1 INFO-VAX	Thu, 19 Jun 2003	Volume 2003 : Issue 337       Contents:/ Re: Another unix/linux cert ... but not on VMS! / Re: Another unix/linux cert ... but not on VMS! / Re: Another unix/linux cert ... but not on VMS! % Re: Backup Software for VMS and Linux % Re: Backup Software for VMS and Linux " Cable for BA356 Deskside ExpansionL Re: Change owner of the file from unpriv account with Control access allowed Re: cxx performance / Re: DCL$AL_TAB_VEC equivalent for OpenVMS Alpha / Re: DCL$AL_TAB_VEC equivalent for OpenVMS Alpha / Re: DCL$AL_TAB_VEC equivalent for OpenVMS Alpha / Re: DCL$AL_TAB_VEC equivalent for OpenVMS Alpha / Re: Digital Terminal Servers 8 Port only USD350 , Re: Do's and don'ts about cleaning keyboards, Re: Do's and don'ts about cleaning keyboards, Re: Do's and don'ts about cleaning keyboards0 Re: Executables started by Java do not stay open Re: Firewall for VMS / TRU64$ Re: HP preps 1.8GHz AMD Athlon 64 PC  Re: HP to port VMS to Oopsteron?  Re: HP to port VMS to Oopsteron?  Re: HP to port VMS to Oopsteron?/ Re: IBM won't buy VMS ... has NIH syndrome too! / Re: IBM won't buy VMS ... has NIH syndrome too! + Re: Installation problem with OpenVMS 7.3.1 + Re: Installation problem with OpenVMS 7.3.1 + Re: Installation problem with OpenVMS 7.3.1  Re: lib$convert_date_string * Re: Max length of Nodename in OpenVMS 7.31* Re: Max length of Nodename in OpenVMS 7.31 Mozilla on old hardware  Re: Mozilla on old hardware  Re: Mozilla on old hardware  Re: Mozilla on old hardware  Re: Mozilla on old hardware  Re: Mozilla on old hardware & Re: NFS Client to MS-Services for Unix# Re: PIPE output defined to a symbol # Re: PIPE output defined to a symbol # Re: PIPE output defined to a symbol @ Re: SEC:U MIT-MAGIC-COOKIE-1, Motif 1.3 and HP TCP/IP XDM (long)& Re: Speed of IDE controllers in DS10 ? Re: Sun to be the next DEC!  Re: VMS Freeware CDs Re: VMS Freeware CDs3 Re: You say you want apps for VMS?  Here they come! @ Re: [ENABLE AUTOSTART] How to determine the state of AUTOSTART ?C Re: [OT] Future of Sun in doubt ?, was: Re: Sun to be the next DEC!   F ----------------------------------------------------------------------    Date: 18 Jun 2003 13:06:42 -0700( From: bob@instantwhip.com (Bob Ceculski)8 Subject: Re: Another unix/linux cert ... but not on VMS!= Message-ID: <d7791aa1.0306181206.2e0a8a8e@posting.google.com>   ^ david20@alpha2.mdx.ac.uk (David Webb) wrote in message news:<bcpk6p$ikn$1@aquila.mdx.ac.uk>... > 0 > You do know what PGP and GNuPG are don't you ?H > Nothing to do with cgis or web services - think public key encryption. >   D yes I know what PGP is, but I can avoid it using other proven secure products/methods ...   > M > The purveyor web server may be a great server but can you be 100% sure that @ > their isn't some security bug lurking somewhere in that code. M > If someone found one what would you do ? What's process' support policy for P > purveyor - would they guarantee to fix the problem or would they just tell youE > it's unsupported and you should use another more modern webserver ?  >    hint:  you can buy the code!   ------------------------------    Date: 18 Jun 2003 16:15:38 -0700( From: bob@instantwhip.com (Bob Ceculski)8 Subject: Re: Another unix/linux cert ... but not on VMS!= Message-ID: <d7791aa1.0306181515.37d92dba@posting.google.com>   ^ david20@alpha2.mdx.ac.uk (David Webb) wrote in message news:<bcq76i$kib$1@aquila.mdx.ac.uk>... > M > Although I'm glad VMS isn't affected I'm now preparing myself for Bob's - I 0 > told you VMS wouldn't be vulnerable onslaught. >    oh ye of little faith ...    ------------------------------    Date: 18 Jun 2003 16:19:03 -0700( From: bob@instantwhip.com (Bob Ceculski)8 Subject: Re: Another unix/linux cert ... but not on VMS!= Message-ID: <d7791aa1.0306181519.4d2bc7bd@posting.google.com>   o zinser@decus.de (Martin P.J. Zinser) wrote in message news:<220d7b8e.0306180830.1169c105@posting.google.com>... ` > david20@alpha2.mdx.ac.uk (David Webb) wrote in message news:<bcna4i$apc$1@aquila.mdx.ac.uk>... > Hello, > l > > In article <d7791aa1.0306170515.6890946d@posting.google.com>, bob@instantwhip.com (Bob Ceculski) writes:: > > >this would never happen on VMS, but on os's like unix: > > >and linux with such poor security models, it does ... > > > [ > > >http://www.computerworld.com/securitytopics/security/holes/story/0,10801,82197,00.html  > > . > > Are you sure VMS isn't affected by this :- > >  > > " R > > Adobe Systems Inc.'s Acrobat Reader 5.06 is affected by the problem, as is the: > > open-source reader Xpdf 1.01, according to the report. > >  > > "  > >  > > XPDF is available on VMS.  > > S > > Note the problem is to do with the PDF reader calling out to run other programs N > > when it encounters a link and the potential for someone to write maliciousS > > code which would allow them to affect other files OWNED by the USER running the  > > pdf reader.  > >  > > "  > M > Maybe it is worth mentioning that there is a new version of Xpdf (2.02pl1)  L > available since a couple of days already which addresses this problem for  > Unix users.  > K > Also upon examination of the patch I did decide this is not a problem on  L > OpenVMS since the exploit actually does rely on backtick expansion, which L > is not something DCL usually does. Which is also the reason why I have notJ > yet updated the OpenVMS port. Xpdf (and the libraries it uses) certainly> > has some shortcomings (the ones I am aware of are listed on M > http://www.decus.de:8080/www/vms/sw/notes/xpdf.htmlx), but these are worked  > on.  > K > Also if Bob rather wants to run Windows, this is obviously his choice ;-)  >  > Greetings, Martin   ? VMS security saves the day again ... and as for windows, a thin = vt emulation client with ie and outlook express is alot safer ? than putting unknown unix code on a vms box ... as we have seen < vms time after time can bail you out of bugs, but as ucx and? apache proves not every time ... to me a windows client is what  windows does best ...    ------------------------------  % Date: Wed, 18 Jun 2003 20:04:56 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> . Subject: Re: Backup Software for VMS and Linux' Message-ID: <3EF10C38.1DDB8160@fsi.net>     CHANGE username to westes wrote: > N > I was told that there is a VMS application that interfaces to Tivoli StorageM > Manager named ABC (http://www.storserver.com) that integrates with VMS very J > well and would allow you to run the LTO-1 tape drives on the TSM storage > host.   @ Good luck trying to restore it on OpenVMS using only VMS BACKUP!  G See, the problem is that most of these soultions do not (fully) support D /IMAGE backups; therefore, it is impossible to restore a system diskG backup. Instead you must rebuild the VMS system disk from scratch, just E as when restoring the backup server so you can restore the VMS system . from the department of redundancy department).  F You're just making an already difficult task border on the impossible.  H ...not to mention the challenges associated with trying back up multipleD terabytes of OpenVMS storage over a gig-E network inside of a decent backup window.   What's the point?    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------    Date: 18 Jun 2003 21:34:47 -0500+ From: young_r@encompasserve.org (Rob Young) . Subject: Re: Backup Software for VMS and Linux3 Message-ID: <zATa$pmawRW5@eisner.encompasserve.org>   [ In article <3EF10C38.1DDB8160@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes: " > CHANGE username to westes wrote: >>  O >> I was told that there is a VMS application that interfaces to Tivoli Storage N >> Manager named ABC (http://www.storserver.com) that integrates with VMS veryK >> well and would allow you to run the LTO-1 tape drives on the TSM storage  >> host. > B > Good luck trying to restore it on OpenVMS using only VMS BACKUP! >   / 	It doesn't use backup, so that isn't an issue.   I > See, the problem is that most of these soultions do not (fully) support F > /IMAGE backups; therefore, it is impossible to restore a system diskI > backup. Instead you must rebuild the VMS system disk from scratch, just G > as when restoring the backup server so you can restore the VMS system 0 > from the department of redundancy department).  < 	Most of us at a minimum have our system disks mirrored.  I G 	haven't lost or restored a system disk in 17 years, came close back in 0 	the day when the system disk was a single disk.  > 	If that concerned, disks are cheap.  Perform a nightly backupE 	of the system disk to another disk online at another site, whatever.    > H > You're just making an already difficult task border on the impossible. >   9 	No.  The product works and works well.  Additionally, it @ 	is a very good way to slam a very fast backup out.  Faster thanE 	going to an LTO drive.  For the high-end you would create a DiskPool D 	, backup to that.  In turn the DiskPool is migrated to tape for DR.: 	Multiple backups can be written to the disk pool at once.  J > ...not to mention the challenges associated with trying back up multipleF > terabytes of OpenVMS storage over a gig-E network inside of a decent > backup window.  ? 	Most modern networks today are switched end-to-end, backing up $ 	over GigE can be very advantageous.  < 	You can scale up and scale out in multiple dimensions.  For
 	instance:  V http://www.storserver.com/main.cfm?menu=2&submenu=vbm&detail=include/seriessummary.inc  D 	One model there has up to 32 Terabytes online, up to a 1.5 TerabyteB 	DiskPool.  You can scale them larger than that.  That is just one= 	vendor and one box.  If you are sooooo very large, you would ) 	"obviously" need more than a single box.    	But maybe better - by example:   W http://www.computerworld.com/databasetopics/data/datacenter/story/0,10801,82145,00.html   2 	He is standing next to 5 Symmetrixes.  5 of them,9 	with more elsewhere - no doubt.  480 disks in 5 of them. 6 	Tivoli Business Management and at the heart of it, no@ 	doubt TSM for Enterprise backup (surely not Veritas or Legato!)  5 	Now I could no doubt trot out dozens of examples but @ 	it wouldn't be good enough.  But there are folks that are using@ 	it in very large situations.  But of course I am sure there are> 	also folks that insist on doing weekly standalone system disk" 	backups to their TK70 drives too.   >  > What's the point?  >   B 	Enterprise Backup (all the hundreds of servers in your EnterpriseB 	being backed up by a single vendor solution - totally integrated,0 	daily DR export, daily reporting, etc. etc.).    3 	Speaking of difficult, how does that DR drill work E 	when instead of TSM (or equivalent Enterprise Backup solution), you  D 	have to have multiple different backup solutions at SunGuard to do  	the DR drill?   				Rob    ------------------------------  % Date: Wed, 18 Jun 2003 16:42:19 -0400 * From: BoylesA <alan.boyles@mindspring.com>+ Subject: Cable for BA356 Deskside Expansion / Message-ID: <vf1jv54k5g1ffb@corp.supernews.com>   K Anyone know the cable that you can use to attach a BA356-KD and a KZPCM-DA?   = One connector is a 68 pin VHDCI and the other is a 68 pin HD.   Y Anyone know of a good SCSI cable guide online for Alpha's.  I searched the HP website but + stuff like that is hard to find these days.    Thanks,    Alan   ------------------------------  # Date: Thu, 19 Jun 2003 00:31:41 GMT  From: Rob.Buxton@wcc.govt.nzU Subject: Re: Change owner of the file from unpriv account with Control access allowed $ Message-ID: <3ef10430.16666031@news>  D On Wed, 18 Jun 2003 09:46:20 +0500, Valentin Likoum <me@privacy.net> wrote:   >Rob.Buxton@wcc.govt.nz wrote:G >> On Tue, 17 Jun 2003 12:35:54 +0500, Valentin Likoum <me@privacy.net> 	 >> wrote:  >>   >[snip] 
 >>><QUOTE> >>>type of access - Control  >>> > >>>Gives you the right to change the protection code and ACL. : >>>You need to satisfy one of the following conditions to  >>>change the owner: >>> 8 >>>- (1) Hold both the old and the new owner identifier.< >>>- (2) Hold the Resource attribute to the identifier that > >>>owns the object while also being allowed control access to + >>>the object through an ACL on the object. 9 >>>- (3) Qualify as a system user, hold SYSPRV or BYPASS  > >>>privilege, or hold a UIC that matches that of the owner of / >>>the volume containing the file or directory. ? >>>- (4) Hold the GRPPRV privilege while also holding a UIC in  & >>>the same group as the object owner. >>></QUOTE>  >>> ? >>>  Well, I think I satisfy the second condition - my process  @ >>>holds the identifier VALENT with Resource attribute and this ? >>>identifier owns the file and I am allowed Control access to  < >>>the file through OTH_ACC right identifier and ACL on the 4 >>>file. So what's wrong here? Why -SYSTEM-F-NOPRIV?6 >>>  Alpha OVMS 7.3 with approprite patches installed  >>>(excluding SYS5 and SYS6).  >>>  Thank you.  >>>  >>   >> I'm guessing a bit, but... H >> You've got permission to change the protection and/or acl of the fileA >> but you're actually tring to change the ownership of the file.  >>  7 >> So, it's a Privilege issue, not a permissions issue. > >> I think you may need the  CMKRNL privilege to achieve this. >>   > > >   No, accordinly to manual quoted above user can change the < >owner of the object in certain conditions without elevated : >privileges. In my case suggestion of Joseph Huber (thank = >you, Joseph) make sense - I need to have resource attribute  ; >on the target identifier as I could affect disk quotas of  / >that identifier then change file' owner to it.   > Yep, I saw Josephs response later and it seemed to make sense.F I was thinking of the submit command which needs cmkrnl if you want to run under a different user.  >  >--  >Best regards, >  Valentin * >  valentin.likoum at ncc dot volga dot ru >    ------------------------------    Date: 18 Jun 2003 19:32:07 -0700) From: ejohnson@factset.com (Eric Johnson)  Subject: Re: cxx performance< Message-ID: <ef79676b.0306181832.1a3eae3@posting.google.com>  g "John Travell" <john@travell.uk.net> wrote in message news:<bcp8cs$lkj0a$1@ID-120847.news.dfncis.de>...   L > To make the comparison fair, you need to either roll the PC back to NT3.51K > (maybe NT4) or to upgrade VMS to V7.3-1. Even then, there are significant I > differences in compiler technology.  It might be that there really is a K > compiler performance problem, if so, the problem report (cxxc_bugs10188)  / > reported by Ken Block may in future fix this.   I Where could one get more information about problem report cxxc_bugs10188?   
 -eric johnson A ejohnson @ fact-nospam-set----.com (remove the dashes and nospam)    ------------------------------   Date: 18 Jun 03 11:21:47 PST From: mckinneyj@cpva.saic.com 8 Subject: Re: DCL$AL_TAB_VEC equivalent for OpenVMS Alpha( Message-ID: <yMpQ7HxVNJ+7@cpva.saic.com>  8 In article <8ss0fvgq17oe4vclbhs08rntjcm48a51vj@4ax.com>,)  David M Smith <dsmit115@csc.com> writes: M > In 1993 Chris Chiesa posted some code which showed how to invoke a DCL COPY R > command from an executable image. This code relied on the use of an undocumentedE > (as far as I can see) global symbol, DCL$AL_TAB_VEC, the address in 6 > DCLTABLES.EXE of the start of the DCL command table. > ? > For details, see Google Groups and search for DCL$AL_TAB_VEC.  > N > This symbol does not appear to be defined / available on OpenVMS Alpha. Does? > anyone know of an equivalent symbol, or technique, for Alpha? K > ------------------------------------------------------------------------- K > David M. Smith 302.391.8533                       dsmit115 at csc dot com K > Computer Sciences Corporation     (Opinions are those of the writer only) K > -------------------------------------------------------------------------   F "technique" - I assume that your goal is to $IMGACT COPY or some otherC program? And I assume that MACRO's ok? this might be enought to get C you going... I can give you a working example of $IMGACTing if this > doesn't do it for you. It was rather lengthy (650+ lines) so ID refrained from posting. If you'd like the entire program it email to  !         mckinneyj AT saic DOT com   = Anyway, here're (all, I hope) the pertinent symbols and code.   ,         $CLIDEF                 ;CLI symbols8         $IACDEF                 ;Image Activator symbols         .IF DEFINED, ALPHAG         $EIHADEF                ;Image Activator Transfer Array symbols 5         $EIHDDEF                ;Image Header symbols          .IFFG         $IHADEF                 ;Image Activator Transfer Array symbols 5         $IHDDEF                 ;Image Header symbols 
         .ENDC 5         $PSLDEF                 ;Process mode symbols   N home_range:                             ; Address range called program "wants"J         .IF DEFINED, ALPHA              ; to inhabit (assume default base)N         .address ^x00010000             ;  AXP start at %x10000 (2nd AXP page)P         .address ^x00FFFFFF             ; more than we'd ever need (32639 pages)         .IFFL         .address ^x00000200             ;  VAX start at %x200 (2nd VAX page)P         .address ^x00FFFFFF             ; more than we'd ever need (32766 pages)O         .ENDC                           ; $IMGACT will only take what is needed   N extdefault:     .ascid  \SYS$SYSTEM:.EXE\       ; command image default string? image_dsc:      .ascid  \COPY\                  ; command image   F img_header_buf:                         ; buffer contains image headerF                 .blkb   block           ;  of program to be $IMGACT'edE vm_actual:                              ; "actual" address range into E                 .blka                   ;   which called program gets >                 .blka                   ;   mapped at run time    # imgact_command:         .blkb   132    imgact_command_desc:, imgact_command_len:     .word   command_size-                         .byte   dsc$k_dtype_t -                         .byte   dsc$k_class_s 1 imgact_command_line_addr: .address imgact_command         K                                         ; let DCL parse the command line... C         pushl   g^CTL$AG_CLITABLE       ; address of command tables G         pushaq  imgact_command_desc     ; address of command descriptor ;         calls   #2,g^CLI$DCL_PARSE      ; check the command :         bsbw    error_check             ; check for errors    G         movq    extdefault,-(sp)        ; build descriptor on the stack >         movl    sp,r0                   ;  then point r0 thereK         $IMGACT_S -                     ; since we'll not call another .EXE L                 NAME=image_dsc,-        ; we needn't worry about cleaning upN                 DFLNAM=(r0),-           ; (recovering the VA space) afterwards'                 HDRBUF=img_header_buf,- '                 IMGCTL=#<IAC$M_MERGE>,- "                 INADR=home_range,-"                 RETADR=vm_actual,-"                 ACMODE=#PSL$C_USERE         addl    #8,sp                   ; remove temp desc from stack :         bsbw    error_check             ; check for errors         $IMGFIX_S :         bsbw    error_check             ; check for errors     ;+F ; When calling a program there should be six arguments on the stack...0 ;       4(AP)  = address of image transfer array0 ;       8(AP)  = address of CLI callback routine/ ;       12(AP) = address of image header buffer 5 ;       16(AP) = address of image filespec descriptor 5 ;       20(AP) = address of image linker option flags 0 ;       24(AP) = address of image run time flags ;-  !         movaq   img_header_buf,r4 :         movl    (r4),r4                 ; the image header?         pushl   #0                      ; ignore run time flags          .IF DEFINED, ALPHA=         pushl   EIHD$L_LNKFLAGS(r4)     ; linker option flags          .IFF=         pushl   IHD$L_LNKFLAGS(r4)      ; linker option flags 
         .ENDC =         pushaq  image_dsc               ; image file filespec 6         pushal  img_header_buf          ; image header>         pushl   CTL$AL_CLICALBK         ; CLI callback routine         .IF DEFINED, ALPHAC         movl    EIHD$L_ACTIVOFF(r4),r0  ; offset to activation data H         movab   EIHA$L_TFRADR1(r4)[r0],r1 ; offset to trans vector array         .IFFC         movzwl  IHD$W_ACTIVOFF(r4),r0   ; offset to activation data G         movab   IHA$L_TFRADR1(r4)[r0],r1 ; offset to trans vector array 
         .ENDC C         pushl   r1                      ; the transfer vector array C         calls   #6,@(r1)                ;  call image's entry point 9         pushl   r0                      ; save the status    ------------------------------  # Date: Wed, 18 Jun 2003 19:09:26 GMT " From:   VAXman-  @SendSpamHere.ORG8 Subject: Re: DCL$AL_TAB_VEC equivalent for OpenVMS Alpha0 Message-ID: <00A21934.127236A2@SendSpamHere.ORG>  H In article <yMpQ7HxVNJ+7@cpva.saic.com>, mckinneyj@cpva.saic.com writes:9 >In article <8ss0fvgq17oe4vclbhs08rntjcm48a51vj@4ax.com>, * > David M Smith <dsmit115@csc.com> writes:N >> In 1993 Chris Chiesa posted some code which showed how to invoke a DCL COPYS >> command from an executable image. This code relied on the use of an undocumented F >> (as far as I can see) global symbol, DCL$AL_TAB_VEC, the address in7 >> DCLTABLES.EXE of the start of the DCL command table.  >>  @ >> For details, see Google Groups and search for DCL$AL_TAB_VEC. >>  O >> This symbol does not appear to be defined / available on OpenVMS Alpha. Does @ >> anyone know of an equivalent symbol, or technique, for Alpha?L >> -------------------------------------------------------------------------L >> David M. Smith 302.391.8533                       dsmit115 at csc dot comL >> Computer Sciences Corporation     (Opinions are those of the writer only)L >> ------------------------------------------------------------------------- > G >"technique" - I assume that your goal is to $IMGACT COPY or some other D >program? And I assume that MACRO's ok? this might be enought to getD >you going... I can give you a working example of $IMGACTing if this? >doesn't do it for you. It was rather lengthy (650+ lines) so I E >refrained from posting. If you'd like the entire program it email to  > " >        mckinneyj AT saic DOT com > > >Anyway, here're (all, I hope) the pertinent symbols and code. > - >        $CLIDEF                 ;CLI symbols 9 >        $IACDEF                 ;Image Activator symbols  >        .IF DEFINED, ALPHA H >        $EIHADEF                ;Image Activator Transfer Array symbols6 >        $EIHDDEF                ;Image Header symbols
 >        .IFF H >        $IHADEF                 ;Image Activator Transfer Array symbols6 >        $IHDDEF                 ;Image Header symbols >        .ENDC6 >        $PSLDEF                 ;Process mode symbols > O >home_range:                             ; Address range called program "wants" K >        .IF DEFINED, ALPHA              ; to inhabit (assume default base) O >        .address ^x00010000             ;  AXP start at %x10000 (2nd AXP page) Q >        .address ^x00FFFFFF             ; more than we'd ever need (32639 pages) 
 >        .IFF M >        .address ^x00000200             ;  VAX start at %x200 (2nd VAX page) Q >        .address ^x00FFFFFF             ; more than we'd ever need (32766 pages) P >        .ENDC                           ; $IMGACT will only take what is needed > O >extdefault:     .ascid  \SYS$SYSTEM:.EXE\       ; command image default string @ >image_dsc:      .ascid  \COPY\                  ; command image > G >img_header_buf:                         ; buffer contains image header G >                .blkb   block           ;  of program to be $IMGACT'ed F >vm_actual:                              ; "actual" address range intoF >                .blka                   ;   which called program gets? >                .blka                   ;   mapped at run time  >  > $ >imgact_command:         .blkb   132 >  >imgact_command_desc: - >imgact_command_len:     .word   command_size . >                        .byte   dsc$k_dtype_t. >                        .byte   dsc$k_class_s2 >imgact_command_line_addr: .address imgact_command >  >  >  > L >                                        ; let DCL parse the command line...D >        pushl   g^CTL$AG_CLITABLE       ; address of command tablesH >        pushaq  imgact_command_desc     ; address of command descriptor< >        calls   #2,g^CLI$DCL_PARSE      ; check the command; >        bsbw    error_check             ; check for errors  >  > H >        movq    extdefault,-(sp)        ; build descriptor on the stack? >        movl    sp,r0                   ;  then point r0 there L >        $IMGACT_S -                     ; since we'll not call another .EXEM >                NAME=image_dsc,-        ; we needn't worry about cleaning up O >                DFLNAM=(r0),-           ; (recovering the VA space) afterwards ( >                HDRBUF=img_header_buf,-( >                IMGCTL=#<IAC$M_MERGE>,-# >                INADR=home_range,- # >                RETADR=vm_actual,- # >                ACMODE=#PSL$C_USER F >        addl    #8,sp                   ; remove temp desc from stack; >        bsbw    error_check             ; check for errors  >        $IMGFIX_S; >        bsbw    error_check             ; check for errors  >  >  >;+ G >; When calling a program there should be six arguments on the stack... 1 >;       4(AP)  = address of image transfer array 1 >;       8(AP)  = address of CLI callback routine 0 >;       12(AP) = address of image header buffer6 >;       16(AP) = address of image filespec descriptor6 >;       20(AP) = address of image linker option flags1 >;       24(AP) = address of image run time flags  >;-  > " >        movaq   img_header_buf,r4; >        movl    (r4),r4                 ; the image header @ >        pushl   #0                      ; ignore run time flags >        .IF DEFINED, ALPHA > >        pushl   EIHD$L_LNKFLAGS(r4)     ; linker option flags
 >        .IFF > >        pushl   IHD$L_LNKFLAGS(r4)      ; linker option flags >        .ENDC> >        pushaq  image_dsc               ; image file filespec7 >        pushal  img_header_buf          ; image header ? >        pushl   CTL$AL_CLICALBK         ; CLI callback routine  >        .IF DEFINED, ALPHA D >        movl    EIHD$L_ACTIVOFF(r4),r0  ; offset to activation dataI >        movab   EIHA$L_TFRADR1(r4)[r0],r1 ; offset to trans vector array 
 >        .IFF D >        movzwl  IHD$W_ACTIVOFF(r4),r0   ; offset to activation dataH >        movab   IHA$L_TFRADR1(r4)[r0],r1 ; offset to trans vector array >        .ENDCD >        pushl   r1                      ; the transfer vector arrayD >        calls   #6,@(r1)                ;  call image's entry point: >        pushl   r0                      ; save the status    I No, he's looking for DCL$AL_TBL_VEC!  This happens to be the base address I of the DCLTABLES.EXE when loaded.  I've already forwarded a solution with ; $IMGACT_S which works in the Chris Chiesa example to David.  --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM              5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------   Date: 18 Jun 03 13:08:46 PST From: mckinneyj@cpva.saic.com 8 Subject: Re: DCL$AL_TAB_VEC equivalent for OpenVMS Alpha( Message-ID: <STCoauGKBF0w@cpva.saic.com>  0 In article <00A21934.127236A2@SendSpamHere.ORG>,%    VAXman-  @SendSpamHere.ORG writes: J > In article <yMpQ7HxVNJ+7@cpva.saic.com>, mckinneyj@cpva.saic.com writes:: >>In article <8ss0fvgq17oe4vclbhs08rntjcm48a51vj@4ax.com>,+ >> David M Smith <dsmit115@csc.com> writes: O >>> In 1993 Chris Chiesa posted some code which showed how to invoke a DCL COPY T >>> command from an executable image. This code relied on the use of an undocumentedG >>> (as far as I can see) global symbol, DCL$AL_TAB_VEC, the address in 8 >>> DCLTABLES.EXE of the start of the DCL command table. >>> A >>> For details, see Google Groups and search for DCL$AL_TAB_VEC.  >>> P >>> This symbol does not appear to be defined / available on OpenVMS Alpha. DoesA >>> anyone know of an equivalent symbol, or technique, for Alpha? M >>> ------------------------------------------------------------------------- M >>> David M. Smith 302.391.8533                       dsmit115 at csc dot com M >>> Computer Sciences Corporation     (Opinions are those of the writer only) M >>> -------------------------------------------------------------------------  >>H >>"technique" - I assume that your goal is to $IMGACT COPY or some otherE >>program? And I assume that MACRO's ok? this might be enought to get E >>you going... I can give you a working example of $IMGACTing if this @ >>doesn't do it for you. It was rather lengthy (650+ lines) so IF >>refrained from posting. If you'd like the entire program it email to >># >>        mckinneyj AT saic DOT com  >>? >>Anyway, here're (all, I hope) the pertinent symbols and code.  >>. >>        $CLIDEF                 ;CLI symbols: >>        $IACDEF                 ;Image Activator symbols >>        .IF DEFINED, ALPHAI >>        $EIHADEF                ;Image Activator Transfer Array symbols 7 >>        $EIHDDEF                ;Image Header symbols  >>        .IFFI >>        $IHADEF                 ;Image Activator Transfer Array symbols 7 >>        $IHDDEF                 ;Image Header symbols  >>        .ENDC 7 >>        $PSLDEF                 ;Process mode symbols  >>P >>home_range:                             ; Address range called program "wants"L >>        .IF DEFINED, ALPHA              ; to inhabit (assume default base)P >>        .address ^x00010000             ;  AXP start at %x10000 (2nd AXP page)R >>        .address ^x00FFFFFF             ; more than we'd ever need (32639 pages) >>        .IFFN >>        .address ^x00000200             ;  VAX start at %x200 (2nd VAX page)R >>        .address ^x00FFFFFF             ; more than we'd ever need (32766 pages)Q >>        .ENDC                           ; $IMGACT will only take what is needed  >>P >>extdefault:     .ascid  \SYS$SYSTEM:.EXE\       ; command image default stringA >>image_dsc:      .ascid  \COPY\                  ; command image  >>H >>img_header_buf:                         ; buffer contains image headerH >>                .blkb   block           ;  of program to be $IMGACT'edG >>vm_actual:                              ; "actual" address range into G >>                .blka                   ;   which called program gets @ >>                .blka                   ;   mapped at run time >> >>% >>imgact_command:         .blkb   132  >> >>imgact_command_desc:. >>imgact_command_len:     .word   command_size/ >>                        .byte   dsc$k_dtype_t / >>                        .byte   dsc$k_class_s 3 >>imgact_command_line_addr: .address imgact_command  >> >> >> >>M >>                                        ; let DCL parse the command line... E >>        pushl   g^CTL$AG_CLITABLE       ; address of command tables I >>        pushaq  imgact_command_desc     ; address of command descriptor = >>        calls   #2,g^CLI$DCL_PARSE      ; check the command < >>        bsbw    error_check             ; check for errors >> >>I >>        movq    extdefault,-(sp)        ; build descriptor on the stack @ >>        movl    sp,r0                   ;  then point r0 thereM >>        $IMGACT_S -                     ; since we'll not call another .EXE N >>                NAME=image_dsc,-        ; we needn't worry about cleaning upP >>                DFLNAM=(r0),-           ; (recovering the VA space) afterwards) >>                HDRBUF=img_header_buf,- ) >>                IMGCTL=#<IAC$M_MERGE>,- $ >>                INADR=home_range,-$ >>                RETADR=vm_actual,-$ >>                ACMODE=#PSL$C_USERG >>        addl    #8,sp                   ; remove temp desc from stack < >>        bsbw    error_check             ; check for errors >>        $IMGFIX_S < >>        bsbw    error_check             ; check for errors >> >> >>;+H >>; When calling a program there should be six arguments on the stack...2 >>;       4(AP)  = address of image transfer array2 >>;       8(AP)  = address of CLI callback routine1 >>;       12(AP) = address of image header buffer 7 >>;       16(AP) = address of image filespec descriptor 7 >>;       20(AP) = address of image linker option flags 2 >>;       24(AP) = address of image run time flags >>;- >># >>        movaq   img_header_buf,r4t< >>        movl    (r4),r4                 ; the image headerA >>        pushl   #0                      ; ignore run time flagso >>        .IF DEFINED, ALPHA? >>        pushl   EIHD$L_LNKFLAGS(r4)     ; linker option flagsl >>        .IFF? >>        pushl   IHD$L_LNKFLAGS(r4)      ; linker option flagsw >>        .ENDCs? >>        pushaq  image_dsc               ; image file filespec 8 >>        pushal  img_header_buf          ; image header@ >>        pushl   CTL$AL_CLICALBK         ; CLI callback routine >>        .IF DEFINED, ALPHAE >>        movl    EIHD$L_ACTIVOFF(r4),r0  ; offset to activation dataaJ >>        movab   EIHA$L_TFRADR1(r4)[r0],r1 ; offset to trans vector array >>        .IFFE >>        movzwl  IHD$W_ACTIVOFF(r4),r0   ; offset to activation datanI >>        movab   IHA$L_TFRADR1(r4)[r0],r1 ; offset to trans vector array6 >>        .ENDCoE >>        pushl   r1                      ; the transfer vector arrayVE >>        calls   #6,@(r1)                ;  call image's entry point ; >>        pushl   r0                      ; save the statusn >  > K > No, he's looking for DCL$AL_TBL_VEC!  This happens to be the base addresslK > of the DCLTABLES.EXE when loaded.  I've already forwarded a solution withg    D Which is at CTL$AG_CLITABLE - where the above code points, right? Or am I completely missing this?     = > $IMGACT_S which works in the Chris Chiesa example to David.  > --Q > VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMe
 >            e7 >   "Well my son, life is like a beanstalk, isn't it?" E >    -- i - Jima   ------------------------------  # Date: Wed, 18 Jun 2003 22:26:40 GMTE# From: hoff@hp.nospam (Hoff Hoffman))8 Subject: Re: DCL$AL_TAB_VEC equivalent for OpenVMS Alpha2 Message-ID: <AG5Ia.2902$IU1.2648@news.cpqcorp.net>  a In article <8ss0fvgq17oe4vclbhs08rntjcm48a51vj@4ax.com>, David M Smith <dsmit115@csc.com> writes:hL :In 1993 Chris Chiesa posted some code which showed how to invoke a DCL COPY# :command from an executable image. -  F   The callable convert API can often be used here, and provides a good/   workaround for the missing callable copy API.l  N  ---------------------------- #include <rtfaq.h> -----------------------------J       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.comw   ------------------------------  % Date: Wed, 18 Jun 2003 18:01:35 -0400u, From: "David Turner" <dbturner@islandco.com>8 Subject: Re: Digital Terminal Servers 8 Port only USD350/ Message-ID: <vf1oj7572pf7dd@news.supernews.com>     correction DSRVW-ZA ( 8 x RS232)   DT   ------------------------------  # Date: Wed, 18 Jun 2003 21:52:42 GMTr# From: "John Smith" <a@nonymous.com>r5 Subject: Re: Do's and don'ts about cleaning keyboardsyI Message-ID: <Ka5Ia.130736$G_.119392@news02.bloor.is.net.cable.rogers.com>e  7 Coffee with milk has ruined one of my laptop keyboards.-    @ "Bochnik, William J" <William_Bochnik@acml.com> wrote in messageF news:2D75787AAF09C64481BDFD89113BE6D50199B65E@ac2kama0102.ac.lp.acml.c om...o/ > pulverized water - now I've heard everything.@ >tE > btw what are the 2 worst things I've seen in a keyboard?  Coca colaa andw > pineapple juice. >m > -----Original Message-----3 > From: JF Mezei [mailto:jfmezei.spamnot@istop.com]o' > Sent: Tuesday, June 17, 2003 11:33 PMv > To: Info-VAX@Mvb.Saic.Comh7 > Subject: Re: Do's and don'ts about cleaning keyboards- >  >e > "David J. Dachtera" wrote:D > > The petroleum distillates it contains can do considerable damage to1 > > plastic and rubber products. Not recommended.6 >5C > You should note that Digital had won environmental awards when ite
 developped. > a way to clear PCB boards with... WATER !!!! >g< > (I assume it may have been finely pulverised water at high
 pressure).+ > -----------------------------------------l< > The information contained in this transmission may containD privileged and confidential information and is intended only for the= use of the person(s) named above. If you are not the intendedtB recipient, or an employee or agent responsible for delivering this= message to the intended recipient, any review, dissemination,9= distribution or duplication of this communication is strictly-E prohibited. If you are not the intended recipient, please contact thed< sender immediately by reply e-mail and destroy all copies ofA > the original message. Please note that we do not accept account ? orders and/or instructions by e-mail, and therefore will not be = responsible for carrying out such orders and/or instructions.n >n >h   ------------------------------  % Date: Wed, 18 Jun 2003 19:53:47 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> 5 Subject: Re: Do's and don'ts about cleaning keyboardst' Message-ID: <3EF1099B.561FD737@fsi.net>o   Bob Koehler wrote: > V > In article <3EF0295A.EDB72B77@127.0.0.1>, Nic Clews <sendspamhere@127.0.0.1> writes: > > = > > Clean pencil erasers are also very good contact cleaners.r > G >    So is Freon, but then it has other issues.  A can of it used to bei' >    a standard part of many toolboxes.y  G Can you still get trichloroethance-based contact/pot cleaner (aerosol)?o   -- f David J. Dachtera  dba DJE Systemsy http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/s   ------------------------------  % Date: Thu, 19 Jun 2003 00:44:19 -0500r/ From: Chris Scheers <chris@applied-synergy.com>c5 Subject: Re: Do's and don'ts about cleaning keyboardst3 Message-ID: <3EF14DB3.353F4544@applied-synergy.com>a   Paul Sture wrote:n > I > My heartful thanks to the person who recommended putting my LK keyboard. > through the dishwasher.e >  > It is beatifully clean now.s > ( > But a pity it doesn't work any more...    @ From reports I have heard, using the dishwasher is about a 50/50F proposition.  (Make sure that you DON'T use any soap or softner.  JustD use water.)  However, if the keyboard can't be cleaned and you don't2 have any other options, 50/50 is better than zero.  F It can take a long time for the keyboard to dry.  The recommendation I, have been given is to let it sit for a week.  G -----------------------------------------------------------------------a$ Chris Scheers, Applied Synergy, Inc.  C Voice: 817-237-3360            Internet: chris@applied-synergy.com o   Fax: 817-237-3074t   ------------------------------    Date: 18 Jun 2003 12:44:08 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)s9 Subject: Re: Executables started by Java do not stay openk3 Message-ID: <5pZT1DPfQbrR@eisner.encompasserve.org>   e In article <vf0qd7c1urvjef@corp.supernews.com>, "Tim Aisthorpe-Buckley" <Tim_A_B@Hotmail.com> writes:n >  >     Java source extract: >         Process p =lE > Runtime.getRuntime().exec("aluser:[iris.btirissocket]runlogo.com");  >   A    This is probably starting the processes as subprocesses of thesE process which is running the Java code.  If so, SHOW SYSTEM will lists# them with an S in the right column.r  F    On VMS, when a parent process exits all it's children processes areC killed.  You need to either keep the parent process around or start-D the other processes as detached processes.  See the documentation onA sys$creprc for how to run sys$system:loginout to start a detachedL process.   ------------------------------  # Date: Wed, 18 Jun 2003 22:20:42 GMT1# From: hoff@hp.nospam (Hoff Hoffman) % Subject: Re: Firewall for VMS / TRU64N2 Message-ID: <_A5Ia.2901$IU1.1874@news.cpqcorp.net>   ..  H   I'd not use Solaris as a firewall, either -- running a general-purposeI   operating system as a firewall is not something that I would choose to  J   do, regardless of the options and availabilies and filters and platform.  I   Can a general-purpose system be secured sufficiently?  Very likely yes, H   at least for the more secure platforms.  Is the effort and the support:   time involved cost-effective?  I don't believe it to be.  G   Running a firewall requires a support effort, and -- with small-scale F   firewalls available with prices starting around US$100 -- not a taskF   that I could reasonably justify choosing to perform myself, as it isH   an ONGOING support task and one that is not without substantial risks.  H   Corporate IP firewalls with stateful inspection and filtering and DDoSG   defenses and subscription services and mail-scanning capabilities aregI   obviously somewhat more expensive, but are still readily available from    security products vendors. d  G   I here refer to the requirement that the firewall must be reviewed asdG   users and software are added and removed and as packages are upgraded A   -- the up-front part of configuring ipchains or other firewall eD   mechanisms is the easy part, and some of the newer GUIs make theseC   tasks even easier.  Where this gets expensive is with KEEPING theiI   system secure.  Regardless of whether the platform is Solaris, OpenVMS,    Windows, or otherwise. n  B   I've seen suggestions of removing chunks off the general-purposeB   box, and this is certainly feasible -- but there are still costsC   involved with maintaining and with upgrading the firewall that ist.   on this locally-produced firewall appliance.  F   A dedicated firewall appliance eliminates much of the support effortD   discussed above, and provides me with a dedicated solution where I@   can also contract for vendor support and upgrades as required.  F   This is all a question of priorities, and particularly one of costs.  H   This decision is a no-brainer for most implementations and most sites.  N  ---------------------------- #include <rtfaq.h> -----------------------------J       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.comv   ------------------------------  # Date: Wed, 18 Jun 2003 18:46:39 GMTr# From: "John Smith" <a@nonymous.com>a- Subject: Re: HP preps 1.8GHz AMD Athlon 64 PCfH Message-ID: <js2Ia.14629$111.14238@news04.bloor.is.net.cable.rogers.com>  7 "JF Mezei" <jfmezei.spamnot@istop.com> wrote in messageu# news:3EF094CC.1193B4D4@istop.com...  > Simon Clubley wrote:B > > If this product is actually released (for all we know it might just be 3 > > a tactic to get better discounts out of Intel),r >nD > The article did speculate that this PDF document was the result of internalD > testing of sample AMD chips. Not a released product. There are two ways tot > look into it:R >iF > -HP would have freedom to choose AMD over Intel and is investigating the new B > AMD chip to see if it woudl fit nicely in its product portfolio. >tB > -HP's ties with Intel are too strong, but HP is still looking at what the AMDC > can do so that HP can better prepare to combat, with Intel chips,c
 the advent	 > of AMD.s > F > I suspect that all manufacturers will be toying with some sample AMD/ > chips/systems over the next couple of months.K >n > B > > HP are starting to get worried about the possibility that IA64 might , > > not achieve a significant market share ? > B > It is a given that IA64 won't get significant market share. Even Inteln? > admitted some time ago that IA64 would be limited to high end- systems and not1 > make it to workstations etc. >.? > > However, dismissing the Inquirer reference still leaves thep question ofiD > > VMS on AMD64, which is a valid one _IF_ HP start releasing AMD64
 systems in > > significant quantities.n >k > E > This can't happen (officially) until there is some sort of official1D > announcement from Intel that development of IA64 will stop after X generation.nE > I suspect that contractual obligations between Compaq and Intel forr thetF > porting of VMS to IA64 would make it hard for HP to abandon IA64. So thel= > decision will have to come from Intel who will then give HP  permission to usee > Intel's 64 bit 8086. >-D > Right now, IA64 is no threath to Power or Sparc. But if Intel were to ditchC > IA64 and go for 64 bit 8086, then I suspect that Intel would haveL far greatertC > success in getting a foothold into enterprise computing. But thati larger. > market would be split between Intel and AMD.    D All this chip intrigue along without advertising of HP's love of VMSE (however conditional it may be) continues to make it simple for ISV'sjB to drop support for VMS, or to not consider porting their existing products to VMS.  B Of course HP *could* hire back the Alpha team on 3-year guaranteedF no-cut contracts at twice what they are making at Intel and still comeC out ahead, as long as HP also went cap-in-hand and down on one kneelE before the Lord of Redmond and begged forgiveness for the failings ofhB Capellas with respect to this haste in killing 64-bit Windows (and applications) on Alpha.    ------------------------------  % Date: Wed, 18 Jun 2003 14:38:10 -0400e* From: JF Mezei <jfmezei.spamnot@istop.com>) Subject: Re: HP to port VMS to Oopsteron?h) Message-ID: <3EF0B189.5773CD76@istop.com>u   David Webb wrote:iI > Since noone has yet been able to buy any VMS on IA64 production systems I > replacing the port to IA64 with a port to x86-64 now would be feasible.r  N I'd say until HP sells the first commercial VMS on IA64, the canning of VMS onK the unwanted IA64 is possible. But the minute you have it out at one payingeG customer, you are committed to supporting it for 5 years (or whatever).a  H My guess is that Intel's 64 bit 8086 would include lockstep (which AMD'sM doesn't correct ?) which would allow Tandem to port to the 8086. So in a way,mK the intel 8086 would remain somewhat proprietary for high end systems while N AMD would battle for the low end and midrange with a chip lacking certain highF end features. (unless AMD decides to match Intel feature for feature).   ------------------------------  % Date: Wed, 18 Jun 2003 20:21:17 -0500s1 From: "David J. Dachtera" <djesys.nospam@fsi.net>.) Subject: Re: HP to port VMS to Oopsteron?>' Message-ID: <3EF1100D.15E372DD@fsi.net>-   rob kas wrote: > 7 > "Bob Ceculski" <bob@instantwhip.com> wrote in message 9 > news:d7791aa1.0306180510.1a0c2e80@posting.google.com... 2 > > that's what the inquirer says might happen ... > >e- > > http://www.theinquirer.net/?article=10056r > & >    Did you read the AMDZone Article? > K >    There is presently a belief that there would be insufficient market tofI > justify the cost involved in porting OpenVMS to systems using the Intelr > IA-32 architecture.e  D ...which, of course, make no sense (never has, never will) since the= vast majority of server systems in the world today are IA32. 1  7 > In addition to the direct costs involved in any port,LJ > each maintainer of a product or a package for OpenVMS has to justify theG > port to "OpenVMS Pentium" or to OpenVMS Itanium, akin to the requireddF > justifications for a product port from OpenVMS VAX to OpenVMS Alpha.   See the above.  $ > But yes, it would be nice to have.  A Profits "would be nice to have, also". HP's silence onthe OPenVMSdH advertising issue speaks volumes more than one can say in a Usenet post.  H > And yes, OpenVMS Engineering is well aware of AMD Opteron/Hammer, too.  C ...and soon may be told to port to it, since Itanic flounders on...s   -- p David J. Dachtera  dba DJE SystemsV http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/-   ------------------------------    Date: 18 Jun 2003 22:41:47 -0700. From: mistdragon@zdnetonebox.com (mist dragon)) Subject: Re: HP to port VMS to Oopsteron?r= Message-ID: <7500353b.0306182141.39a8f692@posting.google.com>-  F Or maybe HP has decided that it could support the AMD64 on desktop andC on workstations because other players like IBM have committed on itgD and it does not compete with IA64 servers that Intel has targeted to@ high-end. This is likely as HP is asking whether customers wouldA _really_ like to have iVMS workstations and I doubt people would.r  C Following on this, IA32 scales well on 2, less on 4 but overall, it * leave IA32 and Opteron to 1-4 cpu servers.  F And as for going upwards, Opteron is targeted to 1-8 processors and it) may be that HP also leaves this to AMD64.   D So IA64 would be targeted to systems from 9 CPU's and up. This makes sense.  C However, it may not be so good for VMS, because most of the base is  less than 9 CPU boxes....a  F As for porting VMS to Opsteron I doubt it wery much because porting isD very expensive and doing that would postpone port for few more yearsD and anyway, if 9 CPU's and up is HP's strategy then why would they ?   Mist   ------------------------------  % Date: Wed, 18 Jun 2003 22:24:14 -0400l* From: "Bill Todd" <billtodd@metrocast.net>8 Subject: Re: IBM won't buy VMS ... has NIH syndrome too!2 Message-ID: <5_GdncS3NttJg2yjXTWJhw@metrocast.net>  3 "jlsue" <jefflsxxxz@sbcglobal.net> wrote in message 2 news:ka81fvsppgj7blgu696ejr8lsqsseku9fo@4ax.com...J > On Wed, 18 Jun 2003 01:34:18 -0400, "Bill Todd" <billtodd@metrocast.net> > wrote: >  > > : > >"JF Mezei" <jfmezei.spamnot@istop.com> wrote in message& > >news:3EEFD077.CF9310E4@istop.com... > >> John Smith wrote:F > >> > a) hp has bashed 'proprietary o/s'es as 'bad' for so long, that, > >> > advertising VMS makes them lose face. > >>! > >> Why then did HP buy Compaq ?  > >oJ > >To give a couple of failed CEOs another couple of years of tenure whileK > >things sorted themselves out.  And perhaps to get its PC and IA32 serverCD > >business (since clearly both CEOs believed that size, rather than profits,
 > >mattered).b > >  > K > Hey, don't forget storage.  Dec/Cpq had great storage, which HP inherited 
 > as well.  L True - though it's not clear just how much of that storage customer base wasJ specific to VMS, Tru64, and NSK (for example, most of the service revenues' were connected to VMS, Tru64, and NSK).   F But HP actually does seem to be trying to develop and sell the storageE potential that it bought with Compaq, unlike most of the rest of thathD acquisition.  Perhaps that's because HP had a fair amount of storageH expertise on its own and can appreciate what it's got there more than inD other areas (such as operating systems, where it seems to think thatG Windows, HP-UX, and Linux are or soon will be the right answers for all  needs).    - bill   ------------------------------  % Date: Wed, 18 Jun 2003 22:40:58 -0400r* From: JF Mezei <jfmezei.spamnot@istop.com>8 Subject: Re: IBM won't buy VMS ... has NIH syndrome too!) Message-ID: <3EF12293.D62230AE@istop.com>a  M > > Hey, don't forget storage.  Dec/Cpq had great storage, which HP inherited, > > as well.  K Would storage be considered a hardware or software solution if it had to be + classified as either software or hardware ?   = VMS is definitely hardware, especially now that Alpha's dead.-  J Do storage arrays use a Digital develop OS as controller ? Has HP kept theM original Digital software, or are its new products all derived from HP-legacy  storage components ?   ------------------------------  # Date: Wed, 18 Jun 2003 17:54:24 GMT & From: jlsue <jefflsxxxz@sbcglobal.net>4 Subject: Re: Installation problem with OpenVMS 7.3.18 Message-ID: <mp91fvcu215tq9egoqfvronjm01fuf95vh@4ax.com>  E On Wed, 18 Jun 2003 13:36:48 GMT, Eric Bruno <eric@ebruno.org> wrote:   H >Looks like I am just going to have to get another box which is offical I >supported list.  Island has some reburbished boxes and I will just keep i >an eye on e-bay.  >   E reburbished - sent back (or banished) to the suburbs where you belong    ------------------------------  # Date: Thu, 19 Jun 2003 00:24:31 GMTl From: Rob.Buxton@wcc.govt.nz4 Subject: Re: Installation problem with OpenVMS 7.3.1$ Message-ID: <3ef102b5.16286750@news>  B On Wed, 18 Jun 2003 17:54:24 GMT, jlsue <jefflsxxxz@sbcglobal.net> wrote:  F >On Wed, 18 Jun 2003 13:36:48 GMT, Eric Bruno <eric@ebruno.org> wrote: >eI >>Looks like I am just going to have to get another box which is offical  J >>supported list.  Island has some reburbished boxes and I will just keep  >>an eye on e-bay. >> >cF >reburbished - sent back (or banished) to the suburbs where you belong >  Excellent...   ------------------------------  % Date: Wed, 18 Jun 2003 23:39:04 -0400n  From: John Santos <JOHN@egh.com>4 Subject: Re: Installation problem with OpenVMS 7.3.15 Message-ID: <1030618232849.2835B-100000@Ives.egh.com>   & On Wed, 18 Jun 2003, Eric Bruno wrote:  D > Technically the 168LX boards are unsupported for OpenVMS. But haveF > worked with 7.2.  It just may not work. Tru64, Linux and FreeBSD allD > run with a problem. So it's most likely a hardware issue with someF > component most likely the SCSI controller since I have tried severalH > different hard disks and two different CD-ROM drives.  The common itemM > is the SCSI controller. The ethernet card is a tulip so should be ok there.8  D Are you sure the CD itself is okay?  Maybe there is a problem in the middle of one of the savesets?  B I have never tried to do "$ANA/DISK/READCHECK" on a CD, so I'm notF sure it is possible, but if you can get to $$$ (Option 7 after bootingB the CD), you could give this a try.  (I'm pretty sure ERRFMT, etc.D aren't running in the stand-alone environment, since they would haveC no where to log stuff, so maybe their absence is masking a hardwarey	 problem.)o  E I don't see how you could get as far as you did if there is somethingw fundamentally wrong.   > I > Looks like I am just going to have to get another box which is offical -J > supported list.  Island has some reburbished boxes and I will just keep  > an eye on e-bay. >  > Thanks for your help.C >  >  > Thanks for help. >  > Charlie Hammond wrote:L > > Eric, I do not see anything obvious from this.  My guess is that you areL > > running out of some resource that is causing the system to hang and alsoN > > blocking any indication of the problem.  But you probably guessed as much. > > L > > Unless someone else has any suggestions, I sugest that you escalate this1 > > formally through your normal support contact.f > >  > > P > > In article <3EEF78FD.60209@ebruno.org>, Eric Bruno <eric@ebruno.org> writes: > > : > >>Here is last file lines of the output when it freezes: > >>I > >>%PCSI-I-DELFIL, deleted DKA0:[000000.PCSI$WRK26.][SYS.HELP]CDSA.HLP;1  > >>%PSCI-I-CREFILE, created b? > >>DISK$ALPHASYS:[VMS$COMMON,][SYSLIB]CDSA$DAALPROXY_SHR.EXE;1  > >>%PSCI-I-CREFILE, created r@ > >>DISK$ALPHASYS:[VMS$COMMON,][SYSLIB]CDSA$DUMMYMULTI_SAR.ESW;1 > >>%PSCI-I-CREFILE, created D@ > >>DISK$ALPHASYS:[VMS$COMMON,][SYSLIB]CDSA$DUMMYMULTI_SAR.EXE;1 > >>%PSCI-I-CREFILE, created g? > >>DISK$ALPHASYS:[VMS$COMMON,][SYSLIB]CDSA$EAYCSP300_SHR.EXE;1  > >> > >> > >>Charlie Hammond wrote: > >>- > >>>In article <3EEEA595.20407@ebruno.org>, r) > >>>Eric Bruno <eric@ebruno.org> writes:  > >>>  > >>>t > >>>i3 > >>>>I have an Aspen Durango II (164LX) Alpha box.03 > >>>>I am having problem installing OpenVMS 7.3.1.  > >>>  > >>>..  > >>>F > >>> 2 > >>>>The install starts and gets to 60% point and: > >>>>just freezes no message no response at the keyboard. > >>>  > >>>..( > >>>  > >>> 0 > >>>>Is there way to get log or trace info ...? > >>>  > >>>e	 > >>>Yes.g > >>>   > >>>Select the DCL option, then > >>>$7 > >>>    $$$ DEFINE/SYSTEM AXPVMS$PCSI_LOG_TRACE "BOTH"  > >>>and; > >>>    $$$ DEFINE/SYSTEM AXPVMS$PCSI_EXECUTE_VERIFY "YES"e > >>> L > >>>LOGOUT from DCL to return to the main menu, then [re]do the insallation > >>>or upgrade. > >>>eL > >>>This will generate a *LOT* of output.  The last lines before the "hang"J > >>>will give an indicatin of where exactly the process is "stuck".  CopyC > >>>them _exactly_ and post them -- 5-10 lines is probably enough.c > >>>r > >> > >> > >  >  >  >    --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  # Date: Wed, 18 Jun 2003 17:59:23 GMT.& From: jlsue <jefflsxxxz@sbcglobal.net>$ Subject: Re: lib$convert_date_string8 Message-ID: <t0a1fvg81293qqfe12mokqrqr20jp25nls@4ax.com>  1 On Tue, 17 Jun 2003 09:40:22 +0100, Chris Sharman # <chris.sharman@sorry.nospam> wrote:;  H >Been playing around with lib$convert_date_string & lib$dt_input_format = >logical, in an effort to get a more forgiving input routine.e > G >Can't even figure out how to do what DCL does, & accept 2 AND 4 digit  H >years. I can pick one or the other (Z2 or Y4), but apparently not both. >Am I missing something ?  >vG >As for allowing every illiterate user's random selection of separator vF >and month format, no chance. You say "please enter dd-mmm-yyyy", and G >they use /, ., space, numeric months, etc. Best I've done is about 10  J >different format strings to try. The application can figure out which to F >try by checking the first non-digit (that's the separator), then the / >character after that is either numeric or not.  > E >Be nice to have a "figure it out" mode - the only ambiguity is with  H >numeric months when you need to rule on the field order: dmy or mdy or 
 >whatever.  I Chris, you could change your application to use lib$get_input and use therF DCL command parser instead (linking it with the properly compiled .cldJ file, of course).  I've written applications this way to relieve me of the; drudgery of writing painfully complex syntax checking code.   9 It's one of the things I love best about VMS programming.    ------------------------------  # Date: Wed, 18 Jun 2003 17:51:35 GMTi& From: jlsue <jefflsxxxz@sbcglobal.net>3 Subject: Re: Max length of Nodename in OpenVMS 7.31 8 Message-ID: <7l91fv0o89ckftoar5gncavm3s5uil2jtk@4ax.com>  : On Wed, 18 Jun 2003 16:22:05 +0100, Dave Brennan <> wrote:  ; >Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> wrote:i >eC >>>   Variable name lengths have nothing to do with data lengths.  f >>J >>Of course; my fault for not thinking before posting.  Once, years ago, IH >>heard someone cry "long live Fortran" when he heard of the 6-characterD >>limit on SCSNODE and briefly remembering this was the origin of myF >>off-the-cuff remark earlier in this thread.  Of course, as you pointH >>out, the length of the name of the variable has nothing to do with the >>contents of that variable. N > F >This may be true but you flushed a load of FORTRAN programmers out of >the woods with that one!$   Hey!  I resemble that remark!t  D I'd rather whip together a Fortran program over a C program any day.   ------------------------------  % Date: Thu, 19 Jun 2003 08:32:48 +1000 1 From: Paddy O'Brien <paddy.o'brien@tg.nsw.gov.au> 3 Subject: Re: Max length of Nodename in OpenVMS 7.31e* Message-ID: <3EF0E890.50005@tg.nsw.gov.au>   jlsue wrote:< > On Wed, 18 Jun 2003 16:22:05 +0100, Dave Brennan <> wrote: >  > < >>Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> wrote: >> >>C >>>>  Variable name lengths have nothing to do with data lengths.    >>>rK >>>Of course; my fault for not thinking before posting.  Once, years ago, IvI >>>heard someone cry "long live Fortran" when he heard of the 6-characterTE >>>limit on SCSNODE and briefly remembering this was the origin of my G >>>off-the-cuff remark earlier in this thread.  Of course, as you point6I >>>out, the length of the name of the variable has nothing to do with thet >>>contents of that variable.  >>G >>This may be true but you flushed a load of FORTRAN programmers out ofG >>the woods with that one! >  >  > Hey!  I resemble that remark!7 > F > I'd rather whip together a Fortran program over a C program any day.  F I do too for the intensive number crunching parts of my applications. E Modern whizz-bang user interfaces that everyone thinks they want are o! doable, but probably easier in C.   F And I'm happy that the majority of my workload is still with Fortran, # and has been for the past 35 years.A   Regards, Paddy      G ***********************************************************************o  C "This electronic message and any attachments may contain privileged > and confidential information intended only for the use of the B addressees named above.  If you are not the intended recipient of C this email, please delete the message and any attachment and advise-B the sender.  You are hereby notified that any use, dissemination, 7 distribution, reproduction of this email is prohibited.   A If you have received the email in error, please notify TransGrid -A immediately.  Any views expressed in this email are those of the  = individual sender except where the sender expressly and with oC authority states them to be the views of TransGrid.  TransGrid usese> virus scanning software but excludes any liability for viruses contained in any attachment.  < Please note the email address for TransGrid personnel is now$ firstname.lastname@transgrid.com.au"  G ***********************************************************************    ------------------------------  + Date: Wed, 18 Jun 2003 20:29:21 +0000 (UTC)CP From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)  Subject: Mozilla on old hardware$ Message-ID: <bcqi31$vk4$1@online.de>  I At home, I have a 3000/600 with 192 MB and a 255/233 with 64 MB.  I know  H that the requirements/recommendations/suggestions for Mozilla are for a G more powerful machine, particularly for more memory.  What will happen  B if I try it nevertheless?  Is it worth a try?  At the moment, I'm I running it on a machine a thousand miles away with the display set to my  H local machine, over a DSL line, so for that reason the current setup is - a bit sluggish anyway (it IS usable, though).s  D (A related question, independent of Mozilla: I still have 7.2-1 and ; things run fine.  Will 7.3-1 be possible with just 64 MB?) e   ------------------------------  % Date: Wed, 18 Jun 2003 22:56:46 +02000$ From: Felix Deichmann <f.dei@web.de>$ Subject: Re: Mozilla on old hardware/ Message-ID: <bcqjm9$38r$02$1@news.t-online.com>(   Phillip Helbig wrote:EK > At home, I have a 3000/600 with 192 MB and a 255/233 with 64 MB.  I know >J > that the requirements/recommendations/suggestions for Mozilla are for a I > more powerful machine, particularly for more memory.  What will happen  / > if I try it nevertheless?  Is it worth a try?r  G I tried it on a DEC 3000/400 with 64 MB. Unfortunately it is very slow g
 and unusable.C  F > (A related question, independent of Mozilla: I still have 7.2-1 and = > things run fine.  Will 7.3-1 be possible with just 64 MB?) d  & SH SYS on the machine mentioned above:   OpenVMS V7.3-1  on node THETIS       SH MEM:>   Physical Memory Usage (pages)3  ) Total        Free      In Use    Modifieda*   8192        2091        5592         509  ( (Basic system with a basic TCPIP config)     Regards    Felixb   ------------------------------  % Date: Wed, 18 Jun 2003 17:01:40 -0400 $ From: "rob kas" <news@paychoice.com>$ Subject: Re: Mozilla on old hardware/ Message-ID: <vf1kpnbeh44b96@corp.supernews.com>I    No It's REALLY slow           Rob     J > At home, I have a 3000/600 with 192 MB and a 255/233 with 64 MB.  I knowI > that the requirements/recommendations/suggestions for Mozilla are for awH > more powerful machine, particularly for more memory.  What will happenC > if I try it nevertheless?  Is it worth a try?  At the moment, I'moJ > running it on a machine a thousand miles away with the display set to myI > local machine, over a DSL line, so for that reason the current setup is / > a bit sluggish anyway (it IS usable, though).c >nE > (A related question, independent of Mozilla: I still have 7.2-1 ands< > things run fine.  Will 7.3-1 be possible with just 64 MB?)   ------------------------------  % Date: Wed, 18 Jun 2003 17:21:55 -0400n* From: JF Mezei <jfmezei.spamnot@istop.com>$ Subject: Re: Mozilla on old hardware) Message-ID: <3EF0D7DF.6D2EC3A9@istop.com>    rob kas wrote: >  >  No It's REALLY slow  C why is mozilla so slow and hungry for resources ? Is it of a design D incompatible with VMS, or just written inefficiently to begin with ?  F Seems to me that Mozilla might be a big driver for 64 bit desktops :-)   ------------------------------  % Date: Thu, 19 Jun 2003 00:01:32 +0200d$ From: Felix Deichmann <f.dei@web.de>$ Subject: Re: Mozilla on old hardware/ Message-ID: <bcqnfm$pme$06$1@news.t-online.com>c   JF Mezei wrote: E > why is mozilla so slow and hungry for resources ? Is it of a designaF > incompatible with VMS, or just written inefficiently to begin with ?  C Programs on Alpha are always quite large when compared to the same >E programs on VAX because of the Alpha's RISC design. That could be an   explanation.  H > Seems to me that Mozilla might be a big driver for 64 bit desktops :-)  > I bet Mozilla on a VAX would run well with 64 MB of memory :-)  C Why is there only a Mozilla port for Alpha? Why is Compaq's ported    source of Mozilla not available?   ------------------------------  # Date: Wed, 18 Jun 2003 22:49:44 GMT$# From: hoff@hp.nospam (Hoff Hoffman)u$ Subject: Re: Mozilla on old hardware1 Message-ID: <c06Ia.2908$IU1.126@news.cpqcorp.net>   w In article <bcqi31$vk4$1@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:zJ :At home, I have a 3000/600 with 192 MB and a 255/233 with 64 MB.  I know I :that the requirements/recommendations/suggestions for Mozilla are for a  H :more powerful machine, particularly for more memory.  What will happen / :if I try it nevertheless?  Is it worth a try? m  D   Mozilla will function, but will be very slow -- your definition ofD   "slow" and "usable" may vary from mine of course, but having triedD   this configuration with an AlphaStation 255/300 with 192 MB, I didC   not wish to continue with that particular effort.  But after somehC   Alpha hardware upgrades, I now run Mozilla full-time and its very A   nice -- I'm in the midst of downloading the current Mozilla 1.4    baselevel right now.  E   My own general recommendations for Mozilla are for an EV56 or later F   Alpha processor, and for 256 MB (or more) of system physical memory.  E :(A related question, independent of Mozilla: I still have 7.2-1 and 2< :things run fine.  Will 7.3-1 be possible with just 64 MB?)   F   It's gonna be tight -- I expect it will work, but you'll be spendingG   a non-trivial portion of yoyur I/O and memory bandwidths with paging.j    N  ---------------------------- #include <rtfaq.h> -----------------------------J       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com-   ------------------------------   Date: 18 Jun 2003 20:23:35 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)/ Subject: Re: NFS Client to MS-Services for Unix-5 Message-ID: <bcqho7$lohjl$2@ID-135708.news.dfncis.de>-  ; OK, I got my testbed up and have NFS working to a Unix box.[ When I try to mount from VMS+   ucx mount dnfs2: /host=host1 /path="/NFS"1  $ I get no device available for DNFS2:  ! What did I miss setting this up??l   bill   -- rJ 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: Wed, 18 Jun 2003 20:16:34 +0000@. From: Graham Burley <100625.30@compuserve.com>, Subject: Re: PIPE output defined to a symbol. Message-ID: <3EF0C8A2.781A1257@compuserve.com>   John Brandon wrote:a > I > I used the examples provided earlier in this thread to do a performance.F > comparison between the PIPE (NEW) and /OUTPUT to disk (OLD) methods.   [snip]  O > In this example, the NEW method will have the following impact on performance- > over the OLD method: > 1 > Increase        [Buffered I/O count]    by 230% 0 > Decrease        [Direct I/O count]      by 25%0 > Decrease        [Page faults]           by 55%0 > Decrease        [Charged CPU time]      by 47% > M > Overall a reduction in system resources for said commands.  Your milage maye > vary.    [snip]  @ Surely you forgot to account for the PIPEs subprocesses, the NEW/ method creates 4 subprocesses, from one sample:o  <                   DirIO   BufIO  Faults     CPU      Elapsed  < OLD                  28      58     195      10  00:00:00.46  < NEW Parent            6      62       5       2  00:00:00.17/ NEW Children (4)      7     475     454       9r=                   -------------------------------------------t< NEW Total            13     537     459      11  00:00:00.17  < Change              -15    +479    +264      +1 -00:00:00.29  = Looks to me like NEW uses more system resources than OLD, buta> because OLD does disk IO it takes longer, moving the temporary% files to a RAM disk proves the point:i  < OLD to RAM disk      24      58     242       8  00:00:00.08    > For info I used this procedure to measure the runs, it creates> a couple of symbols to help extract the accounting records for the PIPEs subprocesses.     E $!TEST.COM - collect some resource usage info for a DCL procedure runn, $! P1 specifies the command procedure to run $! $ if P1 .eqs. "" then $exit %x2 / $ user = f$edit(f$getjpi("","USERNAME"),"TRIM")t $ dio_0 = f$getjpi("","DIRIO") $ bio_0 = f$getjpi("","BUFIO")! $ pgf_0 = f$getjpi("","PAGEFLTS")o $ cpu_0 = f$getjpi("","CPUTIM")a $ start = f$time() $ @'p1'. $ end = f$time() $ dio_1 = f$getjpi("","DIRIO") $ bio_1 = f$getjpi("","BUFIO")! $ pgf_1 = f$getjpi("","PAGEFLTS")s $ cpu_1 = f$getjpi("","CPUTIM")n $ t0 = f$cvtime(start,,"TIME")) $ elap = f$cvtime("''end'-''t0'",,"TIME")yB $ write sys$output "   DirIO   BufIO  Faults     CPU      Elapsed"2 $ write sys$output f$fao("!8SL!8SL!8SL!8SL  !AS" -K         , dio_1 - dio_0, bio_1 - bio_0, pgf_1 - pgf_0, cpu_1 - cpu_0, elap)P $!H $! Define symbols to account for any subprocesses, it will take a littleE $! while for these to hit the accounting data file. Assumes that userdA $! isn't doing much else on the system (check Owner PIDs). AddingdK $! /BEFORE="''end'" will cause some records to be missed. Adding /OWNER=pidnF $! doesn't help because a PIPE subprocess creates subprocesses itself. $!2 $ test_acc == "acc/sinc=""''start'""/user=''user'"I $ test_acc_sum == test_acc + "/summ/rep=(rec,dir,buf,fault,proc,elapsed)"p     Graham   ------------------------------    Date: 18 Jun 2003 15:27:37 -0500 From: briggs@encompasserve.org, Subject: Re: PIPE output defined to a symbol3 Message-ID: <NPVF5oY5Udcg@eisner.encompasserve.org>r  _ In article <3EF0C8A2.781A1257@compuserve.com>, Graham Burley <100625.30@compuserve.com> writes:mB > Surely you forgot to account for the PIPEs subprocesses, the NEW1 > method creates 4 subprocesses, from one sample:-  
 $ pipe ( -  show memory | -%  search sys$pipe "Private Memory" | -6  ( -  read sys$pipe $tmp$ ; -   define/job/nolog $tmp$ &$tmp$ -  ) -  )   That's three subprocesses.   1 for show mem 1 for search sys$pipem3 1 for the sequence of read sys$pipe and define /jobr   Accounting confirms this.s   	John Briggs   ------------------------------  % Date: Wed, 18 Jun 2003 23:40:06 +0000n. From: Graham Burley <100625.30@compuserve.com>, Subject: Re: PIPE output defined to a symbol. Message-ID: <3EF0F856.45DBDAC3@compuserve.com>   briggs@encompasserve.org wrote:e > a > In article <3EF0C8A2.781A1257@compuserve.com>, Graham Burley <100625.30@compuserve.com> writes: D > > Surely you forgot to account for the PIPEs subprocesses, the NEW3 > > method creates 4 subprocesses, from one sample:o >  > $ pipe ( - >  show memory | -' >  search sys$pipe "Private Memory" | -e >  ( - >  read sys$pipe $tmp$ ; -" >  define/job/nolog $tmp$ &$tmp$ - >  ) - >  ) >  > That's three subprocesses. >  > 1 for show mem > 1 for search sys$pipe 5 > 1 for the sequence of read sys$pipe and define /job- >  > Accounting confirms this.a >  >         John Briggs4  < Yes, 3 makes sense, but I still see 4. What's going on here?  G I changed GET_MEMORY_NEW.COM, I changed the search string (I don't have G "Private" in SHOW MEMORY) and added the /NOSYMBOL/NOLOGICAL qualifiers.r   $ type get_memory_new.com  $!> $! this is the NEW method to get a SEARCH string into a symbol $! $ pipe /nosym/nolog ( -i          show memory | -*          search sys$pipe "Main Memory" | -          ( -"            read sys$pipe $tmp$ ; -*            define/job/nolog $tmp$ &$tmp$ -          ) -        )F $ private_memory = f$element(0,")",f$element(1,"(",f$trnlnm("$tmp$")))C $ write sys$output "This server has ''private_memory' total memory"    $ @test get_memory_new% This server has 128.00Mb total memoryt-    DirIO   BufIO  Faults     CPU      Elapsed -        6      62       9       0  00:00:00.20    $ sho sym test_acc:   TEST_ACC == "acc/sinc="18-JUN-2003 22:06:54.51"/user=GB"   $ test_acc/full/out=a.a $ $ sea a.a process,owner,start,finish SUBPROCESS Process TerminationO Account:           GB                Finish time:       18-JUN-2003 22:06:54.70eO Process ID:        000001B3          Start time:        18-JUN-2003 22:06:54.62tO Owner ID:          000001B1          Elapsed time:                0 00:00:00.08aO Terminal name:                       Processor time:              0 00:00:00.03  SUBPROCESS Process TerminationO Account:           GB                Finish time:       18-JUN-2003 22:06:54.70sO Process ID:        000001B4          Start time:        18-JUN-2003 22:06:54.64 O Owner ID:          000001B1          Elapsed time:                0 00:00:00.06iO Terminal name:                       Processor time:              0 00:00:00.01i SUBPROCESS Process TerminationO Account:           GB                Finish time:       18-JUN-2003 22:06:54.70 O Process ID:        000001B2          Start time:        18-JUN-2003 22:06:54.60eO Owner ID:          000001B1          Elapsed time:                0 00:00:00.10 O Terminal name:                       Processor time:              0 00:00:00.02. SUBPROCESS Process TerminationO Account:           GB                Finish time:       18-JUN-2003 22:06:54.71 O Process ID:        000001B1          Start time:        18-JUN-2003 22:06:54.52tO Owner ID:          000000BF          Elapsed time:                0 00:00:00.19SO Terminal name:                       Processor time:              0 00:00:00.05n  % $ write sys$output f$getjpi("","PID")f 000000BF   Looking at the process owners:>  000000BF --> 000001B1 --> ( 0000001B2, 0000001B3, 0000001B4 )     Graham   ------------------------------  % Date: Wed, 18 Jun 2003 14:40:41 -0400.* From: JF Mezei <jfmezei.spamnot@istop.com>I Subject: Re: SEC:U MIT-MAGIC-COOKIE-1, Motif 1.3 and HP TCP/IP XDM (long).) Message-ID: <3EF0B21F.30273AAB@istop.com>3   Fred Kleinsorge wrote:K > THERE WAS NO SUPPORT IN XLIB FOR IT, SO XDM DIDN'T INCLUDE IT!  NOW THEREl@ > IS, AND IN SOME FUTURE RELEASE IT SHOULD BE CHANGED TO USE IT.  J I am curious. Why with the XDM magic cookies authentication be included in@ X-lib ?  These happen well before an "x" functions are started.   N Also, note that the current incarnation ( 5.3) does support magic-cookies, but not the MIT-MAGIC-COOKIE.    ------------------------------  # Date: Wed, 18 Jun 2003 21:41:15 GMTi# From: hoff@hp.nospam (Hoff Hoffman) / Subject: Re: Speed of IDE controllers in DS10 ?o2 Message-ID: <%%4Ia.2897$IU1.1816@news.cpqcorp.net>  E In article <3EDECA1F.70805@home.nl>, Dirk Munk <munk@home.nl> writes: 0 :You were right Hoff, it is 16 MHz or 33 MB/sec.P :Today I received two new DS10 systems, and a accompanying paper stated this 16 P :MHz. That means that it is a hardware related matter, or could it be a (fixed) N :firmware setting ? It would be nice if we could set it to 33 MHz or 66 MB/sec	 :since...t  F   AFAIK, this is "simply" a case of updating the DQDRIVER driver code.  C   Source code of an earlier version is available on the Freeware.     H   I don't know off-hand if there are plans to rework DQDRIVER for fasterC   ATA access and faster transfer modes, though the DQDRIVER code is F   certainly seeking active maintenance and various extension.  (I haveF   tweaked DQDRIVER to allow a particular and unsupported ATA device toI   work, and other engineers have been making far more extensive changes.)h  H   I don't believe that the firmware is in the way here, but I don't knowG   that for certain -- I *think* this is specifically DQDRIVER involved.i  N  ---------------------------- #include <rtfaq.h> -----------------------------J       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.comN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.comr   ------------------------------    Date: 18 Jun 2003 16:15:42 -0700) From: getzler15@yahoo.com (Heinz Getzler)o$ Subject: Re: Sun to be the next DEC!= Message-ID: <a58727ec.0306181515.1101c8ce@posting.google.com>l  m bob@instantwhip.com (Bob Ceculski) wrote in message news:<d7791aa1.0306160915.37f9047b@posting.google.com>...P5 > gee Andrew, I hope this means we willn't be hearing 5 > from you anymore :) ... there is still time to join ! > the OpenVMS Itanium sales team!a > M > http://techupdate.zdnet.com/techupdate/stories/main/0,14179,2914032,00.htmlg  D In many ways Sun's problems are similar to HP's. Sun is primarilly aF hardware  company. In the past few years Sun has made some attempts to> become more of a software company with it's Sun One products &: Java.... Sun's big problem is that it is still paranoid ofD Linux............IBM on the otherhand is not afraid of Linux and hasE embraced it more than any other company. If Sun were smart they wouldo push Java to the Linux market.F Another big problems is that a most of their solutions are designed toF support the Solaris platform only. N-1 is Sun's data center management= solution that is designed to support Sun hardware only. IBM'seE On-demand stategy is a solution designed to support multiple OS's andr	 hardware.r   Profitabliity:  ? Now Sun can claim that they have multiple quarters of continousDA profitablity and 5 billion in cash. But it's uncertain if this is.3 enough to keep them going over the long term.....??p   ------------------------------  % Date: Wed, 18 Jun 2003 16:21:54 -0400d, From: "warren sander" <warren.sander@hp.com> Subject: Re: VMS Freeware CDs , Message-ID: <3ef0ce5d$1@usenet01.boi.hp.com>  I if you all want a 'honking big' downloadable file of the freeware cd(s) Iv
 can do it.H What format do you want (backup save set compressed w/spool?), zip file?   -warren    -- 0K ---------------------------------------------------------------------------.7 Warren Sander                   WW E-Marketing (HP.COM)8B Hewlett-Packard Company         Work:  warren.sander@remove.hp.comK 200 Forest Street MR01-3/K8     Personal: sander.ma.ultranet@remove.rcn.comy. Marlboro, MA 01752              (508) 467-48755    My opinions are my own and I only speak for myselfk*          Read http://www.hp.com/go/openvmsK ---------------------------------------------------------------------------f    . "John Smith" <a@nonymous.com> wrote in messageC news:7PNHa.129895$3Sm.40260@news01.bloor.is.net.cable.rogers.com...  > . > <VAXman- @SendSpamHere.ORG> wrote in message, > news:00A21860.36AE95A1@SendSpamHere.ORG..., > > In article <bcnfrr$roa$2@pcls4.std.com>,; > moroney@world.std.spaamtrap.com (Michael Moroney) writes:eG > > >I can host files if necessary.  I have a DSL line with no transfer. > limit. > > >-- 
 > > >-Mike > >oF > > FYI, I've had the freeware CDs V1 through V5 hosted at my site for > some > > time and I've got a T1.  >a >vG > So all that's required is someplace to list your site, Michael's, andn! > others who are willing to host./ >m >  >.   ------------------------------  # Date: Thu, 19 Jun 2003 05:23:45 GMT ) From: Andrew Balaam <abalaam@yahoo.co.uk>V Subject: Re: VMS Freeware CDsf3 Message-ID: <20030619.5234500.580405386@imagnu.geo>y  6 >>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<<  F On 18/06/03, 21:21:54, "warren sander" <warren.sander@hp.com> wrote=20 regarding Re: VMS Freeware CDs:o    I > if you all want a 'honking big' downloadable file of the freeware cd(s=  ) Iy > can do it.I > What format do you want (backup save set compressed w/spool?), zip fil=  e?  	 > -warrene   > -- >=20I ------------------------------------------------------------------------=e ---o9 > Warren Sander                   WW E-Marketing (HP.COM)sD > Hewlett-Packard Company         Work:  warren.sander@remove.hp.com. > 200 Forest Street MR01-3/K8     Personal:=20! sander.ma.ultranet@remove.rcn.com 0 > Marlboro, MA 01752              (508) 467-48757 >    My opinions are my own and I only speak for myself , >          Read http://www.hp.com/go/openvms >=20I ------------------------------------------------------------------------=  ---e    0 > "John Smith" <a@nonymous.com> wrote in messageE > news:7PNHa.129895$3Sm.40260@news01.bloor.is.net.cable.rogers.com...n > >u0 > > <VAXman- @SendSpamHere.ORG> wrote in message. > > news:00A21860.36AE95A1@SendSpamHere.ORG.... > > > In article <bcnfrr$roa$2@pcls4.std.com>,= > > moroney@world.std.spaamtrap.com (Michael Moroney) writes:-I > > > >I can host files if necessary.  I have a DSL line with no transfe=d re
 > > limit.	 > > > >--R > > > >-Mike > > >.I > > > FYI, I've had the freeware CDs V1 through V5 hosted at my site for=>   > > some > > > time and I've got a T1.e > >e > > I > > So all that's required is someplace to list your site, Michael's, an=  d3# > > others who are willing to host.r > >. > >  > >t  I A gzipped image backup with a published block size would be good. That=20nI way, someone could create their own CD, or just extract the files from=20s it.e   ------------------------------    Date: 18 Jun 2003 22:18:37 -0700. From: mistdragon@zdnetonebox.com (mist dragon)< Subject: Re: You say you want apps for VMS?  Here they come!= Message-ID: <7500353b.0306182118.4355f59c@posting.google.com>l  O > So that article says that Oracle will port some software to VMS-IA64. Will ittK > port ALL of its Oracle software to VMS ?  If not, will it port all of the 9 > software presently available on VMS-Alpha to VMS-IA64 ?h  B AlphaVMS port is database only with non-Oracle standard installer.A Even managing tools are on Windows NT (Note NT, not win2k or xp).SF Afaik there is no question that when AlphaVMS is dropped, its replacedF with equivalent product on IA32 in order to support current customers.  9 Please note that AlphaVMS still has some software (OracleeC Applications), that's EOL, e.g. not being upgraded, but still sold.a  B So yes, it will port all it has and is not EOL that is in fact its database :)h  < Afaik the port has long ago moved away from main business toD cheap-labor country and does not present the mainstream interests ofF Oracle that does unix first and vms 90 days later when unix is stable.   Mist   ------------------------------  # Date: Wed, 18 Jun 2003 23:37:05 GMTo6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)I Subject: Re: [ENABLE AUTOSTART] How to determine the state of AUTOSTART ?t3 Message-ID: <BI6Ia.10407$RM6.134907@news.chello.at>   W In article <d1111de.0306180036.4da3d129@posting.google.com>, sam@ratex.dk (Sam) writes:i >Try the following:i >a  >$ help lexical f$getqui example >M >and look at example number 4. >maybe it is what you need.   G No. I need a SHOW QUEUE/MANAGER/FULL which shows the state of AUTOSTARTeE for every queue manager and every node in the cluster. And I need thePJ corresponding item code for F$GETQUI ("DISPLAY_MANAGER", ...) as expected./ If it is in memory, it should be displayable...x  A I try to reproduce a symptom I saw recently where a newly created3F /AUTOSTART_ON=(node1::mumble,node2::mumble,node3::mumble) queue (whichC also was /STARTed) didn't autostart until I did a ENABLE AUTOSTART.oM Strange thing was, AUTOSTART was already enabled on all nodes in the cluster.dD At least the other umpteenhundred autostart queues believed this ;-)   -- e Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialistw E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Wed, 18 Jun 2003 14:42:37 -0400e* From: JF Mezei <jfmezei.spamnot@istop.com>L Subject: Re: [OT] Future of Sun in doubt ?, was: Re: Sun to be the next DEC!) Message-ID: <3EF0B293.FDDDAE20@istop.com>i   Marty Kuhrt wrote:B > Something like... "Over the last ten years Gartner has made 3000> > predictions, of which 2900 have been WRONG.  Why would _YOU_% > trust anything Gartner has to say?"a  J You don't understand. Gartner doesn't predict the future. It helps justifyJ decisions made by CIOs (IT directors) by producing reports that back theirL decisions for choosing such and such technology. Gartner is not independant.N It is paid by its customers. Gartner isn't going to go tell its customers they are making the wrong decisions.E   ------------------------------   End of INFO-VAX 2003.337 ************************