1 INFO-VAX	Sun, 03 Dec 2000	Volume 2000 : Issue 674       Contents: "EXECUTE" Command  Re: "EXECUTE" Command  Re: "EXECUTE" Command  Re: "EXECUTE" Command * AlphaStation 255/300 and large disk drives3 Re: Charon-VAX Hobbyist Edition - Printout problems ; Re: Compaq Secure Web Server for OpenVMS - access violation ; Re: Compaq Secure Web Server for OpenVMS - access violation  Re: Connecting a CD-writer.. Re: Connecting a CD-writer.. Re: Connecting a CD-writer.. Re: Connecting a CD-writer..' Getting X11/motif to run on new machine ! Re: How fast is a SCSI connection  Re: Q22-bus programming  Re: SW Raid 310 and VMS " Re: Tape Drive replacement options0 US Constitution (was Technology of US elections)  F ----------------------------------------------------------------------  % Date: Sat, 02 Dec 2000 17:09:14 -0600 7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>  Subject: "EXECUTE" Command- Message-ID: <3A29811A.9182F893@earthlink.net>   
 Hi, Folks,  D In recent threads we've again covered the issue of not being able to3 communicate with HSx controllers from a batch job.    F In one thread, someone mentioned SYS$EXAMPLES:ALPHA_LOGGER.C. Now, I'mE about as far from a C coder as you can get. However, at first glance, H this looks like it will spawn a subprocess with a pseudo-terminal as theF control terminal, then pass to the subprocess data that it receives onG its own TT: device (AGAIN!!! GRRRR!!!), receiving on a separate channel H the output from the subprocess which it then displays on the TT: device.  H Looks to me like one could fairly easily modiyf this to read from a fileE or terminal add a CLI$ interface to this program to provide a new DCL  command:   $ EXEC[UTE] [command_procedure]  	/INPUT=dcl_procedure_filespec 	/OUTPUT=target_filespec  D I'm not currently a C coder, but I'd consider taking this on - whichC means don't hold your breath: it could be next summer (2001) before E anything usable comes about. My hobbyist licenses will go through the 6 expire/renew cycle come spring time - that won't help.  C If you want this sooner, go for it! My vision is that it could work  something like this:  : $ EXECUTE/INPUT=dcl_procedure_filespec/OUTPUT=log_filespec - or -4 $ EXECUTE/OUTPUT=log_filespec dcl_procedure_filespec  D ...where "dcl_procedure_filespec" is, for example, a file that might look like this: / $ SET HOST/DUP/SERV=MSCP$DUP/TASK=CLI node_name  ADD DISK DISKptl p t l INIT DISKptl ADD UNIT Dptl DISKptl  EXIT  	 ...where: > 	"node_name" is the name of an HSD or HSJ controller (or DSSI          target) (for example) ( 	"p" is the SCSI port number of a device. 	"t" is the associated target on the SCSI port2 	"l" is the associated LUN of the SCSI port target  E This might even be useful for those occasions where SET HOST/DTE from H with a DCL proc. (whether in batch or not) could solve certain problems.   Whaddaya think?    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/   F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.    ------------------------------   Date: 2 Dec 2000 19:55:24 -0500 / From: jordan@lisa.gemair.com (Jordan Henderson)  Subject: Re: "EXECUTE" Command* Message-ID: <90c5ls$qjk$1@lisa.gemair.com>  ? Interesting...  If I understand the requirement correctly, this > is much the same functionality that "expect" gives one for the@ tcl environment in Unix.  I don't believe that "expect" has been< ported to VMS, but I could be surprised.  For information on> "expect" see http://expect.nist.gov/.  Hmmm, hadn't looked at < this in some time, and I note this particularly interesting ? item from the Expect FAQ: http://expect.nist.gov/FAQ.html#q33 .   7 There's also an expect.pm module for Perl with similar  : functionality, but I doubt that it runs on VMSPerl without significant porting work.   : BTW, I'm not trying to discourage the reinvention of this ; particular wheel on VMS.  The above utilities probably have 9 a number of distinctly Unix features and heritage and it  9 may well be that such a thing redone for and in VMS would 6 be more useful.  An example of this would be the MGFTP9 server/client, which I consider to be among the most full & featured FTP packages around anywhere.  - In article <3A29811A.9182F893@earthlink.net>, 6 David J. Dachtera <djesys.nospam@earthlink.net> wrote: >Hi, Folks,  > E >In recent threads we've again covered the issue of not being able to 4 >communicate with HSx controllers from a batch job.  > G >In one thread, someone mentioned SYS$EXAMPLES:ALPHA_LOGGER.C. Now, I'm F >about as far from a C coder as you can get. However, at first glance,I >this looks like it will spawn a subprocess with a pseudo-terminal as the G >control terminal, then pass to the subprocess data that it receives on H >its own TT: device (AGAIN!!! GRRRR!!!), receiving on a separate channelI >the output from the subprocess which it then displays on the TT: device.  > I >Looks to me like one could fairly easily modiyf this to read from a file F >or terminal add a CLI$ interface to this program to provide a new DCL	 >command:  >   >$ EXEC[UTE] [command_procedure] >	/INPUT=dcl_procedure_filespec  >	/OUTPUT=target_filespec  > E >I'm not currently a C coder, but I'd consider taking this on - which D >means don't hold your breath: it could be next summer (2001) beforeF >anything usable comes about. My hobbyist licenses will go through the7 >expire/renew cycle come spring time - that won't help.  > D >If you want this sooner, go for it! My vision is that it could work >something like this:  > ; >$ EXECUTE/INPUT=dcl_procedure_filespec/OUTPUT=log_filespec  >- or - 5 >$ EXECUTE/OUTPUT=log_filespec dcl_procedure_filespec  > E >...where "dcl_procedure_filespec" is, for example, a file that might  >look like this:0 >$ SET HOST/DUP/SERV=MSCP$DUP/TASK=CLI node_name >ADD DISK DISKptl p t l 
 >INIT DISKptl  >ADD UNIT Dptl DISKptl >EXIT  > 
 >...where:? >	"node_name" is the name of an HSD or HSJ controller (or DSSI   >        target) (for example)) >	"p" is the SCSI port number of a device / >	"t" is the associated target on the SCSI port 3 >	"l" is the associated LUN of the SCSI port target  > F >This might even be useful for those occasions where SET HOST/DTE fromI >with a DCL proc. (whether in batch or not) could solve certain problems.  >   C People are constantly asking for this functionality here in various E guises.  How many times have people asked how to automate a telnet or B serial connection and people point out that kermit does this quite" well with it's scripting language?  B The functionality you are proposing is more general in that it can@ be extended to regular CTERM-based "SET HOST" and to interative 	 programs.   @ While I do suggest that rewriting the "expect" functionality for? VMS might be a good idea, I wouldn't suggest that you wouldn't  ? examine this prior art for ideas.  For example, you'll want to  B support some sort of provision to allow the interaction to respond> to "SET TERM/INQ" functionality so as to appropriately set the> terminal type so that input to screen oriented (SMG$) programs would be convenient.    = It _might_ actually be a good idea into looking into what it  ? would take to bring up expect/tcl in VMS, as you would have the < tcl scripting language available in this environment to help with complex automation tasks.  ? Another option would be writing your utility in such a way that < DCL scripts could interact with it easily.  Perhaps allowing< the input and output from EXEC to be mailboxes that could be7 read and written from a DCL script while it is running.    >Whaddaya think? >   < Well, I do call myself a C programmer and would love to have6 such a thing, but my tuits might be in short supply...  ? On the other hand, perhaps I do have some more time now that it B seems that my constant debate opponent has fled comp.os.vms... :-)   >--  >David J. Dachtera >dba DJE Systems >http://www.djesys.com/  > ; >Unofficial Affordable OpenVMS Home Page and Message Board:   >http://www.djesys.com/vms/soho/ > G >This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings  >is to be expected.  > A >Feel free to exercise your rights of free speech and expression.  > G >However, attacks against individual posters, or groups of posters, are  >strongly discouraged.   -Jordan Henderson  jordan@greenapple.com    ------------------------------  % Date: Sat, 02 Dec 2000 21:12:31 -0600 7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>  Subject: Re: "EXECUTE" Command- Message-ID: <3A29BA1F.EEA6ED11@earthlink.net>    Jordan Henderson wrote:  > A > Interesting...  If I understand the requirement correctly, this @ > is much the same functionality that "expect" gives one for the > tcl environment in Unix.   Not familiar with it.    8 > There's also an expect.pm module for Perl with similar< > functionality, but I doubt that it runs on VMSPerl without > significant porting work.   F The problem there, of course, is that to obtain some relatively simple< functionality, one would need to be install/debug perl, then install/debug expect.pm, ...  H Seems a lot to go through as opposed to "installing" a single .EXE and a single .CLD   ; > BTW, I'm not trying to discourage the reinvention of this  > particular wheel on VMS.    7 Perhaps a simplification, rather than a "re-invention".    E > People are constantly asking for this functionality here in various G > guises.  How many times have people asked how to automate a telnet or D > serial connection and people point out that kermit does this quite$ > well with it's scripting language?  @ Here again, we're taking a "sledge-o-matic" to smack a mosquito.   D > The functionality you are proposing is more general in that it canA > be extended to regular CTERM-based "SET HOST" and to interative  > programs.   , CTERM, /LAT, /TELNET, /DTE, /HSC, /DUP, etc.   B > While I do suggest that rewriting the "expect" functionality for@ > VMS might be a good idea, I wouldn't suggest that you wouldn't@ > examine this prior art for ideas.  For example, you'll want toD > support some sort of provision to allow the interaction to respond@ > to "SET TERM/INQ" functionality so as to appropriately set the@ > terminal type so that input to screen oriented (SMG$) programs > would be convenient.  D My primary focus here is doing things in batch that are prevented by< programs like HSCPAD, DTEPAD (not to mention both LOGGER andF ALPHA_LOGGER) which can only handle a terminal device as input "out ofF the box". SMG would likely not be a major consideration in such cases.   > > It _might_ actually be a good idea into looking into what itA > would take to bring up expect/tcl in VMS, as you would have the > > tcl scripting language available in this environment to help  > with complex automation tasks.  H Same thing again - too much work to accomplish a relatively simple task.  A > Another option would be writing your utility in such a way that > > DCL scripts could interact with it easily.  Perhaps allowing> > the input and output from EXEC to be mailboxes that could be9 > read and written from a DCL script while it is running.   H Properly written, yes, EXECUTE.EXE probably shouldn't need to care aboutH such things and, in fact, should purposely be written to accomodate such* support. Gotta support PIPEs, after all...   > >Whaddaya think? > >  > > > Well, I do call myself a C programmer and would love to have8 > such a thing, but my tuits might be in short supply...   Round, square or otherwise?   A > On the other hand, perhaps I do have some more time now that it D > seems that my constant debate opponent has fled comp.os.vms... :-)  C I'm actually more "fluent" in VAX/DEC BASIC, and most fluent in DCL B since prior to OVMS hobbyist I've not had BASIC available for many years.  G Still, I'd rather this be done in C since it seems more ubiquitous than  BASIC at this point.  F ...of course MACRO32 stills ships with every OVMS distro (an assembler on VAX, a compiler on Alpha).    Hhmmm... Dunno...    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/   F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.    ------------------------------   Date: 2 Dec 2000 23:01:27 -0500 / From: jordan@lisa.gemair.com (Jordan Henderson)  Subject: Re: "EXECUTE" Command* Message-ID: <90cgin$3h0$1@lisa.gemair.com>  - In article <3A29BA1F.EEA6ED11@earthlink.net>, 6 David J. Dachtera <djesys.nospam@earthlink.net> wrote: >[snip]  >>  ? >> Well, I do call myself a C programmer and would love to have 9 >> such a thing, but my tuits might be in short supply...  >  >Round, square or otherwise? >   2 I'm rather round and so are the tuits that I lack.  B >> On the other hand, perhaps I do have some more time now that itE >> seems that my constant debate opponent has fled comp.os.vms... :-)  > D >I'm actually more "fluent" in VAX/DEC BASIC, and most fluent in DCLC >since prior to OVMS hobbyist I've not had BASIC available for many  >years.  > H >Still, I'd rather this be done in C since it seems more ubiquitous than >BASIC at this point.  > G >...of course MACRO32 stills ships with every OVMS distro (an assembler  >on VAX, a compiler on Alpha). >   B I don't want to add too much to the task, but there is BLISS also.  & Been meaning to learn BLISS someday...   >Hhmmm... Dunno... >  >--  >David J. Dachtera >dba DJE Systems >http://www.djesys.com/  > ; >Unofficial Affordable OpenVMS Home Page and Message Board:   >http://www.djesys.com/vms/soho/ > G >This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings  >is to be expected.  > A >Feel free to exercise your rights of free speech and expression.  > G >However, attacks against individual posters, or groups of posters, are  >strongly discouraged.   -Jordan Henderson  jordan@greenapple.com    ------------------------------    Date: 02 Dec 2000 22:04:59 -0800( From: Javier Henderson <javier@kjsl.com>3 Subject: AlphaStation 255/300 and large disk drives - Message-ID: <86d7faox9w.fsf@cartero.kjsl.com>   ? 	I'm trying to find what's the size limit for disk drives on an D AlphaStation 255/300 with VMS 7.2. I currently have two 4 GB drives,> both of which have been spinning for entirely too long and are, starting to log too many errors for comfort.  > 	This is a hobby system, so buying genuine RZ series is out ofF the question. And finding "small" 4GB drives locally has been a trying adventure...  - 	No shortage of 15GB and above drives though.   @ 	Any idea if these would work on this beast? I BTFW (browsed the< fine website) at Compaq but didn't find anything relevant...   -jav   ------------------------------  $ Date: Sat, 2 Dec 2000 21:56:48 -00004 From: "John D. Peedle" <john@peedle.freeserve.co.uk>< Subject: Re: Charon-VAX Hobbyist Edition - Printout problems. Message-ID: <90br75$ats$1@news8.svr.pol.co.uk>  < Robert Alan Byer <byer@mail.ourservers.net> wrote in message+ news:oOYF9sLFtgnb@cartman.ourservers.net...  > > : > > That's not very fair - how are we supposed to learn if' > > we can't lay hands on the real kit?  > > F > > If you really upset him he might end up choosing ... dare I say it > > ... Solaris? > >  > 5 > At least you didn't say the "L" word, being "Linux"  > , > Ewww, I feel unclean now just typing it :}   Go on then... Why?   -- John D. Peedle RHCE - so I'm Biased! G Registered Linux User 167460 - trying to learn VMS on a Charon Emulator & because I've only got DSSI on my VAXen" and they haven't got an OS on them ... large tears descend    ------------------------------  # Date: Sat, 02 Dec 2000 23:01:36 GMT  From: jgessling@my-deja.com D Subject: Re: Compaq Secure Web Server for OpenVMS - access violation) Message-ID: <90bv0f$t83$1@nnrp1.deja.com>   + In article <VA.000001ac.1dceb595@sture.ch>, ? > According to the Accounting utility, the final status code osa	 10000024,0C > text %SYSTEM-F-NOPRIV, isufficient privilege or object protectiono > violation. >q  @ I'm not totaly sure, since I didn't work on it too much.  But myE experience was that the apache account needed lost of privs to reallyr& work right.  That's what I did anyway.   Jimg    & Sent via Deja.com http://www.deja.com/ Before you buy.i   ------------------------------   Date: 2 Dec 2000 23:37:34 -0500-* From: young_r@eisner.decus.org (Rob Young)D Subject: Re: Compaq Secure Web Server for OpenVMS - access violation+ Message-ID: <GsLvAE+e8u2E@eisner.decus.org>-  N In article <VA.000001ac.1dceb595@sture.ch>, Paul Sture <paul@sture.ch> writes: > Problem Report >  > This on Alpha VMS V7.2-1H1.k > G > 1. After installing the CSWS PCSI kit dated 28-SEP-2000 12:03:41, andnE > configuring it according to the Release Notes, running the startup f? > shows that a process is created, but this process disappears   > immediately. > I > According to the Accounting utility, the final status code os 10000024,fC > text %SYSTEM-F-NOPRIV, isufficient privilege or object protectionl > violation. > F > But Analyze/Audit shows no file access violations; only a successfulJ > READ,WRITE access of SYSUAF by that process. I do have the audit set to I > report all file access violations, which leads me to believe that CSWS i% > itself is setting this exit status.  > G > 2. After installing the PERL PCSI kit V0505-3A1-1, dated 29-SEP-2000,rD > and setting the logicals etc according to the installation notes, / > trying to invoke PERL produces the following:N >  > $ perl1 > %DCL-W-ACTIMAGE, error activating image PERLSHRu > -CLI-E-IMGNAME, image file7 > $2$DKA0:[APACHE_KITS.PERL5_005_03.][000000]PERL.EXE;1h< > -SYSTEM-F-SHRIDMISMAT, ident mismatch with shareable image > $!B > $! Have I set the symbols/logicals up correctly? Yes I think so: > $ show log/fu perl_roothA >    "PERL_ROOT" [exec] = "DATA_DISK:[APACHE_KITS.PERL5_005_03.]"u  > [concealed] (LNM$SYSTEM_TABLE) > $ show log ubs_disk/fu9 >    "DATA_DISK" [exec] = "$2$DKA0:" [concealed,terminal]l > (LNM$SYSTEM_TABLE) > $ sho sym perl) >   PERL == "$PERL_ROOT:[000000]PERL.EXE"m > $ show log perlshr@ >    "PERLSHR" = "PERL_ROOT:[000000]PERL.EXE" (LNM$SYSTEM_TABLE)    . 	Shouldn't PERLSHR be pointing to PERLSHR.EXE?   			Rob    * > $ dir/da/siz PERL_ROOT:[000000]PERL*.EXE >  > Directory PERL_ROOT:[000000] > 7 > PERL.EXE;1                 6  29-SEP-2000 09:58:44.17b7 > PERLSHR.EXE;1           2151  29-SEP-2000 09:58:40.68  >   > Total of 2 files, 2157 blocks. >  >  >  >  > ___h > Paul Sture
 > Switzerlandh >    ------------------------------  % Date: Sat, 02 Dec 2000 15:41:07 -0600o7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>-% Subject: Re: Connecting a CD-writer.. - Message-ID: <3A296C73.92F5E6D9@earthlink.net>l   "Richard B. Gilbert" wrote:  > I >         What has "Open Source" to do with the question?  If you had theoK > complete source to VMS, what would you do with it?  I don't think I couldsH > begin to understand tens of millions of lines of BLISS, Macro, C, etc.G > Given the quality of the code and the Unix culture's attitudes toward-J > documentation I suspect that the Unix source would be even less useful!!  H The UN*X-land folks might feel somewhat differently. I've seen more thanC one comment in other newsgroups (including comp.sys.dec) indicatingtF attitudes which would indicate a certain attitude emanating from those1 who do understand a good portion of the sources. i  F My point was that if you don't like something about *BSD or Linux, youF can change it, no matter how deep in the kernel it might be, driver orG otherwise, because the sources and the build procedures are provided inn the distribution.o  C We don't have the luxury of making customizations to DKDRIVER in anoF effort to make some less retrictive compromises than were made by OVMSD Engr.'g. This one of the many sticking points that keep OpenVMS fromD gaining greater acceptance, or even from retaining the acceptance itF currently holds. This would not be a problem in OpenSource land, given. suitable competence with (re-)writing drivers.   --   David J. Dachteraa dba DJE Systems  http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/t  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.o   ------------------------------  % Date: Sat, 02 Dec 2000 21:18:51 -0500S2 From: "Richard B. Gilbert" <DRAGON@compuserve.com>% Subject: Re: Connecting a CD-writer..g7 Message-ID: <200012022119_MC2-BCE7-19AD@compuserve.com>a  <         Are you *sure* that you want people customizing VMS?           I don't!  G         One reason that Unix is hell to work with is that everybody hasaH been customizing it for years.  No two Unixes work quite the same way! =  D Even successive releases from the same vendor tend to have major andJ incompatible changes.  Experience the thrill of upgrading IRIX 4.x to IRI= XoH 5.x and not being able to find your man files anymore.  Yep!  They movedE them.  A Unix wizard would have no trouble finding the pointer to thee: pointer to the new location, but how about the rest of us?  J         For that matter, consider the joys of MS Windows; any version up = toC Win98.  Every vendor felt free to replace system files with updatede! versions.  Ever hear of dll hell?'  G         Once people are free to replace DKDRIVER with a custom version,lH software that would run on any VMS system will no longer run on some VMSJ systems.  Do you really want to see a file system corrupted because of an=  J erroneous modification to DKDRIVER?  Is it worth it just to be able to ru= n J hardware that is, quite frankly, broken (violates the SCSI standard, fail= sD: to implement it fully or fails to implement it correctly)?  J         We've even had some problems with new guys in VMS Engineering who=  J didn't understand the "VMS Philosophy".  Remember the incompatible change= sdJ to callable backup at V7.1?Just a case of somebody inserting some things = inJ alphabetical order that should have been tacked on at the end!  They didn= 't1 know, or possibly, didn't think it was important.h  7         I think I'm glad that VMS is not "open source"!e  + Message text written by "David J. Dachtera"t >> =  J >         What has "Open Source" to do with the question?  If you had the=  J > complete source to VMS, what would you do with it?  I don't think I cou= ldH > begin to understand tens of millions of lines of BLISS, Macro, C, etc.G > Given the quality of the code and the Unix culture's attitudes towardnJ > documentation I suspect that the Unix source would be even less useful!= !f  H The UN*X-land folks might feel somewhat differently. I've seen more thanC one comment in other newsgroups (including comp.sys.dec) indicatingoF attitudes which would indicate a certain attitude emanating from those2 who do understand a good portion of the sources. =    F My point was that if you don't like something about *BSD or Linux, youF can change it, no matter how deep in the kernel it might be, driver orG otherwise, because the sources and the build procedures are provided inh the distribution.t  C We don't have the luxury of making customizations to DKDRIVER in antF effort to make some less retrictive compromises than were made by OVMSD Engr.'g. This one of the many sticking points that keep OpenVMS fromD gaining greater acceptance, or even from retaining the acceptance itF currently holds. This would not be a problem in OpenSource land, given. suitable competence with (re-)writing drivers. <    ------------------------------  % Date: Sat, 02 Dec 2000 21:54:18 -0600t7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>i% Subject: Re: Connecting a CD-writer..e- Message-ID: <3A29C3EA.F5A5DA6D@earthlink.net>J   "Richard B. Gilbert" wrote:h > > >         Are you *sure* that you want people customizing VMS? >  >         I don't! > I >         One reason that Unix is hell to work with is that everybody has,H > been customizing it for years.  No two Unixes work quite the same way!F > Even successive releases from the same vendor tend to have major andL > incompatible changes.  Experience the thrill of upgrading IRIX 4.x to IRIXJ > 5.x and not being able to find your man files anymore.  Yep!  They movedG > them.  A Unix wizard would have no trouble finding the pointer to the < > pointer to the new location, but how about the rest of us?  A This is frequently seen as a difference between Linux distros, orw between *BSD and Linux..   < >         For that matter, consider the joys of MS Windows;   ( I wasn't aware that there are any... ;-)   > any version up tomE > Win98.  Every vendor felt free to replace system files with updated # > versions.  Ever hear of dll hell?i  E Been there. Have the T-shirt, the coffee mug, the cap, the back-pack,n ...k   I >         Once people are free to replace DKDRIVER with a custom version,uJ > software that would run on any VMS system will no longer run on some VMS
 > systems.  F Not sure how that would be possible, but it seems a small price to payD compared to losing one VMS site after another that could easily haveD been saved by slashing the cost of SCSI devices down to a reasonable (competitive) level!  E If it means not replacing VMS with some inferior system because I wasi< able to get a disk subsystem for $8,500 instead $85,000, ...  D ...and before anyone goes off on a "what you pay for" tirade, what IG would get is what the company was willing to pay for, regardless of the E quality. Some quality gear can be had for considerably less than whatn Compaq charges.p  B >  Do you really want to see a file system corrupted because of an& > erroneous modification to DKDRIVER?   D Remember - any "erroneous" problems would be local to a single site.E OVMS Engr.'g would still retain control of the "officially supported"e version.  B It's also important to remember that if there were any SCSI driverH issues the system would likely be inoperative or uselessly unreliable ifG that were in fact the case, or the anomalies would be local the disk(s) H which were accomodated by the driver revisions (giving the hackers a lot of credit here, I know).  9 >         I think I'm glad that VMS is not "open source"!t  t@ Up until recently, so was I. As VMS systems become more and moreD endangered, we need to take action which is more and more radical inC order to prevent OVMS from continuing to lose ground in a wholesaler manner.s  H I've suggested to Rich Marcello more than once since CETS2000 (and a few< times before then) that *ALL* OpenVMS accounts be considered
 "endangered".    -- n David J. Dachterae dba DJE Systemse http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/.  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.    ------------------------------  % Date: Sun, 03 Dec 2000 00:30:19 -0500 2 From: "Richard B. Gilbert" <DRAGON@compuserve.com>% Subject: Re: Connecting a CD-writer..p7 Message-ID: <200012030030_MC2-BCFA-5E34@compuserve.com>   E         $85,000 buys you a whopping big RAID array; something like antJ ESA12000.  If you need something like this, it's highly unlikely that you= 'dJ populate it with random SCSI disks from "Dirt Cheap Drives, Inc."  Even i= fhF you did, the drives would first have to satisfy the HSZ80 controllers;* DKDRIVER wouldn't have much to do with it.  J         Even $8,500 would buy you a SCSI host adapter, cable, shelf, powe= ryH supplies and six, eighteen Gb drives which is a pretty respectable "disk
 subsystem"  E         I doubt very much that sites abandon VMS because SCSI disk is D expensive.  You can get Alphas with IDE <obligatory retching noises>J controllers and disks if you REALLY want them.  I think sites abandon VMS=  C because NT or Linux can do what they want for far less hardware andwC software cost.  If you run VMS, you run it because it gives you the J reliability, availability, and scalability that neither NT nor any Unix c= an quite manage yet.a  J         I very much doubt that VMS will *ever* sell for the prices that N= Tr or LInux systems do.  + Message text written by "David J. Dachtera"aF >If it means not replacing VMS with some inferior system because I was= able to get a disk subsystem for $8,500 instead $85,000, ...<h   ------------------------------  % Date: Sat, 02 Dec 2000 19:19:48 +0000i$ From: Grant Schoep <grant@storm.com>0 Subject: Getting X11/motif to run on new machine; Message-ID: <4.2.0.58.20001202185859.00d345b0@192.69.85.13>,  J I'm trying to get OpenVMS 7.2 running on an Alphastation 200 4/233. But I . can't get it to recognize its graphics device.  I I new to doing admin work on VMS, I just wanted to play with it a bit at t. home and learn more about it. So bear with me.  I I just got the OpenVMS 7.2 media and installed it.  I added my licenses, nL and after the next boot, the machine comes up with with the following error:  1 DECW$DEVICE=I=NODEVICE, no graphics devices founde  K I wonder if my graphics device is not supported in OpenVMS 7.2?(The reason .J I got it for free was that it was no longer supported in the new releases  of Digital Unix.)p  K I'm not sure how to tell what kind of card it is, from the >>> show config   it looks like a PV-PCI.L  L When the system is up and I do a show device, I see a device GUA0:, which I , think may be the card, but it says disabled.  J I may have not added the correct licenses, what licenses should I need to  run it? I I added the DW-MOTIF license, any others that I should need? Until I get  E the network running, I need to type all my licenses in by hand, so I e( haven't added them all from my file yet.  ) Anything ideas of how to proceed? Thanks.s -grant  K ---------------------------------------------------------------------------- Grant Schoep, grant@storm.coma Software Developer L3 Com - Storm Control Systems Basingstoke, UKo   ------------------------------    Date: 03 Dec 2000 11:38:06 +0800/ From: Paul Repacholi <prep@k9.prep.synonet.com> * Subject: Re: How fast is a SCSI connection0 Message-ID: <87pujagoo1.fsf@k9.prep.synonet.com>  4 "Richard B. Gilbert" <DRAGON@compuserve.com> writes:  , > It depends!  (God I love to be helpful!!!)   You're welcome ;)   L > Narrow SCSI-2 can be either 5Mb/second or 10Mb/s.  Wide is either 10 or 2= > 0.J >  Ultra wide SCSI, I believe, is 20Mb/s or 40.  The difference is whether0 > transfers are "asynchronous" or "synchronous".# > Synchronous transfers are faster.n > L > Friedhelm Schmidt, in "The SCSI Bus and IDE Interface" offers the followi= > ng > table ( > Transfer Rate   8-bit   16-bit  32-bit& > Asynchronous    3Mb     6Mb     12Mb& > Synchronous     5Mb     10Mb    20Mb3 > Synchronous     10Mb    20Mb    40Mb  (Fast SCSI)-  = The fly in the mix is arbitration. This completes at the whim>B of the SLOWEST devise on the bus. CDs are notorious for doing this@ badly and VERY slowly. Figures ove a few mS have been mentioned. Yes, Milli-seconds...     dJ > I believe that the PCI bus is 133Mb/sec (subject to correction by anyone( > who knows what he is talking about!) =  ? See line 1. :)  PCI can be 32 or 64 bits wide, and run at 33 ora= 25 MHz. Or 66MHz for 64 bit, plus all sorts of real speeds inj between.  9 The hardest limit on PCI is parity over the bus. This can : be done on a 64 bus now, but the process needed is hot andA expensive. PCI-X seems to speed up by handwaving parity checking.  So take care there...   9 As with SCSI, re-arbitration of the bus will cost, so the-9 peak numbers will never be seen in practise. Also you cans; hit the wall with PCI to system limits. Quite a few systems = do system-bus-PCI-(PCIs). You avoid the contention by puttinga; devices across several PCIs, but the all feed into a singleo) onboard PCI that can limit the bandwidth.s   ~paule( Who was on the PCI SIG 5 or so years ago   ------------------------------  # Date: Sun, 03 Dec 2000 01:51:26 GMT 0 From: Timothy Stark <sword7@grace.speakeasy.org>  Subject: Re: Q22-bus programming8 Message-ID: <yGhW5.36228$IP1.1148618@news1.giganews.com>  . Tim Llewellyn <tim.llewellyn@bbc.co.uk> wrote:D > I spent some time in a past life with the "KA630 CPU Module Users'> > Guide" which did have some very detailed info on the Q22bus C > implementation of the MicroVAX II. As this stuff is very hardware H > specific I suspect you might need different code for different Q22 busE > implementations, which presumably is why the Charon VAX emulator is> > MicroVAX II only.e  J Hmmm, I now understand it.  I have VAX architecture handbook (1st Edition)G that mention both Unibus and Q22-Bus interface.  Also, I am waiting forhB my package from US post parcel because I found and ordered VAX/VMSH internals and data structures v5.2 on half.com (Ebay).  Now I am lookingH for complete Q22-Bus functional specification with all perphs like tape,: disk, ethernet, etc. to complete my own MicroVAX emulator.  
 Thank you!   -- Tim Stark   -- w, Timothy Stark	<><	Inet: sword7@speakeasy.orgJ --------------------------------------------------------------------------F "For God so loved the world, that he gave his only begotten Son, that H whosoever believeth in him should not perish, but have everlasting life.. Amen." -- John 3:16 (King James Version Bible)   ------------------------------  # Date: Sat, 02 Dec 2000 19:55:02 GMT % From: Uwe Zessin <zessin@my-deja.com>o  Subject: Re: SW Raid 310 and VMS) Message-ID: <90bk2k$lau$1@nnrp1.deja.com>C  + In article <VA.000001aa.1d98d279@sture.ch>,S   paul@sture.ch wrote:= > In article <907i9g$j6m$1@nnrp1.deja.com>, Uwe Zessin wrote:b) > > From: Uwe Zessin <zessin@my-deja.com>/ > > Newsgroups: comp.os.vmst$ > > Subject: Re: SW Raid 310 and VMS' > > Date: Fri, 01 Dec 2000 07:00:00 GMTC > >T4 > > In article <905rl6$2m4$1@aquila.news.mdx.ac.uk>,% > >   david20@alpha2.mdx.ac.uk wrote:SG > > > In article <3A25A7BA.D7F355C3@earthlink.net>, "David J. Dachtera"h) > > <djesys.nospam@earthlink.net> writes:] > > > >Mike Price wrote: > > > >>G > > > >> > I don't think HSZTERM is supported any longer. As I wrote inlH > > > >> > another reply, I beleive Dave Gudewicz does not have a RA310,; > > > >> > but a RA3000. That one does not run HSOF or ACS. 
 > > > >> > > > > >> > > [snip]6 > > > >> > > > > > >1+ > > > >Do not try to us SWCC - *PERIOD* !!!0 > > > >0+ > > > >> HSZTERm is not offically supported  > > > > G > > > >As are all of the known ways to communicate with HSx controllers-H > > > >from a batch job. Not sure why that is, or what can be done aboutH > > > >it, but it remains one of the many obstacles to getting OpenVMS +) > > > >StorageWorks into more sites, IMO.l > > >cC > > > Agreed. Compaq really need to either fully support HSZTERM orw6 > > > provide something with equivalent functionality. > > B > > Then you must have missed one of my previous responses in thisF > > thread, although part of it is still here - the first unattributed, > > paragraph above. Here is the info again: > >t> > > There is a new product (= costs money) for the most recent> > > controllers (HSZ70 and up), the SANworks Command Scripter:? > >     http://www.compaq.com/products/sanworks/sanmgt/scs.htmli >u+ > Thanks for the pointer. It requires SWCC.c  @ From reading the quickspecs I understand that it uses the SWCC's _agent_ on the host system.   D > > > Although there are undoubtedly some people who want to controlD > > > their storage from a PC there are far more who want/need to beE > > > able to connect directly from the VMS system to the controller.  > >rF > > My understanding - I haven't worked with the scripter - is that it > > offers both. >E6 > Which raises the question: "Both at the same time?".  @ You mean issuing commands from a VMS system and a management PC?  H I am not sure. It might be possible that the agent[*] serializes access, anyone knows for sure?  A [*] There is only one agent allowed to have access to the storage   subsystem at a time.h   --
 Uwe Zessin    & Sent via Deja.com http://www.deja.com/ Before you buy.w   ------------------------------  % Date: Sat, 02 Dec 2000 12:23:43 -0700 + From: "Barry Treahy, Jr." <Treahy@mmaz.com> + Subject: Re: Tape Drive replacement optionsi( Message-ID: <3A294C3F.8B70790A@mmaz.com>  L I just replaced a Winchester Systems Flash DAT, after having it in daily useG for 9 years.  It ran flawless, without tape damage, and then one day itg' couldn't read or write a single bit :-)   H I too replaced it with a Sony SDT-7000 which has been working perfectly.  K I know that 4mm takes a bruising because of the thin media, but it has done1: good by me.  Never had a problem with backups or restores.   Barrya  * yyyc186.illegaltospam_@flashcom.net wrote:  L > In <4%iV5.2579$Ei1.159309@bgtnsc05-news.ops.worldnet.att.net>, on 11/30/00 >cA >    at 02:32 AM, "Robert Meyer" <meyerra@worldnet.att.net> said:  >vI > I have a Sony SDT-7000 tape drive in my PC-164 running the Hobbiest VMSt$ > license.  Seems to work just fine. >p > Roland >4H > >Our Alpha server is on it's 3rd tape drive from Compaq. The tapes areK > >new, and Compaq sends a Digital service person.   I have an external 4mmrG > >HP DAT DDS II tape drive.  Does the Alpha recognize an HP, and if it > > >does, can I get drivers to connect, or am I limited to onlyK > >Compaq/Digital.  I also have an older DLT III.  If I connect this onto aaI > >SCSI Channel, will the system recognize, if not are drivers available? 3 > >Thank you in advance for your time and patience.a >o > --= > -----------------------------------------------------------oF > yyyc186@flashcom.net              To Respond delete ".illegaltospam"8 >                             MR/2 Internet Cruiser 2.2a: >                             For a Microsoft free univers= > -----------------------------------------------------------i   --  ? Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIOt  A E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028o   ------------------------------  $ Date: Sat, 2 Dec 2000 17:57:23 -0800* From: "Jack Peacock" <peacock@simconv.com>9 Subject: US Constitution (was Technology of US elections)n5 Message-ID: <nNhW5.1042$c52.116845@news1.primary.net>   A "Martin Knoblauch" <martin.knoblauch@compaq.com> wrote in message"0 news:908952$m42$1@mailint03.im.hou.compaq.com...G >  Just interested. I am seeing this 100M votes number a lot. I kind ofs feelG > this is to high. How many US-Americas aer there? How many of them are B > eligible/registered to vote? Given a 50% participation, the 100M numberG > would mean 200M eligible voters. What is the age distribution (peopleu belowt > voting age vs. all people)?  >u@ For the benefit of those outside the US who are curious (I know,H off-topic), or those in the US who were unfortunate enough to take their) high school civics class in recent years:.  H I believe the US official count is around 265 million, perhaps closer toH 270M if you count illegal aliens.  I don't know the age distribution butD disallowing children, aliens, illiterate public school graduates andH felons I would estimate the eligible voters at about 175-200 million, so3 the 100M who voted is right around your 50% figure.e  > However, raw numbers are not the whole story.  The US is not aH monolithic country but 50 separate states plus territories and a federalH district (Washington DC) in a federation.  In fact the original documentC which formed the US was called the Articles of Confederation, laterrC superceded by the current Constitution.  Presidential elections areoC designed to balance population and the rights of individual states,eF which is why there is an Electoral College and each state is given theG right to govern their own election procedures.  This is to ensure smallhH states, such as Nevada where I live, do not become completely irrelevantE because the population is relatively small.  You also see this in how C the Legislative branch is designed, one house by population, one by  number of states.   H The reason there are so many different vote counting systems is that theF federal government is specifically prohibited by the Constitution fromD mandating a single standard.  Each state is free to choose their ownF system, a choice which is usually left to individual counties where itC becomes a matter of economics.  The punch card system is relativelyyD inexpensive compared to electronic systems.  The county where I liveB used the punch cards until the population became too large and theA county had to switch to electronic machines.  Contrary to all theoD publicity there were never any complaints about the cards, even whenF elections were very close.  However, I always checked the card to make: sure the holes were clean before I put it in the vote box.  A It is not uncommon for presidents to receive less than 50% of theeA popular vote...Clinton did it twice.  All that matters is that hebG carries enough states to reach the magic number of electoral votes (270wE this year, a simple majority based on the total of 535 for states + 3lG for DC).  It isn't even that rare that a winner doesn't get the highestiA popular vote count, it's happened three times before.  And if thesB Florida vote is left out, neither candidate will win the electoralF college, in which case the Constitution specifies the election will goE to the House of Representatives (the newly elected one, as of JanuaryeG 6th), who must select the winner from the top 5 candidates who received H electoral votes (in this case only 2).  This happened in the election ofC 1804, there were three candidates none of whom won a majority.  The C House eventually declared Thomas Jefferson the winner.  If CongressgF cannot decide on a winner, then on January 20th, when Clinton and GoreA must leave, by law of succession the Speaker of the House, DenniseD Hasturt, becomes the president.  When Nixon resigned and Ford becameE President, the Speaker of the House, at that time Carl Albert, became A the defacto Vice-President until Nelson Rockefeller was sworn in.e  G And by the way, all this happens independently of the US Supreme Court.nE The power to elect the President is a Constitutional power granted toeD Congress (and state legislatures) alone; it is not a law the SupremeF Court can interpret or change, in much the same sense that the SupremeC Court could not rule on the validity of impeachment charges against D Clinton a few years ago.  The Court, Congress, and the President areC separate and equal branches of government, none can dictate how theh other two operate.    Jack Peacockt   ------------------------------   End of INFO-VAX 2000.674 ************************