1 INFO-VAX	Sat, 16 Oct 2004	Volume 2004 : Issue 575       Contents:  Re: DFU for Itanium, or sources?% Re: DS10L SCSI card - recommendations % Re: DS10L SCSI card - recommendations " ERROR: unable to write header file& Re: ERROR: unable to write header file# Re: Latest on Windoze Navy software # Re: Latest on Windoze Navy software # Re: Latest on Windoze Navy software # Re: Latest on Windoze Navy software  Re: OT: Sun's fighting chance " Re: SYSUAF => export passwords ??? VMS for the blind  Re: VMS for the blind  Re: VMS for the blind   F ----------------------------------------------------------------------  + Date: Sat, 16 Oct 2004 15:38:30 +0000 (UTC) 6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)) Subject: Re: DFU for Itanium, or sources? 0 Message-ID: <newscache$y9oo5i$ity$1@news.sil.at>  l In article <1097847440.229214.299170@z14g2000cwz.googlegroups.com>, "Galen" <gspamtackett@yahoo.com> writes:F >I have searched the web for a DFU version for the Itanium but haven'tG >turned up anything. Is there one available anywhere? Or alternatively,   >are sources for V3.0 available?  N DFU V3.1 (coming soon with OpenVMS V8.2) will contain the Itanic version, too.D As you might already know, support for the VAX ended with DFU V2.7A.F Ton DORLAND (prename.surname@hp.com) didn't release the sources so you. won't find them (except you are inside HPQ)...   --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Sat, 16 Oct 2004 18:31:36 +1000 # From: "Gremlin" <not-here@all.mate> . Subject: Re: DS10L SCSI card - recommendations- Message-ID: <4170dc67@duster.adelaide.on.net>   J Thanks for that.  I need something from which I can boot as the DS10L has C only room for 1 disk if you want to keep the CD/floppy, which I do.   < "Joseph Huber" <huber@NOBODY-mppmu.mpg.de> wrote in message % news:OkjF7Fcf67IZ@vms.mppmu.mpg.de... : > In article <4170928a@duster.adelaide.on.net>, "Gremlin"  > <not-here@all.mate> writes: I >> Rather than using IDE (perhaps slow as in an earlier thread), could I   >> have G >> recommendations for a SCSI card for a 36/72Gb disk - third party if   >> possible L >> as I want to get one in the next day or so and can't access "real" Compaq >> cards.  Adaptec?  > I > I can't tell if it works in a DS10L, but in a XP1000 I have an Adaptec   > 29160 D > (a single port UW-160 card) working with VMS 7.3, after adding the2 > configuration in sys$system:sys$user_config.dat: >  > ! Adaptec SCSI card 29160 # > device          = "Adaptec 29160"  >  name          = PK $ >  driver        = sys$pkadriver.exe >  adapter       = PCIH >  id            = 0x00809005      !ID may vary, check with pci_bus_list >  boot_class    = DK & >  boot_flags    = HW_CTRL_LTR, UNIT_0# >  flags         = SCSI, PORT, BOOT  > end_device > : > Unsupported, but much cheaper than a supported HP cards.+ > (Booting via this interface not tested !)  >  > --  8 >   Joseph Huber, Muenchen  http://www.huber-joseph.de/    ------------------------------    Date: 16 Oct 2004 10:29:44 +0200. From: huber@NOBODY-mppmu.mpg.de (Joseph Huber). Subject: Re: DS10L SCSI card - recommendations+ Message-ID: <OkjF7Fcf67IZ@vms.mppmu.mpg.de>   S In article <4170928a@duster.adelaide.on.net>, "Gremlin" <not-here@all.mate> writes: M > Rather than using IDE (perhaps slow as in an earlier thread), could I have  O > recommendations for a SCSI card for a 36/72Gb disk - third party if possible  L > as I want to get one in the next day or so and can't access "real" Compaq  > cards.  Adaptec?  L I can't tell if it works in a DS10L, but in a XP1000 I have an Adaptec 29160B (a single port UW-160 card) working with VMS 7.3, after adding the0 configuration in sys$system:sys$user_config.dat:   ! Adaptec SCSI card 29160 ! device          = "Adaptec 29160"    name          = PK#   driver        = sys$pkadriver.exe    adapter       = PCI G   id            = 0x00809005      !ID may vary, check with pci_bus_list    boot_class    = DK%   boot_flags    = HW_CTRL_LTR, UNIT_0 "   flags         = SCSI, PORT, BOOT
 end_device  8 Unsupported, but much cheaper than a supported HP cards.) (Booting via this interface not tested !)    --  6    Joseph Huber, Muenchen  http://www.huber-joseph.de/   ------------------------------  % Date: Sat, 16 Oct 2004 20:08:25 +1000 # From: "Gremlin" <not-here@all.mate> + Subject: ERROR: unable to write header file - Message-ID: <4170f319@duster.adelaide.on.net>   I VMS 7.2-2, DS10L, 20Gb Seagate IDE drive, standard DQdriver, boots fine,  I works well, using about 12Gb, when I shutdown, after stopping all queues  2 etc. and then a long time at a blank screen I get:  : "Error writing 2 blocks, buffer = FFFFFFFF.808B1E00, VBN 18  Status code 00000054, 00000000 (Section = Header block)  6 Unable to write header, sump will probably be unusable  : Error writing 32 blocks, buffer = FFFFFFFF.809A0000, VBN 15 Status code 00000054, 00000000 (Section = Error logs)   5 SYS$ERRLOG.DMP is to small to save error log buffers"   M So, is the first message indicating that VMS is bitching about a disk bigger  I than 2Gb?  Or is there something else??  I have already run autogen with  K testfiles and genparams so I know that it has calculated the sizes to it's  % needs for the dump file and the rest.     All ideas appreciated - thanks.    ------------------------------  % Date: Sat, 16 Oct 2004 13:48:20 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> / Subject: Re: ERROR: unable to write header file , Message-ID: <41715EE2.AC6EA923@teksavvy.com>   Gremlin wrote: > , > VMS 7.2-2, DS10L, 20Gb Seagate IDE drive, < > "Error writing 2 blocks, buffer = FFFFFFFF.808B1E00, VBN 1: >  Status code 00000054, 00000000 (Section = Header block)< > Error writing 32 blocks, buffer = FFFFFFFF.809A0000, VBN 17 > Status code 00000054, 00000000 (Section = Error logs)  > N > So, is the first message indicating that VMS is bitching about a disk bigger > than 2Gb?   J I doubt that a DS10 with VMS 7.2-2 would have problems with a 2 gig disk.   K Seems to me more of a problem with the drive or the way it was initialized.   ; If you do SHOW DEV D/FULL does it show an error count > 0 ?   M Can you get to the $ sign ? If you try $PURGE device:[000000...]*.*/log  what 
 does it say ?   M If the physical device is healthy, it could be something in the way the drive < was initialized, with an indexf.sys perhaps not big enough.   @ SHOW DEV disk:/FULL tells you the maximum number of files on it.P $DIR dev:[000000...]*.*;*/grand  will tell you how many files are on your drive.   ------------------------------   Date: 16 Oct 2004 07:08:03 GMT2 From: "Dave Weatherall" <djw-nothere@nospam.nohow>, Subject: Re: Latest on Windoze Navy software? Message-ID: <DTiotGxQ0bj6-pn2-0eYcZdZMNjgd@dave2_os2.home.ours>   ? On Fri, 15 Oct 2004 02:32:30 UTC, Kilgallen@SpamCop.net (Larry   Kilgallen) wrote:   p > In article <2t8isdF1pp61rU1@uni-berlin.de>, Tim Sneddon <first-initiallastname@bsddotinfomedia.com.au> writes: > J > > For languages that don't have string data types (C, BLISS, MACRO) it'sK > > just a matter of convention. Most CRTL functions expect null-terminated & > > and so this has become the "norm". > : > Perhaps for those who use C, but not for the rest of us.  ? Well CP/M had the $-terminated string (can"t remember the BDOS  3 function number) , which always made me wonder :-).   8 Descriptor's aren't fool-proof BTW. They are safer than C null-terminated strings but they can be abused or damaged by other  
 code errors.     --   Cheers - Dave.   ------------------------------   Date: 16 Oct 2004 07:08:05 GMT2 From: "Dave Weatherall" <djw-nothere@nospam.nohow>, Subject: Re: Latest on Windoze Navy software? Message-ID: <DTiotGxQ0bj6-pn2-X9sQCGhGd1PC@dave2_os2.home.ours>   " On Fri, 15 Oct 2004 13:35:18 UTC, < koehler@eisner.nospam.encompasserve.org (Bob Koehler) wrote:  ^ > In article <416EF105.ED2B8A9D@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes: > > Tom Wade wrote: I > >> Yes, you can compensate for this with good programming practice, but O > >> interfaces using string descriptors were always a much safer way to do it.  > > 0 > > Does any language use descriptors natively ? > F >    The Fortran, and I believe COBOL, compilers for VMS certainly do.E >    Using descriptors in Macro-32 and -64 is more a matter of choice F >    than difficulty, since both have .ascid and can easily manipulate, >    descriptors in one or two instructions. > I >    Fortran CHARACTER type (which is really a string type) is passed by  I >    descriptor, this both meets behaviour specified by the ANSI Fortran  M >    standard (required behaviour implies knowing the location and allocated  G >    size of CHARACTER variables) and make it easy to interface to VMS  I >    routines.  I believe some other Fortran compilers pass CHARACTER as  : >    two arguments in order to meet the required behavior. > H >    I've never really understood why DEC decided to use counted stringsI >    instead of descriptors in PASCAL, other than I think that's what the F >    reference model uses.  I think all PASCAL requirements could haveE >    been met if strings were passed by descriptor, and it would have F >    made the interface to VMS routines as simple as it is in Fortran.  D I had the same thought on the few occasions I have had to interface D Fortran with Pascal. The resulting Pascal prototypes look awful (or E awe-fully complex)  but they do work.  I always had the feeling that  F the counted string was a way of dealing with allocated/reserved length v current length.    --   Cheers - Dave.   ------------------------------   Date: 16 Oct 2004 13:40:58 GMT( From: bill@cs.uofs.edu (Bill Gunshannon), Subject: Re: Latest on Windoze Navy software+ Message-ID: <2tcmnaF1tsscfU1@uni-berlin.de>   3 In article <rqbG5FVHC1UK@eisner.encompasserve.org>, 0 	Kilgallen@SpamCop.net (Larry Kilgallen) writes:X > In article <2ta684F1snbanU1@uni-berlin.de>, bill@cs.uofs.edu (Bill Gunshannon) writes:6 >> In article <+DHyy5CYV6xG@eisner.encompasserve.org>,3 >> 	Kilgallen@SpamCop.net (Larry Kilgallen) writes: Z >>> In article <2t8h0sF1tmvr6U2@uni-berlin.de>, bill@cs.uofs.edu (Bill Gunshannon) writes: >>> D >>>> Since I have been at this University I have watched the starterD >>>> language go from Pascal to Ada to Java with later courses beingC >>>> done using Modula, C, C++.  The students also get a smattering D >>>> of COBOL.  You know what.  The beginners make the same mistakes= >>>> using Java that ehy made using Ada and Pascal before it.  >>> D >>> Except that in higher level languages those errors are caught at& >>> compile-time rather than run-time. >>  G >> Well, I didn't say what those errors were. but as you brought it up. G >> I have yet to see a compiler that can determine at compile time that H >> a bad "for" loop or an index going negative is going cause an out-of-I >> bounds array reference.  Yes, that is one of the most common mistakes.  > A > It may be a common mistake in C, but not in Ada where one says:  > 4 >     for EACH_CHAR in reverse CHAR_ARRAY'range loop, >         if CHAR_ARRAY ( EACH_CHAR ) /= ' 'B >             return CHAR_ARRAY ( CHAR_ARRAY'first .. EACH_CHAR ); >         end IF;  >     end loop;  >     return CHAR_ARRAY;  @ Where did I say they were looking for a character in a character@ array?  There are numerous other cases where one uses a for loopA to wander through an array.  Getting it wrong is a common student B mistake.  It's a logic error and as such because things like startC -point and end-point are determined at runtime there is no way that $ the compiler can possibly know them.   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: 16 Oct 2004 13:45:33 GMT( From: bill@cs.uofs.edu (Bill Gunshannon), Subject: Re: Latest on Windoze Navy software+ Message-ID: <2tcmvtF1tsscfU2@uni-berlin.de>   - In article <87is9bv4wy.fsf@prep.synonet.com>,  	prep@prep.synonet.com writes: > E > OK, for the idiots ;) out there, I'll repeat my comp.arch challenge  > of some time ago.  > 2 > Write any non-trivial function in C. It must be, > 
 > Correct.6 > Totally conformant with any standard implementation. > Portable.   ? Is this the old "you can have any two of the three" engineering  problem?  A There is so much undefined how could anyone accept the challenge?  What is non-trivial? What is "correct"? Conformant is easy. < How portable it might is dependant on how non-trivial it is.  = I have an even better challenge.  Re-write Unix (or Linux for > that matter) in Ada.  Then we can compare them for size, speed and functionality.   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: 16 Oct 2004 07:08:06 GMT2 From: "Dave Weatherall" <djw-nothere@nospam.nohow>& Subject: Re: OT: Sun's fighting chance? Message-ID: <DTiotGxQ0bj6-pn2-Rfd964jkqu5J@dave2_os2.home.ours>   2 On Fri, 15 Oct 2004 11:21:28 UTC, Andrew Harrison * <andrew_remove__harrison@sun__.com> wrote:  F > Its no secret why Linux is pretty Solaris like, why it uses PAM, XFNF > NFS, NIS, etc and thats because of the code donated by Sun and Sun's, > insistence on using documented interfaces.  F In the book 'Rebel Code', I was reading just before getting up, Linux B is reported as using the SunOS Man pages to derive the interfaces F used. He had tried to get the documentation for the Posix standard butE that was only available if one paid for it. His university had SunOS  F running so he used that machine to find out how the interfaces should  look in Linux .     --   Cheers - Dave.   ------------------------------  + Date: Sat, 16 Oct 2004 15:38:51 +0000 (UTC)  From: david20@alpha2.mdx.ac.uk+ Subject: Re: SYSUAF => export passwords ??? ) Message-ID: <ckrfab$i2o$1@news.mdx.ac.uk>   c In article <yCqomdjrzsx6@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes: h >In article <ckoqh8$757$1@news.tiscali.fr>, "Alexandre Mongin" <NOSPAM_alexandre.mongin@csgv.fr> writes: > J >> I would like to export all user's passwords stored in the sysuaf from aO >> first machine up to a second one, where the accounts are already existing...  >>  4 >> Is this possible through DCL or some free tools ? >   J There is a public domain tool called SYSUAF which has the ability to take N encrypted passwords from one node and apply them to the corresponding accounts on a remote node.  From the SYSUAF.DOC file   " M SYSUAF.BAS is the source code for the main program (written in VAX BASIC V3).  SYSUAF.DOC is this file.7 SYSUAF.EXE is the executable file for the main program. L TRANSFER_PWD_TO_NODE.BAS is the source code for the password-copying programO   which is executed by the command file created by the SYSUAF.EXE report option     "L" (written in VAX BASIC V3).H TRANSFER_PWD_TO_NODE.EXE is the executable file for the password-copying program L   (which must reside in a directory pointed to by the logical name:   TOOLS:E    in order to be available to the SYSUAF.COM command file created by     SYSUAF.EXE)  L You do not need the TRANSFER_PWD_TO_NODE.* files if you never select the "L" option on the SYSUAF menu.  O You should not need to recompile or relink either of the executables unless you L need to change the features of the program.  (Instructions for compiling andI linking SYSUAF are contained at the beginning of the source code).  These E programs have been tested on VMS V4.6, V4.7 and V5.1, V5.2, and V5.3.   " Special note on report option "K":  I         Report option "K" is used to copy all SYSUAF.DAT data (except for M         any encrypted password data - which is handled by the "L" option) for M         selected users (based upon the responses to the other SYSUAF prompts) N         from one node's SYSUAF.DAT to another node's SYSUAF.DAT.  It does thisN         by creating a command file, which when executed, will copy the accountG         data.  You will be prompted as to whether you want to copy only N         usernames which do not have a matching username (as yet) on the remoteO         node or whether to copy data for all selected usernames (based upon the /         responses to the other SYSUAF prompts).   M         Normally, you would run this option and then run option "L" to create H         the two SYSUAF.COM command files which will then be subsequently         executed, as follows:   J                 1) Copy the SYSUAF.COM command file produced by the reportM                    option "K" to the remote node and execute it on the remote J                    node to add the required new accounts (less passwords).  N                 2) Then execute the command file produced by the report optionG                    "L" on the local node to copy the encrypted password M                    information from the local node's accounts to the accounts L                    on the remote node.  The command file will prompt you for=                    the name of the remote node to be updated.   M         Besides the usual SYSPRV (or equiv) privilege that is required to run K         this utility on the local node, you will also need a proxy from the J         local node to a similarly-privileged account on the remote node in
         order M         for the program to open both nodes' SYSUAF.DAT files for the transfer          operation.  " Special note on report option "L":  L         Report option "L" is used to copy the encrypted password informationN         from one node's SYSUAF.DAT to another node's SYSUAF.DAT.  It does thisL         by creating a command file, which when executed, will copy the data.J         You will be prompted as to whether you want to copy only passwords
         whose O         usernames do not have matching usernames (as yet) on the remote node or K         whether to copy passwords for all selected accounts (based upon the /         responses to the other SYSUAF prompts).   I         Note that the command file created by this option will attempt to M         execute a utility called TRANSFER_PWD_TO_NODE.EXE which is located in N         a logically-named directory called TOOLS:  (You must define the TOOLS:N         logical name and ensure that the executable resides in that directory,-         prior to executing the command file).   N         This report option does not affect the local node's SYSUAF.DAT file inF         any way; Only the remote node's encrypted password data in its#         SYSUAF.DAT file is updated.    "   L This is a fairly old program but I have a copy which has been recompiled for Alpha available at  & ftp://www.axp.mdx.ac.uk/vms/sysuaf.zip  * if you can't find a version anywhere else.    9 I've never used the password copying part of the program.   M Hence if you try this I'd make sure I took a good backup of the remote sysuaf  first just in case.     D >You could, of course, copy entire records with the available tools. > M >> In other words, would it be possible to synchronize the passwords from one  >> machine with the other ?  > I >That can be fairly dangerous, as someone who gained control of the first ? >machine could then readily gain control of the second machine.   K True enough. However single-sign-on still seems to be a big goal of lots of O companies. I'd think synchronised passwords between two VMS systems is probably C a lot safer than synchronising the VMS system with windows systems.     
 David Webb Security Team Leader CCSS Middlesex University   ------------------------------  % Date: Sat, 16 Oct 2004 10:16:12 -0700 % From: "Zachary" <Z_kline@hotmail.com>  Subject: VMS for the blind+ Message-ID: <2td3arF1v9hu2U1@uni-berlin.de>    Hi, M I've recently setup an account on a public access VMS system.  I'm wondering  J how accessible VMS would be for the blind computer user, or if anyone has 3 successfully used it with a screen reading package. @ In future, I may install openVMS if I can get something to work. Thanks,  Zachary Kline    ------------------------------  % Date: Sat, 16 Oct 2004 12:25:32 -0500 * From: Roy Goodman <rlgoodman@acnospam.org> Subject: Re: VMS for the blind9 Message-ID: <Xns95848871A3B8Crlgoodmannyc@216.196.97.142>   ( "Zachary" <Z_kline@hotmail.com> wrote in$ news:2td3arF1v9hu2U1@uni-berlin.de:    > Hi, D > I've recently setup an account on a public access VMS system.  I'mG > wondering how accessible VMS would be for the blind computer user, or C > if anyone has successfully used it with a screen reading package. B > In future, I may install openVMS if I can get something to work.	 > Thanks,  > Zachary Kline  >  >  >   = Do they (whoever "they" are) still sell DECtalk and DECvoice?    Roy $ (remove "nospa" from Email to reply)   ------------------------------  % Date: Sat, 16 Oct 2004 10:49:09 -0700 % From: "Zachary" <Z_kline@hotmail.com>  Subject: Re: VMS for the blind+ Message-ID: <2td58lF1t84kmU1@uni-berlin.de>   8 "Roy Goodman" <rlgoodman@acnospam.org> wrote in message 3 news:Xns95848871A3B8Crlgoodmannyc@216.196.97.142... * > "Zachary" <Z_kline@hotmail.com> wrote in% > news:2td3arF1v9hu2U1@uni-berlin.de:  >  >> Hi,E >> I've recently setup an account on a public access VMS system.  I'm H >> wondering how accessible VMS would be for the blind computer user, orD >> if anyone has successfully used it with a screen reading package.C >> In future, I may install openVMS if I can get something to work. 
 >> Thanks, >> Zachary Kline >> >> >> > ? > Do they (whoever "they" are) still sell DECtalk and DECvoice?  >  > Roy & > (remove "nospa" from Email to reply)M I have no idea.  I got myself a Dectalk Express back in about 1995 or 96.  I  K have no idea whether it is supported.  I guess I should know that, because  E it has recently stopped working.  If it wouldn't violate any kind of  F license, I'm tempted to take the thing apart and look at the internal  speaker.   ------------------------------   End of INFO-VAX 2004.575 ************************