1 INFO-VAX	Wed, 06 Dec 2000	Volume 2000 : Issue 680       Contents:5 Re: =?iso-8859-1?Q?HSZ70_Raid_controller=B4s_manual?=  Archive? or some node help?  Re: Archive? or some node help?   Re: Boing boing boing... splash?; Re: Compaq Secure Web Server for OpenVMS - access violation  Re: Connecting a CD-writer.. Re: Connecting a CD-writer.. DEC BASIC : relative files Re: DEC BASIC : relative files Re: DEC BASIC : relative files Re: DEC BASIC : relative files Re: educational licenses' Re: ES40/HSG80 VMS installation problem 1 Excess RA disks, HSC controllers, etc., available * Getting dump files from detached processes. Re: Getting dump files from detached processes. Re: Getting dump files from detached processes. Re: Getting dump files from detached processes Hardcopy Docs...6 Help: Unable to send mail to recipient outside domain. Re: how to sort a CSV file Re: how to sort a CSV file Re: how to sort a CSV file. Re: I need macro32 code for string replacement. Re: I need macro32 code for string replacement. Re: I need macro32 code for string replacement. Re: I need macro32 code for string replacement8 Re: Installing OpenVMS on AXPPci33 - No SYS$CPU_ROUTINES, Re: Limiting =?iso-8859-1?Q?user=B4s?= login Re: Mozilla M18 i005 Re: Mozilla M18 i005 Re: Mozilla M18 i005% Re: My interpretation of CSC response   Re: Need help, thanks in advance Re: New OpenVMS  Education site  Re: New OpenVMS  Education site  Re: New OpenVMS  Education site  OCR VMS software Re: OCR VMS software Re: OpenVMS Items  Re: OpenVMS Items , Re: Problems installing MQ Series on VMS 6.2 raid system needed Re: raid system needed Re: raid system neededE Sun new cluster comparison ommitted VMS (but VMS was their benchmark) $ Re: TCPware/Purveyor Problem Part #2$ Re: TCPware/Purveyor Problem Part #2 Re: Technology of US elections$ Tester for FW re-submission of Spell Re: Two hosts sharing RA system  Re: US Constitution  Re: US Constitution 4 Re: US Constitution (was Technology of US elections), Re: Use the e-store button off the home page, Re: Use the e-store button off the home page RE: Very weird DCL behavior  Re: Very weird DCL behavior  Re: Very weird DCL behavior  Re: Very weird DCL behavior  Re: VMS and NT integration?????  Re: VMS vs. Tru64 Unix* Re: Who =?iso-8859-1?Q?=B4s?= the wizard ?* Re: Who =?iso-8859-1?Q?=B4s?= the wizard ? Re: Zip Help!!!   F ----------------------------------------------------------------------  # Date: Tue, 05 Dec 2000 18:47:43 GMT % From: Uwe Zessin <zessin@my-deja.com> > Subject: Re: =?iso-8859-1?Q?HSZ70_Raid_controller=B4s_manual?=) Message-ID: <90jd8c$h26$1@nnrp1.deja.com>   5 In article <001201c05ede$b797bf60$c9a4e7c8@valdemir>, 4   "Valdemir J. Santos" <valdemir-@uol.com.br> wrote:. > This is a multi-part message in MIME format.  @ Please don't post in HTML, a simple ASCII message is sufficient.  D > Where I can get HSZ70 raid controller=B4s manual in the internet ?  ) "HSx Array Controller Documentation page" ; http://www.compaq.com/products/storageworks/array-and-scsi-  controllers/HSxuserdocs.html   --
 Uwe Zessin3 (If you want to send mail, please use user "zessin" / who lives at "decus.decus.de", not my-deja.com)     & Sent via Deja.com http://www.deja.com/ Before you buy.    ------------------------------  $ Date: Tue, 5 Dec 2000 17:25:53 -07000 From: "rmiddleton" <teamx@teamx.box-o-phear.org>$ Subject: Archive? or some node help?3 Message-ID: <2KfX5.1482$yA1.203882@news.uswest.net>   L Anyway, running an ancient as400 alpha with 6.2, connected to a second node,J a DecServer90M.  I had to re-create a Port (LTAxxx) and queue etc.  I needI to move the port from the Primary Node to the DecServer node.  I have set E the port's node name to point to the DecServer90 with SET PORT LTA113 L /NODE=DSR90M and such but when looking at the node's under LATCP the serviceL shows up on the AS400.  Am I just missing something simple?  Sorry to botherL you with something i would find easy under a Unix and if there is an archive could someone point me to it.    thanks,    Rick  D also if the question doesn't provide enough info please let me know.   ------------------------------  % Date: Tue, 05 Dec 2000 21:43:35 -0600 7 From: "David J. Dachtera" <djesys.nospam@earthlink.net> ( Subject: Re: Archive? or some node help?- Message-ID: <3A2DB5E7.FD87F4ED@earthlink.net>    rmiddleton wrote:  > N > Anyway, running an ancient as400 alpha with 6.2, connected to a second node,L > a DecServer90M.  I had to re-create a Port (LTAxxx) and queue etc.  I needK > to move the port from the Primary Node to the DecServer node.  I have set G > the port's node name to point to the DecServer90 with SET PORT LTA113 N > /NODE=DSR90M and such but when looking at the node's under LATCP the serviceN > shows up on the AS400.  Am I just missing something simple?  Sorry to botherN > you with something i would find easy under a Unix and if there is an archive > could someone point me to it.  > 	 > thanks,  >  > Rick > F > also if the question doesn't provide enough info please let me know.  I O.k. There's not enough here to go on, and what is here is bit nebulous.    H Can you post the output from the following commands (substitute names as needed):   $ MCR LATCP SHOW PORT LTAxxx   $ SHOW QUEUE/FULL queue_name  C What port on the DECserver90M needs to be associated with which LTA  device?   - Let's keep it simple and see where we can go.    --   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: Tue, 05 Dec 2000 19:02:24 +0000   From: steven.reece@QUINTILES.COM) Subject: Re: Boing boing boing... splash? H Message-ID: <OFD62DCC16.A450A137-ON802569AC.00685238@qedi.quintiles.com>  I Both are small, perfectly formed, ideal for purpose but we just wish they < were a little bigger (i.e. more widely used/known about) ???  8 Chris Smith (chriss at mufasa dot puserv dot com) wrote:J >>>You can kick it or hit it really hard, and if it worked before, there's a  good chance it still will?  G After being locked in a vault for years, aside from being a little less E flashy it would have a chance of still functioning (bouncing, in this  case)?   <<<Trimmed for brevity>>>  Anything else?<<<    ------------------------------  % Date: Tue, 05 Dec 2000 19:34:00 +0100   From: Paul Sture <paul@sture.ch>D Subject: Re: Compaq Secure Web Server for OpenVMS - access violation+ Message-ID: <VA.000001b8.0528a321@sture.ch>   G In article <90giqd$r6b$1@mailint03.im.hou.compaq.com>, Gaitan D'Antoni   wrote:5 > From: "Gaitan D'Antoni" <gaitan.dantoni@compaq.com>  > Newsgroups: comp.os.vms F > Subject: Re: Compaq Secure Web Server for OpenVMS - access violation& > Date: Mon, 4 Dec 2000 12:04:30 -0500 > ! > Paul Sture wrote in message ...  > >Problem Report  > >  > >This on Alpha VMS V7.2-1H1. > > H > >1. After installing the CSWS PCSI kit dated 28-SEP-2000 12:03:41, andE > >configuring it according to the Release Notes, running the startup ? > >shows that a process is created, but this process disappears  > >immediately.  > > J > >According to the Accounting utility, the final status code os 10000024,D > >text %SYSTEM-F-NOPRIV, isufficient privilege or object protection
 > >violation.  > > G > >But Analyze/Audit shows no file access violations; only a successful J > >READ,WRITE access of SYSUAF by that process. I do have the audit set toI > >report all file access violations, which leads me to believe that CSWS & > >itself is setting this exit status. > < > Check APACHE$ROOT:[LOGS]ERROR_LOG. for any error messages. >  This gives the same message.  M The fix to the problem was to clean up sylogin.com, which was executing some   development environment stuff. ___ 
 Paul Sture Switzerland    ------------------------------   Date: 5 Dec 2000 14:00:27 -0500 * From: kuhrt@eisner.decus.org (Marty Kuhrt)% Subject: Re: Connecting a CD-writer.. + Message-ID: <AfSC+qGH4puc@eisner.decus.org>   g In article <3A296C73.92F5E6D9@earthlink.net>, "David J. Dachtera" <djesys.nospam@earthlink.net> writes:  > "Richard B. Gilbert" wrote:  >>  J >>         What has "Open Source" to do with the question?  If you had theL >> complete source to VMS, what would you do with it?  I don't think I couldI >> begin to understand tens of millions of lines of BLISS, Macro, C, etc. H >> Given the quality of the code and the Unix culture's attitudes towardK >> documentation I suspect that the Unix source would be even less useful!!  > E > We don't have the luxury of making customizations to DKDRIVER in an H > effort to make some less retrictive compromises than were made by OVMS  > You can write your own DKDRIVER and implement it, if you wish.< I was involved in a project that did just that a while back.? Don't expect VMS Engineering to help if it goes blooey, though.    ------------------------------  % Date: Tue, 05 Dec 2000 18:31:24 -0600 7 From: "David J. Dachtera" <djesys.nospam@earthlink.net> % Subject: Re: Connecting a CD-writer.. - Message-ID: <3A2D88DC.68E8379B@earthlink.net>    Marty Kuhrt wrote: > i > In article <3A296C73.92F5E6D9@earthlink.net>, "David J. Dachtera" <djesys.nospam@earthlink.net> writes:  > > "Richard B. Gilbert" wrote:  > >>L > >>         What has "Open Source" to do with the question?  If you had theN > >> complete source to VMS, what would you do with it?  I don't think I couldK > >> begin to understand tens of millions of lines of BLISS, Macro, C, etc. J > >> Given the quality of the code and the Unix culture's attitudes towardM > >> documentation I suspect that the Unix source would be even less useful!!  > > G > > We don't have the luxury of making customizations to DKDRIVER in an J > > effort to make some less retrictive compromises than were made by OVMS > @ > You can write your own DKDRIVER and implement it, if you wish.> > I was involved in a project that did just that a while back.  - Have you any code you can post or contribute?   A > Don't expect VMS Engineering to help if it goes blooey, though.    That goes without saying...    --   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: Tue, 5 Dec 2000 23:52:07 +00001 From: Steve Wright <usenet@wrightnet.demon.co.uk> # Subject: DEC BASIC : relative files 4 Message-ID: <FmGeIlAn+XL6Ew6L@wrightnet.demon.co.uk>  H We need to know how to open an existing relative file and write a record; to it without having to specify the specific record number.   H If we create the file (or open an empty file), we can issue PUT commandsH without specifying a record number. This writes records 1, 2, 3 ..., and so on.  G However if we open an existing file that already has a record number 1, F any attempt to PUT without a record number results in error 153-Record already exists.   F The manual suggests that after a PUT operation the next record pointerE is advanced to the next record. This does not appear to the case when C error 153 is encountered. The manual implies that reissuing the PUT F would attempt to write the record to the next record slot in the file.C However this is not true. The next record pointer appears not to be E advanced and any attempt to PUT without record number will attempt to  create record 1 again.  A What we would like to do is to write the record to the end of the H relative file. There does not appear to be an easy way of doing this. ItG would be nice if we could use the ACCESS APPEND clause, but this is one  relevant for sequential files.  E Unfortunately we cannot use a sequential file as our program needs to ? delete records on occasions and that option is not available to 1 sequential files (apart from SCRATCH to the end).   
 Any ideas? --   Steve Wright   ------------------------------  % Date: Wed, 06 Dec 2000 09:51:45 +0800 4 From: Dave Sneddon - bigpond <dbsneddon@bigpond.com>' Subject: Re: DEC BASIC : relative files + Message-ID: <3A2D9BB1.6F6FBD33@bigpond.com>    Steve Wright wrote:  >   
 	... snip ...    C > What we would like to do is to write the record to the end of the J > relative file. There does not appear to be an easy way of doing this. ItI > would be nice if we could use the ACCESS APPEND clause, but this is one   > relevant for sequential files. > G > Unfortunately we cannot use a sequential file as our program needs to A > delete records on occasions and that option is not available to 3 > sequential files (apart from SCRATCH to the end).   A I believe a fixed length sequential file can be accessed as per a @ relative file.  This would allow you to open "access append" and/ still access records directly by record number.  >  > Any ideas? > -- > Steve Wright     --   Regards, Dave. I ------------------------------------------------------------------------- I David B Sneddon (dbs)    VMS Systems Programmer     dbsneddon@bigpond.com I DBS software at ...   http://www.users.bigpond.com/dbsneddon/software.htm I "Life is what happens to you while you're busy making other plans" Lennon    ------------------------------  % Date: Tue, 05 Dec 2000 19:44:58 -0500 ) From: yyyc186.illegaltospam_@flashcom.net ' Subject: Re: DEC BASIC : relative files 9 Message-ID: <3a2d8ef1$2$lllp186$mr2ice@news.flashcom.com>   9 In <FmGeIlAn+XL6Ew6L@wrightnet.demon.co.uk>, on 12/05/00  A    at 11:52 PM, Steve Wright <usenet@wrightnet.demon.co.uk> said:   J Stupid question?  Why don't you use an RMS Indexed file?  Simply declare aK field or series of fields on your existing record layout as a primary key.  H You can then do keyed hits for specific records, delete records, and add* records simply by using the PUT statement.  G Relative files aren't for use by those that don't understand how to use H them.  You physically must read/write specific records to the file.  TheI file normally needs to be pre-extended...been a long time, but I think it E used to need to be fixed in size as it would not automatically extend > itself...but that could have been specific to one client site.  G Normally relative files have the following layout:  Record 1 = control, J Record 2-n = data.  In a multi-user application/environment a user wishingB to write new records to the file you open the file on two seperateJ channels.  The first channel is used soley for the purpose of manipulatingI the control record.  You lock the control record prior to writing the new H record, then PUT and unlock the control once the new record is written. F The control record contains, at a minimum Next_Avail and Max_Rec_CountH values.  Lots of people shortcut this approach and it generally leads to	 disaster.   J If you are planning on "deleting" records from your relative file you willJ need to write a PACK function (like DBASE used to have, because it had the
 same problem)   G My humble advice to you would be to use an Indexed file since that will J give you the freedom of manipulation you want.  You cannot do what you are" trying to do with a relative file.   Roland    I >We need to know how to open an existing relative file and write a record < >to it without having to specify the specific record number.  I >If we create the file (or open an empty file), we can issue PUT commands I >without specifying a record number. This writes records 1, 2, 3 ..., and  >so on.   H >However if we open an existing file that already has a record number 1,G >any attempt to PUT without a record number results in error 153-Record  >already exists.  J >The manual suggests that after a PUT operation the next record pointer isI >advanced to the next record. This does not appear to the case when error D >153 is encountered. The manual implies that reissuing the PUT wouldI >attempt to write the record to the next record slot in the file. HowevermI >this is not true. The next record pointer appears not to be advanced andeI >any attempt to PUT without record number will attempt to create record 1a >again.l  B >What we would like to do is to write the record to the end of theI >relative file. There does not appear to be an easy way of doing this. IteH >would be nice if we could use the ACCESS APPEND clause, but this is one >relevant for sequential files.b  F >Unfortunately we cannot use a sequential file as our program needs to@ >delete records on occasions and that option is not available to2 >sequential files (apart from SCRATCH to the end).   >Any ideas?n   -- f; -----------------------------------------------------------tD yyyc186@flashcom.net              To Respond delete ".illegaltospam"6                             MR/2 Internet Cruiser 2.2a8                             For a Microsoft free univers; -----------------------------------------------------------    ------------------------------  % Date: Tue, 05 Dec 2000 21:37:51 -0600 7 From: "David J. Dachtera" <djesys.nospam@earthlink.net> ' Subject: Re: DEC BASIC : relative files - Message-ID: <3A2DB48F.F40A8861@earthlink.net>    Steve Wright wrote:y > J > We need to know how to open an existing relative file and write a record= > to it without having to specify the specific record number.     My memory is a bit fuzzy, but...  A The only way I ever knew of to do that was for the file stream to H already be positioned at EOF (OPEN for ACCESS APPEND?). Then a PUT would succeed, IIRC.  @ Remember also, that (if I'm not mistaken) deleted records can beE re-written in REL files. If time/overhead are not a big problem, OPENCG for ACCESS MODIFY and you can read (GET) records sequentially until yourG get a "?Record has been deleted" error, then use that record for a PUT.t   --   David J. Dachterao dba DJE Systemst http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/e  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.4   ------------------------------  $ Date: Wed, 6 Dec 2000 10:58:18 +1030A From: "Geoff Roberts" <geoffrobx@stmarksx.ppx.catholicx.edux.aux>4! Subject: Re: educational licenses-1 Message-ID: <kKfX5.2014$Vu1.44528@ozemail.com.au>c  - <steven.reece@quintiles.com> wrote in message@B news:OF41AF6716.8C0ADC13-ON802569AC.00449100@qedi.quintiles.com...   <snip>  G Thanks, had a look.  I was afraid it might be too good to be true.  Andm it was.  Single user? @ Yeah that'll work well on our Vax 6000.   Since CSLG is rampant,- overpriced, overkill for a single VMS system, H especially and old one, (probably reasonable if you have a roomful), butF not for just one web/mail/proxy/ftp server box.  Sadly this new freebyG is so completely inadequate as to be essentially useless.  Perhaps they D believe schools use VMS workstations in a classroom environment?  OrF that high school students have a Vaxstation on their desk at home?   IF wish, but Billware is so rampant we have to teach in that environment,H Office xxxx won't run on VMS, and that's what the real world has on mostH of its desktops, so it's only in a server or cluster that such a licenseF is any real use, and single user is not viable in that environment.  IA suppose I *could* use it to update my VS4000-90 to the latest VMSoH version, really must scrounge a monitor cable and mouse for that.  Ahem,E another thing, there is no mention of where to get the media, perhaps E the Hobbyist distribution could be used and the edu license installedr* over the hobbyist one?  Sounds reasonable.  G That said, I fear I may be guilty of looking a gift horse in the mouth.yE There have been huge strides recently in making VMS more available to E those without the budget of a major corporation, the Hobbyist programoD particularly is a major achievement.  Thank you for that Compaq, andC thank you for at least making an effort to help the edu crowd, it'seB certainly well intentioned though perhaps not as useful as you may think.  D  Our Vax will fumble along with it's VMS 6.0/Multinet 3.2 Rev B. for another 5 years or so.F  Then we'll retire it (again - it was donated in 96 when it's original) owner had no further use for it.)  Maybe.d  E > Incidently, the new design for the website/OpenVMS portal is great.   + Yes, very nice and quite slick.  Good work.c  @ > pity that it isn't served by a cluster of GS320s.........  :-)  # Yes, now that would be something...i  H > Mind you, if Compaq would like to buy a pair of AlphaServer 8400s from ther > UK to run it.........w  H LOL.  Wouldn't mind one myself, but the PTB would go into cardiac arrest at the price tag.    Cheers  
 Geoff Roberts  Computer Systems Manager Saint Mark's College Port Pirie,e South Australiae6 geoffrob at stmarks dot pp dot catholic dot edu dot au ICQ: 1970476   ------------------------------  # Date: Tue, 05 Dec 2000 22:42:21 GMT  From: Dirk Munk <munk@home.nl>0 Subject: Re: ES40/HSG80 VMS installation problem' Message-ID: <3A2D6F4C.BEF7283E@home.nl>i   Per Berg Nissen wrote: > D > We have a problem with installation of OpenVMS version 7.2-1 on an > AlphaServer ES40, > configured with disks on a HSG80 (MA8000).N > When we boot the OpenVMS Operation System CDROM version 7.2-1 we can not see > any disks on the HSG80.lJ > If we install OpenVMS 7.2-1 on an internal disk and supply the following	 > patchescM > (VMS_PCSI, VMS721_UPDATE, VMS721_FIBRE_SCSI) we can see the external disks:v  G There is also a important shadow patch that maybe applicable if you use- VMS shadowing.H Furthermore, don't forget to set the identifier for the console lun too,, VMS uses this to communicate with the HSG80.  $ set this_controller identifier = nn   C You will see a device GGnn in VMS after you have made this setting.o   > + > Anyone seen this kind of problem before ?e3 > How do we solve this problem in a supported way ??  D As others pointed out, this is normal for the 7.2-1 distribution CD.   > 	 > Regards 	 >    /Pera   ------------------------------   Date: 5 Dec 2000 12:30:06 PST?T From: Fairfield@SLAC.Stanford.EDU (Ken Fairfield; SLAC: 650-926-2924; FAX: 926-3515): Subject: Excess RA disks, HSC controllers, etc., available3 Message-ID: <Hk5KH7dPJsrb@mccdev.slac.stanford.edu>s  H         We've just retired the  following  "ancient"  equipment from ourH     VMScluster.   It is most likely destined for the scrap heap, via our     Salvage department.s  H      3  HSC-70  CI Storage controllers   2x(6 K.sdi + 2 K.si) requestorsH                                        + 1x(7 K.sdi + 1 K.si) requestors     56  RA90    Disksa      4  RA92      "       6  RA72      "o      6  RA73      ":     "      8  SA600      Racks for disks       1  DECarray     "    "    "        1  8-port Star Couplerl    H     In addition, there are  enough  "SDA"  cables  (I  think that's whatH     these are called) to connect all the disks to the HSC's, all in veryH     good  condition.   I  ocld probably come up with three  sets  of  CIH     cables as well, but I'd rather not (a problem of pulling  them  backH     out  from where they're currently living!).  All equipment excludingH     the Star Coupler were in service and running normally until I turned     them off last Monday.   H         We'd be happy to have "someone" take these off our hands so theyH     don't just get scrapped.  You  will/would  have to pay shipping from      the address in my signature.  '         If interested, contact me ASAP.                Thanks, Kenl -- sM  Kenneth H. Fairfield            |  Internet: Fairfield@SLC.Slac.Stanford.Edur:  SLAC, 2575 Sand Hill Rd, MS 46  |  Voice:    650-926-2924:  Menlo Park, CA  94025           |  FAX:      650-926-3515N  -----------------------------------------------------------------------------B  These opinions are mine, not SLAC's, Stanford's, nor the DOE's...   ------------------------------  # Date: Tue, 05 Dec 2000 19:33:45 GMT. From: david_dawkins@my-deja.comn3 Subject: Getting dump files from detached processesl) Message-ID: <90jfup$jdt$1@nnrp1.deja.com>m    [OpenVMS 7.1-2 on Alpha cluster]  E   We have a process that runs detached for days on end, and then diestC unexpectedly.  This program is a mixture of C, C++ and Fortran. ItsoB basic functions is to read an RMS file and send records via TCP/IP to another process.   8   I have found that we can run the process with /DUMP to7 produce a .DMP file for use with ANALYZE/PROCESS_IMAGE.s  >   However, I have also found that while we can compile all the= modules with "/NOOPT/DEBUG=ALL", we must link with "/NODEBUG"a' if we wish to run the process detached.   A   The resultant dump file can be analyzed, but it does not appear ? possible to relate it to the source code while in the debugger,d@ so it is basically impossible to perform a post-mortem this way.  ,   The ANA/PROC help docs indicate that using  6     $ ANA/PROC/IMAGE=APP_LINKED_WITH_DEBUG.EXE APP.DMP  ; should do the trick, but although this works with my simple:A test applications, it does not work with the problem application;h2 the analyze process itself produces a stack trace:   > Creating a subprocess.# > Condition signalled to take dump:p >e > %SYSTEM-F-ACCVIO ... > C > %DEBUG-I-CANTCREATEMAIN, could not create the debugger subprocesse2 > -LIB-F-NOCLI, no CLI present to perform function1 > %DEBUG-I-SHRPRC, debugger sharing users processdF > %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual address=4 > FFFFFFFFA11D0010, PC=FFFFFFFFA11D0010, PS=0000001BF > %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual address=4 > 000000000000000C, PC=0000000000000012, PS=00000003    
 Questions:  @ 1. We can send a command to the process that will induce a crash> by evaluating (*(int*)1); when we do this we get a stack traceA in the file pointed at by /ERROR and we get a .DMP file. However,fA the last genuine crash didn't produce either of these files; whateA we can do to find out where and why this program is "going away"?   D 2. What's the best way for us to produce a dump file from a detached* process that will be useful when analyzed?  E 3. Is it possible to run a program detached that has been linked withnE "/debug"?   My test program, which should crash immediately, seems toh- go into a loop burning CPU when started with:t  , 	RUN/NODEBUG/DETACH/ERROR=ERROR.LOG TEST.EXE  ! When stopped, error.log contains:   B > %DEBUG-I-CANTCREATEMAIN, cold not create the debugger subprocess2 > -LIB-F-NOCLI, no CLI present to perform function1 > %DEBUG-I-SHRPRC, debugger sharing users processc2 > %DEBUG-F-NOINPAVAIL. input objects not availableC > -RMS-F-DEV, error in device name of inappropriate device type forD > operation.  
 Program is% > int main() { printf( "%s\n", 1 ); }o  / It stack traces nicely when run non-detached :)   > Any pointers, tips, advice, explanations, gratefully received.   Regardst
 David Dawkinsn    & Sent via Deja.com http://www.deja.com/ Before you buy.g   ------------------------------   Date: 5 Dec 2000 20:32:08 GMTo2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)7 Subject: Re: Getting dump files from detached processesu6 Message-ID: <90jjc8$iua$1@mailint03.im.hou.compaq.com>  K In article <90jfup$jdt$1@nnrp1.deja.com>, david_dawkins@my-deja.com writes:e! :[OpenVMS 7.1-2 on Alpha cluster]  :eF :  We have a process that runs detached for days on end, and then diesD :unexpectedly.  This program is a mixture of C, C++ and Fortran. ItsC :basic functions is to read an RMS file and send records via TCP/IPe :to another process.  D   Watch the virtual memory and the process quotas -- ever-increasing$   values would tend to imply a leak.  G   I tend to prefer to set up a shared log file (see the FAQ) opened andeH   operated by the application, and I've found that SET FILE/END_OF_FILE >   can sometimes help -- if the file is not correctly closed...  9 :  I have found that we can run the process with /DUMP toi8 :produce a .DMP file for use with ANALYZE/PROCESS_IMAGE. ..  B :  The resultant dump file can be analyzed, but it does not appear@ :possible to relate it to the source code while in the debugger,A :so it is basically impossible to perform a post-mortem this way.m  A   Generate full listings files, and full maps.  Using the virtualSC   address in the map, you can find the module, and using the module !   listing, you can find the code.i  A :1. We can send a command to the process that will induce a crashe? :by evaluating (*(int*)1); when we do this we get a stack tracewB :in the file pointed at by /ERROR and we get a .DMP file. However,B :the last genuine crash didn't produce either of these files; whatB :we can do to find out where and why this program is "going away"?  B   Make sure you create the detached process with /AUTHORIZE -- see-   Ask The Wizard topic 3681 for some details.h  E :2. What's the best way for us to produce a dump file from a detachede+ :process that will be useful when analyzed?   G   I tend to prefer to run the process "heavy" when debugging, that is,  %   running LOGINOUT to get DCL loaded.   D   Alternatively, you can use DBG$INPUT and DBG$OUTPUT logical names,C   or you can create the process with a WSA-class workstation devicePF   as its output device, or you can create a DECterm and specify it as E   the input and output device for the detached process.  There are a wI   couple of other techniques for invoking the process with the debugger, -   too...   F :3. Is it possible to run a program detached that has been linked with
 :"/debug"?  I   Yes.  One option involves setting up a DBG$INIT file containing a "go" .   command or similar such.  F   You can also patch clear the "linked with debug" flag that is storedF   in the image header.  (Yes, you can PATCH on OpenVMS Alpha.  Use theF   ZAP tool in the Freeware RMS_TOOLS area.  Use ZAP to read the block,D   dump yoursekf into the debugger from ZAP, decode/encode/patch the D   block, exit the debugger back into ZAP, and then write the block.)  ? :Any pointers, tips, advice, explanations, gratefully received.y  J   I strongly encourage any non-trivial program include its own integrated I   debugging support.  (Yes, I also recommend the use of the debugger, andd,   I regularly use both techniques together.)  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Tue, 05 Dec 2000 22:29:08 GMTi From: david_dawkins@my-deja.coms7 Subject: Re: Getting dump files from detached processes ) Message-ID: <90jq79$sva$1@nnrp1.deja.com>c  H > :  We have a process that runs detached for days on end, and then diesF > :unexpectedly.  This program is a mixture of C, C++ and Fortran. ItsE > :basic functions is to read an RMS file and send records via TCP/IPo > :to another process. >eF >   Watch the virtual memory and the process quotas -- ever-increasing& >   values would tend to imply a leak.A OK, the sysadmins eliminated this problem. Don't appear to be anyl quota leaks.    C > :1. We can send a command to the process that will induce a crash A > :by evaluating (*(int*)1); when we do this we get a stack tracedD > :in the file pointed at by /ERROR and we get a .DMP file. However,D > :the last genuine crash didn't produce either of these files; whatD > :we can do to find out where and why this program is "going away"? >eD >   Make sure you create the detached process with /AUTHORIZE -- see/ >   Ask The Wizard topic 3681 for some details.l  = but we start the process in the same way both times; how wille? authorize change anything?  when we send the "crash" command it A creates .DMP files beautifully. when it dies it produces nothing.a  @ When we send the "crash" command, we induce an ACCVIO;  is thereD a condition that can cause the *same* program (started the same way)7 to die and not produce a .DMP file, and no stack trace?h  ; It seems that if authorize fixed anything, then the "crash".? test would not work either. (I did check the wizrd in this too)t  G > :2. What's the best way for us to produce a dump file from a detachedu- > :process that will be useful when analyzed?v >e? >   I tend to prefer to run the process "heavy" when debugging, 0 >   that is, running LOGINOUT to get DCL loaded.  : How does this get you a dump file with full debug symbols?3 In any case, we aren't even getting a dump file :-(i  A > :Any pointers, tips, advice, explanations, gratefully received.t > @ >   I strongly encourage any non-trivial program include its ownA >   integrated debugging support.  (Yes, I also recommend the usenC >   of the debugger, and I regularly use both techniques together.)a  @ I agree. But at this stage, regular diagnostics, a command queryA interface etc haven't been able to narrow down where we should betA looking. The debugger is of no use to us until the thing crashes. ? All we can do at this stage is leave the process running in the > foreground in a debugger.  I just can't believe that VMS can't> produce a dump file from a detached file that is useful in the	 debugger.0  A I'm wondering if there are signal handlers, etc that we can checkeC on, to make sure that we haven't blocked particular exceptions that F would ordinarily be caught by the VMS$CreateDumpFileHandler() function? (if you see what I mean).  I already disabled our main block ofd signal handlers.  7 Thanks very much for your help so far. I appreciate it.    Regards,
 David Dawkinsh    & Sent via Deja.com http://www.deja.com/ Before you buy.t   ------------------------------  % Date: Tue, 05 Dec 2000 21:29:06 -0600d7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>k7 Subject: Re: Getting dump files from detached processesi- Message-ID: <3A2DB282.C44E0971@earthlink.net>C    david_dawkins@my-deja.com wrote:
 > Hoff wrote:a > >nA > >   I tend to prefer to run the process "heavy" when debugging,o2 > >   that is, running LOGINOUT to get DCL loaded. > < > How does this get you a dump file with full debug symbols?  F It lets you use RUN/NODEBUG in the procedure you pass to LOGINOUT. You6 run LOGINOUT with RUN/DUMP which enables process dump.  5 > In any case, we aren't even getting a dump file :-(r  D That should do it for you. I can't post during the day, but a lot of8 others can. Post again if you need help setting this up.   --   David J. Dachtera- dba DJE Systemse http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/a  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.e   ------------------------------  $ Date: Tue, 5 Dec 2000 20:44:20 -0500" From: "Hal Kuff" <Kuff@Tessco.Com> Subject: Hardcopy Docs... . Message-ID: <3a2d9bfb_1@goliath.newsfeeds.com>  9 Anyone know if hardcopy manuals will be avaialble in 7.3?m        > -----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----A http://www.newsfeeds.com - The #1 Newsgroup Service in the World! > -----==  Over 80,000 Newsgroups - 16 Different Servers! =-----   ------------------------------  % Date: Wed, 06 Dec 2000 13:25:32 +0800l5 From: Netsurfer <netsurfer@sentosa.singaporemail.com>,? Subject: Help: Unable to send mail to recipient outside domain.s8 Message-ID: <9dir2tcfabjhupad3fs1el5j5r1e1c9cj9@4ax.com>  1 I got this unique problem, which i can't resolve.    I am using OpenVMS 7.1  C I am able to send mails from my DEC Alpha4000 to Microsoft ExchangeeE SMTP Server, but I can't send to those email addresses not related too our office domain.  6 Assumption: my machine full IP is dec02.myoffice.com, 5 and full IP of MS Exhange Server is mail.myoffice.comh   ==== SMTP Configuration=                                                       Options A Initial interval:   0 00:30:00.00  Address_max:    16 NOEIGHT_BITe; Retry interval:     0 01:00:00.00  Hop_count_max:  16 RELAYsA Maximum interval:   3 00:00:00.00                     TOP_HEADERS   t@ Timeout       Initial    Mail    Receipt    Data    Terminate   < Send:            5        5        5          3         10   Receive:         5  i% Alternate gateway:  mail.myoffice.com  General gateway:    not definedl  u Substitute domain:  not definedi  Zone:               myoffice.com  r Postmaster:         UCX_SMTP? Log file:           SYS$SPECIFIC:[UCX_SMTP]UCX$SMTP_LOGFILE.LOGh  u0 Generic queue       Queues   Participating nodes                      V" UCX$SMTP_DEC02_00      1     DEC02 ========  C With above setting, I had no problems sending any SMTP mails to anySE users "@myoffice.com", but not anyone else (ie. john.doe@external.comu etc..)  F But if i were to set the "Substitute domain" setting to "myoffice.com"F I am able to send to any external emails addresses except for users inD my exchange server, all mails with "@myoffice.com" will be processedD within the DEC machine, not the Exchange Server, returning errors as above:  ' ---- Transcript of session follows ----d: %UCX-E-SMTP_SNDERROR, Error detected while sending mail to netsurfer@DEC02.myoffice.com# -MAIL-E-NOSUCHUSR, no such user !AS>    , Can someone help me to resolve this problem.          D Regards,  	 Netsurferm        ====R For any personal email replies, please remove " sentosa. " from my E-mail address.   ------------------------------  # Date: Tue, 05 Dec 2000 21:09:10 GMT.+ From: Craig A. Berry <calepine@my-deja.com> # Subject: Re: how to sort a CSV file ) Message-ID: <90jlhg$ol3$1@nnrp1.deja.com>p  ) In article <90j7o2$bob$1@nnrp1.deja.com>,e.   Bru, Pierre <Pierre.Bru@spotimage.fr> wrote: > hello, >BC > is there a way to sort a Comma Separated Values file (well, maybe G > another separator) with the standart sort utlity ? (of course, fieldsr > are not fixed length...)  C GNU sort does this quite handily, though I don't know if there is ahG working version on VMS (Hmm, something for the to-do list).  If this is G a once-off thing, I'd consider moving the file to some other machine tokE do the manipulations with GNU sort, Excel (as Jim suggested), Access, F Filemaker, or some other CSV-aware application.  On VMS, I'd look intoC DCL (as Hoff suggested) or Perl (which has a CSV module available).-E TPU editor macros can sometimes work wonders, and I wouldn't overlookKD Datatrieve or an RDBMS if you have them available since you might beD able to import the data into one of these and export it in the order	 you want.n    & Sent via Deja.com http://www.deja.com/ Before you buy.i   ------------------------------    Date: 06 Dec 2000 05:35:05 +0800/ From: Paul Repacholi <prep@k9.prep.synonet.com>e# Subject: Re: how to sort a CSV filen0 Message-ID: <87vgsyr1py.fsf@k9.prep.synonet.com>   jbecker@ui.urban.org writes:  G > Nice wish-list item though: SORT /DELIMITER="," where /KEY's POSITION A > keyword becomes a field number instead of a character position.0  B Especially since this is the second or third time this has come up
 in few weeks.e   <heresy staked and inflamed >e     ~paulS   ------------------------------   Date: 6 Dec 2000 00:28 -0400 From: hein@eps.zko.dec.c*m# Subject: Re: how to sort a CSV file % Message-ID: <6DEC200000280544@miasys>e  Y In article <90j7o2$bob$1@nnrp1.deja.com>, Bru, Pierre <Pierre.Bru@spotimage.fr> writes...  >hello,r > B >is there a way to sort a Comma Separated Values file (well, maybeF >another separator) with the standart sort utlity ? (of course, fields >are not fixed length...)d  I I'd recommend picking up GAWK and/or PERL to readily deal with CSV files.g) Or... horror of horrors, use a Unix sort.s  L Hmmm, would the renewed Posix efforts bring us a variable field driven sort?   fwiw,  	Hein.   ------------------------------   Date: 5 Dec 2000 14:53:42 -0500n From: briggs@eisner.decus.org.7 Subject: Re: I need macro32 code for string replacement2+ Message-ID: <XobMGpLGAYTe@eisner.decus.org>A  P In article <90iufa$fpqs$1@as121.tel.hr>, "IdrEASY" <IdrEASY@bigfoot.com> writes:" > I have some code for changecase.L > With some modifications (I don't know almost anything about macro32) I had > conversionG > of all ASCII characters from 32 to 127 (left side of ASCII table) and: > similar 176 to 255K > (right side of ASCII table), but it does not transform from right to leftS > side.  > F > For example I want replace ASCII character 200 to ASCII character @.  D I think you mean to say that you want to replace ASCII character 192 with ASCII character @.n  = And I think you mean to translate between 32-127 and 160-255.p  5 It's not clear why you would want to do such a thing.t  % One way is to hard code a case table:i  : 	; Assume source character in R0, do translation in place. 	cmpb	r0,#32 	blssu	donet
 	cmpb	r0,#128h
 	bgequ	10$
 	bisb	#128,r0t	 	brw	done" 10$:
 	cmpb	r0,#160h 	blssu	doner
 	bicb	#128,r0  done:h 	e% Another way is to use a lookup table:i   	.psect	data,noexe,quade translation_table: 	.byte	0 	.byte	1 	.byte	2 	...	 	.byte	31b
 	.byte	160
 	.byte	161 	...
 	.byte	255
 	.byte	128
 	.byte	129 	...
 	.byte	159	 	.byte	32t	 	.byte	33  	...
 	.byte	127   	.psect	code,exe8 	; First argument is static string descriptor for source> 	; Second argument is static string descriptor for destination! 	.entry	translate,^m<R2,R3,R4,R5>z/ 	movl	4(ap),r2	; Source string descriptor in R2a4 	movl	8(ap),r3	; Destination string descriptor in R32 	movtc	@r2,@4(r2),#32,translation_table,@r3,@4(r3) 	movl	#ss$_normal,r0 	ret   ------------------------------  $ Date: Tue, 5 Dec 2000 23:41:01 +0100% From: "IdrEASY" <IdrEASY@bigfoot.com> 7 Subject: Re: I need macro32 code for string replacements( Message-ID: <90jr88$14ug$1@as121.tel.hr>   Sorry I was in big hurry.P  J @ was only for example, because I must have several masks for replacement,L because Croatian keyboard under OpenVMS and VT420 terminal have local Slavic
 charactersK on ^ ] \ [ @ (mask 32) ~ } | [ ` (old ASCII or CROSCII code table) and when)C I'm transferring some datas in text format from SQL database I haveRK characters with ASCII values bigger than 127 and not with mask 32, but withi 116,106 and so on.  J With mentioned MACRO32 code I can convert for example ~ in ^ with mask 32,& but not ASCII(200) in ^ with mask 116.    I will post that code, tomorrow.   ------------------------------  $ Date: Tue, 5 Dec 2000 23:42:33 +0100% From: "IdrEASY" <IdrEASY@bigfoot.com>/7 Subject: Re: I need macro32 code for string replacement,( Message-ID: <90jrb3$12vv$1@as121.tel.hr>  K Thank you, I will try combine this with my code which is similar with firstd example that you sent me.    ------------------------------  % Date: Tue, 05 Dec 2000 19:38:51 -0500N) From: yyyc186.illegaltospam_@flashcom.net 7 Subject: Re: I need macro32 code for string replacementG9 Message-ID: <3a2d8ab3$1$lllp186$mr2ice@news.flashcom.com>r  ; In <90jatl$gut$1@mailint03.im.hou.compaq.com>, on 12/05/00 oB    at 06:07 PM, hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) said:    I >In article <009F422A.3C5D641D@SendSpamHere.ORG>, system@SendSpamHere.ORGt4 >(Brian Schenkenberger, VAXman-) writes: :In articleJ ><90iufa$fpqs$1@as121.tel.hr>, "IdrEASY" <IdrEASY@bigfoot.com> writes: :>I >have some code for changecase.e  9 >  There are existing OpenVMS RTL calls for translations.f  M >:>With some modifications (I don't know almost anything about macro32) I had O >:>conversion of all ASCII characters from 32 to 127 (left side of ASCII table).G >:>and similar 176 to 255 (right side of ASCII table), but it does not r% >:>transform from right to left side.d >:>eG >:>For example I want replace ASCII character 200 to ASCII character @.r >:
 >:Why "@"?  1 >  And why Macro32?  Why not some other language?   9 Because that is the class they are currently enrolled in.    -- t; -----------------------------------------------------------sD yyyc186@flashcom.net              To Respond delete ".illegaltospam"6                             MR/2 Internet Cruiser 2.2a8                             For a Microsoft free univers; -----------------------------------------------------------    ------------------------------    Date: 06 Dec 2000 05:49:51 +0800/ From: Paul Repacholi <prep@k9.prep.synonet.com>.A Subject: Re: Installing OpenVMS on AXPPci33 - No SYS$CPU_ROUTINESe0 Message-ID: <87puj6r11c.fsf@k9.prep.synonet.com>  4 hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) writes:  X > In article <90ijue$9l5$1@mordred.cc.jyu.fi>, Osmo Kujala <kujala@tukki.jyu.fi> writes: > :By the way there is :A > :literal HWRPB_CPU_TYPE$K_MANNEQUIN = 3; !  Mannequin simulatoru8 > :Does anyone know if OpenVMS 7.* can run in Mannequin? >  > G >   I have no idea if one of these boxes is still running.  I know thateG >   we have not tested OpenVMS on the simulators in a looonnnnggg time.e  + Is that a simulator, or the ADU EV3 system?u   ~Paul    ------------------------------  % Date: Tue, 05 Dec 2000 20:06:16 -0600"7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>s5 Subject: Re: Limiting =?iso-8859-1?Q?user=B4s?= logino- Message-ID: <3A2D9F18.CE05DFBA@earthlink.net>    > "Valdemir J. Santos" wrote:d > D > Im needing limit users login to only one login by time. example: >  > username: test > password:xxxxxxxxg >  > welcome to alpha1....O >  > alpha1>  set host 0  >  > username: test > password: xxxxxxxx > , > you dont have permission to login again ! >  > control returned to alpha1	 > alpha1>  > 2 >  Is there any Authorizes flag to enable this  ?  C Well, yeah. Others have mentioned the MaxJobs parameter in the UAF.T  ' Something you could try in SYLOGIN.COM:f  2 $ IF	(F$GETJPI( 0, "MODE" ) .EQS. "INTERACTIVE") -8 .AND.	(F$EDIT( F$GETJPI( 0, "PRCNAM"), "TRIM" )  .NES. -- 	F$EDIT( F$GETJPI( 0, "USERNAME" ), "TRIM" ))p $ THEN@ $	WRITE SYS$OUTPUT " You dont have permission to login again !" $	LOGOUT $ ENDIF    --   David J. Dachterae dba DJE Systemsu http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/a  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.    ------------------------------  $ Date: Tue, 5 Dec 2000 16:12:11 -0500% From: "Brian Tillman" <tillman_brian>m Subject: Re: Mozilla M18 i005n$ Message-ID: <3a2d5972$1@news.si.com>  . >VAX doesn't have upcalls support, that's why.  L Many threaded applications don't require upcalls and they don't run any less well.r --A Brian Tillman                   Internet: tillman_brian at si.comaA Smiths Industries, Inc.                   tillman at swdev.si.coms= 3290 Patterson Ave. SE, MS      Addresses modified to prevents< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  # Date: Tue, 05 Dec 2000 21:44:05 GMTs' From: Colin Blake <colin@theblakes.com>e Subject: Re: Mozilla M18 i005n- Message-ID: <3A2D60F6.97CF2104@theblakes.com>y   Brian Tillman wrote:  0 > >VAX doesn't have upcalls support, that's why. >aN > Many threaded applications don't require upcalls and they don't run any less > well.   G Unfortunately the Netscape NSPR layer code doesn't run without upcalls.    ------------------------------  # Date: Tue, 05 Dec 2000 22:40:49 GMTi= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)e Subject: Re: Mozilla M18 i005u0 Message-ID: <009F426A.F7F2B5F6@SendSpamHere.ORG>  W In article <3A2D60F6.97CF2104@theblakes.com>, Colin Blake <colin@theblakes.com> writes:e >Brian Tillman wrote:Z >-1 >> >VAX doesn't have upcalls support, that's why.s >>O >> Many threaded applications don't require upcalls and they don't run any lesse >> well. >hH >Unfortunately the Netscape NSPR layer code doesn't run without upcalls. >i  . Unfortunately ... Netscape ... doesn't run ...   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMa            eO city, n., 1. a place where trees are cut down and streets are named after them.t   ------------------------------  % Date: Tue, 05 Dec 2000 16:34:52 -0500n' From: "Dale A. Marcy" <dqm@y12.doe.gov>h. Subject: Re: My interpretation of CSC response+ Message-ID: <3A2D5F7B.9A99B1BD@y12.doe.gov>    Kenneth,  M      Thanks for the information.  I will have to dig into the code to see howfL the code uses that information to determine if it can be eliminated.  Again,I this is code that was written before my time, so I do not know all of thetH details.  The CSC did not mention that option to me, they only mentioned$ removing the JPI$M_NO_TARGET_INSWAP.  
 Dale A. Marcyd. Science Applications International Corporation   Kenneth Randell wrote: > 
 > Hello -- > K > Just some clarification (hopefully).  The following FORTRAN code from CSC2 > reproduces the problem;rD > however, if you change the JPI$_RIGHTS_SIZE to something else, say > JPI$_ASTEN, theaN > problem also goes away, even if you leave in the JPI$M_NO_TARGET_INSWAP.  If > you G > remove the JPI$M_NO_TARGET_INSWAP and leave the JPI$_RIGHTS_SIZE, thee	 > problemeK > also goes away, so if you can modify your 'idle process killer' to ignore  > JPI$_RIGHTS_SIZE,; > I think you will be all set. > 
 > Ken Randellr >  > C       $ ! TEST.FOR > C       $ FORTRAN TEST > C       $ LINK TESTr > C       $ RUN TEST > Cm% >         Program         Getjpi_Test  >         Implicit        None > % >         Include         '($Jpidef)'($ >         Include         '($Ssdef)'( >         Include         '($SysSrvNam)' > B >         Integer*4       Getjpi_Control_Flags, Process_ID, Status3 >         Integer*4       Rights_Size, Image_Length3 > # >         Character       Image*256D > % >         Structure       /Item_List/a >          Union >           Mape2 >            Integer*2    Buffer_Length, Item_Code? >            Integer*4    Buffer_Address, Return_Length_Addressl >           End Maph >           Map\" >            Integer*4    End_List >           End Map) >          End Union >         End Structure  > 3 >         Record  /Item_List/             Getjpi(4)a > 7 >         Getjpi_Control_Flags = JPI$M_NO_TARGET_INSWAP > >         Getjpi(1).Buffer_Length=sizeof(Getjpi_Control_Flags)7 >         Getjpi(1).Item_Code=JPI$_GETJPI_CONTROL_FLAGSs= >         Getjpi(1).Buffer_Address=%loc(Getjpi_Control_Flags)-+ >         Getjpi(1).Return_Length_Address=03 > / >         Getjpi(2).Buffer_Length=sizeof(Image)c+ >         Getjpi(2).Item_Code=JPI$_IMAGNAMEm. >         Getjpi(2).Buffer_Address=%Loc(Image)< >         Getjpi(2).Return_Length_Address=%Loc(Image_Length) > 5 >         Getjpi(3).Buffer_Length=sizeof(Rights_Size)-. >         Getjpi(3).Item_Code=JPI$_RIGHTS_SIZE4 >         Getjpi(3).Buffer_Address=%Loc(Rights_Size)+ >         Getjpi(3).Return_Length_Address=0n >  >         Getjpi(4).End_List=0 > 3 >          Process_ID=-1                  !Wildcard1 >  >          Status=Ss$_Normal > / >          Do While(Status .NE. Ss$_NoMoreProc)s8 >           Status=Sys$Getjpiw(, Process_ID,, Getjpi,,,); >           If(Status .EQ. Ss$_NoMoreProc)Then    !All donek >             Goto 300 >           EndIf0 > 9 >           If(.NOT. Status)Call Lib$Signal(%Val(Status))e8 >           PRINT 200, Process_ID, Image(1:Image_Length) > 5 > 200       FORMAT (' PID=', Z8, ', image="', A, '"')e > 300      EndDo
 >         Ends   ------------------------------  % Date: Tue, 05 Dec 2000 18:47:22 -0600m7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>6) Subject: Re: Need help, thanks in advance - Message-ID: <3A2D8C9A.49F8F324@earthlink.net>e   "Philip J. Lewis" wrote: > ; > Legacy is what Unix people say about VMS (and all IBMs op  > , ( > but they never did know how to count !K > Legacy is what MS people say about ALL other opersating systems, but they- > never could spell. > L > It is always derogatory when used in the systems sense, the implication isF > that one has inherited these systems (a legacy) and is now forced toN > maintain them while newer and, according to the writer/speaker/pointy haired/ > one, ipso facto better options are available.- > G > The word is also used in database and other IT technology areas.  TherM > meaning is the same, it is always derogatory and in ALMOST ALL instances it  > is not accurate. > < > I would humbly suggest you drop it from your IT vocabulary  H Just my opinion, and perhaps it shows my age (XLVI), but I remember whenF a legacy was like - make that was - your inheritance: the fortune leftG to you by your fore-bearers, something to be proud of, a well-spring ofaG value, a treasure, a wealth of knowledge, everything that was built andi) worked for by those who went before us...e  > Have I really outlived the world which brought me to this day?   -- o David J. Dachtera  dba DJE Systemse http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/0  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: 5 Dec 2000 21:06:33 GMTx2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)( Subject: Re: New OpenVMS  Education site, Message-ID: <90jlcp$1sq@gap.cco.caltech.edu>  n In article <90j336$etq$1@mailint03.im.hou.compaq.com>, "Sue Skonetski" <susan.skonetski@compaq.nospam> writes:H >The new web site for the new OpenVMS Educational License Program is now >available at: >o >http://www.openvmsedu.com >F <SNIP> > L >Please let your educational partners and customers know about this excitingM >new program.  And if you represent an educational institution, we invite you  >to try the site and sign up!m >i  - Exciting?  How about massively disappointing?   D Strip away the hype and all that remains is a roughly 2000 line DCL H procedure that loads PAKs.  There is NOTHING ELSE in it.  And these PAKsK only allow a form of usage (single user VMS machines) that is unheard of inoN academia, rendering it unsuitable for pretty much everybody who currently runsJ VMS in this environment.  This educational license program doesn't provideI a way to obtain the actual programs, nor any hardware deals, nor any formaI of support, nor any migration path off the expensive and creaky CSLG/ESL, E and etc. etc.  All points which we raised when this program was first , discussed and which Compaq chose to ignore.   E It does provide an OpenVMS base license, but unless I'm reading theirnC convoluted legalese incorrectly it CANNOT be combined with the CSLG_  on a multiuser system because:    H    "Upon agreement to these terms Compaq Computer Corporation ("COMPAQ")G     grants you a limited right to use the Software Program Version (theP8     "Software") specified above by a single individual."6                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^  I So while the CSLG would provide a multiuser license, the 2nd user runningkI on top of the education base license would violate the license agreement.e   A different section says:   2   "However, the Software may be copied and used onA    multiple computers that are under your control in a classroom eK    environment, including use by multiple students who have agreed to thesed4    License Terms solely for instructional purposes."  K Oh right.  I defy the people at Compaq to identify a single college or K-12 F school ANYWHERE on the face of the Earth that has a room with multipleJ single user VMS workstations set up for "instructional purposes".  ClearlyG this license was written by a PC lawyer who had not a CLUE about what an( multiuser OS is or what VMS is used for.  > At least they set the price correctly - since it is WORTHLESS.  3 Rich Marcello had best don his asbestos suit now...h   Regards,   David Mathog mathog@seqaxp.bio.caltech.edui? Manager, sequence analysis facility, biology division, Caltech .   ------------------------------    Date: 06 Dec 2000 05:58:50 +0800/ From: Paul Repacholi <prep@k9.prep.synonet.com>a( Subject: Re: New OpenVMS  Education site0 Message-ID: <87n1ear0md.fsf@k9.prep.synonet.com>  3 nothome@spammers.are.scum (Malcolm Dunnett) writes:I     ...k  O >    I'm sorry to sound so cynical, but I'm really disappointed in this program K > After waiting almost a year from the first suggestions it was coming it'seM > disheartening to see that it really doesn't do much for an institution. TheLO > only thing wrong with the CSLG was that it doesn't include base licenses, butyI > rather than fix that Compaq chose to develop this program. This programtN > doesn't even address that problem. It's clear from the stuff above that it'sK > not permitted to use the base PAK from this program along with multi-usertI > PAKS from CSLG to achieve a multi-user system ( that would violate the e& > "single, named individual" clause.)   A At least it now extends to all edu institutions, and specificallytE include students. But, I too feel that adding this and base licencingbA to CSLG would be a much better move. Gateshit software is winningt> hand over fist in high schools here. And it's telling. A largeH local VMS/Alpha shop is planning to go to 'Newer Technologies' and Unix.1 BTW, where _has_ Andrew been the last week or so?    ~Pauld   ------------------------------  % Date: Tue, 05 Dec 2000 21:52:57 -0600e7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>l( Subject: Re: New OpenVMS  Education site- Message-ID: <3A2DB819.B47D0F04@earthlink.net>a   Paul Repacholi wrote:u3 > BTW, where _has_ Andrew been the last week or so?h   Sshhh!!! Don't jinx it!p   --   David J. Dachteraa 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.t   ------------------------------  % Date: Tue, 05 Dec 2000 12:27:19 -0700 - From: "Rowell, Bradley" <browell@state.mt.us>  Subject: OCR VMS software @ Message-ID: <1245D1C0C039D411933708002BB29C644B2751@DOAISD02003>  I Does anyone know of any Alpha-VMS optical character recognition software?   & -------------------------------------- Bradley G. Rowells$ Montana Department of Transportation' ---------------------------------------t   ------------------------------   Date: 5 Dec 2000 22:22:00 GMT-2 From: mathog@seqaxp.bio.caltech.edu (David Mathog) Subject: Re: OCR VMS softwareg, Message-ID: <90jpq8$55k@gap.cco.caltech.edu>  p In article <1245D1C0C039D411933708002BB29C644B2751@DOAISD02003>, "Rowell, Bradley" <browell@state.mt.us> writes: >eJ >Does anyone know of any Alpha-VMS optical character recognition software? >i  F There are some OCR packages in an early state of development listed onI www.sourceforge.net and the linux Access-HOWTO mentions xocr.  You'd have K to port one of these, but there's nothing intrinsically Unixy about OCR, sofK it should be possible.  On the other hand, there are a few notes around on lI the web indicating that xocr doesn't work very well, so you may be better A off using one of the commercial OCR packages for Windows or Mac. W   David Mathog mathog@seqaxp.bio.caltech.eduf? Manager, sequence analysis facility, biology division, Caltech     ------------------------------  $ Date: Tue, 5 Dec 2000 13:54:30 -06001 From: "Dave Gudewicz" <david.gudewicz@abbott.com>t Subject: Re: OpenVMS Items8 Message-ID: <90jgvg$11q$1@fizban.fizban.pprd.abbott.com>  ( Here's what I got when I tried the link:  < This site is only accessible through the link pages found at7 WWW.OPENVMS.COMPAQ.COM or WWW.OPENVMS.DIGITAL.COM (app),   Dave...t  @ "Sue Skonetski" <susan.skonetski@compaq.nospam> wrote in message0 news:90j0h0$e7k$1@mailint03.im.hou.compaq.com...L > Just wanted to let you know that you can now purchase OpenVMS Items off of& > the OpenVMS web site.  Please visit. >  > L http://www.e-softwarefactory.com/acb/showprod.cfm?&DID=30&CATID=194&ObjectGr > oup_ID=833 >e > This list will be growing. >0 > suec >t >x   ------------------------------   Date: 5 Dec 2000 21:01:59 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: OpenVMS Items6 Message-ID: <90jl47$j43$1@mailint03.im.hou.compaq.com>  l In article <90jgvg$11q$1@fizban.fizban.pprd.abbott.com>, "Dave Gudewicz" <david.gudewicz@abbott.com> writes:) :Here's what I got when I tried the link:  : < :This site [http://www.e-softwarefactory.com/ -srh] is only C :accessible through the link pages found at WWW.OPENVMS.COMPAQ.COM   :or WWW.OPENVMS.DIGITAL.COM...    H   I am told that this is the expected (though a little weird) behaviour.J   In other words, please visit www.openvms.compaq.com to get to the goods.    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  $ Date: Tue, 5 Dec 2000 14:48:20 -0500  From: John Santos <JOHN@egh.com>5 Subject: Re: Problems installing MQ Series on VMS 6.2 5 Message-ID: <1001205144125.2913A-100000@Ives.egh.com>a  % On 5 Dec 2000, Larry Kilgallen wrote:S  R > In article <90icbn$mcm$1@nnrp1.deja.com>, Dave Barlow <uzko@my-deja.com> writes: > I > > We have MQ Series V2.2 running on a number of VAXen under VMS 6.2. WeeI > > tried restarting the MQ queue manager on a developement node but thislJ > > ACCVIO'ed with an OPCDEC error message. Our internal MQ series support > I > ACCVIO and OPCDEC are two separate faults.  Are you saying you got themy; > both in the same signal, or were they successive events ?  > J > > and developers could find no problems with their sotware so we decided > > to re-install MQ Series. > J > That is a Microsoftian approach to the problem, and rarely helps on VMS. > D > > After deleting the existing installation the install fails with, > > : > > %VMSINSTAL-I-RESTORE, Restoring product save set B ...: > > %VMSINSTAL-I-RESTORE, Restoring product save set D ...+ > > %SYSTEM-F-BADPARAM, bad parameter valuegF > > %VMSINSTAL-F-UNEXPECTED, Installation terminated due to unexpected
 > > event. > > I > > Running VMSINSTAL with verify enabled, after editing kitinstal.com, Is+ > > found that the install script is doing,a > > # > > $MC VMI$KWD:SETDOMAIN_D 1000002C  C One one of my 3 VMS nodes, "MC" does not equate to "MCR".  However, J this give "%DCL-W-ABVERB, ambiguous command verb - supply more characters"  H It is a sign of sloppiness, though.  Command files eschew abbreviations.; (I've seen DEC/Compaq .com files that use "MC" also, IIRC.)a    F > > Which fails with %SYSTEM-F-BADPARAM, bad parameter value. Manually' > > running this executable also fails.2 > G > Please tell the vendor of SETDOMAIN_D.EXE to learn how to use chained:8 > messages, such that the error would be something like: > . > 	%SETDOMAIN-F-CALLENQ, error calling SYS$ENQ5 > 	-SETDOMAIN-I-CONTEXT, during setup for WOMBAT modeh* > 	-SYSTEM-F-BADPARAM, bad parameter value > 1 > That would help them solve your problem sooner.l   Agreed.i   > K > > Any one seen this specific error or know a solution. Thanks in advance.d > O > Even if someone does know an answer, please get the vendor to fix their code.-  E I don't know, but am involved in a back-burner project to upgrade oneeG customer from MQSeries V1.5 to V2.2.  Anyone ever done this?  Are therejH any tools to help?  (Although the programming interface to MQSeries V2.2F looks upward compatible with V1.5, the management interface appears toI be entirely different.  The design concepts don't even appear to map fromf one to the other!)   -- o John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Tue, 05 Dec 2000 15:09:27 -0500i. From: Hank Vander Waal <hvanderw@novagate.com> Subject: raid system needed-8 Message-ID: <000c01c05ef7$46a89400$2c96a8c6@mscmain.com>  E I have a customer with 1000a 5/400 that needs to add an external raidrJ system.  He has about 15 drives that are in the SBB drive  containers thatJ he would like to use.  Can someone tell me what raid box I can get for him ?? (running VMS 7.1)g   Hank Vander Waal   ------------------------------   Date: 5 Dec 2000 21:32:47 GMTj2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: raid system neededv6 Message-ID: <90jmtv$jjh$1@mailint03.im.hou.compaq.com>  i In article <000c01c05ef7$46a89400$2c96a8c6@mscmain.com>, Hank Vander Waal <hvanderw@novagate.com> writes:.F :I have a customer with 1000a 5/400 that needs to add an external raidK :system.  He has about 15 drives that are in the SBB drive  containers thatmK :he would like to use.  Can someone tell me what raid box I can get for himl :??  (running VMS 7.1)  J   The goal here being the simple creation of a RAID configuration :-), or E   might there be some more specific background information available?t  G   You can (of course) use OpenVMS Alpha V7.1 (or better, OpenVMS Alpha sL   V7.1-2) with host-based volume shadowing, with no extra hardware required.K   This assumes you can get to OpenVMS Alpha V7.1-2 with ECOs, and you have  H   disks with the same numbers of blocks, and you acquire the appropriateJ   shadowing licenses for your site from those licenses that are available.  G   Given the likely vintage of these (unspecified) older disks, I'd also I   look at replacing the disks with a new StorageWorks configuration with mL   an HSZ or an RA-series controller configuration.  This based on increases G   in capacity, a reduction in the number of spindles required, improvedfJ   latency and bandwidth, and the ease of access to the newer StorageWorks K   RAID controllers and shelving, and the ease of installing the new storagez:   configuration (rather than retrofitting into the old)...  H   If you want to pursue this, you will have to specifically identify theK   disks and the current StorageWorks configuration.  Also some idea if this4K   will be in-board (backplane RAID), host-based RAID, external (HSZ-series) 9   RAID, and what sort of host controller(s) are in use.  a  I   I am here rather obviously assuming that the "SBB" is a reference to a e    StorageWorks Building Block...   N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  $ Date: Wed, 6 Dec 2000 14:32:43 +13009 From: "Antony Wardle" <antony.wardle@nnnoospam.met.co.nz>  Subject: Re: raid system needed=1 Message-ID: <kEgX5.2043$Vu1.45151@ozemail.com.au>I  . I think that the only supported box is a ba352  + Well that is what we've got. Can't remembers how many slots it has.  % We also use shadowing for redundancy.f  - We don't use raid, just jbods and a 3 channel$- mylec controller. Seems to be limited to 32gbs and seven logicals disks  	 any help?.   antony          ; "Hank Vander Waal" <hvanderw@novagate.com> wrote in messagee2 news:000c01c05ef7$46a89400$2c96a8c6@mscmain.com...G > I have a customer with 1000a 5/400 that needs to add an external raid L > system.  He has about 15 drives that are in the SBB drive  containers thatL > he would like to use.  Can someone tell me what raid box I can get for him > ?? > (running VMS 7.1). >. > Hank Vander Waal >( >l >a   ------------------------------  % Date: Tue, 05 Dec 2000 20:41:32 -0500  From: Ray <lists@aik.tec.sc.us>gN Subject: Sun new cluster comparison ommitted VMS (but VMS was their benchmark), Message-ID: <3A2D994C.B6085CB@aik.tec.sc.us>  1 http://www.theregister.co.uk/content/2/15251.htmlt  A                  "So this was some omission, we noted. "Who?" was.A                  Scott McNealy's response to our Compaq question,iE                  although Sun cluster boss and Solaris VP Andy IngramvF                  had more to add: Compaq didn't figure in Sun's sights@                  in the dot.com enterprise he told The Register,B                  although VMS Clusters had served as the benchmark2                  for Sun's engineers. Funny that."   ------------------------------  # Date: Tue, 05 Dec 2000 22:16:29 GMT 3 From: corbett@epp.ne.mediaone.net (Michael Corbett)r- Subject: Re: TCPware/Purveyor Problem Part #2o8 Message-ID: <slrn92quug.k7a.corbett@epp.ne.mediaone.net>  J On Tue, 05 Dec 2000 11:45:56 GMT, Neil Rieck <n.rieck@sympatico.ca> wrote: >O.K.t >cF >So after some hacking I've discovered that the only way to get SYSUAFL >authorization to work with "sample_auth_vax.dll" is to write a batch job toG >update file "Purveyor$Hidden.GAC" for the associated "virtual server".o >uH >Does anyone know if this solution is correct? I just assumed that there; >would be a special (seventh?) entry one could make in file 0 >"Purveyor$Hidden.GAC" like "Allow-SYSUAF-only". >a  F In the .GAC file you want to have an ACL that allows any authenticated
 user access -    Allow-user: +,*,*,Defaultd  I The key is the "+" in the username field it means any authenticated user.p   -- dK +-------------------------------------------------------------------------+ D Michael Corbett                           Email: Corbett@process.com   ------------------------------  # Date: Wed, 06 Dec 2000 01:56:25 GMT ) From: "Neil Rieck" <n.rieck@sympatico.ca>y- Subject: Re: TCPware/Purveyor Problem Part #2-; Message-ID: <d1hX5.34195$J32.1170926@news20.bellglobal.com>a   Thanks for the prompt response.RH Is the plus sign in the ACL documented anywhere? (web or hardcopy docs?)  
 Neil Rieck  @ "Michael Corbett" <corbett@epp.ne.mediaone.net> wrote in message2 news:slrn92quug.k7a.corbett@epp.ne.mediaone.net...L > On Tue, 05 Dec 2000 11:45:56 GMT, Neil Rieck <n.rieck@sympatico.ca> wrote: > >O.K.w > >nH > >So after some hacking I've discovered that the only way to get SYSUAFK > >authorization to work with "sample_auth_vax.dll" is to write a batch jobi toI > >update file "Purveyor$Hidden.GAC" for the associated "virtual server".g > >wJ > >Does anyone know if this solution is correct? I just assumed that there= > >would be a special (seventh?) entry one could make in file 2 > >"Purveyor$Hidden.GAC" like "Allow-SYSUAF-only". > >  >uH > In the .GAC file you want to have an ACL that allows any authenticated > user access -o >r > Allow-user: +,*,*,Defaultn > K > The key is the "+" in the username field it means any authenticated user.i >  > -- >oK +-------------------------------------------------------------------------+ F > Michael Corbett                           Email: Corbett@process.com   ------------------------------  # Date: Tue, 05 Dec 2000 23:03:15 GMT  From: unixguys@my-deja.com' Subject: Re: Technology of US elections ) Message-ID: <90js7e$uk6$1@nnrp1.deja.com>v  ' 12/05/00; (Reuters): Buckingham Palace.m  ; HRH Queen Elizabeth II,announced today that in the light ofp: America's "failure to elect a President and thus to govern5 itself"..."America's independence is hereby revoked".a  F Speaking through Sir Robert Lowdney, Lord Chamberlain of Her Majesty'sG Court of Parliamentary and Political Exchanges, she proclaimed that shei; will forthwith resume her monarchal duties over all states,uE commonwealths and other territories. "except Utah, which she does nota fancy."e   >>>END>>>Reuters>>>  >>>60>>>    & Sent via Deja.com http://www.deja.com/ Before you buy.l   ------------------------------    Date: 06 Dec 2000 09:01:33 +0800/ From: Paul Repacholi <prep@k9.prep.synonet.com>g- Subject: Tester for FW re-submission of Spellt0 Message-ID: <87vgsy4b2q.fsf@k9.prep.synonet.com>  - This is a call for testers for my updated and , re-packaged Spell, originally from the Decus submission.w  * The new copy is a PROD and forget, I hope.( One PCSI file, installs into sys$system,/ sys$share, sys$help, and updates the dcltables.   . You will need 7.2/7.2-1 to keep the RTLs happy2 I'd expect, but if you think Pascal and an earlier+ version will work, let me know how it went.l   TNX in advance,a   ~Paulu  * PS, due to mailer/user braindamage, please' remove the 'k9' from the address beforec
 replieing.   ------------------------------  $ Date: Wed, 6 Dec 2000 00:25:19 +01000 From: "Taki Pioy" <nospam@localhost.spamwarn.zz>( Subject: Re: Two hosts sharing RA system4 Message-ID: <XReX5.561$r42.1233@nntpserver.swip.net>  - <steven.reece@quintiles.com> wrote in messageaB news:OF4DEEF704.F536E4D6-ON802569AC.0040F54C@qedi.quintiles.com... > L > One of the clusters uses three HSZ70 pairs whilst the other uses a pair of@ > HSZ50s and a pair of HSZ40s.  They are each strung between two AlphaServersL > using either KZPBAs in the hosts (on the HSZ70s) or KZPSAs (on the 40s and > 50s).S >wJ > Each of the hosts has a Y-cable on the adapter card so that the host can beH > powered down and the SCSI cable disconnected without breaking the SCSI bus. >'   Something like this? :-)1                                     ------ ------m$   (term) ||----Y      Y      Y----||               KZP    HSZ    KZPt  B Hmm... Does it matter if the HSZ goes in the middle or at one end?  > Does Compaq, or somewhere else perhaps, have any documentation8 on shared SCSI "issues", also for non-clustered systems?  H > I've not done any work with shared systems, although reading the HSZ70J > manuals it does appear to support this to some extent.  I'm not sure howD > well it works.  It also comments on Passthrough devices where SCSI (snip)  4 Aren't there other ways to "reserve" disks to hosts?   Regards  /Fad   ------------------------------  $ Date: Tue, 5 Dec 2000 23:37:27 -00004 From: "John D. Peedle" <john@peedle.freeserve.co.uk> Subject: Re: US Constitution. Message-ID: <90ju87$ubt$1@news8.svr.pol.co.uk>  < The Americans saved our bacon. There is no question of that.   What upsets people is:'    They waited so long to get involved.iI    They milked us (the British) of everything they could before they did.rI    Their unjustified cocky atitude - if we had the natural resources that:
 the US has...n*    We were expected to be sooooo grateful.  K America should remember that it needs the rest of the world a lot more than K we need it. It needs us to sell to, to buy from, and, just occasionally, touB gently remind it that things were not *quite* like that after all.  I During 19 years in the Military, I worked with US Forces on more than onebI occasion. Their logistic support is second to none. Their chaps are keen,DL enthusiastic (if a little undisciplined!) and dedicated. (We still beat them/ every time we play war games with them though!)-  D All that said, I do not agree with Nicholas Monsarrat who said " TheL Americans are not a great race, there's just a lot of them". They are great, usually.  7 Isn't this way off topic? (good conversation though...).   John D. Peedle   ------------------------------    Date: 06 Dec 2000 05:43:48 +0800/ From: Paul Repacholi <prep@k9.prep.synonet.com>4 Subject: Re: US Constitution0 Message-ID: <87sno2r1bf.fsf@k9.prep.synonet.com>  # Shane.F.Smith@Healthnet.com writes:p  M > I have full respect for everyone who fought, regardless of nationality. I'm M > not claiming America didn't make a massive contribution. My problem is withcF > the way it's all been portrayed since - neglecting to give credit toM > everyone else. The American tendancy to forget that Britain fought, and thehD > Russians, and the Canadians, and the exiled Polish, and the French > resistance, and many more.  % More Australians died in WWI than US.c  1 > I object to only American soldiers appearing in J > saving private Ryan, U571 supposedly capturing the first enigma machine,K > the heartfelt but mistaken belief that America won the war single handed.5A > There are British graveyards in France, too. Don't forget that.o  < The first info on Enigma came from the Poles. Then a special5 snatch op was set up to 'obtain' on from the factory.n  B BTW, is there any North American or Caribean country the US hasn't invaded?   ~PaulP   ------------------------------  # Date: Tue, 05 Dec 2000 19:54:24 GMTo" From: falk@arc.ab.ca (Alfred Falk)= Subject: Re: US Constitution (was Technology of US elections).2 Message-ID: <90018AE38falkarcabca@205.233.108.180>  5 rdeininger@mindspring.com (Robert Deininger) wrote in B <rdeininger-0512000204370001@user-2ivecjc.dialup.mindspring.com>:   6 >In article <3A2C7904.F4C89E87@videotron.ca>, JF Mezei' ><jfmezei.spamnot@videotron.ca> wrote: xI >> In Canada's recent election, there was one seat which was contested, a G >> recount was done and the runner-up ended up winning the seat. It was A >> done in the matter of a couple of days, no lawsuits, no courtseA >> involved. Why ? Because there are clear national standards andi+ >> procedures on how this is accomplished. I >tC >There are pretty clear standards in FL also.  A certain faction isyH >seeking to change those standards.  No, the standards are not perfectly: >uniform across the state, but they are well-established.  > H >Of course, we had many close elections this year, and many of them wereG >recounted.  Virtually all _could_ have been taken to court, but almost-F >none were.  I think the reasons have more to do with the character ofC >the folks involved than with details of the laws and regulations. z  J When an election is close, automatic recounts are triggered.  In addition,E a candidate can request the courts to order a re-count.  There are a i- couple of seats being contested as of Dec. 4.4  D >> It seems to me that the US election systems are like the tower ofF >> Babel. A huge collection of different and incompatible systems, andF >> when something breaks, all hell breaks loose because nobody can seeF >> the big picture, not even the supreme court, since everything is so >> localised.  >b# >Different, yes.  Incompatible, no.C >rI >What fraction of ballots in Canada are typically undervoted, or spoiled,  >or otherwise uncountable? i  K http://www.election.ca/ has 1997 results, indicating 1.4% rejection, up to  I 3.4% in Quebec and New Brunswick.  I suppose this year's will be similar.dH They don't break down reasons for rejection but it it would include any J ballot that has something other than "one X in an appropriate location".  * It wouldn't include ballots not deposited.   @ ----------------------------------------------------------------A   A L B E R T A         Alfred Falk               falk@arc.ab.ca P@ R E S E A R C H         Information Systems Dept   (780)450-5185+   C O U N C I L         250 Karl Clark Road 1                         Edmonton, Alberta, Canadat http://www.arc.ab.ca/   T6N 1E4   http://www.arc.ab.ca/staff/falk/   ------------------------------  % Date: Tue, 05 Dec 2000 20:12:05 -0600o7 From: "David J. Dachtera" <djesys.nospam@earthlink.net> 5 Subject: Re: Use the e-store button off the home pagea- Message-ID: <3A2DA075.D7D05454@earthlink.net>    Sue Skonetski wrote: > K > Please use www.openvms.compaq.com and select the shop the OpenVMS e-store    I counted some nine cookies!   -- b David J. Dachtera  dba DJE Systemsn http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/g  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: Tue, 05 Dec 2000 21:06:35 -0600a7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>e5 Subject: Re: Use the e-store button off the home page - Message-ID: <3A2DAD3B.AAFBEA7C@earthlink.net>t   Sue Skonetski wrote: > K > Please use www.openvms.compaq.com and select the shop the OpenVMS e-storel  @ *DEFINITELY* not priced for small business, and *WAY* out of the1 hobbyists's league (money-wise) for OLD and SPLs.p  ? Cut some of those prices by 80% to 90% and *MAYBE* we can talk!p  G I can't say "close, but no cigar" - this isn't even in the same galaxy!   = Exactly what part of "affordable" does Compaq not understand?-   -- a David J. Dachtera0 dba DJE Systems  http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/c  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.7   ------------------------------  % Date: Tue, 05 Dec 2000 15:59:37 -050004 From: "Bochnik, William J" <BochnikWJ@BERNSTEIN.COM>$ Subject: RE: Very weird DCL behaviorJ Message-ID: <2B37459189B0D211BE710000F8EF9D850890894E@nts0147.beehive.com>  J sync isnt going to help if he's got more than 1 tape drive (but all thingsJ are possible with properly hacked dcl).  If he has only 1 tape drive (or 17 man program per tape drive) then SYNC is the way to go.    > -----Original Message-----J > From:	hoffman@xdelta.zko.dec.nospam [SMTP:hoffman@xdelta.zko.dec.nospam]* > Sent:	Tuesday, December 05, 2000 1:17 PM > To:	Info-VAX@mvb.saic.comt& > Subject:	Re: Very weird DCL behavior >  > C > In article <o19X5.8166$Cn4.198087@weber.videotron.net>, "Syltrem"   > <syltrem@videotron.ca> writes: >  > :The main procedurepE > :has a list of things to do which it dispatches (via a spawn/nowaite
 > command)$ > :to the next available tape drive. > ..H > :When one backup is finiished, the spawned procedure notifies the mainL > :process that it has finished so that another job can be sent to that tape > :drive (am I clear enough?). > K >   I'd probably either use SYNCH and separate batch jobs, or (if you have kJ >   four tape drives), a four-slot batch queue and ALLOCATE mmdu:,mmdu,...J >   or ALLOCATE/GENERIC, as this can greatly simplify the synchronization  >   involved.  c > L > :Upon reading the mailbox, the main procedure sometimes (once every 100 or1 > :200 times it does the READ command) hits errorD- > :%RMS-E-RSA, record stream currently activeaG > :when (or right after) reading the mailbox. I do not know why this iseL > :happening, and the weird thing about this is that the .LOG shows commands, > :that do not exist in the .COM file, like: > D >   Looks like something is getting scrozzled deep within DCL, or inC >   the code between OpenVMS and the mailbox driver -- DCL can onlyhB >   use mailboxes with some additional assistance, so you are, um,< >   boldly going where few DCL programmers have gone before. >    > :Can someone help? > K >   Please acquire and apply the current ECOs for whatever OpenVMS release SH >   and platform might be in use here.  And seriously consider using the7 >   SYNCH and/or ALLOC and the four-slot batch queue...: > 4 >  --------------------------- pure personal opinion > ---------------------------y1 >    Hoff (Stephen) Hoffman   OpenVMS Engineeringm > hoffman#xdelta.zko.dec.com   ------------------------------  $ Date: Tue, 5 Dec 2000 16:11:47 -0500& From: "Syltrem" <syltrem@videotron.ca>$ Subject: Re: Very weird DCL behavior7 Message-ID: <ZPcX5.8666$Cn4.207476@weber.videotron.net>o  J The problem with SYNCH is that the backup processes are detached processesK (I mean, not *real* detached, but separate from my main process). The way IeG do it, I mount the tape in the main procedure, and backups append to it J (BACK/NOREW) without having to rewind the tape after every backup command.  I The SYNCH method would take at least twice as long to complete everythingeJ because of the time required to rewind the tape, then for the next process; to go at logical end-of-tape befors actually doing its job.-   This is OVMS Alpha 7.2-1   Syltrem0    ? "Hoff Hoffman" <hoffman@xdelta.zko.dec.nospam> wrote in messagec0 news:90jbf7$gut$3@mailint03.im.hou.compaq.com... >cC > In article <o19X5.8166$Cn4.198087@weber.videotron.net>, "Syltrem"e <syltrem@videotron.ca> writes: >t > :The main procedureoE > :has a list of things to do which it dispatches (via a spawn/nowaitM command)$ > :to the next available tape drive. > ..H > :When one backup is finiished, the spawned procedure notifies the mainL > :process that it has finished so that another job can be sent to that tape > :drive (am I clear enough?). >eJ >   I'd probably either use SYNCH and separate batch jobs, or (if you haveJ >   four tape drives), a four-slot batch queue and ALLOCATE mmdu:,mmdu,...I >   or ALLOCATE/GENERIC, as this can greatly simplify the synchronizationm
 >   involved.c >lL > :Upon reading the mailbox, the main procedure sometimes (once every 100 or1 > :200 times it does the READ command) hits error5- > :%RMS-E-RSA, record stream currently activepG > :when (or right after) reading the mailbox. I do not know why this is L > :happening, and the weird thing about this is that the .LOG shows commands, > :that do not exist in the .COM file, like: >rD >   Looks like something is getting scrozzled deep within DCL, or inC >   the code between OpenVMS and the mailbox driver -- DCL can only B >   use mailboxes with some additional assistance, so you are, um,< >   boldly going where few DCL programmers have gone before. >  > :Can someone help? >sJ >   Please acquire and apply the current ECOs for whatever OpenVMS releaseH >   and platform might be in use here.  And seriously consider using the7 >   SYNCH and/or ALLOC and the four-slot batch queue...- >-, >  --------------------------- pure personal# opinion ---------------------------c1 >    Hoff (Stephen) Hoffman   OpenVMS Engineering0 hoffman#xdelta.zko.dec.com >    ------------------------------   Date: 5 Dec 2000 23:08:45 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)$ Subject: Re: Very weird DCL behavior6 Message-ID: <90jsht$knb$1@mailint03.im.hou.compaq.com>  ` In article <ZPcX5.8666$Cn4.207476@weber.videotron.net>, "Syltrem" <syltrem@videotron.ca> writes:K :The problem with SYNCH is that the backup processes are detached processes C :(I mean, not *real* detached, but separate from my main process).    C   "Subprocess" is the term that you seek here, and you are sharing hA   the tape MOUNT itself among multiple members of the job tree...h  >   Why not have one process (job) tree for each tape drive?  OrA   alternatively, you can gang together tape drives with one batcho#   job, and alternate media access.)I  
 :...The way ItH :do it, I mount the tape in the main procedure, and backups append to itK :(BACK/NOREW) without having to rewind the tape after every backup command.e  A   Ah, more details on the particular problem you seek to solve...    Always useful...  J :The SYNCH method would take at least twice as long to complete everythingK :because of the time required to rewind the tape, then for the next process-< :to go at logical end-of-tape befors actually doing its job.  H   SYNCHRONIZE itself is to synchronize the completion of the batch jobs,G   and does not have a direct effect on the tape position.  You are hereuI   refering to the operation of (and sharing of) the tape mount operationsa,   as processes are logged in and logged out.  I   Depending on the tape drive used, take a look at the fast-skip support.'   :This is OVMS Alpha 7.2-1n  	   Okfine.   J   I would guess that there is something odd about the DCL interaction withK   the mailbox and the timer -- in the part of the example DCL code that wasnK   not posted, I will assume you have opened the mailbox with /SHARE, and I  K   will further assume you have exactly one application reading the mailbox ""   and zero or more writers active.  M   Have you considered using an indexed file for the work pending?  (It might oK   be a little cleaner to implement and debug, and you can store completion 9*   status information in the file records.)  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Tue, 05 Dec 2000 21:12:30 -0600e7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>a$ Subject: Re: Very weird DCL behavior- Message-ID: <3A2DAE9E.60C5527B@earthlink.net>e   Hoff Hoffman wrote:  > b > In article <ZPcX5.8666$Cn4.207476@weber.videotron.net>, "Syltrem" <syltrem@videotron.ca> writes:M > :The problem with SYNCH is that the backup processes are detached processeseD > :(I mean, not *real* detached, but separate from my main process). > D >   "Subprocess" is the term that you seek here, and you are sharingC >   the tape MOUNT itself among multiple members of the job tree...  > @ >   Why not have one process (job) tree for each tape drive?  OrC >   alternatively, you can gang together tape drives with one batch8% >   job, and alternate media access.)   H That's sort of what I did for a former site. I'd have no problem sharingG the code with anyone - they're cool with it. It's a bit complex in it'slF full-blown form (dispatching backup jobs, synchronization, post-backupG tasks, shadow/mirror-set split and rejoin, etc.), but it works reliablya" (as surprising as that may sound).   -- e David J. DachteraS dba DJE Systemst 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.t   ------------------------------  $ Date: Tue, 5 Dec 2000 16:20:08 -0500% From: "Brian Tillman" <tillman_brian>p( Subject: Re: VMS and NT integration?????$ Message-ID: <3a2d5b53$1@news.si.com>  I >Hardly large systems taking up tons of space. I'd think the system unitsv cansK >probably sit on a shelf in a rack fairly easily (certainly a Microvax 3100,; >model 80 could - can't remember the sizes of VAX 4000's ).l  F I've got two 400-505As and an Alpha 1000 in a single full height rack. --A Brian Tillman                   Internet: tillman_brian at si.comeA Smiths Industries, Inc.                   tillman at swdev.si.com = 3290 Patterson Ave. SE, MS      Addresses modified to preventh< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------    Date: 06 Dec 2000 02:22:44 +0800/ From: Paul Repacholi <prep@k9.prep.synonet.com>a Subject: Re: VMS vs. Tru64 Unixl0 Message-ID: <874s0isp6z.fsf@k9.prep.synonet.com>   hein@eps.ko.dec.c*m writes:    OK, some questions if I may...  G > Unix has RADicalied/NUMAtied its UBC. For 'absolute' performance UnixbG > applcation implementors use 'RAW' IO also avoiding a shared (central)nJ > buffer. Unix (currently) has the additional benefit of using a RAD-local5 > adapter for it's IO if an access path is available.   ; What do you mean by RAD? I take it this is moving to a mored< shared-memory SMP system than the current 'common-everthing'> approach. Each CPU rns its 'own' IO devices and does buffering into CPU-local memory?   > ? > >Also, tests I've run have shown 20:1 advantage to VMS vs DU4c= > >in heavy paging. This was with FFTs, so don't expect it tot > >at all repressent a DB case.i > 5 > Production system are not supposed to page or swap.n  . Yes, but there is the small matter of price...  M > No matter which one is better or worse, Test showing a magnitude differenceAJ > in end user performance on the same box are 'suspect' in the very least.H > Either they uncovered a anomaly which is probably fixable or they just: > show un-equal tuning/coding/setup skills for the tester.  > :) The results stunned me. I expected DU would have reasonable; paging. I've no doubt DUs performance could be improved bute< all the tweeks I did just moved things a percent or two. But' my DU skills are pretty non-existant...u  9 Grab the FFTW code and run the speed test. Shove the size ; up till most of the data is paging and compare. FFTs access ) is about as horrid as it gets for paging.o   ~Paulf   ------------------------------  % Date: Tue, 05 Dec 2000 13:47:38 -0500l5 From: Forrest Kenney <Forrest.Kenney@compaq.com.doom>n3 Subject: Re: Who =?iso-8859-1?Q?=B4s?= the wizard ?t/ Message-ID: <3A2D384A.285B022E@compaq.com.doom>-  H     The wizard and his minions made an attempt to find out the source of( the current photo and who is in it.  TheH wizard was unable to come up with an answer.  The wizard is working on a! more up to date photo for the webm> page but cannot commit to a release date for this new picture.     Forrest KenneyF No I am not the wizard but I have helped the wizard every now and then   ------------------------------  % Date: Tue, 05 Dec 2000 18:48:37 -0600n7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>>3 Subject: Re: Who =?iso-8859-1?Q?=B4s?= the wizard ?t- Message-ID: <3A2D8CE5.3F356F7E@earthlink.net>u  & "Brian Schenkenberger, VAXman-" wrote: > [snip]( > decus_geeks_annotated.gif to be exact.   The Wizard's familiars, eh?w   -- s David J. Dachterat dba DJE Systemst http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/r  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: Tue, 05 Dec 2000 19:07:48 -0600e7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>t Subject: Re: Zip Help!!!- Message-ID: <3A2D9164.CE16179A@earthlink.net>    Dave O'Brien wrote:u > H > Speaking of zip problems, does anyone know of a size limitation on theK > input file when creating a zip file?  I haven't figured out exactly whereeC > the limit is, but is seems to be somewhere in the neighborhood ofh > 2GB.  Here is an example:o  F The file size would be circa. %7fffffff, the maximum value that can be expressed in a signed longword.l  d > [snip] > Any ideas?  H The lastest-and-greatest, V2.3 supports up to 4GB files (%Xffffffff, theG maximum file size that can be expressed in an unsigned longword). You'dt, need UNZIP V5.4-1 to unpack such an archive.   Get it at (no, not Porky's):+ ftp://ftp.freesoftware.com/pub/infozip/VMS/   C ...but, I'll repeat once again that there are, IMHO, better ways tolG archive such large files and that the CPU time necessary for attemptingiG to compress such huge files is very likely to make it prohibitive: some-* hours on an Alpha, perhaps days on a VAX.   E A recent test showed that a VAX (normal production load) took nine(9)aH times longer to ZIP a "small" (circa. 10MB) file than an Alpha (VAX 4000H Vs. Alpha 2100 4/300 (I think)). That means a nine(9) hour ZIP job on an6 Alpha would run for 81 hours (3 09:00:00.00) on a VAX.  > DLT would be a far better alternative than ZIP, GZIP or BZIP2.   -- a David J. Dachtera0 dba DJE Systems  http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/g  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.i   ------------------------------   End of INFO-VAX 2000.680 ************************