synergy.com>AY Subject: Re: Silly Question:  Any easy workarounds of the limitation of <1Gb Boot disks ot2 Message-ID: <425C9E97.4040709@applied-synergy.com>  ( Christine Ricketts/Andrew Stewart wrote: > Greetings Oh VMS Wizards,o > ; > I am refurbishing two MicroVAX 3100s (model DV-31BT1-A01)/7 > for a mate who lives far from the bright city lights.p; > One to use and another for a complete set of spare parts.e > ; > RZ23 (100MB) disks were being swapped for RZ28s (2Gb) but>< > I couldn't boot from the latter.  From memory, at power up5 > they print KA42?? V1.4 then count down from F to 0.l< > When I typed in BOOT DKAnnn, they echoed back -DKAnnn then@ > printed out an error message.  Sorry, forgot to write it down. > @ > I dragged out later MicroVAX 3100s (a model 80 and a model 40)+ > which happily booted from the same RZ28s. > > From the depths of my memory came the information that early8 > MicroVAXen boot ROMs could not handle >1Gb boot disks. > ; > I can get around this problem fairly easily by scroungingi; > 500MB SCSI disks, either 3.5 inch internally or 5.25 inchl< > RZ5x's externally in SZ12s, but the evil thought struck me& > that perhaps there were others ways. > 5 > 1/ Were there updated Boot ROMs for these machines?e > B > 2/ From "Version 4.4 VAX/VMS Internals and Data Structures" 1988@ > p669 "VMB can prompt for the name of any standalone program toF > be loaded into memory" and "VMB can load the contents of a bootstrap > block from the system disk". > E > Is there an easy way to boot from an RZ23 with a minimum VMS systemeC > then immediately reboot to an RZ28?  The RZ23 wouldn't be a wasten< > as it could have a swap and/or page file(s), or user data. > 1 > Apologies in advance if these are stupid ideas.l  D The problem is that the ROMs in early VAXstation 3100s and MicroVAX G 3100s used SCSI commands that can only address the first 1GB of a disk.f  F If any of the boot files are beyond the first 1GB, these commands can # not locate the correct disk blocks.n  G I don't see anyway to reboot to an RZ28 as the code that wants to boot t5 the RZ28 is still going to use the ROM based drivers.e  D I think this problem affects all VAXstation 3100s and MicroVAX 3100 + models 10 and 20 and (I think) 10e and 20e.d  F AFAIK, updated ROMs were never released for the VAXstation 3100s, but G I've been told that ROMs were available for the MicroVAXes.  I have no e& idea how to get ahold of them, though.  I If you can ensure that all boot files are in the first 1GB of the drive, nH things should work.  But beware of any patches or upgrades that replace  any of these files!1  F Also, SYSDUMP.DMP must be in the first 1GB as I think the ROM drivers + are used to write this file during a crash.a  
 Good luck!   --  G ----------------------------------------------------------------------- $ Chris Scheers, Applied Synergy, Inc.  B Voice: 817-237-3360            Internet: chris@applied-synergy.com    Fax: 817-237-3074   ------------------------------  % Date: Tue, 12 Apr 2005 11:53:26 -07002, From: Ken Fairfield <my.full.name@intel.com>+ Subject: Re: UAF search for last login datey+ Message-ID: <d3h5f6$ir7$1@news01.intel.com>n   flamingomn@hotmail.com wrote:p > hi.n > . > i'm trying not to 'reinvent the wheel' here.F > i'm trying to find a DCL .cOM that will read the sysuaf.lis and pullF > out users that haven't logged in in xdays.  the xdays should be a P1 > parameter. > # > i've got an ALPHA with OVMS7.3-2.EI > i've copied the SCANUAF from decus and was wondering if that would workf > in my enivironment?-  F      There's also Joe Meadow's UAF utility, also available from HunterD   Goatley's file server on www.process.com.  UAF is written C, FWIW,E   and the interface is kinda' awkward, but it does the job.  I use itaI   as the core utility for excatly the sort of thing you're after: lookingtH   for user accounts that haven't been logged in within the last xx days,   etc.   	-Kena -- u6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken FairfieldX! D1C Automation VMS System Supportd" who:   kenneth dot h dot fairfield where: intel dot com   ------------------------------  % Date: Tue, 12 Apr 2005 19:52:24 -0600f( From: "r.m" <nospam@rmendoza.copper.net>+ Subject: Re: UAF search for last login date + Message-ID: <425c7bd8_5@newsfeed.slurp.net>   
 r.m wrote:   > flamingomn@hotmail.com wrote:i >  >> hi. >>/ >> i'm trying not to 'reinvent the wheel' here.eG >> i'm trying to find a DCL .cOM that will read the sysuaf.lis and pull G >> out users that haven't logged in in xdays.  the xdays should be a P1g
 >> parameter.  >>$ >> i've got an ALPHA with OVMS7.3-2.J >> i've copied the SCANUAF from decus and was wondering if that would work >> in my enivironment? >  > K > Is scanuaf written in BASIC?  If so then it is probably the same program dJ > I have been using since about 6.something maybe 7.something, and I last H > used it on a 7.3-1 system a little over a year ago.  Haven't tried it K > since we upgraded to 7.3-2 but I don't see why it wouldn't work.  Not at  0 > work at the moment so can't take a look at it.   Responding to my own post:  B The utility that I have used is called sysuaf, not scanuaf, it is , written in  BASIC and dates from about 1994.   Rene.t   ------------------------------    Date: 12 Apr 2005 16:06:58 -0700 From: jordan@ccs4vms.com0 Subject: Unix - VMS compatible encryption methodC Message-ID: <1113346772.925890.298660@l41g2000cwc.googlegroups.com>h  9 Got two servers on opposite sides of the internet; secure F communications between them is not currently possible.  We want to addG file encryption capabilities so that only encrypted files are sent backSF and forth (probably via FTP).  The catch:  while one server is genericF *BSD of recent vintage, the VMS Alpha system is stuck at V7.1 (and UCXE V4.2) with no immediate upgrade prospects.  Simple obfuscation of the@C data is not sufficient; it needs to be encrypted, though it doesn'tmB look like a public key setup is required since a secret key can be& safely set up and secured at each end.  E The available encryption tools I've found so far (GnuPG and some PerloE tools) seem to want VMS V7.2-1 or later (some much later).  I haven'tn? found if older versions of Perl that ran on V7.1 might have the G libraries to do encryption; ditto with GnuPG which doesn't seem to have ' been ported to such an old VMS release.h  G I know that an old PGP (V2.x?) was available on V6.x VMS versions, evenAE VAX.  Is it possible that if we could get the old PGP on the VMS sideo/ that GnuPG on the unix box would be compatible?   - Thanks for any info or alternate suggestions.s   Rich   ------------------------------    Date: 12 Apr 2005 17:06:36 -0700 From: bob@instantwhip.comI4 Subject: Re: Unix - VMS compatible encryption methodC Message-ID: <1113350796.087696.315010@o13g2000cwo.googlegroups.com>   + we run pgp on vms 7.1 ... people we send toc) run gpg and can denecrypt on both windozet and unix ...   ------------------------------  % Date: Tue, 12 Apr 2005 20:29:54 -0400 # From: "John Smith" <a@nonymous.com>74 Subject: Re: Unix - VMS compatible encryption method, Message-ID: <NcmdnZzbAbHj98HfRVn-tA@igs.net>   jordan@ccs4vms.com wrote: ; > Got two servers on opposite sides of the internet; secureeH > communications between them is not currently possible.  We want to addD > file encryption capabilities so that only encrypted files are sentE > back and forth (probably via FTP).  The catch:  while one server ismG > generic *BSD of recent vintage, the VMS Alpha system is stuck at V7.1o= > (and UCX V4.2) with no immediate upgrade prospects.  SimpleCF > obfuscation of the data is not sufficient; it needs to be encrypted,D > though it doesn't look like a public key setup is required since a: > secret key can be safely set up and secured at each end. >eG > The available encryption tools I've found so far (GnuPG and some PerltG > tools) seem to want VMS V7.2-1 or later (some much later).  I haven'tEA > found if older versions of Perl that ran on V7.1 might have thetD > libraries to do encryption; ditto with GnuPG which doesn't seem to. > have been ported to such an old VMS release. >aD > I know that an old PGP (V2.x?) was available on V6.x VMS versions,G > even VAX.  Is it possible that if we could get the old PGP on the VMS 6 > side that GnuPG on the unix box would be compatible? > / > Thanks for any info or alternate suggestions.C     No VPN or stunnel?   --F OpenVMS - The never advertised operating system with the dwindling ISV base.e   ------------------------------  % Date: Tue, 12 Apr 2005 18:16:27 -0700b From: Z <Z@no.spam>n4 Subject: Re: Unix - VMS compatible encryption method* Message-ID: <Kp_6e.3037$OU6.2636@fe02.lga>   jordan@ccs4vms.com wrote:k; > Got two servers on opposite sides of the internet; secureyH > communications between them is not currently possible.  We want to add   Can't run SSH?  E http://vms.process.com/ftp/ssh/docs/%09%09%09%09ssh_release_notes.txte  F Operating System and Version: OpenVMS VAX V5.5-2, 6.2, 7.0, 7.1, 7.2, @ 7.3, OpenVMS Alpha V6.2, 7.0, 7.1, 7.2, 7.2-1, 7.2-2, 7.3, 7.3-1  1 UCX Version:                  V4.0 ECO5 and latera    ' See also: http://freessh.org/other.htmln    I > file encryption capabilities so that only encrypted files are sent back H > and forth (probably via FTP).  The catch:  while one server is generic   SSH has scp and sftp.    ------------------------------  % Date: Tue, 12 Apr 2005 21:32:13 -04001' From: Dave Froble <davef@tsoft-inc.com>e4 Subject: Re: Unix - VMS compatible encryption method0 Message-ID: <115ot2bdngk1ha1@corp.supernews.com>   jordan@ccs4vms.com wrote:$; > Got two servers on opposite sides of the internet; secure0H > communications between them is not currently possible.  We want to addI > file encryption capabilities so that only encrypted files are sent back H > and forth (probably via FTP).  The catch:  while one server is genericH > *BSD of recent vintage, the VMS Alpha system is stuck at V7.1 (and UCXG > V4.2) with no immediate upgrade prospects.  Simple obfuscation of thesE > data is not sufficient; it needs to be encrypted, though it doesn'thD > look like a public key setup is required since a secret key can be( > safely set up and secured at each end. > G > The available encryption tools I've found so far (GnuPG and some PerliG > tools) seem to want VMS V7.2-1 or later (some much later).  I haven'taA > found if older versions of Perl that ran on V7.1 might have the-I > libraries to do encryption; ditto with GnuPG which doesn't seem to havem) > been ported to such an old VMS release.d > I > I know that an old PGP (V2.x?) was available on V6.x VMS versions, evenpG > VAX.  Is it possible that if we could get the old PGP on the VMS sidee1 > that GnuPG on the unix box would be compatible?. > / > Thanks for any info or alternate suggestions.t >  > Rich >   H I don't know whether it will work with the versions you're running, but 6 have you looked at the products from Process Software?  H There's also router based VPN, and another poster has suggested VPN and  stunnel.  - Why no upgrade prospects?  7.1 is rather old..   Dave   ------------------------------  % Date: Tue, 12 Apr 2005 23:33:27 -0500n6 From: "Craig A. Berry" <craigberry@mac.com.spamfooler>4 Subject: Re: Unix - VMS compatible encryption methodD Message-ID: <craigberry-FAE9D3.23332712042005@news.isp.giganews.com>  C In article <1113346772.925890.298660@l41g2000cwc.googlegroups.com>,T  jordan@ccs4vms.com wrote:  G > The available encryption tools I've found so far (GnuPG and some PerlyG > tools) seem to want VMS V7.2-1 or later (some much later).  I haven't$A > found if older versions of Perl that ran on V7.1 might have thee > libraries to do encryption  G What Perl tools in particular?  If you're talking about Digest::MD5, I  D believe it became part of the Perl core in 5.8.0.  I use Perl 5.8.4 C (5.8.6 is current) on OVMS Alpha v7.1 on an almost daily basis.  I nH should be clear that it's Perl 5.8.4 that I run regularly on OVMS v7.1, H not Digest::MD5, though IIRC Digest::MD5 built and passed all its tests  without any problems.n  @ It's probable that binary kits of Perl recent enough to include H Digest::MD5 are not readily available, though it can easily be added to F any installation of Perl if you don't mind building it yourself.  Are  you lacking a C compiler?b   ------------------------------  # Date: Tue, 12 Apr 2005 17:38:35 GMT-& From: John Reagan <john.reagan@hp.com> Subject: Re: VMS 8.2 or 7.3-2c2 Message-ID: <vIT6e.3614$%P1.2158@news.cpqcorp.net>    Peter 'EPLAN' LANGSTOEGER wrote:   > / > 	states BASIC V1.5A valid for V6.2 til V7.3-2W0 > 	states DECset V12.5 valid for V7.3 til V7.3-2 > F > What word is the most official one ? Service ? Support ? Marketing ?F > I personally can live unsupported, but if we have a critical problemA > it might be too late to discuss the supported/unsupported issueyB > with service then (I don't however expect critical problems withC > development tools, so exactly this discussion here might be moot)  >  > G > It would be good, if there were a support matrix the other way round,pK > means which versions (note the plural) of a product are supported/working   > on a specific OpenVMS version. >   H The "problem" with convincing the product management folks of providing @ the extra information is the difference between "supported" and G "working".  I've exchanged several emails with my product manager this | morning on this topic.  B Given the schedules, I actually never fully tested Pascal V5.8 on F OpenVMS V8.2.  Without that full qualification cycle, she is hesitant D about claiming "support".  Her definition of "support" means "we've A fully tested it and it works as expected".  Other definitions of FH "support" might be "we strongly think it will work and if it doesn't we I will make it work".  Is it work my time to go back and fully test 7 or 8 tE different versions of Pascal (or any other layered product) on newer nD versions of OpenVMS?  It is even more complicated with Pascal since F fresh installations might require a newer version for STARLET.PAS but 2 the compiler portion would install/work just fine.  C I'll keep emailing her to see if there is some way to provide more > information.   -- s John Reagan / HP Pascal/{A|I}MACRO for OpenVMS Project Leader  Hewlett-Packard CompanyD   ------------------------------   End of INFO-VAX 2005.205 ************************