1 INFO-VAX	Sun, 27 Nov 2005	Volume 2005 : Issue 661       Contents:) Re: DECwindows on a graphics-less machine  Re: moving Apache... Powerstorm 4D20 manual" question about set file/attributes& Re: question about set file/attributes& Re: question about set file/attributes& Re: question about set file/attributes& Re: question about set file/attributes& Re: question about set file/attributes& Re: question about set file/attributes Re: System services and DCL  Re: System services and DCL , Re: Vaxstation 3100 1gig limit & MOP booting, Re: Vaxstation 3100 1gig limit & MOP booting, Re: Vaxstation 3100 1gig limit & MOP booting, Re: Vaxstation 3100 1gig limit & MOP booting, Re: Vaxstation 3100 1gig limit & MOP bootingM Re: Weird screen color ... greenish screens likely card-video incompatibility   F ----------------------------------------------------------------------  # Date: Sun, 27 Nov 2005 03:19:24 GMT   From: John Santos <john@egh.com>2 Subject: Re: DECwindows on a graphics-less machine* Message-ID: <0B9if.4116$AB2.1870@trnddc08>   JF Mezei wrote: D > If I have a server that is without graphics card, is there a magicF > incantation during startup to cause DECwindows software to started /I > INSTALLed, but no X-server started ? (since there is no graphics head).  > A > Essentially, I'd like to be able to do SET DISPLAY and run decw L > applications on that graphics-less node to be  displayed on a workstation. > F > I had once tried this on my all mightly Microvax II, but the startupB > procedure complained bitterly about a missing graphics card, andD > subsequent SET DISPLAY complains about a no-such-device available. > E > Is it just a question of manually doing a SYSGEN CONNECT command to # > create the WSA0 template device ?   B I think you just need to install DECWindows (layered product + O/SC support in in VMS install), set WINDOW_SYSTEM = 1 in modparams.dat, A autogen and reboot.   (WINDOW_SYSTEM is dynamic, but VMS probably ! only looks at it during startup.)   > It's been a very long time since I installed it but I have two; headless VAXes that both run DECWindows apps fine this way. ? I don't remember if installing DECWindows automatically defines @ WINDOW_SYSTEM, or if this was an extra manual step.  I'm sure it is in the installation manual.   --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Sat, 26 Nov 2005 13:58:04 -0500 ' From: Dave Froble <davef@tsoft-inc.com>  Subject: Re: moving Apache... 0 Message-ID: <11ohc00pnlau2db@corp.supernews.com>   Neil Rieck wrote: 7 > "Neil Rieck" <n.rieck@sympatico.ca> wrote in message  5 > news:XwYhf.7780$e43.458495@news20.bellglobal.com...  >  >>Execution phase starting ... >>9 >>The following product will be installed to destination:  >>; >>DEC AXPVMS VMS732_UPDATE V5.0 DISK$ALPHASYS:[VMS$COMMON.]  >> >>? >>%PCSI-E-INSVOLSPC, insufficient space on volume DISK$ALPHASYS  >>H >>-PCSI-E-REQVOLSPC, 405085 blocks required; 339260 blocks free; -65825  >>blocks net >>F >>Terminating is strongly recommended. Do you want to terminate? [YES] >> >>########## >>M >>We need to get Apache off our system disk (development + production). This  M >>weekend I was able to get OpenVMS-7.3-2 onto our development system (after  I >>some extreme purging of our system disk) but I was not able to install   >>UPDATE-5.  >> >>We must do this on KAWC09 + >>KAWC15 should be kept in sync with KAWC09 L >>Because Apache content is on the system disk, we are not backing up those  >>files. >>
 >>Neil @ home  >> > M > I'm losing my mind. Please disregard the previous post. It was meant to go   > to my colleagues at work.   E Yes, some may have wondered why you were assigning us this task.  :-)    --  4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com 170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  % Date: Sat, 26 Nov 2005 20:27:24 -0600 E From: Staffan Tjernstrom <thisisafakeaddress@staffan.tjernstrom.name>  Subject: Powerstorm 4D20 manual D Message-ID: <pan.2005.11.27.02.27.14.574079@staffan.tjernstrom.name>  D I wonder if anyone might happen to have an electronic version of the Powerstorm 4D20 documentation?  H The link http://www.support.compaq.com/open3d/sw4d20.htm doesn't seem to be returning anything.   ------------------------------  + Date: Sat, 26 Nov 2005 18:56:20 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)+ Subject: question about set file/attributes $ Message-ID: <dmab4k$ruu$1@online.de>  & I have two files, one (call it T) has   J Record format:      VFC, 2 byte header, maximum 0 bytes, longest 158 bytes/ Record attributes:  Print file carriage control   " and is produced via @FOO.COM/OUT=T   and the other (call it W) has   G Record format:      Variable length, maximum 0 bytes, longest 348 bytes 4 Record attributes:  Carriage return carriage control   and is produced via SEA/OUT=W.  H What I want to do is append W to T and access it via the OSU WWW server.  G OSU serves up either file fine by itself.  However, if I append W to T, G then there are HUGE spacings between the lines when viewed via HTTP, or C only the first line of W is seen, somewhat garbled (though not when H viewing the file with TYPE or with EDT).  I've tried using SET FILE/ATTRB to set W's attributes to be like those of T, or vice versa, before& appending, but neither seems to work.   C Can I do what I want with SET FILE/ATTR?  If not, would ANA/RMS and  CONV/FDL help?  I I suppose I could TYPE each file and redirect the output, but that seems  
 inelegant.   ------------------------------  % Date: Sat, 26 Nov 2005 20:10:25 +0100 & From: Paul Sture <paul.sture@decus.ch>/ Subject: Re: question about set file/attributes + Message-ID: <3urq90FvnvpjU1@individual.net>   / Phillip Helbig---remove CLOTHES to reply wrote: ( > I have two files, one (call it T) has  > L > Record format:      VFC, 2 byte header, maximum 0 bytes, longest 158 bytes1 > Record attributes:  Print file carriage control  > $ > and is produced via @FOO.COM/OUT=T >   > and the other (call it W) has  > I > Record format:      Variable length, maximum 0 bytes, longest 348 bytes 6 > Record attributes:  Carriage return carriage control >   > and is produced via SEA/OUT=W. > J > What I want to do is append W to T and access it via the OSU WWW server. > I > OSU serves up either file fine by itself.  However, if I append W to T, I > then there are HUGE spacings between the lines when viewed via HTTP, or E > only the first line of W is seen, somewhat garbled (though not when J > viewing the file with TYPE or with EDT).  I've tried using SET FILE/ATTRD > to set W's attributes to be like those of T, or vice versa, before( > appending, but neither seems to work.  > E > Can I do what I want with SET FILE/ATTR?  If not, would ANA/RMS and  > CONV/FDL help? > K > I suppose I could TYPE each file and redirect the output, but that seems   > inelegant. >    Try:   $ @FOO.COM/OUT=T: $ convert/fdl=nl: T *	! convert to variable instead of VFC   ------------------------------  + Date: Sat, 26 Nov 2005 22:14:11 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)/ Subject: Re: question about set file/attributes $ Message-ID: <dmamnj$dqt$1@online.de>  6 In article <3urq90FvnvpjU1@individual.net>, Paul Sture <paul.sture@decus.ch> writes:    > $ @FOO.COM/OUT=T< > $ convert/fdl=nl: T *	! convert to variable instead of VFC  1 Why can't I do the same thing with SET FILE/ATTR?    ------------------------------  % Date: Sat, 26 Nov 2005 23:42:34 +0100 & From: Paul Sture <paul.sture@decus.ch>/ Subject: Re: question about set file/attributes , Message-ID: <3us6mpF12q7n3U1@individual.net>  / Phillip Helbig---remove CLOTHES to reply wrote: 8 > In article <3urq90FvnvpjU1@individual.net>, Paul Sture  > <paul.sture@decus.ch> writes:  >  >  >>$ @FOO.COM/OUT=T< >>$ convert/fdl=nl: T *	! convert to variable instead of VFC >  > 3 > Why can't I do the same thing with SET FILE/ATTR?  >   D Do a DUMP of the output file before and after to see the difference.  F The VFC file output from @FOO/OUT contains the record length in bytes D 0-1 then 2 bytes of VFC info before each record's data, whereas the E converted file only has the 2 byte record length before the record's  G data. You'll see that the converted file's records are 2 bytes shorter   due to stripping the VFC info.   ------------------------------    Date: 26 Nov 2005 16:01:10 -0800" From: chris_doran@postmaster.co.uk/ Subject: Re: question about set file/attributes C Message-ID: <1133049670.584182.277800@f14g2000cwb.googlegroups.com>   / Phillip Helbig---remove CLOTHES to reply wrote: F > In article <dmamnj$dqt$1@online.de>, helbig@astro.multiCLOTHESvax.de4 > (Phillip Helbig---remove CLOTHES to reply) writes: > : > > In article <3urq90FvnvpjU1@individual.net>, Paul Sture! > > <paul.sture@decus.ch> writes:  > >  > > > $ @FOO.COM/OUT=T@ > > > $ convert/fdl=nl: T *	! convert to variable instead of VFC > > 5 > > Why can't I do the same thing with SET FILE/ATTR?  > > > Experimenting some more, I find that convert/fdl=nl: T T and5 > SET FILE/ATTR=(RAT:CR,RFM:VAR) T both work, whereas E > SET FIL/ATTR=(RAT:PRN,RFM:VFC) F doesn't.  I guess this makes sense G > since the last command would use the first two bytes as the length of H > the record, giving me the bogus extremely long lines (and chopping offI > the first two characters in the display).  So, obviously, I should have + > been converting T instead of F all along.  > G > However, CONVERT looks better than SET FIL/ATTR since it doesn't give 8 > the "incompatible attributes" warning.  DIR/FULL afterG > SET FILE/ATTR=(RAT:CR,RFM:VAR) shows that they are the same as for F. E > However, CONVERT decreases the USED size of T from 59 to 57 blocks.  >  > Why the decrease in size?  > J > Why the "incompatible attributes" warning when the attributes themselves > are the same?  > H > Why can't I do what I want with SET FILE/ATTR?  I realise that CONVERTE > is more powerful, but I would have thought that SET FILE/ATTR would ( > provide a clean solution in this case.  G SET FILE/ATTR only fiddles with the file header bytes that describe how F the records are written; the records themselves are unchanged. CONVERTB works much harder and reformats them if necessary, in your exampleE removing the VFC bytes that Paul has explained appear in each record. E If you just use SET FILE for VFC to non-VFC, the VFC bytes remain and E will appear as funny characters at the start of each record (possibly E invisible controls) if you just TYPE it. If you SET FILE from non-VFC F to VFC, the first two characters in each record will be treated as VFCC instructions and all sorts of strange things will happen, including  multiple blank lines.   B SET FILE/ATTR can handle some things like embedded CRs and/or LFs,G because this "stream" format was used in early DEC systems like DOS and E RT-11, but it can't help in all cases, e.g. there isn't a file header ; option that says "ignore the first 2 bytes of each record".   D But if you had a file that had been through a transfer that had keptC the VFC bytes but lost the fact that they were there, then SET FILE D would be your friend because it could change the file header to tell readers they were there.  C Because you can really screw up a file by changing header bits, DEC C were initially reluctant to provide SET FILE/ATTR, but they gave in . after the appearance of a DECUS utility, FILE.   Chris    ------------------------------  % Date: Sat, 26 Nov 2005 19:22:17 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> / Subject: Re: question about set file/attributes + Message-ID: <4388FC30.FAAAEC8@teksavvy.com>   # chris_doran@postmaster.co.uk wrote: E > Because you can really screw up a file by changing header bits, DEC E > were initially reluctant to provide SET FILE/ATTR, but they gave in 0 > after the appearance of a DECUS utility, FILE.  E As I recall, there was a VERY long period before SET FILE/ATTRIB came D in. FILE was extremely useful and often needed when exchanging filesE with foreign systems with utilities that didn't quite create the file 7 properly on VMS (eg: attributes not matching contents).    ------------------------------  % Date: Sat, 26 Nov 2005 19:20:15 -0500 . From: JF Mezei <jfmezei.spamnot@vaxination.ca>/ Subject: Re: question about set file/attributes - Message-ID: <4388FBB6.B5B6D34B@vaxination.ca>   / Phillip Helbig---remove CLOTHES to reply wrote: H > Why can't I do what I want with SET FILE/ATTR?  I realise that CONVERTE > is more powerful, but I would have thought that SET FILE/ATTR would ( > provide a clean solution in this case.    F On my system, the APPEND command issues a warning of incompatible fileP formats. And while DUMP/RECORD does separate the records properly, TYPE doesn't.  F It is definitely a bug/feature of APPEND. At least it warns that there is a problem with the file.    The simple solution is:   / CONVERT/APPEND search_output.TXT  at_output.txt   B This provides appropriate conversion of the search output into the! format of the @command/output=xxx     F To answer the above question: SET FILE/ATTRIB only changes the way RMS@ parses the file contents, it doesn't change the contents. If theH contents were improperly added and half the file has one style of recordE and the other half of file has different style, then no amount of SET E FILE/ATTRIB will fix it. You need a utlit like CONVERT that reads the G data from the bad file and write a new file with unified content format ( (and attributes that match that format).   ------------------------------  % Date: Sat, 26 Nov 2005 17:10:25 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> $ Subject: Re: System services and DCL, Message-ID: <4388DD51.3D647EAB@teksavvy.com>   Larry Kilgallen wrote:N > And in fact "selection of" is the way that lexical functions are implemented > currently.  F But with the addition or one lexical function per VMS release, it willE take a long time before the full selection of applicable routines are  available as lexicals.   ------------------------------  % Date: Sat, 26 Nov 2005 22:37:16 -0500 ' From: Dave Froble <davef@tsoft-inc.com> $ Subject: Re: System services and DCL0 Message-ID: <11oiadgsj08ue8f@corp.supernews.com>   JF Mezei wrote:  > Larry Kilgallen wrote: > N >>And in fact "selection of" is the way that lexical functions are implemented >>currently. >  > H > But with the addition or one lexical function per VMS release, it willG > take a long time before the full selection of applicable routines are  > available as lexicals.  B If you have a particular need, write your own.  That way, it does  exactly what you want.   --  4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com 170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------   Date: 26 Nov 2005 19:27:36 GMTK From: moroney@world.std.spaamtrap-dot-com.no-spam.invalid (Michael Moroney) 5 Subject: Re: Vaxstation 3100 1gig limit & MOP booting F Message-ID: <4388b728$0$99950$892e7fe2@authen.yellow.readfreenews.net>  4 "Stanley F. Quayle" <squayle@insight.rr.com> writes:  , > On 26 Nov 2005 at 2:14, John Santos wrote:3 > Remember the 1GB limit *ONLY* applies to booting.  > E > The issue is crashing, not booting.  If the system bugchecks, it'll C > trash the system disk.  It's all explained in the VMS FAQ (off of  > www.hp.com/go/vms).  > A The issue is crashing AND booting.  The console code can read and  write D only to the first 1GB of the disk.  If it tries to read in a portionC of VMS that's not in the first 1GB, it'll read the wrong blocks and E likely won't boot.  If the dumpfile isn't the first 1GB, a crash dump D will write to the wrong blocks and trash the drive (or at least many files).    ------------------------------  % Date: Sat, 26 Nov 2005 17:17:10 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 5 Subject: Re: Vaxstation 3100 1gig limit & MOP booting , Message-ID: <4388DEE6.E32B8976@teksavvy.com>   Dave Froble wrote:D > If you really want a VAXstation, then look for a VAXstation 4000.   ) I already have a 3100. Hence my question.   H > I'm wondering what is the smallest size SCSI disk still in production.A > All those older disk drives will quit working, sooner or later.   H Hence my question. Newer SCSI drives don't work on VAXes anyways becauseA VAXes only went to SCSI 2 50 pin original connectors, none of the E currently weird and incompatible stuff with wide, differential, wide& 2 differential and whatever they will think of next.  E So finding spare drives now while still available would be the way to F go. The Seagate 1gig I currently have was new. I was able to find some6 place that still had some stock of these "old" drives.  E Can someone confirm that once VMS has booted, the 1gig limit does not D apply until the system crashes ? I had read that because the disk isC mounted by the firmware, VMS does not remount it and the firmware's 2 drivers are used for the life of the VMS instance.   ------------------------------  % Date: Sat, 26 Nov 2005 17:18:21 -0800 # From: "Tom Linden" <tom@kednos.com> 5 Subject: Re: Vaxstation 3100 1gig limit & MOP booting ( Message-ID: <ops0vkkvvzzgicya@hyrrokkin>  . On 26 Nov 2005 19:27:36 GMT, Michael Moroney  < <moroney@world.std.spaamtrap-dot-com.no-spam.invalid> wrote:  6 > "Stanley F. Quayle" <squayle@insight.rr.com> writes: > - >> On 26 Nov 2005 at 2:14, John Santos wrote: 4 >> Remember the 1GB limit *ONLY* applies to booting. >>F >> The issue is crashing, not booting.  If the system bugchecks, it'llD >> trash the system disk.  It's all explained in the VMS FAQ (off of >> www.hp.com/go/vms). >>C > The issue is crashing AND booting.  The console code can read and  > write F > only to the first 1GB of the disk.  If it tries to read in a portionE > of VMS that's not in the first 1GB, it'll read the wrong blocks and G > likely won't boot.  If the dumpfile isn't the first 1GB, a crash dump F > will write to the wrong blocks and trash the drive (or at least many	 > files).  > C This has been commented on many times, the firmware only supports a H 24 bit address, so addresses wrap.  You can actually usea a larger disk,F I have, but you may at some point crash, which is how I d1scovered it.   ------------------------------  + Date: Sun, 27 Nov 2005 03:44:16 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney) 5 Subject: Re: Vaxstation 3100 1gig limit & MOP booting ( Message-ID: <dmba2g$vsf$1@pcls4.std.com>  % "Tom Linden" <tom@kednos.com> writes:   / >On 26 Nov 2005 19:27:36 GMT, Michael Moroney   = ><moroney@world.std.spaamtrap-dot-com.no-spam.invalid> wrote:   7 >> "Stanley F. Quayle" <squayle@insight.rr.com> writes:  >>. >>> On 26 Nov 2005 at 2:14, John Santos wrote:5 >>> Remember the 1GB limit *ONLY* applies to booting.  >>> G >>> The issue is crashing, not booting.  If the system bugchecks, it'll E >>> trash the system disk.  It's all explained in the VMS FAQ (off of  >>> www.hp.com/go/vms).  >>> D >> The issue is crashing AND booting.  The console code can read and >> writeG >> only to the first 1GB of the disk.  If it tries to read in a portion F >> of VMS that's not in the first 1GB, it'll read the wrong blocks andH >> likely won't boot.  If the dumpfile isn't the first 1GB, a crash dumpG >> will write to the wrong blocks and trash the drive (or at least many 
 >> files). >>D >This has been commented on many times, the firmware only supports aI >24 bit address, so addresses wrap.  You can actually usea a larger disk, G >I have, but you may at some point crash, which is how I d1scovered it.   K I stand behind what I wrote.  You are correct about the addresses wrapping, K but exactly what happens depends on _which_ if any files are beyond the 1GB G region.  It may be unbootable (say if SYSBOOT.EXE was out of range), or I it may boot just fine but a crash dump will scribble all over the system  E disk if all the executables are in range but the dump file is out of  H range.  This is apparently what happened to you, and it's also the worst possible outcome.    ------------------------------  + Date: Sun, 27 Nov 2005 03:56:32 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney) 5 Subject: Re: Vaxstation 3100 1gig limit & MOP booting ( Message-ID: <dmbapg$vsf$2@pcls4.std.com>  / JF Mezei <jfmezei.spamnot@teksavvy.com> writes:   F >Can someone confirm that once VMS has booted, the 1gig limit does notE >apply until the system crashes ? I had read that because the disk is D >mounted by the firmware, VMS does not remount it and the firmware's3 >drivers are used for the life of the VMS instance.   H There is no such thing as "mounted by the firmware".  The firmware readsG (and in the case of a crash dump, writes) disk blocks, and that's about H all it does.  Once VMS is up and running at a certain point, it switchesF over to its own driver, which is used exclusively from then on (exceptG for writing crash dumps).  Files accessed before then must be below the D 1GB boundary for Vaxstation 3100s, ones accessed only after then canE be anywhere.  There is some magic that flags the system disk as being F mounted at startup (there's no $ MOUNT/SYSTEM systemdisk: in any .COM  file) but that's irrelevant.   ------------------------------  % Date: Sat, 26 Nov 2005 23:32:09 -0500  From: "Ray" <no@spam.me>V Subject: Re: Weird screen color ... greenish screens likely card-video incompatibility* Message-ID: <5Faif.3642$7p7.1962@fe06.lga>  3 "pbritto" <britto.paulo@gmail.com> wrote in message = news:1133028250.690055.196260@z14g2000cwz.googlegroups.com... @ > I changed the decw$private_server_setup.com, I uncommented theH > decw$default_visual_class = "3,4,0" to decw$default_visual_class = "3"G > in order to set my video to PseudoColor (whatever it means) because I > > was having problems installing this software called VXL. The0 > decw$default_visual_class was commented in theG > decw$private_server_setup.com (in fact it was template before, when I H > uncommented it I rename it .com), but it still green. One thing that IG > noticed, I changed that because this software support told me to, but F > before that, for troubleshooting he asked me to issue a command thatH > said the default video setting was 1024x1248, something very high res,; > but this screen doesn't look this high res in DECWindows.   D The ZLXP-L1 aka PBXGC-A* is 24 planes, 1280x1024, true color, 72 Hz.  + Does your monitor support 1280x1024 @72 Hz?   6 Did you make a typo when setting the video resolution?  6 $ define/exec/system/nolog decw$server_refresh_rate 72J $ define/exec/system/nolog decw$server_pixel_depth  24 !(or 8 if you want) $ decw$xsize_in_pixels == 1280 $ decw$ysize_in_pixels == 1024   ------------------------------   End of INFO-VAX 2005.661 ************************