1 INFO-VAX	Sat, 26 Nov 2005	Volume 2005 : Issue 659       Contents:! Re: $GETDVI suggestions requested ! Re: $GETDVI suggestions requested ! Re: $GETDVI suggestions requested  Re: A small matter of detail... ) ANU News: Erro building REBUILD_ITEM_FILE ) ANU News: Erro building REBUILD_ITEM_FILE - Re: ANU News: Erro building REBUILD_ITEM_FILE - Re: Calculating size of the virtual I/O cache 0 CSWS PHP 1.3 and GD support (PHP Version 4.3.10)4 Re: CSWS PHP 1.3 and GD support (PHP Version 4.3.10)* Re: Decent alpha system for OpenVMS newbie* Re: Decent alpha system for OpenVMS newbie* Re: Decent alpha system for OpenVMS newbie* Re: Decent alpha system for OpenVMS newbie* Re: Decent alpha system for OpenVMS newbie* Re: Decent alpha system for OpenVMS newbie* Re: Decent alpha system for OpenVMS newbie/ Re: HP : Massive strike and protest march today F laughed. cried:::::   VAXen looking for a new home (near Montreal, QC)= Re: Selling Alphaserver 256MB memory modules & other stuff...  Re: System services and DCL  Re: System services and DCL  Re: System services and DCL  Re: System services and DCL  T4 Graphs via TLviz 0 VAXen looking for a new home (near Montreal, QC)( Vaxstation 3100 1gig limit & MOP booting, Re: Vaxstation 3100 1gig limit & MOP booting  F ----------------------------------------------------------------------  % Date: Fri, 25 Nov 2005 14:35:22 -0600 2 From: David J Dachtera <djesys.nospam@comcast.net>* Subject: Re: $GETDVI suggestions requested+ Message-ID: <43877589.7EF7479C@comcast.net>    Rob Brooks wrote:  > 6 > David J Dachtera <djesys.nospam@comcast.net> writes: > K > > Well, if F$MULTIPATH() will help me get a list of paths to a device, is I > > there a way to get the I/O counts per path so I can write some DCL to I > > see how well the system is doing at balancing the I/O load across the $ > > available auto-switchable paths? > Q > Multipath does not automatically switch the paths in order to balance I/O load.   E What I was referring to there is the (documented) behavior where in a G path switch may occur as devices are being MOUNTed based on the factors  documented.   N > It *may* switch paths if there an I/O error on the current path.  I won't goK > into a long description here about the mechanics of how and why we select H > paths.  The "Guidelines to Cluster Configurations" manual has a decent8 > explanation about multipath; it may be in chapter six. > P > There are plans to look into "concurrent multipath", such that I/O can go downK > more than one path at a time.  If that happens, it'll be post-V8.3 in any  > event. > N > To answer your question more fully, we don't keep track of I/O on a pathwideH > basis across a system.  For any given device, one can use $GETDVI withK > dvi$_opcnt and the various path names to get the operation count for each  > path per device.  G However, there is no F$GETDVI() equivalent to this (or, if there is, it F is not effectively documented). One must still get/parse the output of= SHOW DEVICE/FULL to get the operation counts per path, AFAIK.    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  % Date: Fri, 25 Nov 2005 14:36:03 -0600 2 From: David J Dachtera <djesys.nospam@comcast.net>* Subject: Re: $GETDVI suggestions requested+ Message-ID: <438775B3.3348AE55@comcast.net>   
 R Boyd wrote:  >  > Rob, > H > An item that I would dearly like to see -- have wanted since the earlyH > days of clustering is a DVI code that would return the same value thatF > you see in the MNT CNT column output by the DCL SHOW DEVICE command.D > This number represents the number of cluster members that have theH > device mounted.  There have been a number of times where I have wantedB > to use this number and have had to resort to the usual tricks to) > extract it from the SHOW DEVICE output.   $ OH, YESS!!!!! *STRONGLY* seconded!!!   --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  % Date: Fri, 25 Nov 2005 21:31:53 -0800 ( From: Jeff Cameron <roktsci@comcast.net>* Subject: Re: $GETDVI suggestions requested0 Message-ID: <BFAD3349.183B5%roktsci@comcast.net>  L On 11/22/05 5:59 AM, in article 11o691qgj7tr46@corp.supernews.com, "Syltrem"! <syltremzulu@videotron.ca> wrote:    > D > "Guy Peleg" <guy.peleg@remove_this_header@hp.com> wrote in message( > news:4383099a$1@usenet01.boi.hp.com... > F >> V8.3 will add a new lexical function - F$CUNITS to converts variousL >> units, currently it can only handle converting blocks to bytes, using the >> new >> function: >>  8 >> IPL31> write sys$output f$getdvi("dsa0","freeblocks") >> 26431038 A >> IPL31> write sys$output f$cunit(f$getdvi("dsa0","freeblocks")) 
 >> 12.60GB	 >> IPL31>  >>   >  > Guy, > K > Why f$cunits, and not f$CVunits when we already have f$CVtime, f$CVsi and 	 > f$CVui? B > All other Conversion lexicals start with CV... easy to remember. > H > BTW this is most useful ! Doing it in DCL with integers only, has someL > limitations.. although there's supposed to be a f$math or something coming > soon (you talked about this).  > 	 > Thanks! 
 > Syltrem  >  > K There is a problem with displaying number of bytes. In DCL with disk drives L in the size that are now common, the number of bytes can easily overflow the DCL integer symbol type.   ------------------------------  % Date: Sat, 26 Nov 2005 01:55:56 +0800  From: prep@prep.synonet.com ( Subject: Re: A small matter of detail...- Message-ID: <87r7948vdf.fsf@prep.synonet.com>   ) Ken Robinson <kenrbnsn@gmail.com> writes:   C > On 11/23/05, prep@prep.synonet.com <prep@prep.synonet.com> wrote:   < >> I just got a CD kit for Alphas, the DEC04 one and went to= >> have a look see. Well, knowing that CSWING and ISO CDs are A >> not on speaking terms, I copied the openvms directory from the C >> system tools disk to a disk with about 1.7Gb free... And ran out 
 >> of space!!   D >> And when I look at the wreckage, I find 9 or 10 copies of all theI >> files! It seems that the copy moved the entire subtree, then went back B >> and did it all over again, and again... untill the disk filled.  H > What account were you logged in as when you did the copy, and/or whereD > was your current directory? Also, if you were logged into a system@ > that was part of a cluster, how many nodes are in the cluster?  E Default was a scratch directory on a `user' disk, with the [.openvms] E directory created. No searchlists involved. And using my own account. D It would apear that the 10 copy directorys are because they happened/ before the disk filled up. The rest only had 9.   D Note, the `System Tools' CD is an ISO9660 one, not ODS-2 as far as I know.    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------    Date: 25 Nov 2005 11:55:42 -07001 From: nothome@spammers.are.scum (Malcolm Dunnett) 2 Subject: ANU News: Erro building REBUILD_ITEM_FILE, Message-ID: <$mKpXClz81Ng@malvm9.mala.bc.ca>  1 I'm trying to build the REBUILD_ITEM_FILE program . that comes with ANU news, but it wants to link/ against VAXCTRL.EXE, which I don't have ( Alpha 1 VMS 7.3-2). I can link it against VAXCTRL.OBJ but & I get the following undefined symbols:  & %LINK-W-NUDFSYMS, 2 undefined symbols: %LINK-I-UDFSYM,         C$RFI ! %LINK-I-UDFSYM,         STRNNCPY  3 %LINK-W-USEUNDEF, undefined symbol C$RFI referenced )         in psect $LINK$ offset %X00000210 Z         in module RBLDITMFILE file SUPDISK:[ANUNEWS_V62.NEWS_BUILD]REBUILD_ITEM_FILE.OBJ;16 %LINK-W-USEUNDEF, undefined symbol STRNNCPY referenced)         in psect $LINK$ offset %X00000410 Z         in module RBLDITMFILE file SUPDISK:[ANUNEWS_V62.NEWS_BUILD]REBUILD_ITEM_FILE.OBJ;1  # Any ideas where I might find these?    ------------------------------  + Date: Fri, 25 Nov 2005 14:21:13 -0600 (CST) * From: sms@antinode.org (Steven M. Schweda)2 Subject: ANU News: Erro building REBUILD_ITEM_FILE2 Message-ID: <05112514211301_2038971B@antinode.org>  1 From: nothome@spammers.are.scum (Malcolm Dunnett)   3 > I'm trying to build the REBUILD_ITEM_FILE program 0 > that comes with ANU news, but it wants to link1 > against VAXCTRL.EXE, which I don't have ( Alpha 3 > VMS 7.3-2). I can link it against VAXCTRL.OBJ but ( > I get the following undefined symbols:  C    I don't know how the rest of this stuff is built, but the normal @ procedure with DEC/Compaq/HP C is just say LINK and hope for the* best, with no reference to VAXC<anything>.  ( > %LINK-W-NUDFSYMS, 2 undefined symbols:  > %LINK-I-UDFSYM,         C$RFI # > %LINK-I-UDFSYM,         STRNNCPY    ?    I don't know anything about C$RFI, but STRNNCPY looks like a G misspelling of strncpy() (and strncpy() _is_ a real C RTL function).  I F don't see a STRNNCPY anywhere in SYS$LIBRARY (on my Alpha at V7.3-2). 7 Nor C$RFI.  (I don't have a powered VAX at the moment.)   H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode-org     Saint Paul  MN  55105-2547    ------------------------------   Date: 25 Nov 05 17:48:39 EST) From: cook@wvnvms.wvnet.edu (George Cook) 6 Subject: Re: ANU News: Erro building REBUILD_ITEM_FILE! Message-ID: <bFOyQziU3jpX@wvnvms>   ` In article <$mKpXClz81Ng@malvm9.mala.bc.ca>, nothome@spammers.are.scum (Malcolm Dunnett) writes:3 > I'm trying to build the REBUILD_ITEM_FILE program 0 > that comes with ANU news, but it wants to link1 > against VAXCTRL.EXE, which I don't have ( Alpha 3 > VMS 7.3-2). I can link it against VAXCTRL.OBJ but ( > I get the following undefined symbols: > ( > %LINK-W-NUDFSYMS, 2 undefined symbols:  > %LINK-I-UDFSYM,         C$RFI # > %LINK-I-UDFSYM,         STRNNCPY  5 > %LINK-W-USEUNDEF, undefined symbol C$RFI referenced + >         in psect $LINK$ offset %X00000210 \ >         in module RBLDITMFILE file SUPDISK:[ANUNEWS_V62.NEWS_BUILD]REBUILD_ITEM_FILE.OBJ;18 > %LINK-W-USEUNDEF, undefined symbol STRNNCPY referenced+ >         in psect $LINK$ offset %X00000410 \ >         in module RBLDITMFILE file SUPDISK:[ANUNEWS_V62.NEWS_BUILD]REBUILD_ITEM_FILE.OBJ;1 > % > Any ideas where I might find these?   4 You need to link it with the ANUnews NEWSRTL module.     George Cook  WVNET    ------------------------------  % Date: Fri, 25 Nov 2005 15:05:00 -0600 2 From: David J Dachtera <djesys.nospam@comcast.net>6 Subject: Re: Calculating size of the virtual I/O cache+ Message-ID: <43877C7B.D09A6D31@comcast.net>    Ian Miller wrote:  > 2 > >why none of your volumes is in "Full XFC mode".+ > because they have not yet implemented it?   G Well, actually, yes they have. IIRC, it was issued as an ECO for V7.3-1 H and became the default for V7.3-2. See MC SYSMAN HELP SYS_P VCC_FLAGS in V7.3-2 or later.   --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  % Date: Sat, 26 Nov 2005 05:41:26 +1030 * From: Mark Daniel <mark.daniel@vsm.com.au>9 Subject: CSWS PHP 1.3 and GD support (PHP Version 4.3.10) / Message-ID: <43875cea$1@duster.adelaide.on.net>   = If my reporting utility is correct this seems rather limited:         * GD support: yes       * GD Version: 2.0 or higher      * FreeType Support: no       * T1Lib Support: no      * GIF Read Support: no       * GIF Create Support: no       * JPG Support: no      * PNG Support: no      * WBMP Support: no       * XBM Support: no+      * JIS-mapped Japanese Font Support: no   G My understanding of this report would be that I can create graphics to  / my heart's content - I just can't display them.   E Also, and I'm never sure if you guys want to hear about this sort of  G stuff, I consistently get the following from deep down in the sharable  G image (unfortunately it's a non-trivial PHP application - phpMyAdmin),  I way out of my depth.  The rest of the PHP application, and others, works  D quite well.  Don't think it's my fault.  Might help you pinpoint an  obscure issue.   OpenVMS Alpha V8.2! CPQ-AXPVMS-CSWS_PHP-V0103--1.PCSI    > %SYSTEM-F-HPARITH, high performance arithmetic trap, Imask=00000000, Fmask=00000002, summary=02, PC=00000000001E9C94, PS=0000001B P > -SYSTEM-F-FLTINV, floating invalid operation, PC=00000000001E9C94, PS=0000001B1 > %TRACE-F-TRACEBACK, symbolic stack dump follows R >   image    module    routine             line      rel PC           abs PC      Q >  PHPSHR                                     0 0000000000197C94 00000000001E9C94 Q >  PHPSHR                                     0 000000000015E1AC 00000000001B01AC Q >  PHPSHR                                     0 0000000000164F94 00000000001B6F94 Q >  PHPSHR                                     0 00000000001635F4 00000000001B55F4 Q >  PHPSHR                                     0 000000000016417C 00000000001B617C Q >  PHPSHR                                     0 0000000000185088 00000000001D7088 Q >  PHPSHR                                     0 0000000000187C24 00000000001D9C24 Q >  PHPSHR                                     0 0000000000187C24 00000000001D9C24 Q >  PHPSHR                                     0 0000000000174BB8 00000000001C6BB8 Q >  PHPSHR                                     0 0000000000142D54 0000000000194D54 Q >  PHPWASD  PHPWASD  ProcessRequest       43135 0000000000001BEC 0000000000031BEC Q >  PHPWASD  PHPWASD  main                 42757 00000000000005A8 00000000000305A8 Q >  PHPWASD  PHPWASD  __main                   0 000000000000006C 000000000003006C Q >                                             0 FFFFFFFF80325F94 FFFFFFFF80325F94    ------------------------------  % Date: Fri, 25 Nov 2005 21:30:37 -0500 4 From: "Peter Weaver" <newsgroup@weaverconsulting.ca>= Subject: Re: CSWS PHP 1.3 and GD support (PHP Version 4.3.10) : Message-ID: <iNPhf.33053$gK4.977986@news20.bellglobal.com>  7 "Mark Daniel" <mark.daniel@vsm.com.au> wrote in message ) news:43875cea$1@duster.adelaide.on.net... ? > If my reporting utility is correct this seems rather limited:  >  >      * GD support: yes" >      * GD Version: 2.0 or higher >      * FreeType Support: no  >      * T1Lib Support: no >      * GIF Read Support: no  >      * GIF Create Support: no  >      * JPG Support: no >      * PNG Support: no >      * WBMP Support: no  >      * XBM Support: no- >      * JIS-mapped Japanese Font Support: no  >...  I Thank you, I posted the same thing on the VAMP board but nobody responded H with either yes or no, so I started wondering if it was just me. I triedL recompiling PHP_GD using HAVE_LIBJPEG, it compiles with just one Information@ error, but GD_INFO() reports the same thing. If I recompile withH HAVE_LIBFREETYPE I get a lot of errors in GDFT.C and PHP_GD doesn't evenK load. Not being a C programmer I'm not sure how to fix all the errors. BTW: J I have recompiled with the default HAVE_* and PHP_GD works the same as the< original version, so I assume my build procedure is correct.   ------------------------------    Date: 25 Nov 2005 10:49:27 -0800+ From: "jfragemann" <jfragemann@hotmail.com> 3 Subject: Re: Decent alpha system for OpenVMS newbie C Message-ID: <1132944567.879046.190780@g49g2000cwa.googlegroups.com>    Thanks for the replies.   G Hans, I found a couple of these white boxen on ebay. Which one of these A would be preferrable to run OpenVMS and test web server software?   C I understand that these machines need to be patched to run OpenVMS.   ? http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=3D5834413668 ? http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=3D5834417436     The 5000 looks cooler though ;-)   Kind Regards, J=FCrgen   ------------------------------  # Date: Fri, 25 Nov 2005 18:30:56 GMT 9 From: Alan Adams <alan.adams@orchard-way.freeserve.co.uk> 3 Subject: Re: Decent alpha system for OpenVMS newbie ? Message-ID: <c1ef77cf4d.Alan.Adams@orchard-way.freeserve.co.uk>   A In message <1132927692.524319.87190@g47g2000cwa.googlegroups.com> 6           "jfragemann" <jfragemann@hotmail.com> wrote:   > Hello, >  > H Vlems wrote:I > > check Ebay for Digital Server auctions. The Digital Server 3000, 5000 I > > and 7000 series are equivalent to the AlphaServer 800, 1200 and 4100.  > F > Thanks for the suggestion. I can see that there are a couple on ebay > right now. > ? > http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=5834413668 ? > http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=5834417436  > H > Currently, no one seems to be bidding.  Which system would you choose? >  > A Digital Server 5000  > Part number: FR-K8F5W-AA >  >     * Alpha 5/533MHz CPU >     * 128MB Memory >     * 3 x 4GB SCSI Discs >     * Graphics Card 
 >     * CDRom  >     * Floppy Disc  >  > OR >  > A Digital Server 3000  > Part number: FR-K7F4W-WA >  >     * Alpha 500MHz CPU >     * 256MB Memory >     * 4 x 9GB SCSI Discs >     * Mylex Raid Controller  >     * S3 Graphics Card >     * DE500 Ethernet
 >     * CDRom  >     * Floppy Disc  >     * No O/S?  > B > As a newcomer, I have no idea which system is perferrable to runD > OpenVMS on. Also, I understand that they would need to be patched.  L I'm not sure of the relative processor speeds of these two, but the 3000 hasG twice the memory - always a good thing - and the ability to set up RAID F discs. No OS would not be a problem for a hobbyist application, as the# license is free, and the CDs cheap.   F I hope somebody else will enlighten you about the compatability of theB graphics and raid controller with VMS - these can be tricky areas.   Alan   > G > Well, "5000" sounds cooler and better, and the enclosure looks nicer, 7 > but choosing a system based on that would be lame ;-)  > I > Thanks for your help guys. I dunno why more people aren't interested in  > VMS. >  > Regards, Jrgen  >    --  
 Alan Adams& alan.adams@orchard-way.freeserve.co.uk http://www.nckc.org.uk/    ------------------------------  % Date: Fri, 25 Nov 2005 23:04:20 +0100 6 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= <arne@vajhoej.dk>3 Subject: Re: Decent alpha system for OpenVMS newbie : Message-ID: <43878a63$0$118$edfadb0f@dread16.news.tele.dk>   jfragemann wrote: C > Can't wait to get started. Unfortunately it is a bit hard to find I > Alphas here in Europe (Denmark), but I'll just have to keep looking for  > a EV56 or EV6...  = If you go backwards in old posts to 29-10-2005 16:51 you will 2 an Alpha in Denmark being put up for sale on Ebay.  + Occasionally there are also Alpha's at QXL.    You can buy at Ebay DE/UK/US.     You can buy at www.islandco.com.   Arne   ------------------------------  % Date: Sat, 26 Nov 2005 00:02:47 +0100 & From: "H Vlems" <nospam@what.ever.com>3 Subject: Re: Decent alpha system for OpenVMS newbie < Message-ID: <3256c$43879837$513b9a2c$21038@news.versatel.nl>  D The choice depends on what you want. Remember that both systems wereK originally configured for NT, not for VMS nor Tru64. The point here is that C the white box alpha's may have controllers built-in that were never G supported by VMS or in fact had crippled firmware so that VMS and Tru64 K wouldn't run on them. This applies especially for SCSI controllers found in  the 5000 and 7000 models. " Just don't panic now, see below...  K The 3000 has an on-board SCSI controller that drives four slots for SCA (80 L pin ultra wide) SCSI disks and a CD-ROM. Just above the CDROM there's a slotK free for another SCSI-2 device  (50 pin connector and power are available).    Advantages: L - the built-in SCSI controller is supported by VMS and Tru64. The Mylex RAIDL controller is supported by VMS but you need additional software to configure it (I'm not a Mylex fan ;-).K - it has 256 MB, depending on the size of the memory boards you may be able J to install another 256 MB. The 3000 has 8 slots, each four slots must have the same sized boards.H - the graphics card is not mentiones, usually that's an ELSA Gloria (akaK TRIO) card. Decent performance for DECwindows, period (but you don't buy an  alpha to play games, right?). G - fairly low power consumption; electricity is expensive and high power 9 consumption means a lot of heat generated by the machine.    Disadvantages:L - the cpu is an EV5 model, as opposed to the EV56 that's inside the 5305. SoI the 3000 is slower than than the 5000, more than the 33 MHz difference in  clock speed would suggest.J - internal disks need special trays, these trays are unique to the DS 3000G and AS 800 and not easy to get. Occasionally available on eBay, usually + together with a Fujitsu disk, the RZ1CF-CF.   E The 5000 has an internal drive bay that accomodates 7 drives, the SBB L bricks.  Usually configured as a split bus configuration. There's an onboardJ CSI-2 controller that drives the CD-ROM drive that lives in one of the twoH top bays. The other bay is empty and you can put a SCSI-2 disk in there,K power and connector already inside. Most 5000's come with a SCSI controller  that is only supported by NT. H Mine was fitted with a KZPCM (dual UW SCSI and ethernet) to make it work	 with VMS.    Advantages:  - up to 2 GB (4 GB?) memory  - two (fast) cpu's - two power supplies   Disadvantages:% - uses a lot more power than the 3000 A - you may need to shop around for a VMS supported SCSI controller   H The 5000 is my favourite Alpha (then again my favourite VAX is the modelF 8550 so go figure ;-). It's the fastest Alpha you can get for free, if
 you're lucky.   L Before you start bidding on eBay, try Terry Bail (terrybail1@rogers.com). He= has a lot of used DEC gear and is an honest guy to deal with. $ Check my site: home.zonnet.nl/hvlems  
 Good luck, Hans  8 "jfragemann" <jfragemann@hotmail.com> schreef in bericht< news:1132927692.524319.87190@g47g2000cwa.googlegroups.com... Hello,   H Vlems wrote:G > check Ebay for Digital Server auctions. The Digital Server 3000, 5000 G > and 7000 series are equivalent to the AlphaServer 800, 1200 and 4100.   D Thanks for the suggestion. I can see that there are a couple on ebay
 right now.  = http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=5834413668 = http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=5834417436   F Currently, no one seems to be bidding.  Which system would you choose?   A Digital Server 5000  Part number: FR-K8F5W-AA       * Alpha 5/533MHz CPU     * 128MB Memory     * 3 x 4GB SCSI Discs     * Graphics Card      * CDRom      * Floppy Disc    OR   A Digital Server 3000  Part number: FR-K7F4W-WA       * Alpha 500MHz CPU     * 256MB Memory     * 4 x 9GB SCSI Discs     * Mylex Raid Controller      * S3 Graphics Card     * DE500 Ethernet     * CDRom      * Floppy Disc 
     * No O/S?   @ As a newcomer, I have no idea which system is perferrable to runB OpenVMS on. Also, I understand that they would need to be patched.  E Well, "5000" sounds cooler and better, and the enclosure looks nicer, 5 but choosing a system based on that would be lame ;-)   G Thanks for your help guys. I dunno why more people aren't interested in  VMS.   Regards, Jrgen    ------------------------------  % Date: Fri, 25 Nov 2005 19:27:10 -0500 * From: "d b turner" <dbturner@islandco.com>3 Subject: Re: Decent alpha system for OpenVMS newbie : Message-ID: <AXNhf.36390$6y4.33643@bignews3.bellsouth.net>  
 hang on a mo.    Ebay is great but no warranty   & we have alpha systems starting at $250  ) gosh guys ... a little fickle ain't ya???    DT www.islandco.com    7 "jfragemann" <jfragemann@hotmail.com> wrote in message  = news:1132944567.879046.190780@g49g2000cwa.googlegroups.com...  Thanks for the replies.   G Hans, I found a couple of these white boxen on ebay. Which one of these A would be preferrable to run OpenVMS and test web server software?   C I understand that these machines need to be patched to run OpenVMS.   = http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=5834413668 = http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=5834417436     The 5000 looks cooler though ;-)   Kind Regards, Jrgen   ------------------------------  % Date: Fri, 25 Nov 2005 19:30:13 -0500 * From: "d b turner" <dbturner@islandco.com>3 Subject: Re: Decent alpha system for OpenVMS newbie : Message-ID: <r_Nhf.36392$6y4.10072@bignews3.bellsouth.net>   give me a call monday F I'll save a little hassle and hopefully we can get cheap systems going     david  island computers usa 912 447 6622 x201  www.islandco.com          7 "jfragemann" <jfragemann@hotmail.com> wrote in message  < news:1132857490.870477.63490@g44g2000cwa.googlegroups.com... >  > Hi people, > F > I would like to try running a web server on OpenVMS. I have no prior4 > experience with this interesting operating system. > F > I would like a decent alpha system for playing around with and learnD > more. I do not plan on using the system extensively for actual webI > hosting, but merely to learn how to run web servers and various modules 3 > such as SSL and SSH authentication under OpenVMS.  > D > Any suggestions? Also, it would be nice if the system could handleI > Mozilla as well, without too much sweat. I know this could be a problem  > for the older versions.  > 	 > Thanks,  >  > Regards, Jurgen  >    ------------------------------  % Date: Fri, 25 Nov 2005 16:23:29 -0800 # From: "Tom Linden" <tom@kednos.com> 3 Subject: Re: Decent alpha system for OpenVMS newbie ( Message-ID: <ops0tndfhnzgicya@hyrrokkin>  I BTW, David, do you know if there is a way to get a Miata to automatically H power back up after the restoration of Power (following outage) possiblyI through the artifice of a soldering iron?  Not currently running any of    these H (tho still have a GL running Tru64, doesn't have that problem) but was   always3 curious of any way to remedy Digital's design flaw.    Tom   H On Fri, 25 Nov 2005 19:30:13 -0500, d b turner <dbturner@islandco.com>   wrote:   > give me a call monday H > I'll save a little hassle and hopefully we can get cheap systems going >  >  > david  > island computers > usa 912 447 6622 x201  > www.islandco.com >  >  >  >  > 8 > "jfragemann" <jfragemann@hotmail.com> wrote in message> > news:1132857490.870477.63490@g44g2000cwa.googlegroups.com... >>
 >> Hi people,  >>G >> I would like to try running a web server on OpenVMS. I have no prior 5 >> experience with this interesting operating system.  >>G >> I would like a decent alpha system for playing around with and learn E >> more. I do not plan on using the system extensively for actual web J >> hosting, but merely to learn how to run web servers and various modules4 >> such as SSL and SSH authentication under OpenVMS. >>E >> Any suggestions? Also, it would be nice if the system could handle J >> Mozilla as well, without too much sweat. I know this could be a problem >> for the older versions. >>
 >> Thanks, >> >> Regards, Jurgen >> >  >    ------------------------------  % Date: Fri, 25 Nov 2005 14:30:30 -0600 2 From: David J Dachtera <djesys.nospam@comcast.net>8 Subject: Re: HP : Massive strike and protest march today+ Message-ID: <43877466.29BBBCC1@comcast.net>   
 AEF wrote: > [much snippage] 9 > Principles have the problem of unintended consequences.   6 I'm sure the same can be said of a lack of principles.  G All forms of "limits" have "unintended consequences". One paradigm that D most here can related to is the environment imposed by a programmingC language or operating environment: one learns to operate within the F limits imposed by a finite framework. The same can probably be said ofF any forms of rules, limits, legislation, ... what-have-you. One learns to operate within limits.    > I believe it is F > important to address such consequences, even if it means backing off' > from the originating principle a bit.   F At the risk of starting yet another lengthy subthread (which I hope itD won't as we've already been down the "moral absolute" vs. "no rules"D rat-hole), I'd have to evoke Machiavelli at that this point: the end does not justify the means.   D One of the guiding tenets I live by is that "principles are like the@ stars: we set our course by them, but rarely do we attain them".   --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  # Date: Sat, 26 Nov 2005 04:39:11 GMT - From: "nospam@nospam.com" <nospam@nospam.com> O Subject: laughed. cried:::::   VAXen looking for a new home (near Montreal, QC) 8 Message-ID: <avofo1pabha4632h02edirejcsjdu38f45@4ax.com>   Syltrem    First of all - THANK YOUA The trip down memory lane, while painful, was also about the good  times.   At UWO - we used the DEC 10.F At Polycom - we used the Microvax (ok, actually, I did not, but others I worked with did)B At Sealtest (and then Silverwoods as Ault Foods took over the milk6 world), we started with the Vax 11/750 Then the 780's.  ? Strangely enough, VAX became a black box to me after that time. C Really did not care (even to ask) "what is the box called - what is E the model number - what is the version number!" Really - did not care  - it did not seem to matter.  D Oh sure, the version of Cobol seemed important (if then else endif). But the black box?    F So, when I saw your web pages - detailing history - that I lived thru,@ in the words of the great Selma Diamond (tv sitcom: Night Court)  2    "I laughed, I cried, it became part of my life"  = Now, the world programs on these small boxes running windows. % Rebooting every time someone sneezes.   C A vax - I'm told - could run for 15 years - without rebooting, save 7 for maintenance and Eastern Seaboard power failures ;-) B [ This was a quote from a South Florida company that STILL uses an Alpha box - a vax matured?]   E Another person said the demise of the vax / alpha was really due to a 
 numbers game.    IBM shops had 150 programmers.   Vax shops had 5 programmers.C Smarter, faster, all knowledgeable, and totally aware of the entire - company. Altruistic - is that the right word?   & I would work on a vax in a heart beat. If only someone would hire me !    (ask me for my resume !!!)   Jerrold  j1076366@hotmail.com      - On Fri, 25 Nov 2005 15:17:38 -0500, "Syltrem" ! <syltremzulu@videotron.ca> wrote:   I >We have 3 VAX, models 4000-200 (2) and 4000-600 (1) that do not serve a   >purpose anymore. N >There is also one cabinet with 5 DSSI disk and a tape (TK50) drive for those 
 >machines.& >Those are free for someone to pickup.L >They are currently in Marieville (Montreal south-shore near Richelieu) but 8 >could possibly be moved to downtown Montreal if needed. > J >Those are not desktop computers like the 4000-100. You can see a picture @ >here: http://www.mcmanis.com/chuck/computers/vaxen/fastvax.html > ' >If anyone is interested, please write. / >Remember there is no zulu in my email address!  >  >Syltrem >    ------------------------------  % Date: Sat, 26 Nov 2005 00:05:13 +0100 & From: "H Vlems" <nospam@what.ever.com>F Subject: Re: Selling Alphaserver 256MB memory modules & other stuff...< Message-ID: <29a6a$438798c9$513b9a2c$21780@news.versatel.nl>   Where are you in Italy?   & <vanjkos@gmail.com> schreef in bericht; news:1132931880.979976.3260@f14g2000cwb.googlegroups.com... : > You're right.:-) But since we passed to Dell's PoweredgeG > servers,instead of throwing all this stuff away(and having to pay for @ > it),i placed them all in my garage.So i would be happy gettingI > something for them.At least the ram memory and/or the other small/light 	 > things. F > On web sites of companies that sell used/refurbished alpha systems,iE > saw that they still give "much" worth for similar configurations to G > these that i have.Perhaps i should try on Ebay or somewhere else.Also E > because i'd be happy to give something to someone here,but you know F > that not all people are honest,i wouldn't like to send something for1 > free hehe :-).Thanx for your thoughts Hans.Ciao  >    ------------------------------  % Date: Fri, 25 Nov 2005 15:00:18 -0600 2 From: David J Dachtera <djesys.nospam@comcast.net>$ Subject: Re: System services and DCL+ Message-ID: <43877B61.169894D8@comcast.net>    Dave Froble wrote: >  > Larry Kilgallen wrote:` > > In article <4383EB14.DEA607A4@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes: > > J > >>Every now and then, someone asks for a new lexical to emulate a systemI > >>service. Every now and then, the DCL god (Guy Peleg) pop over here to / > >>announce some new addition to the lexicals.  > >>J > >>Instead of manually adding function after function, how feasable would> > >>it be to give DCL the ability to call any system service ? > >>3 > >>eg: F$CALLSYS(routine_name,arg1,arg2,arg3, etc) 2 > >>    F$CALLLIB(routine_name,arg1,arg2,arg3 etc) > >> > >>L > >>And then, you might want F$CALLUSER for user written system services.... > >> > >>" > >>Is this technically feasable ? > >  > > : > > Perhaps you could outline the behavior you expect for: > > ' > >       F$CALLSYS("CMKRNL",arg1,arg2)  > > 7 > > The contents of both arg1 and arg2 are of interest.  >  > Ah, humor! > I > Many system services require the address(s) of a structure with various I > data in the structure.  (Forget what they're called.)  Building such in J > DCL would be, at a minimum, a challenge.  I like your idea, sort of, but > it's not practical.    Well, how 'bout, for example:    $ x = F$SYSCALL( - 	"routine_name", - 	"DESCRIPTOR=symbol_name", - 	"REFERENCE=symbol_name", -  	"VALUE=expression"  		.  		.  		.  	)  G ...where the routine sets up the parameter block based on the arguments  passed:   A The "DESCRIPTOR" keyword tells the lexical function to insert the H appropriate information for a string descriptor into the parameter block at that point.  @ The "REFERENCE" keyword tells the lexical function to insert the@ appropriate information for a the address of a longword into the parameter block at that point.  H The "VALUE" keyword tells the lexical function to insert the appropriateF information for the value of an expression into the parameter block at that point.   G That's more or less how it works in VAX/DEC/HP BASIC, is it not, except > that the programmer provides the passing mechanism explicitly?  B As to JF's inquiry about having a DCL code block run with elevatedE privileges (for example), I could envision extending the CALL keyword 9 and the "@" operator with additional qualifiers, such as    8 $ CALL/PRIV=(priv[_list])/[EXEC|KERN] label p1[ p2[...]]  	 ...where: F o /PRIVILEGE causes the routine or procedure to run with the specifiedE privilege(s) if the process holds or can acquire it(them) (the listed  privilege(s) or SETPRV).E o /EXECUTIVE_MODE and /KERNEL_MODE would operate similarly to the way G they do with the DEFINE or ASSIGN command, executing a CMEXEC or CMKRNL G prior to executing the requested routine/procedure, falling back to the E less privileged access based on the privileges held by the process at * the time the routine/procedure is invoked.   --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  % Date: Fri, 25 Nov 2005 16:23:38 -0500 - From: William Webb <william.w.webb@gmail.com> $ Subject: Re: System services and DCLI Message-ID: <8660a3a10511251323s2d5693e5k6538a4ae563f6180@mail.gmail.com>   @ On 11/25/05, David J Dachtera <djesys.nospam@comcast.net> wrote: > Dave Froble wrote: > >  > > Larry Kilgallen wrote:L > > > In article <4383EB14.DEA607A4@teksavvy.com>, JF Mezei <jfmezei.spamno= t@teksavvy.com> writes:  > > > L > > >>Every now and then, someone asks for a new lexical to emulate a syste= m K > > >>service. Every now and then, the DCL god (Guy Peleg) pop over here to 1 > > >>announce some new addition to the lexicals.  > > >>L > > >>Instead of manually adding function after function, how feasable woul= d @ > > >>it be to give DCL the ability to call any system service ? > > >>5 > > >>eg: F$CALLSYS(routine_name,arg1,arg2,arg3, etc) 4 > > >>    F$CALLLIB(routine_name,arg1,arg2,arg3 etc) > > >> > > >>L > > >>And then, you might want F$CALLUSER for user written system services.= .. > > >> > > >>$ > > >>Is this technically feasable ? > > >  > > > < > > > Perhaps you could outline the behavior you expect for: > > > ) > > >       F$CALLSYS("CMKRNL",arg1,arg2)  > > > 9 > > > The contents of both arg1 and arg2 are of interest.  > >  > > Ah, humor! > > K > > Many system services require the address(s) of a structure with various K > > data in the structure.  (Forget what they're called.)  Building such in L > > DCL would be, at a minimum, a challenge.  I like your idea, sort of, bu= t  > > it's not practical.  >  > Well, how 'bout, for example:  >  > $ x =3D F$SYSCALL( - >        "routine_name", -& >        "DESCRIPTOR=3Dsymbol_name", -% >        "REFERENCE=3Dsymbol_name", -  >        "VALUE=3Dexpression"  >                . >                . >                .
 >        ) > I > ...where the routine sets up the parameter block based on the arguments 	 > passed:  > C > The "DESCRIPTOR" keyword tells the lexical function to insert the J > appropriate information for a string descriptor into the parameter block > at that point. > B > The "REFERENCE" keyword tells the lexical function to insert theB > appropriate information for a the address of a longword into the  > parameter block at that point. > J > The "VALUE" keyword tells the lexical function to insert the appropriateH > information for the value of an expression into the parameter block at
 > that point.  > I > That's more or less how it works in VAX/DEC/HP BASIC, is it not, except @ > that the programmer provides the passing mechanism explicitly? > D > As to JF's inquiry about having a DCL code block run with elevatedG > privileges (for example), I could envision extending the CALL keyword : > and the "@" operator with additional qualifiers, such as > < > $ CALL/PRIV=3D(priv[_list])/[EXEC|KERN] label p1[ p2[...]] >  > ...where: H > o /PRIVILEGE causes the routine or procedure to run with the specifiedG > privilege(s) if the process holds or can acquire it(them) (the listed  > privilege(s) or SETPRV).G > o /EXECUTIVE_MODE and /KERNEL_MODE would operate similarly to the way I > they do with the DEFINE or ASSIGN command, executing a CMEXEC or CMKRNL I > prior to executing the requested routine/procedure, falling back to the G > less privileged access based on the privileges held by the process at , > the time the routine/procedure is invoked. >  > -- > David J Dachtera > dba DJE Systems  > http://www.djesys.com/ > + > Unofficial OpenVMS Hobbyist Support Page: $ > http://www.djesys.com/vms/support/ > * > Unofficial Affordable OpenVMS Home Page:! > http://www.djesys.com/vms/soho/  > $ > Unofficial OpenVMS-IA32 Home Page:! > http://www.djesys.com/vms/ia32/  >  > Coming soon:( > Unofficial OpenVMS Marketing Home Page >   2 Sounds like "Build-your-own-security-holes" to me.   WWWebb   --C NOTE: This email address is only used for noncommerical VMS-related  correspondence. C All unsolicited commercial email will be deemed to be a request for 8 services pursuant to the terms and conditions located at# http://bellsouthpwp.net/w/e/webbww/    ------------------------------    Date: 25 Nov 2005 16:15:49 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) $ Subject: Re: System services and DCL3 Message-ID: <7VWIb1kljR59@eisner.encompasserve.org>   ` In article <43877B61.169894D8@comcast.net>, David J Dachtera <djesys.nospam@comcast.net> writes:   > Well, how 'bout, for example:  >  > $ x = F$SYSCALL( - > 	"routine_name", - > 	"DESCRIPTOR=symbol_name", - > 	"REFERENCE=symbol_name", -  > 	"VALUE=expression"  > 		.  > 		.  > 		.  > 	)  @ At that point, one might as well use a compiled language and getA the performance advantages as well.  By choosing a strongly typed 5 compiled language one even gets detection of defects.    ------------------------------  % Date: Fri, 25 Nov 2005 21:35:08 -0800 ( From: Jeff Cameron <roktsci@comcast.net>$ Subject: Re: System services and DCL0 Message-ID: <BFAD340C.183B6%roktsci@comcast.net>  J On 11/22/05 9:31 PM, in article 4383FE99.AB5E93E1@teksavvy.com, "JF Mezei"% <jfmezei.spamnot@teksavvy.com> wrote:    > Larry Kilgallen wrote:9 >> Perhaps you could outline the behavior you expect for:  >>  ( >>         F$CALLSYS("CMKRNL",arg1,arg2) >>  6 >> The contents of both arg1 and arg2 are of interest. >  > Simple.... > F > arg1: the name of a DCL subroutine label. All statements inside thatI > subroutine until the RETUN will execute in kernel mode :-) A DCL syntax $ > error crashes the whole system :-) > F > arg2: a string containing coma delimited list of arguments which are. > passed to the subroutine as P1, P2, P3 etc). >  > I > Seriously, not all system services/ RTL can logically be mapped to DCL. F > But there might be a lot which could be used. But perhaps individualI > lexical implementation is a better way. Shirley the VMS engineers would  > have discussed this ? I How would you construct complex data structures such as Item lists, which ( are common arguments to system services?   ------------------------------    Date: 25 Nov 2005 13:09:53 -0800) From: "Ken Robinson" <kenrbnsn@gmail.com>  Subject: T4 Graphs via TLvizC Message-ID: <1132952993.920695.264930@g43g2000cwa.googlegroups.com>   G I'm generating T4 Graphs using the program TLviz (V1.5-12). It's a nice F interface except for one problem. There's no explanation or indicationF of what is being displayed. For example, I selected "[MON.DSK]Rsp" forG a particular drive. It showed a spike at a certain time. The problem is = that there are numbers on the left and right, but there is no 6 indication of the units. The numbers go from 0 to 110.   Any help or explanation?   Thanks Ken Robinson   ------------------------------  % Date: Fri, 25 Nov 2005 15:17:38 -0500 * From: "Syltrem" <syltremzulu@videotron.ca>9 Subject: VAXen looking for a new home (near Montreal, QC) 0 Message-ID: <11oesam39nb0aa8@corp.supernews.com>  H We have 3 VAX, models 4000-200 (2) and 4000-600 (1) that do not serve a  purpose anymore.M There is also one cabinet with 5 DSSI disk and a tape (TK50) drive for those  	 machines. % Those are free for someone to pickup. K They are currently in Marieville (Montreal south-shore near Richelieu) but  7 could possibly be moved to downtown Montreal if needed.   I Those are not desktop computers like the 4000-100. You can see a picture  ? here: http://www.mcmanis.com/chuck/computers/vaxen/fastvax.html   & If anyone is interested, please write.. Remember there is no zulu in my email address!   Syltrem    ------------------------------  % Date: Fri, 25 Nov 2005 20:22:17 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 1 Subject: Vaxstation 3100 1gig limit & MOP booting , Message-ID: <4387B8BB.FF52AE25@teksavvy.com>  H I understand that the firmware in the Vaxstation 3100 is limited to 1gigF disks, and that when you boot from the disk, since the disk is mountedK by the firmware code, the 1gig limit exists even once VMS has fully booted.   H But if you boot from ethernet, is it correct to assume that you can then have the disk of any size ?   E How difficult would it be to modify the boot process so that it would ? just mount the vaxstation's disk and point the logicals so that E everything else executes from the disk instead of over the ethernet ?   B Is it correct to state that the first DCL that executes during theG booting process is done from STARTUP.COM ? So would it be a question of F hacking the STARTUP.COM on that system's root on the boot node to justG mount the disk and then invoke the real STARTUP.COM on that drive after B making sure already defined logicals such as SYS$DISK poing to the vaxstation's drive ?   ------------------------------  # Date: Sat, 26 Nov 2005 02:14:17 GMT   From: John Santos <john@egh.com>5 Subject: Re: Vaxstation 3100 1gig limit & MOP booting ' Message-ID: <ZxPhf.197$to5.16@trnddc04>    JF Mezei wrote: J > I understand that the firmware in the Vaxstation 3100 is limited to 1gigH > disks, and that when you boot from the disk, since the disk is mountedM > by the firmware code, the 1gig limit exists even once VMS has fully booted.  > J > But if you boot from ethernet, is it correct to assume that you can then > have the disk of any size ?  > G > How difficult would it be to modify the boot process so that it would A > just mount the vaxstation's disk and point the logicals so that G > everything else executes from the disk instead of over the ethernet ?  > D > Is it correct to state that the first DCL that executes during theI > booting process is done from STARTUP.COM ? So would it be a question of H > hacking the STARTUP.COM on that system's root on the boot node to justI > mount the disk and then invoke the real STARTUP.COM on that drive after D > making sure already defined logicals such as SYS$DISK poing to the > vaxstation's drive ?  D No idea, but you could shadow a local drive with the remotely-servedB boot disk.  You would have to do a complete shadow copy after eachF boot, but once the copy completed, all reads would come from the localD disk.  (Could you then drop the original member from the shadow set,B remount it as a new shadow set and make it available to boot-serveD other nodes?  You would have to dynamically finagle the cluster bootF information for the other nodes.  There might be problems if two nodesF booted at the same time.  Then all I/O would be local to the "too-big" local system disk!)   F I assume your real issue is lack of availability of 1.0 GB and smallerF drives.  (It's fairly easy and cheap to get RZ29 (4.3GB) and even RZ28; (2GB) drives on Ebay, but 1GB and smaller seem to be rare.)   D Remember the 1GB limit *ONLY* applies to booting.  You can mount andB use larger disks.  You could put 2 disks on the VS3100, a 1GB diskE with just VMS (no page or swap or dump files, no user files), and put B the rest on a much larger disk.  I have VAX VMS V7.3 with TCPware,F C and BASIC compilers, full expanded HELP and system libraries, etc.onD an RZ28 (2GB) with 65% free, so it should have no problem fitting on a 1GB disk.   E With V7.3-2 (Alpha only) and later with dissimilar disk shadowing and E adjustable volume size limits, perhaps you could init a large disk to H be 1.0GB, but I don't know if VAX VMS would honor this.  (You might haveE to init the disk on an alpha and then physically move it to the VAX.)   E You could also cheat and init a large disk with a large pre-allocated H file (marked nomove) on the last xx% of the disk, to keep VMS from usingB anything but the 1st 1GB.  You could then initialize and mount theC big file as a virtual disk container with LD or VD, so you wouldn't  really be wasting the space.  F BTW, does anyone know if DOSD (? - dump file off system disk) works onF a VS3100 on a bigger-than-1GB drive, i.e. is DOSD mounted by VMS afterG the real drivers are loaded, vs. normal on-system-disk dump file, which G is accessed by the ROM-based boot drivers?  Not that this should matter A much for a hobbyist system - I don't think HP would be willing to F analyze you crash dumps for you unless you came up with an *extremely* "interesting" crash...   --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------   End of INFO-VAX 2005.659 ************************