1 INFO-VAX	Wed, 29 Nov 2000	Volume 2000 : Issue 667       Contents: 4mm tape drive not working" Re: Batch reports and print queues Re: Broken TK50 / Charon-VAX Hobbyist Edition - Printout problems   DIR command gives an audit entry$ Re: DIR command gives an audit entry' Eg: Propagating a TXN over DECnet CHILD ' Re: generate a shareble image with DECC ' Re: generate a shareble image with DECC   Re: Gigabit Ethernet and Alpha's  Re: Gigabit Ethernet and Alpha'sC Help required - Configure NFS server (VMS) and NFS Client (Unix/NT)  How fast is a SCSI connection ! Re: How fast is a SCSI connection / HSZTERM... status of support of it by Compaq... 3 Re: HSZTERM... status of support of it by Compaq... 3 Re: HSZTERM... status of support of it by Compaq...  Re: MMS clean up utilities Re: More VMS, BSD, and FORTRAN Re: More VMS, BSD, and FORTRAN Re: More VMS, BSD, and FORTRAN Re: More VMS, BSD, and FORTRAN Re: More VMS, BSD, and FORTRAN Re: Mozilla M18 i005 Re: Mozilla M18 i005 Re: Mozilla M18 i005 Re: Mozilla M18 i005 Re: Mozilla M18 i005A RE: New Encompass Survey... PLease Take 10 Minutes to Fill It Out A Re: New Encompass Survey... PLease Take 10 Minutes to Fill It Out A RE: New Encompass Survey... PLease Take 10 Minutes to Fill It Out A Re: New Encompass Survey... PLease Take 10 Minutes to Fill It Out  Re: OpenVMS DHCP Client?$ Propagating a TXN over DECnet parent SNMP and MIB agents  Re: SW Raid 310 and VMS  Re: SW Raid 310 and VMS  Re: SW Raid 310 and VMS  Re: System Service and RTL Re: Technology of US elections Re: Technology of US elections Re: Technology of US elections Re: Technology of US elections Re: Technology of US elections Re: Technology of US elections: Re: TK70: request for data transer fro TK70 to 4mmDat tape Re: VMS and NT integration?????  Re: VMS and NT integration?????  Re: VMS System Backup  Re: VMS System Backup  Re: VMS System Backup  Re: VMS System Backup # What happened to www.graysoft.com?! - Re: Why so much virtual memory for tiny jobs? - Re: Why so much virtual memory for tiny jobs? - Re: Why so much virtual memory for tiny jobs?   F ----------------------------------------------------------------------    Date: 29 Nov 2000 05:34:52 -0800; From: babiarz at endor dot com <babiarz_member@newsguy.com> # Subject: 4mm tape drive not working * Message-ID: <9030ls012ng@edrn.newsguy.com>  D I have a 4mm tape drive that refuses to allow a tape to be inserted.; It was working fine last week, yellow light is now flashing   2 alpha 1000a, vms 7.2-1 current uptime is 159 days." tape drive is ARCHIVE PYTHON 04106  G I hate to take down the system for tape drive, but is there something I ; can do to "reset" the drive without taking the system down.    tia    john   ------------------------------  # Date: Wed, 29 Nov 2000 08:59:22 GMT / From: Mike Price <mike.price@littlewoods.co.uk> + Subject: Re: Batch reports and print queues ) Message-ID: <902gh9$f8r$1@nnrp1.deja.com>   F Check the login.com of your account and the account the batch job runs from. G There may be a check on f$mode (see help on lexicals for more info) for B INTERACTIVE or BATCH. If so the logic in the command file may skipE round some command that redefine the print command or set up logicals F that change the defualt queues. If you know the queue name involved it should be easy to find   Mike --B All opinions are my own and do not necessarily reflect those of my employer    & Sent via Deja.com http://www.deja.com/ Before you buy.    ------------------------------  # Date: Wed, 29 Nov 2000 15:32:44 GMT  From: chris_muller@my-deja.com Subject: Re: Broken TK50( Message-ID: <9037it$qb$1@nnrp1.deja.com>   John,   D Just in case you have a number of TK50's that you'd like migrated to$ other media, please keep us in mind.   regards, Chris Muller www.mullermedia.com  212-344-0474      > In article <MPG.1477e3801ffebb69989698@news.bellatlantic.net>,.   John Santos <john.santos@verizon.net> wrote:/ > In article <rdeininger-1111001451060001@user- C > 2ivear9.dialup.mindspring.com>, rdeininger@mindspring.com says... E > > In article <1001110181326.2092C-100000@Ives.egh.com>, John Santos  <JOHN@egh.com> wrote:  > >  > > D > > > I had to pull power from the drive before removing it from the cabinet F > > > (power cable isn't long enough), so I don't know.  It was amber, and H > > > looked like an LED, but I'm not sure what a photodiode looks like. > > 
 > > <snip> > > H > > > Maybe a problem with the BOT sensing?  There seems to be a slot inF > > > the part of the leader that is permantly attached to the take-upG > > > reel that should be sensed by the same LED/photodiode that senses 6 > > > the hole at EOT, maybe that isn't working right? > > G > > Sounds like a good possibility.  Either the LED, the photodiode, or  the G > > circuits they drive may be sick.  Have you cleaned the dust-bunnies  out  > > around the LED & diode?  > G > Just did that.  Found one bit of fluff under the takeup reel, which I G > removed for the 1st time.  It seems to be working better now, I.e. it E > eventually unloads, though the 1st time I tried it, it spooled back  and G > forth a few times and took several minutes.  Inserted a tape, mounted  it, G > and then did an immediate "$ dismount/unload".  I spun back and forth  for H > about a minute, then seemed to go back to the load point, both red andG > green lights on.  (It looked just like a "$dismount/nounload" at that C > point.)  After waiting a bit, it seemed to be doing nothing, so I D > released the red button, and it wound forward and back a couple of times E > before finally unloading.  Later, I tried to resume my V7.2 upgrade  and B > thinking it wanted the 2nd tape, started up with that one in the drive.H > Turns out it is still in Upgrade Phase 2, which was trying to load theD > Library and optional savesets from the first tape, which I left at work. F > Anyway, after failing to find the save-sets on the 2nd (wrong) tape, itF > successfully unloaded the tape all by it self, which is a good sign.@ > I'll have to retrieve the 1st tape before going any further... > G > > Does SET MAGTAPE/REWIND work ok?  That's not the same as unloading,  and I'm E > > not even certain it stops at the place where the index hole is in  front of@ > > the LED.  But it's something to do 'till the doctor arrives. > > F > > It _might_ still be the tach.  If the tach isn't accurate, I don't think F > > the drive can count inches of tape, and it might get very confused about F > > where it is.  If the tape slips past the tach roller, there may be	 problems. B > > That might happen if the tape tension is wrong, or if the tach turns very hard C > > due to lack of lubrication in its bearings.  But if you take it F apart, I don't know how to restore the alignment, and I don't know how much it matters. > >  > >  > > > I think the EOT H > > > hole is probably ignored when reading.  When it reaches the end ofF > > > a track, the read head probably senses this and initiates a head# > > > reposition to the next track?  > > F > > Dunno.  It would be very interesting to have a look at the service manual > > for these drives.  >  > I've never seen one... >  > --
 > John Santos  >     & Sent via Deja.com http://www.deja.com/ Before you buy.    ------------------------------  # Date: Wed, 29 Nov 2000 13:48:39 GMT 0 From: Timothy Stark <sword7@grace.speakeasy.org>8 Subject: Charon-VAX Hobbyist Edition - Printout problems8 Message-ID: <XO7V5.61975$DG3.1190805@news2.giganews.com>   Hello folks:  J I just upgraded to latest version of Charon-VAX on Windows 2000 and Linux.F I ended up some problems with serial lines. Now I discovered that thisE version has only two serial lines enabled.  That's why I tried to set G vaxprint on TXA7: but this emulator complaint about that.  When I moved H it to TXA1:, a problem disappeared.  I learned that this version is moreE crippled than before! I am not happy with that.  Be warned with that.   G Also, I figured out how to set up print queues on OpenVMS 7.2.  I asked J you for help but no one responsed my pervious postings!  I finally figured9 out when I read comments in SYSTARTUP_VMS.COM. I executed B 'START/QUEUE/MANAGER/NEW' to run queue manager daemon sucessfully.; Then I executed 'INIT/QUEUE/START/ON=TXA1: TXA1_PRINT' and  B 'INIT/QUEUE/START/GENERIC=TXA1_PRINT SYS$PRINT' successfully. Then6 I executed 'PRINT <file>' and it worked so well but...  D I have a problem with printouts but they seems messed up.  They were7 printed double-spaced.  I want single-spaced printouts.   H My printer is HP Deskjet 970Cxi.  Does anyone have experience with that?  
 Thank you!   -- Tim Stark   --  , Timothy Stark	<><	Inet: sword7@speakeasy.orgJ --------------------------------------------------------------------------F "For God so loved the world, that he gave his only begotten Son, that H whosoever believeth in him should not perish, but have everlasting life.. Amen." -- John 3:16 (King James Version Bible)   ------------------------------  % Date: Wed, 29 Nov 2000 11:54:01 +0100 ' From: Theo Jakobus <jakobus@iaf.fhg.de> ) Subject: DIR command gives an audit entry * Message-ID: <3A24EE59.1A32469D@iaf.fhg.de>   Hi!   = Every night I'm saving my files. After backup I do a listing.  Part of the log file: 
 (22:36) $!! (22:36) $! Directory of the tape. 
 (22:36) $!H (22:36) $   MOUNT/CACHE=TAPE_DATA/MEDIA_FORMAT=COMPACTION IAF021$MKA100: BACKUP TAPE 3 %MOUNT-I-MOUNTED, BACKUP mounted on _IAF021$MKA100: ; (22:37) $   DIR/SIZE/DATE/WIDTH=(FILENAME=25,SIZE=10) TAPE:    Directory TAPE:[]   > JAK_001127_F.BCK;1             111045  27-NOV-2000 00:00:00.00    ; The last command produces an entry in the security journal: 4 Security audit (SECURITY) on IAF021, system id: 1045' Auditable event:          Object access 1 Event time:               28-NOV-2000 22:37:37.97 " PID:                      00000932# Process name:             BATCH_497 ! Username:                 JAKOBUS & Process owner:            [PL,JAKOBUS] Image name:               / IAF021$DKA0:[SYS0.SYSCOMMON.][SYSEXE]VMOUNT.EXE   Object class name:        DEVICE* Object name:              _IAF021$MBA3581:& Object owner:             [PL,JAKOBUS]; Object protection:        SYSTEM:W, OWNER:R, GROUP:, WORLD: ! Access requested:         LOGICAL E Status:                   %SYSTEM-F-NOPRIV, insufficient privilege or  object protection violation          ! I would like to understand this.       Regards, --    ; *********************************************************** ; *                                                         * ; *  Theo Jakobus                                           * ; *  Fraunhofer-Institut fuer Angewandte Festkoerperphysik  * ; *  Tullastr. 72                                           * ; *  D-79108 Freiburg                                       * ; *  Germany                                                * ; *  Phone:   +49-(0)761-5159-325                           * ; *  FAX :    +49-(0)761-5159-200                           * ; *  e-mail:  jakobus@iaf.fhg.de                            * ; *  http://www.iaf.fhg.de                                  * ; *                                                         * ; ***********************************************************    ------------------------------    Date: 29 Nov 2000 16:07:10 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de> - Subject: Re: DIR command gives an audit entry H Message-ID: <y4aeai6cm9.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  N It's the MOUNT command that is generating the error, not DIRECTORY. MOUNT doesK a lot of operations internally, and will retry certain operations when they I fail, so it is quite plausible that it encountered some sort of privilege M violation along the way. As MOUNT is trusted software (as trusted as software D can be, I suppose), apart from its curiosity value, I wouldn't care.   	Jan   ------------------------------  # Date: Wed, 29 Nov 2000 11:23:14 GMT  From: richard_maher@my-deja.com 0 Subject: Eg: Propagating a TXN over DECnet CHILD) Message-ID: <902ouv$l6c$1@nnrp1.deja.com>   G $!-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- 	 =-=-=-=-= / $!      MAHER_DECDTM_PROPAGATION_CHILD_NODE.COM  $!8 $!      This command file to be used in conjunction with *_PARENT_NODE.COM  $!D $!      Parent node process is run on node GROAT to which a proxy is requiredG $!-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- 	 =-=-=-=-=  $       Set Verify2 $       DEFINE/PROCESS SQL$DISABLE_CONTEXT "FALSE"' $       SQLMOD == "$SYS$SYSTEM:SQL$MOD"  $!C $!      The code is yours *as is*. I don't offer any support, but I  will answer any C $!      questions (here or via direct e-mail) when/if I can. Enjoy!  $!G $!      NB: This is USER mode, UNpriviledged code! You do *NOT* have to  install C $!      anything. Turn all priviledges *OFF* except (netmbx,tmpmbx)  $! $       set verify  $       gosub   Create_child_cob $       gosub   dist_sql_sqlmod " $       gosub   Create_out_msg_cob $!9 $       cobol/lis                               child.cob ; $       cobol/lis                               out_msg.cob  $!D $!      You need to create a child database in sub-directory [.test] from default directory $!( $       if f$search("test.dir") .eqs. "" $       then" $               create/dir [.test]# $               set default [.test] 1 $               @SQL$SAMPLE:PERSONNEL SQL S NOCDD  $               set default [-] 
 $       endif  $       set verify $!$ $!      Then run the following sql:- $! $       SQL  ATTACH 'FILE [.TEST]PERSONNEL'; ? ALTER TABLE DEPARTMENTS ALTER COLUMN DEPARTMENT_NAME CONSTRAINT ( NO_SPACE CHECK (DEPARTMENT_NAME <> ' '); COMMIT;  EXIT $!+ $!      The SQLMOD is compiled as follows:-  $!' $       SQLMOD DIST_SQL.SQLMOD/CONTEXT= ' (SET_TRANS_RW,UPDATE_DEPT)/CONSTR=IMMED A $!      SQLMOD DIST_SQL.SQLMOD/CONTEXT=(SET_TRANS_RW,UPDATE_DEPT)  $!= $!      You need to do a $MACRO/LIS on the following (call it 
 DIST_DEF.MAR)  $! $       CREATE DIST_DEF.MAR *         .TITLE EXTERNAL SYMBOL DEFINITIONS  &         $PSLDEF                 GLOBAL&         $LNMDEF                 GLOBAL&         $DDTMDEF                GLOBAL&         $DDTMMSGDEF             GLOBAL           .END $! $       MACRO/LIS DIST_DEF.MAR $! $!@ $!      Then you link the CHILD images and create command file:- $!0 $       LINK CHILD, OUT_MSG, DIST_DEF, DIST_SQL, SYS$LIBRARY:SQL$USER/LIB $! $!      Finally:- * $       open/write out sys$login:child.com" $       write out "$run child.exe" $       close out  $! $!; $!      CHILD.COM will be called via PARENT on another node  $! $Exit: $       Exit  H $!---------------------------------------------------------------------- -------- $Create_child_cob: $       create child.cob identification division.+ program-id.    child_rtn with ident "v1.0".  author.        public domain.    data division. working-storage section. * @ 01  out_msg                             pointer value   external out_msg.@ 01  ddtm$m_nowait           pic 9(9)    comp    value   external ddtm$m_nowait.@ 01  ddtm$_integrity         pic 9(9)    comp    value   external ddtm$_integrity.@ 01  lnm$m_case_blind        pic 9(9)    comp    value   external lnm$m_case_blind. @ 01  psl$c_super             pic 9(9)    comp    value   external psl$c_super.@ 01  io$_readvblk            pic 9(9)    comp    value   external
 io$_readvblk. @ 01  io$_writevblk           pic 9(9)    comp    value   external io$_writevblk.@ 01  ss$_bufferovf           pic 9(9)    comp    value   external ss$_bufferovf.@ 01  ss$_linkdiscon          pic 9(9)    comp    value   external ss$_linkdiscon. @ 01  ss$_linkabort           pic 9(9)    comp    value   external ss$_linkabort.@ 01  ss$_remote              pic 9(9)    comp    value   external ss$_remote. @ 01  ss$_normal              pic 9(9)    comp    value   external ss$_normal. @ 01  ss$_abort               pic 9(9)    comp    value   external
 ss$_abort.- 01  sys_status              pic 9(9)    comp.  * C 01  db_spec                 pic x(50)           value   "alias pers  filename [.test]personnel". - 01  sqlcode                 pic 9(9)    comp. 
 01  dtm_iosb. -     03  dtm_iosb_status     pic 9(4)    comp.2#     03                      pic xx. -     03  reason_code         pic 9(9)    comp.rA 01 rdb$message_vector                                   external.t-     03 rdb$lu_num_arguments pic 9(9)    comp.e-     03 rdb$lu_status        pic 9(9)    comp.t@     03 rdb$alu_arguments                        occurs 18 times.-         05 rdb$lu_arguments pic 9(9)    comp. 	 01  iosb. -     03  cond_val            pic 9(4)    comp.A-     03  byte_count          pic 9(4)    comp.M-     03  chan_info           pic 9(9)    comp.I *n' 01  msg_buf                 pic x(256).o- 01  msg_len                 pic 9(4)    comp.q 01  sql_ctx.:     03                      pic 9(9)    comp    value   1.:     03                      pic 9(9)    comp    value   1.;     03                      pic 9(9)    comp    value   16.dC     03  tid                 pic x(16)           value   low-values.R=     03                      pic 9(9)    comp    value   zero.o *l- 01  net_chan                pic 9(4)    comp.1< 01  cmd_eof                 pic x               value   "N"." 01  commit_flag             pic x. 01  letter_sent.&     03  ls_tid              pic x(16).&     03  ls_bid              pic x(16).&     03  new_name            pic x(30).'     03                      pic x(450).M *t 01  node_list.     03  item_string.:         05                  pic 9(4)    comp    value   8.@         05                  pic 9(4)    comp    value   external lnm$_string.A         05                              pointer value   referencee
 node_name.-         05                  pic 9(9)    comp.s-     03                      pic 9(9)    comp.y% 01  node_name               pic x(8).e- 01  node_name_len           pic 9(4)    comp.o *7 procedure division.  kick_off section.i 00.?     perform child_init.a       perform get_cmd.+     perform dist_trans until cmd_eof = "Y".y  &     call "disconnect_db" using sqlcode%     if rdb$lu_status not = ss$_normalh2         call "sys$putmsg" using rdb$message_vector1         call "lib$stop" using by value ss$_abort.c *i fini.e
     stop run.o *  child_init section.- 00.- *+ * Shake hands. *-     call "sys$assign" )         using   by descriptor   "sys$net"a(                 by reference    net_chan'                 by value        0, 0, 0o         giving  sys_status.21     if sys_status not = ss$_normal and ss$_remotet2         call "lib$stop" using by value sys_status. *+; * Find out which node invoked us. We need the node name for,8 * the $start_branch as we are an Authorized Participant.7 * This allows the child task to be called from multiplep * parent nodes transparently.  *-     call "sys$trnlnm"r0         using   by reference    lnm$m_case_blind>                 by descriptor   "LNM$PROCESS_TABLE", "SYS$NET"6                 by reference    psl$c_super, node_list         giving  sys_status.p4     if sys_status not = ss$_normal and ss$_bufferovf2         call "lib$stop" using by value sys_status.  B     inspect node_name tallying node_name_len for characters before
 initial "::".o  :     call "declare_connection_name" using sqlcode, db_spec.%     if rdb$lu_status not = ss$_normalo3          call "sys$putmsg" using rdb$message_vectore2          call "lib$stop" using by value ss$_abort. *k dist_trans section.v 00.      move ls_tid to tid.d       call "sys$start_branchw"$         using   by value        0, 0(                 by reference    dtm_iosb$                 by value        0, 0&                 by reference    ls_tid;                 by descriptor   node_name (1:node_name_len)e&                 by reference    ls_bid         giving  sys_status..B     if sys_status = ss$_normal move dtm_iosb_status to sys_status.A     if sys_status not = ss$_normal call "lib$stop" using by value, sys_status.r *w     perform the_update.4 *4 *+/ * Send back a zero length record to signal EOF..: * DECnet PhaseIV didn't require an address PhaseV does :-( *-     call "sys$qiow"i:         using   by value        0, net_chan, io$_writevblk$                 by reference    iosb#                 by value        0,02+                 by reference    letter_sentg)                 by value        0,0,0,0,0p         giving  sys_status.r;     if sys_status = ss$_normal move cond_val to sys_status.nA     if sys_status not = ss$_normal call "lib$stop" using by value  sys_status.'       if commit_flag = "Y"          perform commit_trans      else perform abort_trans.h *e fini.n     perform get_cmd. *a get_cmd section. 00.t     call "sys$qiow"s9         using   by value        0, net_chan, io$_readvblkb$                 by reference    iosb#                 by value        0,0y+                 by reference    letter_sents+                 by value        512,0,0,0,0e         giving  sys_status.i;     if sys_status = ss$_normal move cond_val to sys_status.n3     if sys_status = ss$_linkdiscon or ss$_linkabortd         move "Y" to cmd_eofu     else&         if sys_status not = ss$_normal6             call "lib$stop" using by value sys_status. *  the_update section.  00.h/     call "set_trans_rw" using sqlcode, sql_ctx. %     if rdb$lu_status not = ss$_normaln2         call "sys$putmsg" using rdb$message_vector1         call "lib$stop" using by value ss$_abort.n  8     call "update_dept" using sqlcode, new_name, sql_ctx.%     if rdb$lu_status not = ss$_normali         move "N" to commit_flag"         call "sys$putmsg" :                 using   by reference    rdb$message_vector2                         by value        out_msg, 00                         by reference    net_chan"                 giving  sys_status&         if sys_status not = ss$_normal5             call "lib$stop" using by value sys_statuse         end-if     else          move "Y" to commit_flag. *  fini.t *o commit_trans section.p 00.l     display "Ending branch".     call "sys$end_branchw"$         using   by value        0, 0(                 by reference    dtm_iosb$                 by value        0, 0.                 by reference    ls_tid, ls_bid         giving  sys_status. A     if sys_status not = ss$_normal call "lib$stop" using by valuet sys_status.u *a"     if dtm_iosb_status = ss$_abort          call "sys$getmsg"3                 using   by value        reason_code /                         by reference    msg_len /                         by descriptor   msg_bufh+                         by value        0,0v!                 giving sys_statusm'          if sys_status not = ss$_normal 7               call "lib$stop" using by value sys_status           end-if 9          display "Couldn't commit - " msg_buf (1:msg_len) ,     else if dtm_iosb_status not = ss$_normal=               call "lib$stop" using by value dtm_iosb_status.k *  abort_trans section. 00.t#     display "Aborting transaction".r       call "sys$abort_transw"h0         using   by value        0, ddtm$m_nowait(                 by reference    dtm_iosb$                 by value        0, 0&                 by reference    ls_tid/                 by value        ddtm$_integrityb&                 by reference    ls_bid         giving  sys_status.sB     if sys_status = ss$_normal move dtm_iosb_status to sys_status.A     if sys_status not = ss$_normal call "lib$stop" using by value  sys_status.- *- end program child_rtn. $       return  H $!---------------------------------------------------------------------- -------- $Create_out_msg_cob: $       create out_msg.cob identification division. program-id.    out_msg.  data division. working-storage section. *f@ 01  io$_writevblk           pic 9(9)    comp    value   external io$_writevblk.@ 01  ss$_abort               pic 9(9)    comp    value   external
 ss$_abort.@ 01  ss$_normal              pic 9(9)    comp    value   external ss$_normal.t- 01  sys_status              pic 9(9)    comp.e *d	 01  iosb.d-     03  cond_val            pic 9(4)    comp.e-     03  byte_count          pic 9(4)    comp.t-     03  chan_info           pic 9(9)    comp.u *s linkage section.
 01  msg_desc.o-     03  msg_len             pic 9(4)    comp.s-     03  msg_class           pic 9(4)    comp.i0     03  msg_addr                        pointer.- 01  net_chan                pic 9(4)    comp.l *  procedure division         using   msg_desc,_                 net_chan         giving  ss$_abort. 00.T *+F * This $qiow call could just as easily be a call to T3$SEND that wouldE * transfer all of those lovely descriptive Rdb error messages back toe * your VB message window.  *-     call "sys$qiow"u:         using   by value        0, net_chan, io$_writevblk$                 by reference    iosb<                 by value        0,0,msg_addr,msg_len,0,0,0,0         giving  sys_status.-;     if sys_status = ss$_normal move cond_val to sys_status.sA     if sys_status not = ss$_normal call "lib$stop" using by valuee sys_status.        exit program.i   end program out_msg. $       return  H $!---------------------------------------------------------------------- -------- $Dist_sql_sqlmod:0 $       create Dist_sql.sqlmod module    dist_sql language  cobola parameter colons  ( declare pers alias filename mf_personnel  ! procedure declare_connection_namei         sqlcode,!         :db_spec        char(50);h           attach :db_spec;   procedure set_trans_rw         sqlcode;  "         set transaction read write<                 reserving pers.departments for shared write;   procedure update_dept          sqlcode          :dept_name    char(30);F           update pers.departmentst0                 set department_name = :dept_name.                 where department_code= 'PHRN';   procedure disconnect_db          sqlcode;           disconnect default;I $       return    & Sent via Deja.com http://www.deja.com/ Before you buy.    ------------------------------   Date: 29 Nov 2000 14:07:54 GMT7 From: Thomas.Hahnemann@nospam_s-t.de (Thomas Hahnemann) 0 Subject: Re: generate a shareble image with DECC0 Message-ID: <Oozvf8elmJpy-pn2-t5iIz68oW1dd@Tom2>   Thank you for your reply.V0 I've got some sources they are running on Alpha.0 For some testing purposes it would be helpful to7 get them running on a VAX. On Alpha there is no problemY. to link and install a shared image and use the2 global variables from outside ( i.e. switch output7 for some prozesses on and off by setting a global var).o> On VAX all my code is well translated and also linking is done, with the exception of the global variables. < Unfortunatly the "Cookbook" contains not enough VAX details.   Thomas Hahnemann  B On Wed, 29 Nov 2000 00:41:31, hoffman@xdelta.zko.dec.nospam (Hoff  Hoffman) wrote:e   > l > In article <Oozvf8elmJpy-pn2-1zUxCVNcTXwh@Tom2>, Thomas.Hahnemann@nospam_s-t.de (Thomas Hahnemann) writes: >  >   In no particular order...  > / >   I'd use a symbol prefix on all externals...  > A >   Based on your earlier posting, you wanted a shareable image.  C >   What are you up to with the constants (or values)?  Can you use B >   standard calls and pointers to retrieve the information?  Some& >   details on the particular problem? > D >   Based shareable images can be evil, and should be used only with6 >   great care and full knowledge of the implications. > F >   If this was for OpenVMS Alpha (it is not, per earlier discussion),4 >   I'd suggest a look at Ask The Wizard topic 2486. > P >  --------------------------- pure personal opinion ---------------------------N >    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com >    ------------------------------   Date: 29 Nov 2000 17:43:23 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)0 Subject: Re: generate a shareble image with DECC6 Message-ID: <903f7r$t6u$2@mailint03.im.hou.compaq.com>  j In article <Oozvf8elmJpy-pn2-t5iIz68oW1dd@Tom2>, Thomas.Hahnemann@nospam_s-t.de (Thomas Hahnemann) writes:? :On VAX all my code is well translated and also linking is done-- :with the exception of the global variables.  = :Unfortunatly the "Cookbook" contains not enough VAX details.g ..C :On Wed, 29 Nov 2000 00:41:31, hoffman@xdelta.zko.dec.nospam (Hoff ! :Hoffman) wrote:B :>   Based on your earlier posting, you wanted a shareable image. D :>   What are you up to with the constants (or values)?  Can you useC :>   standard calls and pointers to retrieve the information?  Some ' :>   details on the particular problem?T       I'll rephrase:  I   Do you have a concise source code example of what you are trying to do?=  J   I've pointed to source code examples, now you get to point to yours. :-)    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  + Date: Wed, 29 Nov 2000 08:17:07 +0100 (MET)R& From: Rudolf Wingert <win@fom.fgan.de>) Subject: Re: Gigabit Ethernet and Alpha'sa6 Message-ID: <200011290713.IAA14688@sinet1.fom.fgan.de>   Hello,   Paul A. Jacobi wrotes:   >>>cI There is no support in the Alpha SRM console for booting from the Gigabit!H ethernet adapters.  The Gigabit adapter requires a large chuck of deviceE microcode.  Since there is no ROM on the Gigabit PCI card, the devicetK microcode would have to be stored in the same flash ROM as the SRM console,!K then downline loaded to the card.  Most Alpha systems have a limited amountaK of flash ROM space, which is insufficient for both a new console driver andM for device microcode.D  J A future version of the Gigabit ethernet card is supposed include it's ownF flash ROM on the PCI card for the device microcode.  When this becomesI available, boot support will probably be added to the SRM console for theTK future generation product.  I don't have any estimate of when such a future @ generation Gigabit card will become available with boot support. <<<D  G AFAIK is the Gigabit Ethernet adpater starting with OpenVMS 7.2-1 fullyQE supported as a cluster interconnect. Means your answer, that you mustoG have any other normal Ethernet (10/100) to boot? So you have to buy twoAG network interfaces for a proper clusterconfiguration? If it is so, bad,PF very bad. Also what's happen if the new one bootable interface is out?G Will the old one be change by service? Gigabit Ethernet is state of theBE art and 10Gbit ist in sight. Compaq Alphas should support the network $ adapter wether OS we will run on it.E But I can see only a technology stop of the Alpha product line, or do,# you her some whispering on the aer?    Regards Rudolf Wingert   ------------------------------    Date: 29 Nov 2000 15:39:49 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>a) Subject: Re: Gigabit Ethernet and Alpha's!H Message-ID: <y4d7fe6dvu.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  M I would say there is a big difference between supporting a device when the OSrN is running (Gb Ethernet and VMS in this case) and actually loading the OS overK that device (without accessing any - possibly non-existant - local disks).  J It's the latter that isn't currently supported, and for very good reasons.   	Jan   ------------------------------  % Date: Wed, 29 Nov 2000 17:34:26 +0530), From: Shiv <Sivakumar.Thangavelu@oracle.com>L Subject: Help required - Configure NFS server (VMS) and NFS Client (Unix/NT)* Message-ID: <3A24F0CA.F9CBE81B@oracle.com>  , This is a multi-part message in MIME format.& --------------C7A595682ABA76B8060994D9* Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit    Hi,   L I'm trying to access a VMS Disk on my NT machine thru NFS, but i'm unable to succeed in it.  + NFS Server : VMS machine running OS ver 7.2 : Client    : WinNT M/c running Hummingbird NFS Maestro Solo             : Sun Solaris 2.6(  " I did the following on the VMS box  7 ADD MAP "<unix_Style_dir_name>" <VMS_disk_or_directory>a* ADD export "<unix_Style_dir_name>" /host=*> ADD proxy <vms_username> /uid=<vms_uid> /gid=<vms_gid> /host=* /nfs=(outgoing,incoming)    M The exported dir/disk is visible in the Windows explorer of PC (NT m/c). WheniL i try to access it, i get an authentication error. The same applies with the. Solaris client too, when i try to nfs_mount it  L I'll appreciate any inputs/suggestions/documents regarding this. (pointer toA step by step action for an NFS server setup on VMS will be great)   & Get back to me if further info's reqd.     TIA, Shiv    & --------------C7A595682ABA76B8060994D9- Content-Type: text/x-vcard; charset=us-ascii;)   name="Sivakumar.Thangavelu.vcf" Content-Transfer-Encoding: 7bitr" Content-Description: Card for Shiv  Content-Disposition: attachment;$  filename="Sivakumar.Thangavelu.vcf"   begin:vcard  n:Thangavelu;Sivakumar x-mozilla-html:FALSE version:2.1t. email;internet:Sivakumar.Thangavelu@oracle.com title;quoted-printable:<html><body>&nbsp=3B<table BORDER COLS=3D2 WIDTH=3D"100%" ><tr><td><center><!--<b><font color=3D"#000099">Software Specialist</font></b>--><p><a href=3D"http://www.digitalindiasw.com"><img SRC=3D"http://www.egroups.com/files/ALU/Digital-small-logo.gif" height=3D46 width=3D143></a><br><font size=3D-1>Digital Equipment (I) Limited</font></center></td><td><center><b><font color=3D"#000099"><font size=3D+1>Consultant</font></font></b><br><a href=3D"http://www.oracle.com"><img SRC=3D"http: adr;quoted-printable:;;<html>=0D=0A<body>=0D=0A&nbsp=3B=0D=0A<table BORDER COLS=3D2 WIDTH=3D"103%" >=0D=0A<tr>=0D=0A<td>=0D=0A<center>93-A Industrial Suburb=0D=0A<br>Yeshwanthpur II Stage=0D=0A<br>Bangalore - 560 022=0D=0A<p>+91 (80) 3374785=0D=0A<br><a href=3D"mailto:t.sivakumar@digital.com">t.sivakumar@digital.com</a></center>=0D=0A</td>=0D=0A=0D=0A<td>=0D=0A<center>#3, Bannerghatta Road=0D=0A<br>(Near Dairy Circle)=0D=0A<br>Bangalore - 560 029=0D=0A<p>+91 (80) 5528335 x 1056=0D=0A<br><a href=3D"mailto:Si fn:Sivakumar, T 	 end:vcard   ( --------------C7A595682ABA76B8060994D9--   ------------------------------  # Date: Wed, 29 Nov 2000 10:40:54 GMT(/ From: Mike Price <mike.price@littlewoods.co.uk>p& Subject: How fast is a SCSI connection) Message-ID: <902mfl$jbg$1@nnrp1.deja.com>   G Does anyone know the authoratative (sorry about the spelling) source onaD how fast a SCSI bus can run. Obviously the type of disk etc. matters/ but I just need to know the actual SCSI limits.eD If it is dependant on the machine I am mostly interested in an AlphaB 4100 so the connection is via the PCI bus the disks in question onF Ultra, wide and narrow busses (different Alphas)and some are on cachedF Raid system while some are directly connected to a Storage Works shelf  > Also I suppose the limits of the PCI bus might be nice to know   Thanks in advancem   Mike   --B All opinions are my own and do not necessarily reflect those of my employer    & Sent via Deja.com http://www.deja.com/ Before you buy.    ------------------------------   Date: 29 Nov 2000 05:17 CSTl' From: carl@gerg.tamu.edu (Carl Perkins)a* Subject: Re: How fast is a SCSI connection- Message-ID: <29NOV200005170223@gerg.tamu.edu>   3 Mike Price <mike.price@littlewoods.co.uk> writes...EH }Does anyone know the authoratative (sorry about the spelling) source onE }how fast a SCSI bus can run. Obviously the type of disk etc. matters 0 }but I just need to know the actual SCSI limits.E }If it is dependant on the machine I am mostly interested in an AlphanC }4100 so the connection is via the PCI bus the disks in question onsG }Ultra, wide and narrow busses (different Alphas)and some are on cacheduG }Raid system while some are directly connected to a Storage Works shelfr } ? }Also I suppose the limits of the PCI bus might be nice to knowa }  }Thanks in advance }  }Miken  9 As I recall, the theoretical maximum throughput would be:_$ "slow" narrow   = 5 Megabytes/Second fast narrow     = 10 MB/Sf fast wide       = 20 MB/S  ultrafast wide  = 40 MB/Sn ultra2 wide     = 80 MB/S  ultra3 wide     = 160 MB/S  G (As far as I know, nobody ever actually made any of the ultra or fastertH versions in the narrow configuration, just wide - the theoretical narrow) versions of these would be half as fast.)   D Only the last one is faster than 32 bit PCI (which has a theoreticalB throughputof 132 MB/S), but it is still slower then 64 bit PCI (at 264 or so MB/S theoretical).   --- Carl   ------------------------------  % Date: Wed, 29 Nov 2000 07:07:32 -0500 : From: "Koska, John C. (LNG-MBC)" <John.C.Koska@bender.com>8 Subject: HSZTERM... status of support of it by Compaq...K Message-ID: <3D35AD137AAAD411A6BA0008C7B1B12D05BB35@MBCALBEXC03.BENDER.COM>l  8 Anyone know the status of HSZTERM?  Last I heard it was 9 a non-supported piece of software, but often times used;  4 and used quite successfully with HSZ-40s to HSG-80s.   :) jck   ------------------------------  # Date: Wed, 29 Nov 2000 14:04:43 GMTo/ From: Mike Price <mike.price@littlewoods.co.uk>y< Subject: Re: HSZTERM... status of support of it by Compaq...) Message-ID: <9032dm$s69$1@nnrp1.deja.com>   
 In article@ <3D35AD137AAAD411A6BA0008C7B1B12D05BB35@MBCALBEXC03.BENDER.COM>,=   "Koska, John C. (LNG-MBC)" <John.C.Koska@bender.com> wrote:s9 > Anyone know the status of HSZTERM?  Last I heard it wass: > a non-supported piece of software, but often times used;6 > and used quite successfully with HSZ-40s to HSG-80s. >o > :) jck >  > F I logged a problem with the UK CSC early this year which turned out toG be caused by trying to use SWCC and HSZTERM at the same time (causes ansB HSZ70 to hang!!!!!). The response was that I needed to stop eitherE HSZTERM (i.e. don't do a set ho/scsi anymore) or stop SWCC. They veryaG strongly stated that the ONLY supported way to manage or connect to the$& HSZ70 was via SWCC or the serial port. i.e. HSZTERM is unsupported.C however there was a strong implication that they actually liked the 6 command line interface provided by hszterm themselves.- i.e. it works fine but don't mix it with SWCC / I assume the same applies to other raid systemso   Mike     --B All opinions are my own and do not necessarily reflect those of my employer    & Sent via Deja.com http://www.deja.com/ Before you buy.t   ------------------------------  # Date: Wed, 29 Nov 2000 15:24:18 GMTnD From: "Kris G.G. Clippeleyr (aka Qkcl)" <kris.clippeleyr@pandora.be>< Subject: Re: HSZTERM... status of support of it by Compaq...< Message-ID: <Cc9V5.53973$zF6.893875@afrodite.telenet-ops.be>   Hi,   5 Indeed HSZTERM is not officially supported by Compaq.b> And watch out if you install it on a V7.2 machine (as we did).? It messes up the SET command (at least our kit did HSZTERM010). 2 Otherwise it works just fine on our HSZ50 & HSZ70.  
 Greetings,   -- Kris G.G. Clippeleyr (aka Qkcl) & VMS Wizard in training, Guru wanna be.  : Mike Price <mike.price@littlewoods.co.uk> wrote in message# news:9032dm$s69$1@nnrp1.deja.com...  > In articleB > <3D35AD137AAAD411A6BA0008C7B1B12D05BB35@MBCALBEXC03.BENDER.COM>,? >   "Koska, John C. (LNG-MBC)" <John.C.Koska@bender.com> wrote:o; > > Anyone know the status of HSZTERM?  Last I heard it wasn< > > a non-supported piece of software, but often times used;8 > > and used quite successfully with HSZ-40s to HSG-80s. > >n
 > > :) jck > >n > >yH > I logged a problem with the UK CSC early this year which turned out toI > be caused by trying to use SWCC and HSZTERM at the same time (causes andD > HSZ70 to hang!!!!!). The response was that I needed to stop eitherG > HSZTERM (i.e. don't do a set ho/scsi anymore) or stop SWCC. They verysI > strongly stated that the ONLY supported way to manage or connect to theo( > HSZ70 was via SWCC or the serial port. > i.e. HSZTERM is unsupported.E > however there was a strong implication that they actually liked the-8 > command line interface provided by hszterm themselves./ > i.e. it works fine but don't mix it with SWCCo1 > I assume the same applies to other raid systemsn >t > Mike >o >  > --D > All opinions are my own and do not necessarily reflect those of my
 > employer >s >b( > Sent via Deja.com http://www.deja.com/ > Before you buy.b   ------------------------------    Date: 29 Nov 2000 08:23:13 -0500, From: koehler@eisner.decus.org (Bob Koehler)# Subject: Re: MMS clean up utilities + Message-ID: <NPKLQUMtotDr@eisner.decus.org>   U In article <3A23C56B.184B65F1@fmr.com>, Brad Hamilton <Brad.Hamilton@fmr.com> writes:p > Folks, > E > One of our customers wants to know if there are CMS/MMS "utilities"pG > (freeware or otherwise) to assist in "cleaning up" libraries.  He has C > asked the CSC for help already; if anyone knows of any additionalo& > sources of help, please let me know. >   F Well, there's always backup/delete.  Seriously, what kind of "cleaningG up" does he want to do.  I do lots of "cleaning up" in my CMS librariesdG just using CMS commands, but I've no idea if this is what your customer  means.  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporationo= NASA GSFC Flight Software       | Federal Sector, Civil Group0E                                 | please remove ".aspm" when replyingd   ------------------------------  % Date: Wed, 29 Nov 2000 09:08:11 +0100i% From: "Fred Zwarts" <F.Zwarts@KVI.nl>p' Subject: Re: More VMS, BSD, and FORTRAN . Message-ID: <902dha$mgs$1@info.service.rug.nl>  A "Craig D. Lansing (root)" <lansing@beldar.com> wrote in message = $ news:3A2478C6.3D245D22@beldar.com...H > For those of you who may remember, the saga continues.  My knowledge = ofF > VMS is still nil, so I'm looking for a little help here.  As you may > recall.... >=20G > "Please excuse my VMS ignorance (i.e. I've never used it at all), but  > I'm;G > trying to figure out a vendor's position.  As a little background, wepD > have several applications that communicate with each other via BSDD > sockets on an HP-9000 (HP-UX), HP-3000 (MPE), and Windows 9x.  AllI > mesages are ASCII text < .5K in length.  This is all written in C and =e is >=20G > portable across all platforms.  We are looking at replacing some of =c thetH > applications with a product that runs on DEC Alphas written in FORTRAND > (77 I think).  The vendor says that we need to replace the socketsG > communication with serial  (RS-232) connections between the systems =u via:G > a terminal server.  This seems to be a huge step backward.  Is anyone5H > familiar with whether or not sockets will work on a VMS platform?  CanD > this be written in FORTRAN?  I doubt that we can get the vendor toH > change his mind, but I am mostly just curious about whether or not the > capability exists."a >=20A > Well, we've decided to suck it up and write a Serial to SocketspG > interface program (Comm. Server) to run on the existing platforms andlH > keep the existing programs as they are using sockets to communicate toH > the Comm. Server that will in turn handle the serial communications toG > the Alpha via a terminal server.  It dawned on us today that we couldnE > get rid of the hardware failure points if we just wrote the Comm. =U ServerB > to run under VMS since it is using sockets to communicate to theE > non-VMS world.  The question is, is there a way to have the Comm. =a ServerH > running under VMS write to a file, device, etc. that would look like aF > serial connection to the vendor without having to run a wire out oneF > port on the terminal server and back in another port?  Basically, isE > there a way to create a logical / pseudo device that looks like a =s serialF > connection to the vendor, but is actually completely internal to theD > Alpha with no parts / cables to fail or are we more insane than weF > think?  If this seems like a stupid question, keep in mind that we = know# > nothing about VMS and be kind :-)p >=20 > TIA, > Craig- =20-@ VMS has pseudo terminals. See the Guide to I/O User's Reference,D which describes how to create them and how to communicate with them.   ------------------------------  # Date: Wed, 29 Nov 2000 12:09:00 GMTH From: richard_maher@my-deja.comt' Subject: Re: More VMS, BSD, and FORTRAN ) Message-ID: <902rkp$ms1$1@nnrp1.deja.com>    Hi,t  < I don't know if mailboxes solve the problem but I like them.  " >They require fixed-length records  B This is not strictly true with the stream i/o modifiers since 5.5?  3 >Once a message is read from a mailbox, it's gone..l  F A slight variation is that I believe that you can peek at mbx contents without dequeing it.  G > The potential problem comes up when a reader ceases operation.  SincepD > the mailbox is of a fixed size, it can fill up, at which point any@ > process trying to write to it is blocked until some task reads > messages from the mailbox.  G Uni-directional i/o channels allows a writer to detect when a reader is:B no longer there and abort the write. (as well as no resource wait)   Regards Richard Mahert    & Sent via Deja.com http://www.deja.com/ Before you buy.)   ------------------------------  % Date: Wed, 29 Nov 2000 07:35:50 -0500-, From: Howard S Shubs <hshubs@mindspring.com>' Subject: Re: More VMS, BSD, and FORTRAN > Message-ID: <hshubs-50BC3B.07355029112000@news.mindspring.com>  J In article <902rkp$ms1$1@nnrp1.deja.com>, richard_maher@my-deja.com wrote:  C >This is not strictly true with the stream i/o modifiers since 5.5?g  2 Dunno.  I missed 5.5 through 6.0, 6.2 through 7.0.     G >A slight variation is that I believe that you can peek at mbx contentsn >without dequeing it.i  P Last I looked (quite a while ago), this could be done at a cost of extra code.  ! But as I said, it's been a while.r    H >Uni-directional i/o channels allows a writer to detect when a reader isC >no longer there and abort the write. (as well as no resource wait)n   Good.a  O But what the others have been posting makes sense.  A pseudo terminal would be mM a cool thing.  I'm going to have to look into all of this when I get to work   today. -- s Howard S ShubsD "Run in circles, scream and shout!"  "I hope you have good backups!"   ------------------------------  % Date: Wed, 29 Nov 2000 09:19:01 -0500 * From: Chuck Chopp <ChuckChopp@rtfmcsi.com>' Subject: Re: More VMS, BSD, and FORTRANi+ Message-ID: <3A251055.20706658@rtfmcsi.com>    Fred Zwarts wrote:  f > "Craig D. Lansing (root)" <lansing@beldar.com> wrote in message news:3A2478C6.3D245D22@beldar.com...K > > For those of you who may remember, the saga continues.  My knowledge ofaH > > VMS is still nil, so I'm looking for a little help here.  As you may > > recall.... > >bI > > "Please excuse my VMS ignorance (i.e. I've never used it at all), buto > > I'm I > > trying to figure out a vendor's position.  As a little background, we F > > have several applications that communicate with each other via BSDF > > sockets on an HP-9000 (HP-UX), HP-3000 (MPE), and Windows 9x.  AllL > > mesages are ASCII text < .5K in length.  This is all written in C and is > >-K > > portable across all platforms.  We are looking at replacing some of theoJ > > applications with a product that runs on DEC Alphas written in FORTRANF > > (77 I think).  The vendor says that we need to replace the socketsK > > communication with serial  (RS-232) connections between the systems viaeI > > a terminal server.  This seems to be a huge step backward.  Is anyone J > > familiar with whether or not sockets will work on a VMS platform?  CanF > > this be written in FORTRAN?  I doubt that we can get the vendor toJ > > change his mind, but I am mostly just curious about whether or not the > > capability exists."2 > > C > > Well, we've decided to suck it up and write a Serial to SocketsuI > > interface program (Comm. Server) to run on the existing platforms and J > > keep the existing programs as they are using sockets to communicate toJ > > the Comm. Server that will in turn handle the serial communications toI > > the Alpha via a terminal server.  It dawned on us today that we couldiL > > get rid of the hardware failure points if we just wrote the Comm. ServerD > > to run under VMS since it is using sockets to communicate to theL > > non-VMS world.  The question is, is there a way to have the Comm. ServerJ > > running under VMS write to a file, device, etc. that would look like aH > > serial connection to the vendor without having to run a wire out oneH > > port on the terminal server and back in another port?  Basically, isL > > there a way to create a logical / pseudo device that looks like a serialH > > connection to the vendor, but is actually completely internal to theF > > Alpha with no parts / cables to fail or are we more insane than weK > > think?  If this seems like a stupid question, keep in mind that we knowp% > > nothing about VMS and be kind :-)  > >S > > TIA,	 > > Craige >aB > VMS has pseudo terminals. See the Guide to I/O User's Reference,F > which describes how to create them and how to communicate with them.  c Yes, the pseudo terminal programming interface would work OK for this application.  The comm serverv` program would have to create and maintain a number of pseudo terminal devices that the 3rd partyh application could use as if they were real serial ports.  Then the comm server application would need tob manage the TCP socket communications between each pseudo terminal device and some remote TCP port.  e It is possible to do TCP/IP socket communications in Fortran 77 on OpenVMS.  The UCX $QIO programming e interface is usable and it works just fine.  Both TCPWare and Multinet support the $QIO interface fori[ socket programming and I've written Fortran 77 code that works fine with all 3 major TCP/IP9 implementations on OpenVMS.<  h If I knew who this vendor was that is saying that they cannot do socket communications via Fortran 77 one OpenVMS I'd be willing to work with them on a contract basis to help them enhance their product to doA9 socket communications.  Its not too difficult to do this.B     -- Chuck Chopp=  8 ChuckChopp@rtfmcsi.com            http://www.rtfmcsi.com0                                   ICQ # 22321532@ RTFM Consulting Services Inc.     864 801 2795 voice & voicemail2 103 Autumn Hill Road              864 801 2774 fax4 Greer, SC  29651                  800 774 0718 pager7                                   8007740718@skytel.com-   ------------------------------  % Date: Wed, 29 Nov 2000 08:56:09 -06002) From: "John E. Malmberg" <wb8tyw@qsl.net><' Subject: Re: More VMS, BSD, and FORTRANH/ Message-ID: <t2a64uirvmqc87@corp.supernews.com>$  C "Craig D. Lansing (root)" <lansing@beldar.company> wrote in messageu$ news:3A2478C6.3D245D22@beldar.com...I > For those of you who may remember, the saga continues.  My knowledge of F > VMS is still nil, so I'm looking for a little help here.  As you may > recall.... >dG > "Please excuse my VMS ignorance (i.e. I've never used it at all), butt > I'm G > trying to figure out a vendor's position.  As a little background, weaD > have several applications that communicate with each other via BSDD > sockets on an HP-9000 (HP-UX), HP-3000 (MPE), and Windows 9x.  AllJ > mesages are ASCII text < .5K in length.  This is all written in C and is >rI > portable across all platforms.  We are looking at replacing some of theeH > applications with a product that runs on DEC Alphas written in FORTRAND > (77 I think).  The vendor says that we need to replace the socketsI > communication with serial  (RS-232) connections between the systems viaiG > a terminal server.  This seems to be a huge step backward.  Is anyone H > familiar with whether or not sockets will work on a VMS platform?  CanD > this be written in FORTRAN?  I doubt that we can get the vendor toH > change his mind, but I am mostly just curious about whether or not the > capability exists."h  H The "sockets" api is implemented on the VMS platform, as was reported toL your earlier post.  It is also available from FORTRAN by various programmingI methods.  IIRC: The Compaq C manual has a section on how to call Compaq CsF routines including the socket routines from other languages.  There is5 probably a section in the Compaq Fortran manual also.l  @ http://www.openvms.compaq.com:8000/72final/6529/6529profile.html  H To be more precise the "sockets" api is just an interface that a programH uses to send a byte stream over either a TCP or UDP protocol connection.  H It is not important to the TCP/IP connection if either side is using the "sockets" API.  J What matters is what TCP or UDP port you are using, and the actual details of the data that is being sent.n  A > Well, we've decided to suck it up and write a Serial to Sockets2G > interface program (Comm. Server) to run on the existing platforms andtH > keep the existing programs as they are using sockets to communicate toH > the Comm. Server that will in turn handle the serial communications toG > the Alpha via a terminal server.  It dawned on us today that we could-J > get rid of the hardware failure points if we just wrote the Comm. ServerB > to run under VMS since it is using sockets to communicate to the > non-VMS world.  B It is using either TCP or UDP protocol and a specific port number.K "Sockets" are not really relevant for this.  They are just one of many ways0. of sending or receiving data on a TCP/IP port.  : > The question is, is there a way to have the Comm. ServerH > running under VMS write to a file, device, etc. that would look like aF > serial connection to the vendor without having to run a wire out oneF > port on the terminal server and back in another port?  Basically, isJ > there a way to create a logical / pseudo device that looks like a serialF > connection to the vendor, but is actually completely internal to theD > Alpha with no parts / cables to fail or are we more insane than weI > think?  If this seems like a stupid question, keep in mind that we known# > nothing about VMS and be kind :-)s  I As was suggested by another poster, it may be sufficient to just create a I reverse TELNET port that makes an existing TCP/IP port look like a seriala device.t  + http://www.openvms.compaq.com:8000/72final/h/   6526/6526profile_019.html#tel_create_sess_secH  . [URL is wrapped on two lines for printability]  . This may or may not work for your application.   -Johnt wb8tyw@qsl.network   ------------------------------  % Date: Wed, 29 Nov 2000 12:12:01 +0100e' From: Theo Jakobus <jakobus@iaf.fhg.de>  Subject: Re: Mozilla M18 i005e* Message-ID: <3A24F291.735A1712@iaf.fhg.de>   Alain Chappuis wrote:t >  > Hello,, > Is there anybody work with M18 of Mozilla? >   E I'm using MOZILLA every day. I have JAVA in "Preferences disabled". Ii> get JAVA Script errors so there is still some work to be done.C MOZILLA is installed on the system disk. It requests read and write A access for the user to the SYS$COMMON:[MOZILLA] directory. SystemH managers don't like this.p     Regards, --    ; *********************************************************** ; *                                                         *); *  Theo Jakobus                                           *e; *  Fraunhofer-Institut fuer Angewandte Festkoerperphysik  * ; *  Tullastr. 72                                           *s; *  D-79108 Freiburg                                       * ; *  Germany                                                *A; *  Phone:   +49-(0)761-5159-325                           *K; *  FAX :    +49-(0)761-5159-200                           *R; *  e-mail:  jakobus@iaf.fhg.de                            *r; *  http://www.iaf.fhg.de                                  *h; *                                                         *d; ***********************************************************o   ------------------------------  # Date: Wed, 29 Nov 2000 11:28:57 GMTa' From: Colin Blake <colin@theblakes.com>a Subject: Re: Mozilla M18 i005 - Message-ID: <3A24E7E4.9D523D15@theblakes.com>    Theo Jakobus wrote:s  E > MOZILLA is installed on the system disk. It requests read and writehC > access for the user to the SYS$COMMON:[MOZILLA] directory. Systemp > managers don't like this.   E The "user needs write access to the bin area" problem is being workedeI (there are several bugs on this). But remember, installing Mozilla to thehI system disk is just the DEFAULT installation location. If this makes your G system managers squeamish just install to a different location (PRODUCTo INSTALL ./DEST=whatever).    Colin.   ------------------------------  # Date: Wed, 29 Nov 2000 15:11:30 GMT-= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)n Subject: Re: Mozilla M18 i005j0 Message-ID: <009F3D75.3518AE85@SendSpamHere.ORG>  T In article <3A24F291.735A1712@iaf.fhg.de>, Theo Jakobus <jakobus@iaf.fhg.de> writes: >Alain Chappuis wrote: >> p	 >> Hello, - >> Is there anybody work with M18 of Mozilla?r >>   >/F >I'm using MOZILLA every day. I have JAVA in "Preferences disabled". I? >get JAVA Script errors so there is still some work to be done.nD >MOZILLA is installed on the system disk. It requests read and writeB >access for the user to the SYS$COMMON:[MOZILLA] directory. System >managers don't like this.  J How do you start Mozilla?  Everytime I execute MOZILLA.COM, the DECWindowsF server dies and I need to log in via LAT, etc. and restart the server.     --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM             nO city, n., 1. a place where trees are cut down and streets are named after them.    ------------------------------  % Date: Wed, 29 Nov 2000 16:53:16 +0100-= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>  Subject: Re: Mozilla M18 i005Z) Message-ID: <3A25266C.BA912A11@gtech.com>    Theo Jakobus wrote:DG > I'm using MOZILLA every day. I have JAVA in "Preferences disabled". Ig@ > get JAVA Script errors so there is still some work to be done.  ? You are aware of that Java nd JavaScript have nothing in commonk* except the first 4 letters in their name ?   Arne   ------------------------------  % Date: Wed, 29 Nov 2000 18:37:29 +0100 ' From: Theo Jakobus <jakobus@iaf.fhg.de>  Subject: Re: Mozilla M18 i005 * Message-ID: <3A254CE9.7940D0EE@iaf.fhg.de>  $ Brian Schenkenberger, VAXman- wrote: > V > In article <3A24F291.735A1712@iaf.fhg.de>, Theo Jakobus <jakobus@iaf.fhg.de> writes: > >Alain Chappuis wrote: > >> > >> Hello, / > >> Is there anybody work with M18 of Mozilla?n > >> > >=H > >I'm using MOZILLA every day. I have JAVA in "Preferences disabled". IA > >get JAVA Script errors so there is still some work to be done.tF > >MOZILLA is installed on the system disk. It requests read and writeD > >access for the user to the SYS$COMMON:[MOZILLA] directory. System > >managers don't like this. > L > How do you start Mozilla?  Everytime I execute MOZILLA.COM, the DECWindowsH > server dies and I need to log in via LAT, etc. and restart the server. >   < First you should analyze SYS$MANAGER:DECW$SERVER_0_ERROR.LOG   I have defined a symbol MOZILLA  $ SHOW SYMBOL MOZILLAe+   MOZILLA == "@SYS$COMMON:[MOZILLA]MOZILLA"  $i  H The command procedure SYS$COMMON:[MOZILLA]MOZILLA.COM was created during the installation.t     Regards, -- m  ; ***********************************************************t; *                                                         * ; *  Theo Jakobus                                           *e; *  Fraunhofer-Institut fuer Angewandte Festkoerperphysik  *r; *  Tullastr. 72                                           * ; *  D-79108 Freiburg                                       *t; *  Germany                                                * ; *  Phone:   +49-(0)761-5159-325                           *v; *  FAX :    +49-(0)761-5159-200                           *v; *  e-mail:  jakobus@iaf.fhg.de                            *m; *  http://www.iaf.fhg.de                                  * ; *                                                         *s; ***********************************************************e   ------------------------------  % Date: Wed, 29 Nov 2000 06:48:07 -0800n1 From: "Farrell, Michael" <MFarrell@voltdelta.com> J Subject: RE: New Encompass Survey... PLease Take 10 Minutes to Fill It OutC Message-ID: <7DF45F22D904D31192EE00805F578DF2018BC5AD@NY_EXCHANGE1>d  A When I clicke on the URL referenced below, this is what happened:w     HTTP Error 404  
 404 Not Found-  I The Web server cannot find the file or script you asked for. Please check_+ the URL to ensure that the path is correct.D  C Please contact the server's administrator if this problem persists.    Mike Farrell   > -----Original Message-----9 > From:	Terry C. Shannon [SMTP:terryshannon@mediaone.net] * > Sent:	Tuesday, November 28, 2000 5:55 PM > To:	Info-VAX@Mvb.Saic.ComsD > Subject:	New Encompass Survey... PLease Take 10 Minutes to Fill It > Oute > 6 > en.com.pass -  to surround;  to contain, to include. > = > Help us help you!  As you know, we've recently rolled out a J > new identity, complete with restated goals, objectives and plans for theF > future. But, for any change to be successful, we need feedback - and > member@ > feedback is the best, most "encompassing" feedback we can get. > J > Take a moment to give us your views on the NEW! Encompass. Participation > is" > fast and easy.  Just click on toL > http://inq30.chicago.sba.com/Encompass_Calibration_Survey/3.0_version_enco > mpG > ass_survey.html and share your views.  Here's a bonus -- complete the< > surveyI > by December 8 and you'll automatically be entered into a drawing to win- > one0' > of three iPAQ Personal Audio Players.a > > > Your opinions will be kept confidential.  All data collectedJ > will be tallied by a third party provider so please share your thoughts. > 
 > RememberL > http://inq30.chicago.sba.com/Encompass_Calibration_Survey/3.0_version_enco > mpH > ass_survey.html - puts you one click away from an iPAQ  Personal AudioF > Player and sharing your views with Encompass leadership.  Let's workI > together to continue to build the outstanding user organization that isw > Encompass! >  >  > Posted by... >  >  > Terry C. Shannon > Consultant and Publisher > Shannon Knows Compaq" > email: terryshannon@mediaone.net > http://www.acersoft.comi& > Web (info on SKC):  www.acersoft.com >  >    ------------------------------  # Date: Wed, 29 Nov 2000 14:58:49 GMTu1 From: "Mark D. Jilson" <jilly@clarityconnect.com>uJ Subject: Re: New Encompass Survey... PLease Take 10 Minutes to Fill It Out2 Message-ID: <3A251B12.6BC93842@clarityconnect.com>  G I had to cut and paste the full link together and it then redirected me5 to0 http://inq30.chicago.sba.com/cgi/qweb.cgi?7NCXZQ   "Farrell, Michael" wrote:i > C > When I clicke on the URL referenced below, this is what happened:  >  > HTTP Error 404 >  > 404 Not Found  > K > The Web server cannot find the file or script you asked for. Please checke- > the URL to ensure that the path is correct.  > E > Please contact the server's administrator if this problem persists.  >  > Mike Farrell >  > > -----Original Message-----; > > From: Terry C. Shannon [SMTP:terryshannon@mediaone.net]c, > > Sent: Tuesday, November 28, 2000 5:55 PM > > To:   Info-VAX@Mvb.Saic.Com(K > > Subject:      New Encompass Survey... PLease Take 10 Minutes to Fill It  > > Outi > >i8 > > en.com.pass -  to surround;  to contain, to include. > >o? > > Help us help you!  As you know, we've recently rolled out ahL > > new identity, complete with restated goals, objectives and plans for theH > > future. But, for any change to be successful, we need feedback - and
 > > memberB > > feedback is the best, most "encompassing" feedback we can get. > >gL > > Take a moment to give us your views on the NEW! Encompass. Participation > > is$ > > fast and easy.  Just click on toN > > http://inq30.chicago.sba.com/Encompass_Calibration_Survey/3.0_version_enco > > mpI > > ass_survey.html and share your views.  Here's a bonus -- complete the 
 > > surveyK > > by December 8 and you'll automatically be entered into a drawing to win  > > oner) > > of three iPAQ Personal Audio Players.  > >e@ > > Your opinions will be kept confidential.  All data collectedL > > will be tallied by a third party provider so please share your thoughts. > >  > > RememberN > > http://inq30.chicago.sba.com/Encompass_Calibration_Survey/3.0_version_enco > > mpJ > > ass_survey.html - puts you one click away from an iPAQ  Personal AudioH > > Player and sharing your views with Encompass leadership.  Let's workK > > together to continue to build the outstanding user organization that isa > > Encompass! > >h > >u > > Posted by... > >i > >n > > Terry C. Shannon > > Consultant and Publisher > > Shannon Knows Compaq$ > > email: terryshannon@mediaone.net > > http://www.acersoft.com ( > > Web (info on SKC):  www.acersoft.com > >o > >k   -- iD Jilly	- Working from Home in the Chemung River Valley - Lockwood, NY0 	- jilly@clarityconnect.com			- Brett Bodine fan. 	- Mark.Jilson@Compaq.com			- since 1975 or so, 	- http://www.jilly.baka.com               -   ------------------------------  % Date: Wed, 29 Nov 2000 11:56:06 -0500r# From: John Vottero <John@MVPSI.com> J Subject: RE: New Encompass Survey... PLease Take 10 Minutes to Fill It OutD Message-ID: <C15945A9D9EFCF11BA8B08002BBF1CCC0CD8E0@berry.mvpsi.com>  E Make sure that you're using the complete URL.  It's long and probablyg wrapped onto a new line.   > -----Original Message-----8 > From: Farrell, Michael [mailto:MFarrell@voltdelta.com], > Sent: Wednesday, November 29, 2000 9:48 AM > To: Info-VAX@Mvb.Saic.ComnH > Subject: RE: New Encompass Survey... PLease Take 10 Minutes to Fill It > Outa >  > C > When I clicke on the URL referenced below, this is what happened:  >  >  > HTTP Error 404 >  > 404 Not Foundm > ? > The Web server cannot find the file or script you asked for. u > Please check- > the URL to ensure that the path is correct.I > E > Please contact the server's administrator if this problem persists.m >  > Mike Farrell >  > > -----Original Message-----; > > From:	Terry C. Shannon [SMTP:terryshannon@mediaone.net]F, > > Sent:	Tuesday, November 28, 2000 5:55 PM > > To:	Info-VAX@Mvb.Saic.Com < > > Subject:	New Encompass Survey... PLease Take 10 Minutes  > to Fill It > > Out  > > 8 > > en.com.pass -  to surround;  to contain, to include. > > ? > > Help us help you!  As you know, we've recently rolled out a ? > > new identity, complete with restated goals, objectives and v > plans for theuH > > future. But, for any change to be successful, we need feedback - and
 > > memberB > > feedback is the best, most "encompassing" feedback we can get. > > ? > > Take a moment to give us your views on the NEW! Encompass. 0 > Participation  > > is$ > > fast and easy.  Just click on to > > @ > http://inq30.chicago.sba.com/Encompass_Calibration_Survey/3.0_ version_enco > mpG > ass_survey.html and share your views.  Here's a bonus -- complete theD > surveyI > by December 8 and you'll automatically be entered into a drawing to wino > oneo' > of three iPAQ Personal Audio Players.r > > > Your opinions will be kept confidential.  All data collectedJ > will be tallied by a third party provider so please share your thoughts. > 
 > RememberL > http://inq30.chicago.sba.com/Encompass_Calibration_Survey/3.0_version_enco > mpH > ass_survey.html - puts you one click away from an iPAQ  Personal AudioF > Player and sharing your views with Encompass leadership.  Let's workI > together to continue to build the outstanding user organization that isa > Encompass! >  >  > Posted by... >  >  > Terry C. Shannon > Consultant and Publisher > Shannon Knows Compaq" > email: terryshannon@mediaone.net > http://www.acersoft.com & > Web (info on SKC):  www.acersoft.com >  >    ------------------------------  % Date: Wed, 29 Nov 2000 17:59:03 +0100t6 From: "Martin P.J. Zinser" <zinser@sysdev.exchange.de>J Subject: Re: New Encompass Survey... PLease Take 10 Minutes to Fill It Out1 Message-ID: <3A2535D7.4090803@sysdev.exchange.de>m   Hello!   Farrell, Michael wrote:u  C > When I clicke on the URL referenced below, this is what happened:p >  >  > HTTP Error 404 >  > 404 Not Foundo > K > The Web server cannot find the file or script you asked for. Please checke- > the URL to ensure that the path is correct.  > E > Please contact the server's administrator if this problem persists.  >   F Note that due to superior formatting linebreaks were inserted into theD URL. Once I've put the pieces together the page did load. It is only& usefull for DECUS US members though...   Greetings, Martint   > Mike Farrell >  >  >> -----Original Message-----o: >> From: Terry C. Shannon [SMTP:terryshannon@mediaone.net]+ >> Sent: Tuesday, November 28, 2000 5:55 PMs >> To:   Info-VAX@Mvb.Saic.ComJ >> Subject:      New Encompass Survey... PLease Take 10 Minutes to Fill It >> Out >> t7 >> en.com.pass -  to surround;  to contain, to include.i >>  > >> Help us help you!  As you know, we've recently rolled out aK >> new identity, complete with restated goals, objectives and plans for the G >> future. But, for any change to be successful, we need feedback - andm	 >> member A >> feedback is the best, most "encompassing" feedback we can get.w >> lK >> Take a moment to give us your views on the NEW! Encompass. Participationr >> isr# >> fast and easy.  Just click on tosM >> http://inq30.chicago.sba.com/Encompass_Calibration_Survey/3.0_version_encoc >> mpaH >> ass_survey.html and share your views.  Here's a bonus -- complete the	 >> surveyeJ >> by December 8 and you'll automatically be entered into a drawing to win >> one( >> of three iPAQ Personal Audio Players. >> n? >> Your opinions will be kept confidential.  All data collectedaK >> will be tallied by a third party provider so please share your thoughts.o >>   >> RemembersM >> http://inq30.chicago.sba.com/Encompass_Calibration_Survey/3.0_version_encoa >> mpuI >> ass_survey.html - puts you one click away from an iPAQ  Personal AudioaG >> Player and sharing your views with Encompass leadership.  Let's work J >> together to continue to build the outstanding user organization that is
 >> Encompass!a >> k >>   >> Posted by...M >>   >> n >> Terry C. Shannone >> Consultant and Publishery >> Shannon Knows Compaqe# >> email: terryshannon@mediaone.nets >> http://www.acersoft.com' >> Web (info on SKC):  www.acersoft.como >> s >> a     -- 0P Dr. Martin P.J. Zinser                                 zinser@sysdev.exchange.de Deutsche Boerse Systems AGL Neue Boersenstr. 1                                     Tel: +49 69 2101 5634L 60487 Frankfurt                                        FAX: +49 69 2101 3411P Germany                                                Private:  zinser@decus.de   ------------------------------   Date: 29 Nov 2000 17:40:08 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)! Subject: Re: OpenVMS DHCP Client?a6 Message-ID: <903f1o$t6u$1@mailint03.im.hou.compaq.com>  k In article <lI%U5.25835$Y6.7976070@news1.mntp1.il.home.com>, "Richard Robbins" <RERobbins@home.com> writes: L :Is there a DHCP client for OpenVMS?  I've stumbled across the documentationG :for establishing an OpenVMS DHCP server, but not seen how to set up an*" :OpenVMS machine as a DHCP client.  G   Assuming TCP/IP Services; not prior to V5.1; currently in field test.a  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Wed, 29 Nov 2000 11:27:07 GMT  From: richard_maher@my-deja.com - Subject: Propagating a TXN over DECnet parent-) Message-ID: <902p68$l8n$1@nnrp1.deja.com>K  G $!-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- 	 =-=-=-=-=.+ $!	MAHER_DECDTM_PROPAGATION_PARENT_NODE.COM: $!3 $!	This command file to be used in conjunction with  *_CHILD_NODE.COM $!> $!	Child node process is run on node GROAT to which a proxy is requiredG $!-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- 	 =-=-=-=-=: $	Set Verify, $	DEFINE/PROCESS SQL$DISABLE_CONTEXT "FALSE"! $	SQLMOD == "$SYS$SYSTEM:SQL$MOD"D $!> $!	The code is yours *as is*. I don't offer any support, but I will answer anyd> $!	questions (here or via direct e-mail) when/if I can. Enjoy! $!B $!	NB: This is USER mode, UNpriviledged code! You do *NOT* have to installs> $!	anything. Turn all priviledges *OFF* except (netmbx,tmpmbx) $! $	set verify $	gosub	Create_parent_cob) $	gosub	dist_sql_sqlmodg $!& $	if	f$getsyi("ARCH_NAME") .eqs. "VAX" $	then $		cobol/lis			parent.cobs $	else( $		cobol/GRANULARITY=BYTE/lis	parent.cob $	endifi $!@ $!	You need to create a parent database in the default database. $!
 $	set noon $	sql-$ drop database filename mf_personnel; exit $!# $	@SQL$SAMPLE:PERSONNEL SQL M NOCDDh $	set verify $!& $!	The SQLMOD is compiled as follows:- $!! $	SQLMOD DIST_SQL.SQLMOD/CONTEXT=r' (SET_TRANS_RW,UPDATE_DEPT)/CONSTR=IMMEDa< $!	SQLMOD DIST_SQL.SQLMOD/CONTEXT=(SET_TRANS_RW,UPDATE_DEPT) $!# $!	Then you link the parent image:-h $!1 $	LINK PARENT, DIST_SQL, SYS$LIBRARY:SQL$USER/LIBu $! $!	Finally:- $!
 $!	RUN PARENTn $!B $!	Enter a couple of department_names (you can check the databases if you want)8 $!	before entering spaces. Enter ^Z if you want to exit. $! $Exit: $	Exit  H $!---------------------------------------------------------------------- -------- $Create_parent_cob:u $	create	parent.coby identification division., program-id.    parent_rtn with ident "v1.0". author.        public domain.t   data division. working-storage section. *c@ 01  io$_readvblk            pic 9(9)    comp    value   external
 io$_readvblk.a@ 01  io$_writevblk           pic 9(9)    comp    value   external io$_writevblk.@ 01  ss$_abort               pic 9(9)    comp    value   external
 ss$_abort.@ 01  ss$_remote              pic 9(9)    comp    value   external ss$_remote. @ 01  ss$_normal              pic 9(9)    comp    value   external ss$_normal. - 01  sys_status              pic 9(9)    comp.d *bC 01  db_spec                 pic x(50)           value   "alias persd filename mf_personnel".t< 01  user_exit               pic x               value   "N"." 01  commit_flag             pic x.' 01  msg_buf                 pic x(256).e- 01  msg_len                 pic 9(4)    comp.J- 01  net_chan                pic 9(4)    comp.l 01  letter_sent.&     03  ls_tid              pic x(16).&     03  ls_bid              pic x(16).&     03  new_name            pic x(30).'     03                      pic x(450).> *i 01  sql_ctx.:     03                      pic 9(9)    comp    value   1.:     03                      pic 9(9)    comp    value   1.;     03                      pic 9(9)    comp    value   16.JC     03  tid                 pic x(16)           value   low-values.A=     03                      pic 9(9)    comp    value   zero. & 01  bid                     pic x(16). *I- 01  sqlcode                 pic 9(9)    comp.t
 01  dtm_iosb. -     03  dtm_iosb_status     pic 9(4)    comp.L#     03                      pic xx. -     03  reason_code         pic 9(9)    comp.t  A 01 rdb$message_vector                                   external.E-     03 rdb$lu_num_arguments pic 9(9)    comp. -     03 rdb$lu_status        pic 9(9)    comp.:@     03 rdb$alu_arguments                        occurs 18 times.-         05 rdb$lu_arguments pic 9(9)    comp.i *t	 01  iosb.d-     03  cond_val            pic 9(4)    comp.*-     03  byte_count          pic 9(4)    comp. -     03  chan_info           pic 9(9)    comp.  *o 01  ncb.8     03  child_node          pic x(5)    value   "groat".A     03                      pic x(14)   value   '::"task=child"'.  *  procedure division.  kick_off section.1 00.e     perform parent_init.  F     display "Enter new department name  : " erase screen no advancing.     accept      new_name                 protected                  reversed&     at end      move "Y" to user_exit.  -     perform dist_trans until user_exit = "Y".   @     call "sys$dassgn" using by value net_chan giving sys_status.A     if sys_status not = ss$_normal call "lib$stop" using by value  sys_status.  ** fini.*
     stop run.* ** parent_init section. 00.e *+ * Create the remote task.- *-     call "sys$assign"0#         using   by descriptor   ncbh(                 by reference    net_chan'                 by value        0, 0, 0          giving  sys_status.g1     if sys_status not = ss$_normal and ss$_remoteE2         call "lib$stop" using by value sys_status.  :     call "declare_connection_name" using sqlcode, db_spec.%     if rdb$lu_status not = ss$_normalf3          call "sys$putmsg" using rdb$message_vectoro2          call "lib$stop" using by value ss$_abort. *e fini.' *d dist_trans section.r 00. 4     move function upper-case (new_name) to new_name. *+ * Start the transaction. *-     call "sys$start_transw" $         using   by value        0, 0(                 by reference    dtm_iosb$                 by value        0, 0#                 by reference    tid          giving  sys_status.dB     if sys_status = ss$_normal move dtm_iosb_status to sys_status.A     if sys_status not = ss$_normal call "lib$stop" using by valueg sys_status.v *+7 * Authorize a branch (participant) for the sub-process.c *-     call "sys$add_branchw"$         using   by value        0, 0(                 by reference    dtm_iosb$                 by value        0, 0#                 by reference    tid *                 by descriptor   child_node#                 by reference    bids         giving  sys_status.oB     if sys_status = ss$_normal move dtm_iosb_status to sys_status.A     if sys_status not = ss$_normal call "lib$stop" using by valuel sys_status.  *+A * Give the remote process what it needs to join in. *PARALLELISM*aC * This could just as easily be a TIP URL. YOU could make it happen!  *-     move tid to ls_tid.p     move bid to ls_bid.h       call "sys$qiow" :         using   by value        0, net_chan, io$_writevblk$                 by reference    iosb#                 by value        0,0a+                 by reference    letter_sentd+                 by value        512,0,0,0,0.         giving  sys_status.o;     if sys_status = ss$_normal move cond_val to sys_status. A     if sys_status not = ss$_normal call "lib$stop" using by valueo sys_status.P *+
 * Enlist Rdb.  *-/     call "set_trans_rw" using sqlcode, sql_ctx.t%     if rdb$lu_status not = ss$_normalS2         call "sys$putmsg" using rdb$message_vector1         call "lib$stop" using by value ss$_abort.u *r8     call "update_dept" using sqlcode, new_name, sql_ctx.!     if rdb$lu_status = ss$_normal          move "Y" to commit_flage     else2         call "sys$putmsg" using rdb$message_vector          move "N" to commit_flag. *+A * Synchronize with the child process as we can't commit until his.D * database work is finished. For this example we don't bother asking@ * the child how its update went. Let the DECdtm 2PC sort it out. *-     perform get_reply.C     perform until byte_count = zeros or sys_status not = ss$_normal 5         display "Child> ", letter_sent (1:byte_count)          perform get_replyr     end-perform.A     if sys_status not = ss$_normal call "lib$stop" using by valuei sys_status.-       if commit_flag = "Y"         perform commit_trans     else         perform abort_trans. *0 fini.M * 9     display "Enter new department name  : " no advancing.p     accept      new_name                 protected                  reversed&     at end      move "Y" to user_exit. *  get_reply section. 00.u     call "sys$qiow" 9         using   by value        0, net_chan, io$_readvblkt$                 by reference    iosb#                 by value        0,0o+                 by reference    letter_sente+                 by value        512,0,0,0,0s         giving  sys_status. ;     if sys_status = ss$_normal move cond_val to sys_status.p *. commit_trans section.  00.a*     display "Committing transaction. . .".       call "sys$end_transw"s$         using   by value        0, 0(                 by reference    dtm_iosb$                 by value        0, 0#                 by reference    tid          giving  sys_status.eA     if sys_status not = ss$_normal call "lib$stop" using by valueo sys_status.  * "     if dtm_iosb_status = ss$_abort          call "sys$getmsg"3                 using   by value        reason_codee/                         by reference    msg_len /                         by descriptor   msg_bufn+                         by value        0,0v"                 giving  sys_status'          if sys_status not = ss$_normald7               call "lib$stop" using by value sys_statusk          end-if 9          display "Couldn't commit - " msg_buf (1:msg_len)a,     else if dtm_iosb_status not = ss$_normal=               call "lib$stop" using by value dtm_iosb_status.r *  abort_trans section. 00. (     display "Aborting transaction. . .".       call "sys$abort_transw"i$         using   by value        0, 0(                 by reference    dtm_iosb$                 by value        0, 0#                 by reference    tidr!                 by value        0a         giving  sys_status.-C     if sys_status = ss$_normal or ss$_abort move dtm_iosb_status to- sys_status.eA     if sys_status not = ss$_normal call "lib$stop" using by valuec sys_status.: *: end program parent_rtn.P $	return  H $!---------------------------------------------------------------------- -------- $Dist_sql_sqlmod:s $	create Dist_sql.sqlmod module    dist_sql language  cobolg parameter colons  ( declare pers alias filename mf_personnel  ! procedure declare_connection_namel         sqlcode,!         :db_spec        char(50);0           attach :db_spec;   procedure set_trans_rw         sqlcode;  "         set transaction read write<                 reserving pers.departments for shared write;   procedure update_dept          sqlcodeE         :dept_name    char(30);            update pers.departmentsf0                 set department_name = :dept_name.                 where department_code= 'PHRN';   procedure disconnect_dbe         sqlcode;           disconnect default;  $	return    & Sent via Deja.com http://www.deja.com/ Before you buy.S   ------------------------------  % Date: Wed, 29 Nov 2000 12:33:15 +01000' From: Theo Jakobus <jakobus@iaf.fhg.de>  Subject: SNMP and MIB agents* Message-ID: <3A24F78B.7A4F7EE3@iaf.fhg.de>   Hi!   F System info: DIGITAL TCP/IP Services for OpenVMS Alpha Version V5.0A - ECO 1eF              on a Digital Personal WorkStation  running OpenVMS V7.2-1     F I enabled in SYS$SYSDEVICE:[TCPIP$SNMP]TCPIP$EXTENSION_MIB_RUN.COM the4 agents SVRCLUSTER_MIB, SVRMGT_MIB and SVRSYSTEM_MIB.% Running these executables results in:i5 %DCL-W-ACTIMAGE, error activating image UCX$ESNMP_SHR  -CLI-E-IMGNAME, image file6 IAF021$DKA0:[SYS0.SYSCOMMON.][SYSLIB]UCX$ESNMP_SHR.EXE: -SYSTEM-F-SHRIDMISMAT, ident mismatch with shareable image $/3 The file SYS$LIBRARY:UCX$ESNMP_SHR.EXE is a copy of.D SYS$LIBRARY:TCPIP$ESNMP_SHR.EXE but the executables are linked using3 UCX$ESNMP_SHR.EXE which explains the error message.   # Are there new MIB agents available?o     Regards, --    ; ***********************************************************i; *                                                         *b; *  Theo Jakobus                                           *g; *  Fraunhofer-Institut fuer Angewandte Festkoerperphysik  *s; *  Tullastr. 72                                           * ; *  D-79108 Freiburg                                       *y; *  Germany                                                *o; *  Phone:   +49-(0)761-5159-325                           *n; *  FAX :    +49-(0)761-5159-200                           * ; *  e-mail:  jakobus@iaf.fhg.de                            *n; *  http://www.iaf.fhg.de                                  *a; *                                                         * ; ***********************************************************:   ------------------------------  # Date: Wed, 29 Nov 2000 07:07:07 GMTj% From: Uwe Zessin <zessin@my-deja.com>   Subject: Re: SW Raid 310 and VMS) Message-ID: <9029up$9u7$1@nnrp1.deja.com>.  8 In article <900vsg$p61$1@fizban.fizban.pprd.abbott.com>,4   "Dave Gudewicz" <david.gudewicz@abbott.com> wrote:H > We have a dual 310 with a 4100 running VMS 7.2-1.  Seems to work well,F > but I don't like the idea of configuring this beast and having to doD > it with a PC that has 2 available com ports.  There aren't many of, > those floating around, especially laptops. >e- > I understand it took a long time to qualifye  C Ah, yes. Then you must be talking about the RA3000 (HSZ22, although  it it not really a HSZ).A That was bought from CMD to use it as a low-end NT RAID box, whene@ others requested more features (other operating systems, cluster	 support).e  C You can connect a terminal (emulator) to one port, hit '<ESC>&' and D you will get a character-oriented menu. The password is described in the manual ;-)  > > "Chris Scheers" <chris@applied-synergy.com> wrote in message >oH news:2935D81F548AC499.4D3A3160FC592068.AC1DD4F1F621523E@lp.airnews.net.. .lA > > How well does the StorageWorks Raid 310 system work with VMS?c  : If I recall correctly, then this is a "real" HSZ, a HSZ20.  B I haven't seen any problems - I once built a small cluster with an5 AlphaServer 400 and an -800 runnng OpenVMS V6.2-1H3.>  D Cache batteries as described in this thread, yes, that is a problem.  D It's the same problem with the HSZ40 - need to remove the controller= module, then pull the cache module with the battery on board.a  D I like the external batteries that came with HSZ50 and up. HSZ70 andE up even allow to connect 2 batteries at the same time for the purposetG of swap (not permanently). The new controller shelves (M2100 and M2200)nC of the modular arrays also have 2 battery slots for each controller>$ (again for swap, not permanent use).  / > > What versions are required? (VAX and Alpha)m   Don't know, sorry.E I have no idea if any HSZ beyound the HSZ10 (yes, 'ten') is supported E on VAX although in-house we have played with a VAX 4000-something andi< a HSZ-70. We could even boot the VAX via a KZQSA under V7.1.   --
 Uwe Zessin    & Sent via Deja.com http://www.deja.com/ Before you buy.i   ------------------------------  # Date: Wed, 29 Nov 2000 07:20:57 GMTT% From: Uwe Zessin <zessin@my-deja.com>l  Subject: Re: SW Raid 310 and VMS) Message-ID: <902aon$ad0$1@nnrp1.deja.com>t  , In article <3A246977.579F7C9@earthlink.net>,:   "David J. Dachtera" <djesys.nospam@earthlink.net> wrote: > David Jones wrote: > > < > > In message <900vsg$p61$1@fizban.fizban.pprd.abbott.com>,9 > >   "Dave Gudewicz" <david.gudewicz@abbott.com> writes:4E > > >We have a dual 310 with a 4100 running VMS 7.2-1.  Seems to workFB > > >well, but I don't like the idea of configuring this beast andE > > >having to do it with a PC that has 2 available com ports.  There > > > >aren't many of those floating around, especially laptops. > >HG > > You don't need a PC to access the serial, just use a serial port on$F > > the Alpha or VAX with "set host/dte".  You can also use HSZTERM toF > > talk to the controller via the SCSI bus, but it may hard to find a > > copy of the distribution.n  D I don't think HSZTERM is supported any longer. As I wrote in anotherC reply, I beleive Dave Gudewicz does not have a RA310, but a RA3000. " That one does not run HSOF or ACS.   > [snip] >oH > I don't think there's much to it beyond SYS$SYSTEM:HSZTERM$SCSIPAD.EXE; > and possibly HSZTERM_USER_GUIDE.PS, but I could be wrong.n  ? I haven't looked in the kit, but I'd expect a .CLD, too because-B OpenVMS Alpha V6.2-1H3 (at least, my system :-) doesn't understand '$ SET HOST/SCSI'.  F There is a new product (= costs money) for the most recent controllers. (HSZ70 and up), the SANworks Command Scripter:;     http://www.compaq.com/products/sanworks/sanmgt/scs.html-   --
 Uwe Zessin    & Sent via Deja.com http://www.deja.com/ Before you buy.e   ------------------------------  # Date: Wed, 29 Nov 2000 09:07:40 GMTc/ From: Mike Price <mike.price@littlewoods.co.uk>u  Subject: Re: SW Raid 310 and VMS) Message-ID: <902h0q$fke$1@nnrp1.deja.com>   F > I don't think HSZTERM is supported any longer. As I wrote in anotherE > reply, I beleive Dave Gudewicz does not have a RA310, but a RA3000. $ > That one does not run HSOF or ACS. >e
 > > [snip] > >o+ > > I don't think there's much to it beyond  SYS$SYSTEM:HSZTERM$SCSIPAD.EXE= > > and possibly HSZTERM_USER_GUIDE.PS, but I could be wrong.y >nA > I haven't looked in the kit, but I'd expect a .CLD, too becauseiD > OpenVMS Alpha V6.2-1H3 (at least, my system :-) doesn't understand > '$ SET HOST/SCSI'. >"   General warning to allB Do not try to use SWCC at the same time as HSZTERM. HSZTERm is notF offically supported anyway but if you try to use the 2 together soonerF ofr later the 2 will send requests to the controller at the same time.G The disks then go into hostunavailable state and you are left having toOE reset the HSZ controllers - happened to us 3 times before we got holdT of the right person at the CSC.ED At that point we blatted SWCC and now use the command line interfaceF provided by HSZTERM - depends whether or not you like GUI crap running on an NT PC I suppose ;-)!  C our other Raid systems are RA3000s - no HSZTERM and no command linee interface - just SWCC :-(s   --B All opinions are my own and do not necessarily reflect those of my employer    & Sent via Deja.com http://www.deja.com/ Before you buy.-   ------------------------------  % Date: Wed, 29 Nov 2000 16:58:21 +0100o= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>e# Subject: Re: System Service and RTLa) Message-ID: <3A25279D.4A84986B@gtech.com>t   Chris Scheers wrote: > Arne Vajhj wrote: > > Chris Scheers wrote:M > > > LIB$SPAWN can copy your DCL context to the new process.  (SYS$CREPRC byeJ > > > itself can not do this.  Actually, there is no documented way (other) > > > than LIB$SPAWN) to do this at all.)  > >  > > I wrote: > > J > > # You find out that LIB$SPAWN can be done in 5 lines, while SYS$CREPRC: > > # may take you 25-250 lines depending on what you want > > L > > 250 lines are not enough for complete LIB$SPAWN emulation, but I can notL > > rigth away think of any specific LIB$SPAWN usage where 250 lines of code. > > can not achieve the desired functionality. > B > LIB$SPAWN does things like copying the current process' symbols,D > logicals, command table, key definitions, notification setting, ONH > CONTROL settings, default directory, prompt string, etc.  This is done< > by special communications between the parent and child DCLH > interpreters.  This communication channel is not available to a normal
 > SYS$CREPRC.  > J > It would be possible to create an initialization command file that wouldI > recreate most of the context and then feed that to SYS$CREPRC, but then D > it would be difficult to get the new DCL to read from the terminal > correctly. > @ > I don't think that you can write a LIB$SPAWN replacement usingF > SYS$CREPRC without using privileged code.  (LIB$SPAWN uses some SUPR
 > mode code.)_   I do not disagree.  D I have just never had any need for LIB$SPAWN where I needed all thisE functionality. Tyoical I only need to fire off one single DCL command , and I do not need any symbols/logicals etc..   Arne   ------------------------------  % Date: Wed, 29 Nov 2000 11:38:24 +0100)/ From: Nigel Arnot <sysmgr@maxwell.ph.kcl.ac.uk> ' Subject: Re: Technology of US electionsb6 Message-ID: <009F3D81.58E3DF0D.2@maxwell.ph.kcl.ac.uk>   Roy Omond writes:   Q > > > Seriously though,  your elections have shown the dangers of poor technology - > > > when combined with poor election rules.  > >hO > >         Poor technology?  But the Canadians take a marker and mark a ballot(N > >         and drop it in a box and all ballots are counted by hand.  SomehowP > >         I don't view that as superior technology especially when we probably3 > >         have to count 10 times as many ballots.  > - > Oh dear, I know I'm going to regret this...  > J > It's maybe not superior *technology*, but it most certainly does seem toL > be a superior *method*.  As to the 10 times issue, just put in 10 times asG > many counters.  Why does everyone seem to forget the KISS principle ?t > J > After all this fiasco, I'm now absolutely convinced that those countriesG > using the manual method have got it right (my own included), and thatoD > the USofA has got it completely wrong (technology for technology's > sake). >   I Don't regret it, you are 100% correct. KISS sums it up. It *is* superior,dF both in terms of simplicity for the voter and because of a subtlety of the manual counting process.  H Manual counting of paper ballots in the case of a very close election isH a process that proceeds by iterative refinement. In the first count, theL ballots are sorted and bundled (usually into 100s). If it's close, a recountL consists of looking for miscounted bundles and misclassified ballots within H a bundle. Each correction is vetted by all parties, and once accepted isH a permanent state change. Each recount removes errors until the count isI to all human purposes perfect, in other words until no further errors cans> be found. This convergence can fail only if one or more of theN counters is corrupt - and by making sure that every time a bundle is recountedH it is recounted by a different counter, that risk can also be controlled and a corrupt counter detected.y  @ If you want to speed up the initial count, use document scanningF technology for the sorting and initial counting (setting it to reject H difficult-to-scan papers for manual assessment). Few elections are closeI enough for any manual recounting to be necessary. But it's the rare close_E elections where the fairless and accuracy of the process is of utmostbG importance. If the process is open to challenge, either legal or moral,i0 the foundations of democracy itself are damaged.  M "It is not sufficient that justice is done, it must also be seen to be done".dG Similarly with elections. Even if highly accurate, voting machines and v9 fully automated counting fail the "seen to be done" test.r  I And by the way, it is far harder to stuff a ballot box with thousands of f@ convincingly different handwritten marks than with thousands of M indistinguishable punched cards. (This isn't innuendo about the USA election.tG Rather an observation of various tyrants' failure to convincingly "fix"b; their re-elections, and another form of "seen to be done".)_   	Yours,s
 		Nigel Arnotl- 		NRA@MAXWELL.PH.KCL.AC.UK                      7 		"In the beginning there was nothing, which exploded."    ------------------------------   Date: 29 Nov 2000 11:55:54 GMT' From: david20@alpha2.mdx.ac.uk (D.Webb) ' Subject: Re: Technology of US electionso0 Message-ID: <902qsa$478$1@aquila.news.mdx.ac.uk>  \ In article <3A244D19.A383A11A@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: >"D.Webb" wrote:I >> A full hand recount of Florida was the only way this could possibly bee1 >> settled with a reasonable amount of certainty.n >LF >I am not sure that a state-wide manual recount would yield comparableJ >adjustements since each county has its own technology and ballot systems.K >Counties where the mechanical error rate would be close to 0 would producefO >very little change with a manual recount. But in the end, it would provide thet" >image of a more legitimate count. > N >But what is wrong with doing a mechanical recount state-wide (which was done)I >and then only doing hand recounts in those counties where the mechanical  >counts are suspect ?  >r  I Probably nothing but I can imagine that there might then be disputes over N whether the mechanical recount is suspect or not. To accomodate this you wouldJ have to accept for recount any county which either party said was suspect L (otherwise you would still end up with disputes later). Much cleaner to just% say you'll do them all to start with.,      O >Had Bush agreed to hand recounts right from the start, the issue would be overtN >with.  Had he agreed to a hand recount, the Bush side would have had far moreL >"constructive" input on how to define a valid ballot, how pregnant a ballot) >would have had to be to be counted etc.   > O >Without Bush's active participation, the definition of a valid ballot was left ; >to Gore and thus open for accusations of biased counting. o >uL >Again, another sign that the election SYSTEM failed because it did not have1 >the rules clearly set out PRIOR to the election.e > K >If the election rules clearly stipulated that in order to be valid, a hole O >must be fully perforated with no chad, then none of these stupid debates would, >have happened.  >i >g >yN >OJ simpson got off scotts free because there was the slightest of doubts thatO >he wasn't the murderer even though everyone knows he did it. Bush won't have a L >"true" win as long as there is any doubts that Gore might actually have wonK >it. And it gets worse if Gove manages to get his votes counted and gains at >lead of only 10 votes.A >t > H >What I do find questionable is the story that some foreign ballots wereL >received after election day without a valid postmark and that the Bush campL >fought to have those validated.  If the military, in order to get the votesK >there quicker, used a different mechanism to transmit the votes, then this M >mechanism should have been validated by the election officials and each voteo= >stamped with some state seal to approve/validate the ballot.g > G >If there is a possibility that a vote was cast AFTER the election and -K >inserted in the stack of foreign ballots, then none of the foreign ballotsmN >without vald postmarks should be accepted. Mind you, with today's technology,I >one should be able to fake a valid postmark with whatever date you want.d > 	 >--------_ >cL >Lets assume, for the sake of discussion, that GOD knew that Gore had won byM >5000 votes, but because of mechanical problems in only one county, that someeI >6000 votes were not counted properly. Suppose that county was heavily inEJ >favour of Gore.  We'd have the results we have today with Bush winning by >about 900 votes.a >eL >Recounting that county would give Bush an extra 100 votes and Gore an extraN >5900 votes. Yep, that would tilt the scale. But would it be unfair to recount >only that one county ?p  N In this hypothetical situation it would obviously be unfair not to count theseK votes since it changes the election results. But since we do not have God'scL view we cannot know that there are not other missed votes in other counties.M Hence to be seen to be fair would require counting of other counties as well.     
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------  % Date: Wed, 29 Nov 2000 12:22:59 +0100t/ From: Nigel Arnot <sysmgr@maxwell.ph.kcl.ac.uk>E' Subject: Re: Technology of US elections 7 Message-ID: <009F3D87.93028EEA.30@maxwell.ph.kcl.ac.uk>H   > O > Sorry, but I have to respond to this as well.  The main point is that Al Gore M > stands there whining that every vote must count.  But in reality he is onlyBN > whining that every vote in counties that primarily voted for him must count.  N Come off it. I saw him on the TV offering to abide by the result of a Florida-I wide hand recount. And I saw Bush declining the offer. "I wanna win", not*  I want to see democracy prevail.  */ > He has purposely thrown out military votes.     H The ones that didn't get postmarked, because somebody stopped doing thatF with military mail without realizing the consequences. Yes, I noticed J that too. A big mistake, revealing that he's no better than his opponent.  "I wanna win too".    G Not being a USA citizen I won't comment any further, beyond saying that I IMO what's going on now is less edifying than nude female mud-wrestling. FA Also rather damaging to the image of the USA outside her borders.:  G (I wonder what would have happened if voting were compulsory and there eH was a vote for "None of the above"? Give the silent majority some teeth.G Perhaps politicians would behave better if they periodically discovered*4 how many didn't think any of them worth voting for!)   	Yours,o
 		Nigel Arnot - 		NRA@MAXWELL.PH.KCL.AC.UK                   u  7 		"In the beginning there was nothing, which exploded."n   ------------------------------  % Date: Wed, 29 Nov 2000 12:29:02 +0100 / From: Nigel Arnot <sysmgr@maxwell.ph.kcl.ac.uk>t' Subject: Re: Technology of US electionsi7 Message-ID: <009F3D88.6B899CF9.34@maxwell.ph.kcl.ac.uk>e   > >fJ > >The method is a tool to acheive a goal. In the case of canada, and manyF > >countries, the old fashioned X on a ballot does the job very nicelyD > >because we only mark a single X on a ballot with perhaps only 5-7G > >choices. (Hey I found out there is a "Marijuana party" in Canada and 4 > >they someone running in our electoral district !) > L > Hey, they drew more votes nationally than the Communist, Canadian Action, 5 > Marxist-Leninist, and Natural Law parties combined!( >   ' So stick that in your pipe, and, er ...i (Sorry, couldn't resist)   	Yours, 
 		Nigel Arnotc- 		NRA@MAXWELL.PH.KCL.AC.UK                   c  7 		"In the beginning there was nothing, which exploded."1   ------------------------------    Date: 29 Nov 2000 08:17:14 -0500, From: koehler@eisner.decus.org (Bob Koehler)' Subject: Re: Technology of US elections,+ Message-ID: <AIjxSoJBM9rD@eisner.decus.org>e  \ In article <3A245198.3C5115EB@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: > "Ebinger . Eric" wrote:h8 >> not a Democracy).  The simplest and cleanest solution8 >> is to call the Florida state legislature together and2 >> have them decide who should receive the Florida >> electoral votes.  > L > Is this something which is already in the laws/constitution ? How would itP > look if a republican state legislature were to choose a republican candidate ?E > (even if the later has a few hundred votes more than his opponent).e  D The Constitution specifies (Article II, Section 1) "Each state shallG appoint, in such manner as the Legislature thereof may direct, a numberAH of electors, ...", other parts of the Constitution guarantee individualsH rights to vote for the President without specifying how the selection ofE electors is tied to the right to vote (which is why it varies amongstQ the states).  G (http://lcweb2.loc.gov/const/constquery.html, a part of the Thomas siteb$ provided by the Library of Congress)  H A Federal law further specifies that the process cannot be changed after@ the election.  The Republicans are trying to make a lot of this,A claiming Gore is trying to change how votes are counted after the H election, but Gore's legal arguments seem to be fairly well concentratedA on were the votes counted as provided by Florida and federal law.c  D The relevent question is, does the Florida Constitution or state lawH provide for further action by the legislature at this time.  I'm bettingE it does, but I have a copy of the Florida Constitution, and of course  IANAL.  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporationd= NASA GSFC Flight Software       | Federal Sector, Civil Group E                                 | please remove ".aspm" when replyingk   ------------------------------   Date: 29 Nov 2000 10:10:47 GMT7 From: woodacre@scala.reading.sgi.com (Michael Woodacre)p' Subject: Re: Technology of US elections . Message-ID: <902kn7$3f7ua$1@fido.engr.sgi.com>  \ In article <3A245020.91377048@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: |> Michael Woodacre wrote:T |> > for 'winning' this election. Given that Florida is a statistical tie, shouldn't@ |> > the Florida electorial college votes be split between them? |> nP |> You can't change the rules of the contest after the contest to decide who theN |> winner is. If the election laws in Florida do not deal with situations of a9 |> statistical tie, then it is the laws which are flawed.c  P Problem is, the election process is flawed, and by a much bigger margin than theO few hundred votes difference between the candidates. The deadlines in place andeT the lawsuits being thrown around mean this will never get sorted out in any rationalQ fashion by the 12th. Florida should be ashamed of their process. How can you pickoW a winner when the margin of error in the counting is larger than the difference between U the candidates? If all 25 votes go to either candidate, it is clearly an illegitimateeP result and the resulting president will never be able to get away from this mess   |> PK |> Who should be the body to decide that instead of sending one block of 25 N |> votesas the current laws state, that they should send 12 votes for Bush, 12H |> votes for Gore and 1 vote for Nader ? (or 13 to bush and 12 to gore). |> "Q |> The florida body who would make that decision would be doing this in a contextTP |> where they would knowingly be deciding the next president of the USA. AnotherQ |> reason why all election laws/rules must be clearly set out BEFORE elections so O |> that after the election, there is no possibility that the election officialstL |> can be seen using their own powers to decide the outcome of the election.   -- E    Michael S. Woodacre d UUCP: woodacre@sgi.com   n Phone: +44 1118 925 7846
 M/S: IGB-3110e Address    : 1530 Lakeside$              Arlington Business Park              Theale, Berkshire              Reading              RG7 4SB              United Kingdom-   ------------------------------  # Date: Wed, 29 Nov 2000 15:39:10 GMT  From: chris_muller@my-deja.comC Subject: Re: TK70: request for data transer fro TK70 to 4mmDat tape-) Message-ID: <9037ut$10i$1@nnrp1.deja.com>0  
 Hi Rainer,  : We can convert tk70 to 4mm and a variety of other formats.     regards, Chris Muller www.mullermedia.com2 212-344-0474    / In article <3A1E4B8C.4C356CF8@erl9.siemens.de>,t6   Rainer Burger <rainer.burger@erl9.siemens.de> wrote:G > We would need to know if anybody still has a TK70 tape drive on a VAXtA > machine and could read the tape content from this TK70 tape andL transfer > it to a 4mm DAT tape. < > The tape has been written under VMS 5.5-2 (maybe also -3). >n6 > All suggestions are welcome. Offers are appreciated. >i > Sincerely  > Rainer Burger  >  >     & Sent via Deja.com http://www.deja.com/ Before you buy.b   ------------------------------  # Date: Wed, 29 Nov 2000 14:35:38 GMT2+ From: Jordan Henderson <jordan@my-deja.com>i( Subject: Re: VMS and NT integration?????) Message-ID: <90347q$to4$1@nnrp1.deja.com>b  + In article <OEH0PzGf9hYD@eisner.decus.org>,s<   Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) wrote:D > If you want to be able to fully maintain VMS from another machine,B > you will need access to the console prompt, such as via an RS232A > connection, direct from your NT machine or via reverse-LAT to ad7 > terminal server.  Terminal servers are cheap on eBay.t  A Is there an inexpensive reverse-LAT solution for NT?  I know thatRE Pathworks has support for this, but I can't offhand think of anythingh else that does.r  @ Does the VMS Hobbyist program include client Pathworks licenses?  C Failing this, is anyone familiar with the LAT support now availables. for Linux?  Can one do reverse LAT from Linux?  > The reason I'm asking is that I've been tempted to get a cheapB terminal server to work with my home machines (VAXen now, possiblyD Alphas in the future) and if I could put them to use with Windows or> Linux then the terminal server would be of even greater value.   > E > As for terminal emulators -- there are many fine terminal emulatorsg: > available for Windows.  None of them are from Microsoft. >.? > In article <3A23ECB7.925502CF@ohio.edu>, "Richard D. Piccard"  <piccard@ohio.edu> writes:F > > Telnet using Kermit-95, not the bogus terminal emulator that comesD > > with.  Yes, I know that isn't what you had in mind, but I really thinkk > > it is the answer.  > >r' > >                                 RDPd > >a > >S > > Ray Phelan wrote:o > >iF > >> Is there a simple way to maintain a VMS box, via an NT machine on thet > >> same network????? > >o > > --F > > ==================================================================F > > Dick Piccard                           Academic Technology ManagerF > > piccard@ohio.edu                                 Computer ServicesF > > http://oak.cats.ohiou.edu/~piccard/                Ohio University > >  > >h >f   -- -Jordan Henderson  jordan@greenapple.comu    & Sent via Deja.com http://www.deja.com/ Before you buy.n   ------------------------------  % Date: Wed, 29 Nov 2000 15:45:03 -0300 ) From: fabio_compaq@ep-bc.petrobras.com.bri( Subject: Re: VMS and NT integration?????L Message-ID: <OFA524AEE5.E61EA99A-ON032569A6.00666644@ep-bc.petrobras.com.br>  H VMS and NT integration ! ! !  Sounds interesting when talking about new=  	 technicaleH developments like SAN integration, etc ... but I think it=B4s waste of = time and H resources trying to put old hardware (LAT based) to work in this new wo= rldl( of TCP/IP connections (Web, and etc...).  E Why dont  develop a version of Windows CE to run in Microvaxes ?  :-)t  C I would like to see a "Metaframe" for OpenVMS. Something like usingh> a Citrix ICA client to access OpenVMS or a VT-Win terminal ...E Would be much more interesting to port Win32 applications to OpenVMS.u   Regards,   FC                    < Jordan Henderson <jordan@my-deja.com> em 29/11/2000 12:35:38H                                                                        =     =20rH                                                                        =     =20dH                                                                        =     =20l    @                                                              =20@                                                              =20@                                                              =20@  Para:    Info-VAX@Mvb.Saic.Com                              =20@                                                              =20@  cc:      (bcc: Fabio dos Santos Cardoso/E-P-BC/Contratada)  =20@                                                              =20@                                                              =20@                                                              =20@  Assunto: Re: VMS and NT integration?????                    =20@                                                              =20           =i    + In article <OEH0PzGf9hYD@eisner.decus.org>, <   Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) wrote:D > If you want to be able to fully maintain VMS from another machine,B > you will need access to the console prompt, such as via an RS232A > connection, direct from your NT machine or via reverse-LAT to aa7 > terminal server.  Terminal servers are cheap on eBay.-  A Is there an inexpensive reverse-LAT solution for NT?  I know thatkE Pathworks has support for this, but I can't offhand think of anythings else that does.w  @ Does the VMS Hobbyist program include client Pathworks licenses?  C Failing this, is anyone familiar with the LAT support now availablea. for Linux?  Can one do reverse LAT from Linux?  > The reason I'm asking is that I've been tempted to get a cheapB terminal server to work with my home machines (VAXen now, possiblyD Alphas in the future) and if I could put them to use with Windows or> Linux then the terminal server would be of even greater value.   >oE > As for terminal emulators -- there are many fine terminal emulatorsi: > available for Windows.  None of them are from Microsoft. >o? > In article <3A23ECB7.925502CF@ohio.edu>, "Richard D. Piccard"e <piccard@ohio.edu> writes:F > > Telnet using Kermit-95, not the bogus terminal emulator that comesD > > with.  Yes, I know that isn't what you had in mind, but I really thinki > > it is the answer.e > >h' > >                                 RDPr > >  > >c > > Ray Phelan wrote:  > >tF > >> Is there a simple way to maintain a VMS box, via an NT machine on thei > >> same network????? > >h > > --J > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=I =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=e9 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3DiF > > Dick Piccard                           Academic Technology ManagerF > > piccard@ohio.edu                                 Computer ServicesF > > http://oak.cats.ohiou.edu/~piccard/                Ohio University > >l > >s >f   -- -Jordan Hendersonn jordan@greenapple.coms    & Sent via Deja.com http://www.deja.com/ Before you buy.S       =d   ------------------------------  # Date: Wed, 29 Nov 2000 08:52:43 GMTn/ From: Mike Price <mike.price@littlewoods.co.uk>  Subject: Re: VMS System Backup) Message-ID: <902g4q$f5n$1@nnrp1.deja.com>   F This is a part of a system disk backup we use (see below). It requiresF a logical of TAPNAM which is equated to the device name of the tape onD the VMS system (you could modify this to use s disk file but I wouldG ALWAYS recommend having a tape backup - borrow a SCSI tape and cable ifg
 you need to!)e  C This is slightly against the perfect solution as any files that arenC being accessed may be dubious on the save set. It is designed to ben, used when the system is as idle as possible.B I believe that the operating system CD distribution media allows aF proper full system disk backup but you need to take the system down to+ do this and run the backup from the consolef       $ init 'tapnam' systap/MED=COMP # $ mount/for/noass 'tapnam'/MED=COMPe	 $ set ver  $ set on
 $ sho time5 $ doback:==backup/med=comp/block=65535/ign=interl/logpD $ DOBACK/LABEL=systap/image sys$sysdevice: 'tapnam'EXCSDKD5.BCK/SAVE
 $ sho time
 $ SET NOON $ DISM 'TAPNAM'r   Hope this helps    Mike   --B All opinions are my own and do not necessarily reflect those of my employer    & Sent via Deja.com http://www.deja.com/ Before you buy.o   ------------------------------   Date: 29 Nov 2000 14:14:41 GMT8 From: hammond@not@peek.ppb.cpqcorp.net (Charlie Hammond) Subject: Re: VMS System Backup6 Message-ID: <90330h$r5d$1@mailint03.im.hou.compaq.com>  w In article <Uj_U5.1137$YS1.92256@bgtnsc04-news.ops.worldnet.att.net>, "Robert Meyer" <meyerra@worldnet.att.net> writes:hK >I support a 6.x VMS system operating on an Alpha server.  Our organizationtJ >has purchased a network backup solution.  Prior to using it I need a full* >system backup and a recovery method.  ...  : Several replys have correctly pointed you to BACKUP/IMAGE.A However, please be aware that BACKUP/IMAGE will backup the entiretE "system disk", including any user data -- i.e. not just the "system".tF If your "backup solution" requires you to restore the "system" withoutM restoring possible stale user data on the system disk this could be an issue.    -- sK     Charlie Hammond -- Compaq Computer Corporation -- Pompano Beach  FL USAiF          (hammond@peek.ppb.cpqcorp.net -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------  # Date: Wed, 29 Nov 2000 15:10:37 GMTa* From: Alan E. Feldman <alan48@my-deja.com> Subject: Re: VMS System Backup) Message-ID: <903696$vef$1@nnrp1.deja.com>t  E In article <Uj_U5.1137$YS1.92256@bgtnsc04-news.ops.worldnet.att.net>,e2   "Robert Meyer" <meyerra@worldnet.att.net> wrote:? > I support a 6.x VMS system operating on an Alpha server.  Our  organizationF > has purchased a network backup solution.  Prior to using it I need a fullF > system backup and a recovery method.  On IBM there is a Mksysb tape. On? > HP/UX there is a make_recovery utility which generates a fullu recovery tape.B > What method is used to backup the full system on VMS?  Thanks in advance foriD > any assistance.  This tape will be done prior to loading a network backup > client.  CYA  ;->. > Thanks in advance.   --  D First make sure that all your network and backup software are on theF system disk. IOW, whatever you need to restore data remotely should be on the SYSTEM disk.t  ? Do a "standalone backup" of the system disk with a command likei  B $ BACKUP/IMAGE/VERIFY DUA0: MKA0:SYSTEM-date.B/LABEL=SYSTEM/REWIND  ? adding any qualifiers that may be appropriate for your hardware A configuration. You'll have to make a standalone backup kit onto anD removable media (prob. tape) first. Then, in case all your disks areF lost, you can boot from the standalone backup kit on the tape and then? resotre the system disk from tape to a new disk. (Boot from theaE standalone backup kit or CD for both system disk saves and restores.).  G With VMS 6.2 and later, you can backup your system disk using the menu-3F driven system on the operating system distribution CD instead of using Standalone Backup.   (For more details, check  H http://www.openvms.compaq.com:8000/72final/6017/6017pro_043.html#6017bac kupc   )w  F Then, in case your system is nuked, you can restore the system disk byB booting standalone/CD, restore the system disk with a command like  ' $ BACKUP/IMAGE MKA0:SYSTEM-date.B DKA0:   G replacing the device names with your actual device names. If the targeti7 disk has a different cluster size, you may benefit fromd adding /TRUNCATE in this step.  D Once you have your SYSTEM disk restored, boot from it, bring up yourG network and backup software, and restore the rest from the remote site!h  B NOTE: If you wish to e-mail me, do NOT use the deja address. It isB broken. Instead, use one of the addresses below, removing the long wrong part first. Thanks.T   -- Disclaimer: JMHO Alan E. Feldman  &-)/ w: afeldman@gfigroup.BorisYeltsinCooksCajun.com85 h: alan48@dellnet.YouCantBelieveEverythingYouRead.como    & Sent via Deja.com http://www.deja.com/ Before you buy.a   ------------------------------  % Date: Wed, 29 Nov 2000 11:55:30 -0500e- From: "Peter Weaver" <peter.weaver@stelco.ca>  Subject: Re: VMS System Backup/ Message-ID: <t2ad9eendnpb6b@corp.supernews.com>s  : "Robert Meyer" <meyerra@worldnet.att.net> wrote in message? news:Uj_U5.1137$YS1.92256@bgtnsc04-news.ops.worldnet.att.net...e? > I support a 6.x VMS system operating on an Alpha server.  Ourp organizationF > has purchased a network backup solution.  Prior to using it I need a full >...  D Other people have answered your backup questions already, but I wantC to throw in a comment or two about network backups. Be careful what F client you have, at least one client out there will not handle certainF conditions that are definitely present on your system drive and may orB may not be present on your other drives. I am not a fan of networkA backups as a replacement for VMS backups but I have had a network & backup solution shoved down my throat.  C If you are going to use a network backup to backup user drives then F that is a good solution for users who need to recover one or two filesF from time to time without bothering an operator (assuming your networkE backup is using a tape robot). But for disaster recovery backups makelF sure you also do a full backup to a tape and store that tape off-site.? In case of a disaster the network backup robot will be too busyeE restoring the Unix and NT toys while you recall your tape and load it - on a SCSI tape drive attached to your system.-   ------------------------------  # Date: Wed, 29 Nov 2000 13:34:23 GMT 0 From: Timothy Stark <sword7@grace.speakeasy.org>, Subject: What happened to www.graysoft.com?!8 Message-ID: <zB7V5.61968$DG3.1190377@news2.giganews.com>   Hello folks:  G Yesterday and today, I tried to access (surf) GrayMatter's web site for H VAX printing software but it is down.  Yesterday, I received that serverC is not found or so.  Today I tried that again but my web browser iso waiting for responses forever.  5 Does anyone have access to its web site successfully?>$ Its web address is www.graysoft.com.  
 Thank you!   -- Tim Stark   -- e, Timothy Stark	<><	Inet: sword7@speakeasy.orgJ --------------------------------------------------------------------------F "For God so loved the world, that he gave his only begotten Son, that H whosoever believeth in him should not perish, but have everlasting life.. Amen." -- John 3:16 (King James Version Bible)   ------------------------------    Date: 29 Nov 2000 08:30:11 -0500, From: koehler@eisner.decus.org (Bob Koehler)6 Subject: Re: Why so much virtual memory for tiny jobs?+ Message-ID: <9m$Y0lDYuRX$@eisner.decus.org>   V In article <901bj8$id8$1@nnrp1.deja.com>, Alan E. Feldman <alan48@my-deja.com> writes:? > Why does it take so much virtual memory to run this tiny job?e >  > $ type show-work.com > $    SHOW WORK > $n > $    SHOW PROCESS/ACCOUNTING > $t > $    EXIT  > 5 > Here is the part of the log file after SYLOGIN.COM:e >    [...]t  C >  Direct I/O count:          26  Peak virtual size:           3077   C Probably depends very much on what is in SYLOGIN.COM and LOGIN.COM.l; Any one image activation could easily map the 3077 pages.  l  < Unless the virtual map becomes so big it causes you to raiseF VIRTUALPAGECNT on a VAX (obsolete on Alpha), what do you care?  MemoryA utilization and page file utilization are most likely much lower.t  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences CorporationE= NASA GSFC Flight Software       | Federal Sector, Civil GrouplE                                 | please remove ".aspm" when replying    ------------------------------  # Date: Wed, 29 Nov 2000 14:03:19 GMT 1 From: "Mark D. Jilson" <jilly@clarityconnect.com>66 Subject: Re: Why so much virtual memory for tiny jobs?1 Message-ID: <3A250E10.1AA7725@clarityconnect.com>r  ? Yes Peak Virtual & Peak Pagefile are actually Peak Virtual.  WeiF documented this in a DSNLink article.  I would think that the pages inE the VA not in the WS would be address space that has been created buteH not used.  If LIB$GETVM is called such that it just allocates a bunch ofH VA space but this space doesn't get touched by the process it will neverH have physical memory allocated to it and thus not show up in the workingF set.  You really need to get a better look at the process and see whatD the PPGCNT, GPGCNT, WSSIZE.  The following is my modification of the' command procedure in the VMS doc set.       C $ ! WORKSET1.COM - Command file to display working set information.L@ $ !       Requires 'WORLD' privilege to display other processes. $!
 $ esc[0,8]=27y $ bold = esc+"[1m" $ normal = esc+"[0m" $ a   = """  $ pid   = "" $ context = "" $ IF p1.NES. "" THEN pid = p1 G $ WRITE sys$output "                                        Working Set- Informa* $ WRITE sys$output "",H $ WRITE sys$output "              PagFl PagFil PagFil  WS    WS    WS    WS  Pa*RG $ WRITE sys$output "Processname   Quota Remain Peak   Extnt Quota Defltb Size in* $ WRITE sys$output ""e $! $ START:( $ IF p1.EQS."" THEN pid = F$PID(context) $ IF pid.EQS."" THEN EXITy $ pid   = a+pid+at& $ username = F$GETJPI('pid,"USERNAME")$ $ IF username.EQS."" THEN GOTO START' $ processname = F$GETJPI('pid,"PRCNAM")c' $ imagename = F$GETJPI('pid,"IMAGNAME")r) $ imagename = F$PARSE(imagename,,,"NAME")n& $ wsdefault = F$GETJPI('pid,"DFWSCNT")$ $ wsquota = F$GETJPI('pid,"WSQUOTA")& $ wsextent = F$GETJPI('pid,"WSEXTENT")" $ wssize = F$GETJPI('pid,"WSSIZE")' $ globalpages = F$GETJPI('pid,"GPGCNT")o( $ processpages = F$GETJPI('pid,"PPGCNT")( $ pagefaults = F$GETJPI('pid,"PAGEFLTS")( $ pgflquota = F$GETJPI('pid,"PGFLQUOTA")( $ pagfilcnt = F$GETJPI('pid,"PAGFILCNT")& $ virtpeak = F$GETJPI('pid,"VIRTPEAK")& $ pages   = globalpages + processpages@ $ text   = F$FAO("!13AS!6(6SL)!6SL''bold'!6SL''normal'!7SL!AS",-      hE processname,pgflquota,pagfilcnt,virtpeak,wsextent,wsquota,wsdefault,- ,       wssize,pages,pagefaults," "+imagename) $ WRITE sys$output textr $ IF p1.NES."" THEN EXIT $ GOTO START   "Alan E. Feldman" wrote: >  > Hello, > ? > Why does it take so much virtual memory to run this tiny job?9 >  > $ type show-work.com > $    SHOW WORK > $o > $    SHOW PROCESS/ACCOUNTING > $r > $    EXITh > 5 > Here is the part of the log file after SYLOGIN.COM:w >  > $    SHOW WORK@ >   Working Set      /Limit= 895   /Quota= 1790    /Extent= 2048I >   Adjustment enabled    Authorized Quota= 1790  Authorized Extent= 2048  > $  > $    SHOW PROCESS/ACCOUNTING > I > 28-NOV-2000 16:39:29.15   User: SYSTEM           Process ID:   00001959 : >                           Node: IDS08            Process > name: "BATCH_80" >  > Accounting information: C >  Buffered I/O count:        38  Peak working set size:        392 C >  Direct I/O count:          26  Peak virtual size:           3077pC >  Page faults:              468  Mounted volumes:                0   >  Images activated:           3+ >  Elapsed CPU time:          0 00:00:00.12 + >  Connect time:              0 00:00:00.98- > $  > $    EXIT2: >   SYSTEM       job terminated at 28-NOV-2000 16:39:29.27 >  >   Accounting information:NG >   Buffered I/O count:       41         Peak working set size:     410eG >   Direct I/O count:         27         Peak page file size:      3077gG >   Page faults:             529         Mounted volumes:             0,H >   Charged CPU time:    0 00:00:00.13   Elapsed time:     0 00:00:01.10 > 1 > [Last 4 lines edited to fit deja screen width.]l > I > Okay. Now the peak working set size is 410 pages. I assume that this isnD > the largest number of pages that were in the working set (which isC > physical memory). The term "working set size" elsewhere means the D > working set size *limit*, which is 895 here. So the terminology is > already inconsistent.O > H > I assume that "Peak page file size" is the same as "Peak virtual size"C > which I also assume is the same as "Peak page file" in ACCOUNTING G > displays which I furthermore assume is the same as "Virtual pages" in H > the SHOW PROC/CONT display. Are these four quantities all the same? Am2 > I missing any others that are the same as these? > I > I would have guessed that this was the peak number of pages in the page=H > file or the peak total virtual memory for the job. But since there wasD > enough room in memory below WSDEFAULT for the job to run, why were7 > there so many (extra) pages in "Peak page file size"?  > H > I always see this behavior for tiny jobs and seek an explanation. (RunG > on a VAX running VMS V5.5-2HF which has the closest to vanilla VMS of H > all the VAXes in my shop. I see the same behavior on our V6.1 and V6.2 > machines.) > G > TIA. (I will look for responses in the newsgroup. *IF* you e-mail me, G > please don't use the deja address; it doesn't work reliably. Instead,=< > use one of the ones below and remove the huge wrong part.) >  > -- > Disclaimer: JMHO > Alan E. Feldman  &-)1 > w: afeldman@gfigroup.BorisYeltsinCooksCajun.coma7 > h: alan48@dellnet.YouCantBelieveEverythingYouRead.comu > ( > Sent via Deja.com http://www.deja.com/ > Before you buy.    -- sD Jilly	- Working from Home in the Chemung River Valley - Lockwood, NY0 	- jilly@clarityconnect.com			- Brett Bodine fan. 	- Mark.Jilson@Compaq.com			- since 1975 or so, 	- http://www.jilly.baka.com               -   ------------------------------  % Date: Wed, 29 Nov 2000 14:27:49 -0000f* From: "Richard Brodie" <R.Brodie@rl.ac.uk>6 Subject: Re: Why so much virtual memory for tiny jobs?+ Message-ID: <9033p4$m20@newton.cc.rl.ac.uk>   [ "Alan E. Feldman" <alan48@my-deja.com> wrote in message news:901bj8$id8$1@nnrp1.deja.com...   ? > Why does it take so much virtual memory to run this tiny job?iC >  Direct I/O count:          26  Peak virtual size:           3077f  G It's a feature of using shared rather than statically linked libraries.u< You get to map a whole load of pages that you don't need but> it's a cheap operation and saves real memory. And with a batchB file, you get to map the DCL interpreter too. Unless I am mistaken: all these count to the total, which is only ~1.5MB anyway.   ------------------------------   End of INFO-VAX 2000.667 ************************