1 INFO-VAX	Wed, 14 Jun 2000	Volume 2000 : Issue 331       Contents: %HELP 	 Re: %HELP  Re: Alternate boot root number Analyze/audit  question  Re: Analyze/audit  question  Re: Analyze/audit  question  Re: Analyze/audit  question  Re: Backup performance" Re: blocks incorrectly marked free C 6.2-007 compilation annoyance / Re: C bashing (was Re: VMS File Caching Futures  Re: Chinese VMS " Re: CMQ/Dec with a sense of Humor?" Re: CMQ/Dec with a sense of Humor?" Re: CMQ/Dec with a sense of Humor?" Re: CMQ/Dec with a sense of Humor?" Re: CMQ/Dec with a sense of Humor?" Re: CMQ/Dec with a sense of Humor?" Re: CMQ/Dec with a sense of Humor?" Re: CMQ/Dec with a sense of Humor?" Re: CMQ/Dec with a sense of Humor? Command syntax for BACKUP  Re: Command syntax for BACKUP * DECnet-Plus and DECnet task to task from C DSM (Was: Re: Fun VMS Facts?) * Re: Files-11 ODS-2 Readability for FreeBSD* Re: Files-11 ODS-2 Readability for FreeBSD Re: Fun VMS Facts? Re: Fun VMS Facts? Re: Fun VMS Facts? Re: Fun VMS Facts? Re: Fun VMS Facts? Re: Fun VMS Facts? Re: Fun VMS Facts? Re: Fun VMS Facts? Re: Fun VMS Facts? Re: Fun VMS Facts? Re: Fun VMS Facts? Re: Fun VMS Facts?- Re: how to stop echo on DS90M telnet listener - Re: how to stop echo on DS90M telnet listener - Re: how to stop echo on DS90M telnet listener $ Job logical name table quota problem( Re: Job logical name table quota problem( Re: Job logical name table quota problem( Re: Job logical name table quota problem Re: MVII heat output" Re: Need VMS media - Austin, Texas. OpenVMS and Windows NT Integration for Dummies2 Re: OpenVMS and Windows NT Integration for Dummies2 Re: OpenVMS and Windows NT Integration for Dummies2 Re: OpenVMS and Windows NT Integration for Dummies2 RE: OpenVMS and Windows NT Integration for Dummies OpenVMS Training on CD-ROM Re: OpenVMS Training on CD-ROM= Problam solved (was Re: Job logical name table quota problem) A Re: Problam solved (was Re: Job logical name table quota problem) A RE: Problam solved (was Re: Job logical name table quota problem) 5 Re: Proxy problem: Why does node:: work but 0:: fail? 9 Re: ramdisk vs. file cache, and the winner is, file cache  Re: Recording CD's on VMS * Removing/Stripping RFC headers from emails* Removing/Stripping RFC headers from emails. Re: Removing/Stripping RFC headers from emails. Re: Removing/Stripping RFC headers from emails Replacing the Quorum Disk  Seaching SYSUAF.DAT  Re: Seaching SYSUAF.DAT  Re: Seaching SYSUAF.DAT  Re: Seaching SYSUAF.DAT  re: Searching SYSUAF.DAT Some HW questions  test Re: VAX on Intel? & Re: VMS File Caching Futures and so on VMS Printing using PCL6  Re: VMS Printing using PCL6  Re: VMS Security features  Re: VMS Security features  Re: VMS Security features  Re: VMS Security features  Re: VMS Security features  Re: VMS Security features  Re: VMS Security features  Re: VMS Security features ; Re: Which cpu does Sun server use: I386, Mip, Ppc or Alpha? ; Re: Which cpu does Sun server use: I386, Mip, Ppc or Alpha? ; Re: Which cpu does Sun server use: I386, Mip, Ppc or Alpha? ; Re: Which cpu does Sun server use: I386, Mip, Ppc or Alpha? ; Re: Which cpu does Sun server use: I386, Mip, Ppc or Alpha? ; Re: Which cpu does Sun server use: I386, Mip, Ppc or Alpha? ; Re: Which cpu does Sun server use: I386, Mip, Ppc or Alpha?  ZLXp-E1/E2/E3 Switch Settings   F ----------------------------------------------------------------------  % Date: Wed, 14 Jun 2000 11:44:09 +0200  From: BastlJan@ipb.cz  Subject: %HELPD Message-ID: <399A9DBB0F52D311A3990000F80415B1039AF783@pre028.ipb.cz>   %HELP    ------------------------------   Date: 14 Jun 2000 14:27:04 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: %HELP6 Message-ID: <8i84no$fp5$2@mailint03.im.hou.compaq.com>  \ In article <399A9DBB0F52D311A3990000F80415B1039AF783@pre028.ipb.cz>, BastlJan@ipb.cz writes: :%HELP  C   Depending on your path of access, you have connected to an email  C   server with a large number of subscribers, or to a newsgroup with    a larger number of readers.   E   For assistance and subscription information and general background  G   on INFO-VAX and the comp.os.vms newsgroup and OpenVMS itself, please  B   see the OpenVMS Frequently Asked Questions (FAQ), available via:  "     http://www.openvms.compaq.com/:     http://www.openvms.digital.com/wizard/openvms_faq.html9     ftp://ftp.digital.com/pub/Digital/dec-faq/OpenVMS.txt 1     ftp://ftp.digital.com/pub/Digital/dec-faq/vms :     ftp://rtfm.mit.edu/pub/usenet/news.answers/dec-faq/vms,     comp.answers and news.answers newsgroups  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 14 Jun 2000 14:23:23 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)' Subject: Re: Alternate boot root number 6 Message-ID: <8i84gr$fp5$1@mailint03.im.hou.compaq.com>  G In article <8i6hdu$oq6$1@nnrp1.deja.com>, WalingK@datacom.co.nz writes: E :I have 2 identical Alpha servers (2000 4/275 ) in a VMS cluster (VMS E :7.1-2). SYSTEMA boots of [SYS0] and SYSTEMB of [SYS1]. SYSTEM B is a H :DR/backup system for SYSTEMA. The plan is that if SYSTEMA goes down forB :a longish period, SYSTEMB will be rebooted from its root, [SYS0]. : D :Is there a method to change the boot root number on SYSTEMB without :using the console?   D   There have been a few hacks that alter the SRM console environmentD   variables, but the officially supported solution involves changingD   the console environment variable settings directly at the console.  B   Alternatively, look at how the systems are currently used by theD   application(s), and find out if you can use the cluster alias, theE   batch queues, the distributed lock manager, and related mechanisms  F   (for instance), and keep both systems and both applications running B   in parallel (getting the most from your current configuration atB   all times), with a graceful (and automatic) degradation to fewerC   node(s) on system hardware failure.  (Why reboot your system(s),     if you don't have to?)  C   Also look seriously at your current configuration and your likely G   failure modes, as both SYSTEMA and SYSTEMB are booting from the same  F   files -- the system-specific roots (usually) contain only the node'sG   name, configuration, and parameters -- thus if you encounter a system G   disk problem, both systems are toast.  Mechanisms such as host-based  F   or controller-based volume shadowing (aka mirroring, RAID) can help E   you should there be a disk hardware failure, but the use of volume  I   shadowing nor multiple roots is not relevent to, say, erroneous system  :   file deletions nor to nefarious system file alterations.  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Wed, 14 Jun 2000 11:24:02 -0500 # From: "JHansen" <jhansen@fnbnd.com>   Subject: Analyze/audit  question2 Message-ID: <TjO15.98$KK1.11523@news.corpcomm.net>  G I would like to have the Analyze/audit...... command run every night to B retrieve audit logging information.  I would like this to pull the( information from the previous day.  i.e.C analyze/audit/since=13-jun-2000.......... would be run on the 14th. B Can this be done in a batch file?  If so how ?  Where would I findG information on how to do this?  I have tried the VMS help, but no luck.  Thanks for your time and help  Jeff   ------------------------------  % Date: Wed, 14 Jun 2000 12:56:01 -0400 & From: Ken Robinson <ksrobin@erenj.com>$ Subject: Re: Analyze/audit  question- Message-ID: <3947B921.D7375E2B@fpe.erenj.com>    JHansen wrote: > I > I would like to have the Analyze/audit...... command run every night to D > retrieve audit logging information.  I would like this to pull the* > information from the previous day.  i.e.E > analyze/audit/since=13-jun-2000.......... would be run on the 14th. D > Can this be done in a batch file?  If so how ?  Where would I findI > information on how to do this?  I have tried the VMS help, but no luck.  > Thanks for your time and help   G What you want to specify here is "/since=yesterday". Look in HELP under  DCL_TIPS Date_Time.  Yes, it can be done in batch.    Ken Robinson! OpenVMS System Manager Consultant  ksrobin@erenj.com    ------------------------------  % Date: Wed, 14 Jun 2000 13:20:36 -0400 * From: "linux" <linuxmtl@-NOSPAM-yahoo.com>$ Subject: Re: Analyze/audit  question+ Message-ID: <8i8ejp$6iu$1@pr0n.openface.ca>    Just try....   analyze/audit/since=yesterday     ; "JHansen" <jhansen@fnbnd.com> a crit dans le message news: ' TjO15.98$KK1.11523@news.corpcomm.net... I > I would like to have the Analyze/audit...... command run every night to D > retrieve audit logging information.  I would like this to pull the* > information from the previous day.  i.e.E > analyze/audit/since=13-jun-2000.......... would be run on the 14th. D > Can this be done in a batch file?  If so how ?  Where would I findI > information on how to do this?  I have tried the VMS help, but no luck.  > Thanks for your time and help  > Jeff >  >    ------------------------------  % Date: Wed, 14 Jun 2000 18:38:06 +0100 * From: "Richard Brodie" <R.Brodie@rl.ac.uk>$ Subject: Re: Analyze/audit  question, Message-ID: <8i8ftt$15d2@newton.cc.rl.ac.uk>  [ "JHansen" <jhansen@fnbnd.com> wrote in message news:TjO15.98$KK1.11523@news.corpcomm.net... I > I would like to have the Analyze/audit...... command run every night to D > retrieve audit logging information.  I would like this to pull theP > information from the previous day.  i.e. 13-jun-2000 would be run on the 14th.  K It's not obvious where you are stuck on. What in particular is the problem?  Here are a few pointers:  B You can specify the previous day as 'YESTERDAY'. See help specify.  H Running it as a batch job only involves creating a one line command fileI consisting of $Analyze/audit...... and using submit to send it to a batch  queue.  D For regularly running a batch job, this has been discussed recently;J either use some cron like scheduling tool or have the job resubmit itself.   ------------------------------  % Date: Wed, 14 Jun 2000 10:17:00 +0100 2 From: Chris Sharman <Chris.Sharman@CCAgroup.co.uk> Subject: Re: Backup performance - Message-ID: <39475B9C.2F332D4@CCAgroup.co.uk>   @ >>We have a full-ish RAID 230 box, and some older disks (due forI >>replacement), all backed up with a TZ87. Advice from here some time ago ) >>has got the backup running pretty well.  >>I >>We do backup & defrag every night, for most of the night, which used to F >>be OK. Increasingly now we're running 24 hours, though, and could do! >>without the performance impact.  >>H >>I wondered if there was some better way - perhaps to have the RAID boxG >>do the work, rather than VMS ? (Having seen how blisteringly fast the C >>RAID box bring a new shadow set member online, without noticeably C >>degrading performance, it looks like a better hardware solution).   H The TZ87 is DLT (3 I think). It runs fine - to the extent that the disksH (or the bus) seem maxed out - we don't have any problem with tape speed.2 We're seeing OPCOM messages about the quorum disk:" timed-out operation to quorum disk  lost "connection" to quorum diskF This is one of the older disks, purely there as a casting vote. I keepE meaning to give that vote to our AlphaStation, purely to resolve this D nightly chatter - we're living with it on the grounds that quorum isG maintained & it's just chatter (unless someone knows different?). We're ; also seeing access to other disks very severely degraded by ' backup/defrag - possibly mainly defrag.  No DB - just RMS. E Currently we rotate around some 17 disks, backing up over half fully, B and the rest incrementally. The slowest backup is the one we do do< disk->disk: the one containing SYSUAF and other vital stuff.  H The problem is our time 'window' isn't really there for much of the year9 any more - no quiet times, not at night, not at weekends.    Chris    ------------------------------  % Date: Wed, 14 Jun 2000 13:25:21 +0100 2 From: Roy Omond <Roy.Omond@BlueBubble.demon.co.uk>+ Subject: Re: blocks incorrectly marked free 6 Message-ID: <394779B1.FA9EA606@BlueBubble.demon.co.uk>   "Jean-Franois Marchal" wrote:   > To summarize,  > ? > 1.- the AES macro example compiles with several "INFORMATION" + >     messages that do not sound very sane.  > C > 2.- the program works ok on an ODS2 disk , but NOT on an ODS5 one = >     (the block #0 is reported as not belonging to any file)  > C > 3.- the block allocation problem survives to disconnection of all * >     PC clients and dismount of the disk. > E > 4.- same for incorrect busy headers that simply appear as incorrect  >     headers for DUMP.  > B > I think I will not be able to identify the source of the problemD > without a correction of the FIND_LBN program to support ODS5, whatD > I'm totaly unable to do. (You may consider it is an incantation to > VAXMAN ;-)  	 Eh bien !   = Consider using DFU to locate files containing a specific LBN. ( Ah DFU - much better than sliced bread !  	 Roy Omond  Blue Bubble Ltd.   ------------------------------   Date: 14 Jun 2000 17:42:14 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)( Subject: C 6.2-007 compilation annoyance, Message-ID: <8i8g5m$shd@gap.cco.caltech.edu>   Anybody else seeing this?     C Of what possible use is the informational message in this example?     $ cc/ver) Compaq C V6.2-007 on OpenVMS Alpha V7.2-1  $! $! $! $ create killme.c  #include <stdlib.h>  #include <stdio.h> int main(void){    (void) printf("Hello\n");  }  $! $! $!8 $ cc/standard=ansi89/prefix=all/warn=(enable=all) killme #include <decc$types.h>  .^@ %CC-I-UNUSEDINCL, This nested include file appears to be unused.E at line number 26 in module STDIO of text library SYS$COMMON:[SYSLIB]  DECC$RTLDEF.TLB;4   5 Programs which generate this message link and run ok. 3 Some .c files give multiple copies of this message.    Regards,   David Mathog mathog@seqaxp.bio.caltech.edu ? Manager, sequence analysis facility, biology division, Caltech     ------------------------------  % Date: Wed, 14 Jun 2000 09:13:54 -0400 ( From: Ed Vogel <edward.vogel@compaq.com>8 Subject: Re: C bashing (was Re: VMS File Caching Futures* Message-ID: <39478512.6932BE82@compaq.com>   Wayne Sewell wrote:    > J > Fantastic!!  I didn't know about this.  Is this only with 6.2 and later? >   D Since V5.7 the C compiler team has made an effort to add code to theG compiler that will detect potential coding errors.  The questcode group A was added in V5.7.   Each release since then has added additional = messages to this group.   Of the three  cases mentioned in my C earlier posting, the only one that is new for V6.2 is the detection  of nested comments.     @ I would suggest that all code be compiled /WARN=ENABLE=QUESTCODE+ at some point during the development cycle.   C We (the Compaq C compiler team) seek input on other coding mistakes A that the compiler could catch.  While you can post them here, the E best method to request a feature is to fill-in the product suggestion  form at:  ; http://www.openvms.digital.com/commercial/c/c_feedback.html     <                                                     Ed VogelH                                                     Compaq C Engineering   ------------------------------   Date: 14 Jun 2000 13:59:01 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: Chinese VMS6 Message-ID: <8i8335$fb1$1@mailint03.im.hou.compaq.com>  _ In article <39468234.8BE01E04@email.sps.mot.com>, Steven Xie <r33300@email.sps.mot.com> writes: H :Have a question regarding to the double byte font display. We are doingC :some testing on VMS V7.2-1. There is an application running on the D :system need display chinese (double byte font). My question is if IG :don't install chinese software (chinese vms and chinese motif) to VMS, E :can I be able to display chinese when I running that application (on ) :some workstation with chinese support)?    E   AFAIK, nope.  The Asian languages variants of OpenVMS provide this.OJ   Base OpenVMS has knowledge of wide characters and Unicode, but probably ;   not enough for what this application needs/expects/wants.o  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 14 Jun 2000 11:28:39 GMT- From: "Phil Tregoning" <ptregoni@esoc.esa.de>s+ Subject: Re: CMQ/Dec with a sense of Humor?u= Message-ID: <01bfd5f2$22c37520$4b53b083@ptregoni.esoc.esa.de>n  ? Antony Wardle <antony.wardle@nospam.met.co.nz> wrote in articlea) <Z8Z05.14109$N4.513845@ozemail.com.au>...o > While booting an alpha, (7.1)e > , > I wasn't sure if I had the hold screen, so+ > I thought that I would try a control T tof > see if there was any life. > " > Well, I got a message up sayine: > " > dead_eater waiting for dead_beef >   7 When the security server starts up, you get the messageD  6 %SECSRV-I-CIASTARTINGUP, breakin detection and evasion processing now starting up  . Which I found amusing the first time I saw it.   Phil   ------------------------------  % Date: Wed, 14 Jun 2000 12:31:48 +0100?  From: steven.reece@quintiles.com+ Subject: Re: CMQ/Dec with a sense of Humor?o> Message-ID: <802568FE.003F70E6.00@qedilc01.qedi.quintiles.com>  6 I believe there is a KGB reference too somewhere......   Phil Tregoning wrote:b: >>>When the security server starts up, you get the message  6 %SECSRV-I-CIASTARTINGUP, breakin detection and evasion processing now starting up  1 Which I found amusing the first time I saw it.<<<D   ------------------------------  % Date: Wed, 14 Jun 2000 08:59:46 -0400 # From: Jim Agnew <agnew@hsc.vcu.edu>e+ Subject: Re: CMQ/Dec with a sense of Humor?i+ Message-ID: <394781C2.5779D86F@hsc.vcu.edu>r  < Kevin G. Barkes!!! author, dcl dialoge in the old dec rag...  ! steven.reece@quintiles.com wrote:i > 8 > I believe there is a KGB reference too somewhere...... >  > Phil Tregoning wrote:o< > >>>When the security server starts up, you get the message > 8 > %SECSRV-I-CIASTARTINGUP, breakin detection and evasion > processing now starting up > 3 > Which I found amusing the first time I saw it.<<<o   ------------------------------   Date: 14 Jun 2000 14:36:32 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)+ Subject: Re: CMQ/Dec with a sense of Humor?g6 Message-ID: <8i859g$fp5$4@mailint03.im.hou.compaq.com>  a In article <802568FE.003F70E6.00@qedilc01.qedi.quintiles.com>, steven.reece@quintiles.com writes: 7 :I believe there is a KGB reference too somewhere......n     KGB: Key Grant Block.i  -   CIA: Compound Intrusion Attempt (or Audit).e  +   NSA: Non-discretionary Security Auditing.a  *   The Microfortnights unit of measurement.  '   FUBAR: Failed Unibus Address Register   C   The famous (paraphrasing here) "Hairy code, like its author" and iB   "if the lights go out, does the shadowset dissolve?" comments in   the source code.  B   The "What city, plez?" comment lurking in the directory services!   SYSAP on older versions of SCS.e  =   More than a few chuckles are lurking in the OpenVMS code...s   :Phil Tregoning wrote:; :>>>When the security server starts up, you get the messagea : : :%SECSRV-I-CIASTARTINGUP, breakin detection and evasion...  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Wed, 14 Jun 2000 15:23:53 GMT  From: d.webb@mdx.ac.uk+ Subject: Re: CMQ/Dec with a sense of Humor?8( Message-ID: <8i881k$1h$1@nnrp1.deja.com>  + In article <394781C2.5779D86F@hsc.vcu.edu>,4&   Jim Agnew <agnew@hsc.vcu.edu> wrote:> > Kevin G. Barkes!!! author, dcl dialoge in the old dec rag... >l# > steven.reece@quintiles.com wrote:e > >A: > > I believe there is a KGB reference too somewhere...... > >.  0 Macro library $KGBDEF  or C header file KGBDEF.H    1 Used to specify attributes of Rights Identifiers.  (Key Grant Block definitions)   - See for instance the $GRANTID system service.s  ! Bit Position     Meaning When Seto  C KGB$V_DYNAMIC    Allows holders of the identifier to remove it frommC                  or add it to the process rights database using they-                  DCL command SET RIGHTS_LIST.t  C KGB$V_NOACCESS   Makes any access rights of the identifier null and/E                  void. This attribute is intended as a modifier for a.?                  resource identifier or the Subsystem attribute:  E KGB$V_RESOURCE   Allows holders of an identifier to charge disk spacenE                  to the identifier. It is used only for file objects.i  ? KGB$V_SUBSYSTEM  Allows holders of the identifier to create andM?                  maintain protected subsystems by assigning the-?                  Subsystem ACE to the application images in thea                  subsystem.)      H So on VMS the CIA look after breakins and the KGB control your rights :)      
 David Webb VMS and Unix team leader CCSS Middlesex University   > > Phil Tregoning wrote:v> > > >>>When the security server starts up, you get the message > >1: > > %SECSRV-I-CIASTARTINGUP, breakin detection and evasion > > processing now starting up > >.5 > > Which I found amusing the first time I saw it.<<<p >o    & Sent via Deja.com http://www.deja.com/ Before you buy.h   ------------------------------  % Date: Wed, 14 Jun 2000 16:36:19 +0100c2 From: Roy Omond <Roy.Omond@BlueBubble.demon.co.uk>+ Subject: Re: CMQ/Dec with a sense of Humor?t6 Message-ID: <3947A673.3C040E11@BlueBubble.demon.co.uk>   Dan Sugalski wrote:   , > At 12:24 AM 6/12/00 -0400, JF Mezei wrote: > >Antony Wardle wrote:w0 > > > I wasn't sure if I had the hold screen, so/ > > > I thought that I would try a control T tol  > > > see if there was any life. > > > & > > > Well, I got a message up sayine: > > >i& > > > dead_eater waiting for dead_beef > >iP > >I know that at least one DEC engineer was a fan of Monthy Python. Sounds likeG > >a reference to a famous sketch (where one is invited to eat his dead  > >mother in law). >aJ > More likely it's looking for a quadword with the value 0xDEADBEEF in it.L > One of the fun words and phrases you can actually spell entirely in hex...  Q I can't remember which system it was (ICL ? Emas ? mid 1970's), but uninitialisedpP memory was set to xF00l  (spells "fool" !).  Used to make reading dumps a little bit-, easier.  Must admit, xDEADBEEF is better :-)  	 Roy Omonde Blue Bubble Ltd.   ------------------------------   Date: 14 Jun 2000 16:09:56 GMT- From: "Phil Tregoning" <ptregoni@esoc.esa.de>t+ Subject: Re: CMQ/Dec with a sense of Humor?o= Message-ID: <01bfd619$6df40ba0$4b53b083@ptregoni.esoc.esa.de>s  = Hoff Hoffman <hoffman@xdelta.zko.dec.nospam> wrote in articlef- <8i859g$fp5$4@mailint03.im.hou.compaq.com>...- >  >   KGB: Key Grant Block.- > / >   CIA: Compound Intrusion Attempt (or Audit).M > - >   NSA: Non-discretionary Security Auditing.- > , >   The Microfortnights unit of measurement. > ) >   FUBAR: Failed Unibus Address Registerd > E >   The famous (paraphrasing here) "Hairy code, like its author" and eD >   "if the lights go out, does the shadowset dissolve?" comments in >   the source code. > D >   The "What city, plez?" comment lurking in the directory services# >   SYSAP on older versions of SCS.  >   > I did hear that the last gasp datagram a node attempts to send@ over SCS right before it crashes consists of the text "AAAAARGH"   Phil   ------------------------------  % Date: Wed, 14 Jun 2000 09:18:46 -0700f From: LBohan@spam_less..dbc.com-+ Subject: Re: CMQ/Dec with a sense of Humor? 8 Message-ID: <dtbfkschdbsb1l6ie6tk2fsqem5acfefrb@4ax.com>  E On Wed, 14 Jun 2000 12:31:48 +0100, steven.reece@quintiles.com wrote:E   >e >e7 >I believe there is a KGB reference too somewhere......e >a >Phil Tregoning wrote:; >>>>When the security server starts up, you get the messageI >u7 >%SECSRV-I-CIASTARTINGUP, breakin detection and evasion  >processing now starting upi >=2 >Which I found amusing the first time I saw it.<<< >   1 %EXECINIT-S-CIAO  is one I got a kick out of ....l   ...d( %LOADER-I-INIT, initializing SYS$NTA.EXE% %LOADER-I-INIT, initializing SSPI.EXEh% %LOADER-I-INIT, initializing ACME.EXEu1 %EXECINIT-I-TMVPTES, setting TMV up with its PTEsn, %EXECINIT-I-SVAPTE, setting up EXE$GL_SVAPTE2 %EXECINIT-I-BLD_LOCKID, building the lock id table9 %EXECINIT-I-BLD_RESHASH, building the resource hash table-< %EXECINIT-I-RESMEMALC, performing reserved memory allocation- %EXECINIT-I-CLRWARM, clearing warm-start flag:- %EXECINIT-I-CLRCOLD, clearing cold-start flagj) %EXECINIT-I-ACTIVESET, joining ACTIVE set / %EXECINIT-S-CIAO, transferring to the schedulerj9 %SWAPPER-I-SHUFFLE, executing SWAPPER initialization codeK2 %SWAPPER-I-SYSDISK, checking status of system disk( %SWAPPER-I-POOL, initializing paged pool0 %SWAPPER-I-INIT, calling initialization routines2 %LOADER-I-INIT, initializing SYSTEM_PRIMITIVES.EXE ...p   ------------------------------  % Date: Wed, 14 Jun 2000 10:29:42 -0700n! From: Shane.F.Smith@Healthnet.comi+ Subject: Re: CMQ/Dec with a sense of Humor?iC Message-ID: <OF645F79B4.6390A6E7-ON882568FE.005FF67D@HEALTHNET.COM>n  J Ok, so it's not a part of VMS, but I have a server here called Doris whoseJ shutdown packet consists of the string "die screaming with sharp things inK your head". It's a reference to a movie title mentioned in the Red Dwarf TV8 series.2   Shanec          ? Phil Tregoning <ptregoni@esoc.esa.de> on 06/14/2000 09:09:56 AMK   To:   Info-VAX@Mvb.Saic.Com  cc:o  , Subject:  Re: CMQ/Dec with a sense of Humor?        = Hoff Hoffman <hoffman@xdelta.zko.dec.nospam> wrote in articlek- <8i859g$fp5$4@mailint03.im.hou.compaq.com>...i >  >   KGB: Key Grant Block.d > / >   CIA: Compound Intrusion Attempt (or Audit).e >u- >   NSA: Non-discretionary Security Auditing.e >e, >   The Microfortnights unit of measurement. >d) >   FUBAR: Failed Unibus Address Registern > D >   The famous (paraphrasing here) "Hairy code, like its author" andD >   "if the lights go out, does the shadowset dissolve?" comments in >   the source code. > D >   The "What city, plez?" comment lurking in the directory services# >   SYSAP on older versions of SCS.h >o  > I did hear that the last gasp datagram a node attempts to send@ over SCS right before it crashes consists of the text "AAAAARGH"   Phil   ------------------------------  % Date: Wed, 14 Jun 2000 14:08:25 +0200g9 From: "John Jenniskens" <j.jenniskens@HumanInference.com> " Subject: Command syntax for BACKUP$ Message-ID: <5BK15.2$p34.115@client>   Hello,  I I'm an unexpierenced OpenVMS user, and I have a question about the backup& command.F I want to make a backup of a directory tree, with all the files in it,K afterwards I want to restore the tree, but in another directory. (The finalm( goal is on the customers OpenVMS system)D In Unix I use the tar command, but since tar is not available on all5 OpenVMS -systems,  I want to use the BACKUP  command.b  E Can anyone give me the syntax for the backup and the restore command.   + I've tried it, using the folowing commands:l	 show defa    HOME:[000000.JOHN]  J I want to backup everything in the [.sushi] directory (so the whole tree), in the file john.sav  * backup [.sushi...]*.*;* john.sav /save_set  1 The  backup /list john.sav /save has this result:o Save set:          JOHN.SAVn Written by:        SYSTEMn" UIC:               [000001,000004]* Date:              14-JUN-2000 12:22:27.45< Command:           BACKUP [.SUSHI...]*.*;* JOHN.SAV/SAVE_SET+ Operating system:  OpenVMS AXP version V7.1  BACKUP version:    V6.2d CPU ID register:   80000000s! Written on:        _AS250$DKB400:l Block size:        32256 Group size:        10r Buffer count:      53   J [JOHN.SUSHI]TEST1.DIR;1                                     1  14-JUN-2000 12:13sJ [JOHN.SUSHI.TEST1]TEST1FILEA.;2                             1  14-JUN-2000 12:16 J [JOHN.SUSHI.TEST1]TEST1FILEB.;1                             1  14-JUN-2000 12:16hJ [JOHN.SUSHI]TEST2.DIR;1                                     1  14-JUN-2000 12:13nJ [JOHN.SUSHI.TEST2]TEST2FILEA.;1                             1  14-JUN-2000 12:17-J [JOHN.SUSHI.TEST2]TEST2FILEB.;1                             1  14-JUN-2000 12:17bJ [JOHN.SUSHI]TEST3.DIR;1                                     1  14-JUN-2000 12:14BJ [JOHN.SUSHI.TEST3]TEST3FILEA.;1                             1  14-JUN-2000 12:17iJ [JOHN.SUSHI.TEST3]TEST3FILEB.;1                             1  14-JUN-2000 12:18e  E When I want to put the files in another directory I used the command:n' backup john.sav/save [.susrest...]*.*;*  The result was:s  %BACKUP-E-OPENOUT, error opening/ HOME:[000000.JOHN.SUSREST.JOHN.SUSHI.TEST1]TESTr 1FILEA.;2 as outputn -RMS-E-DNF, directory not founds" -SYSTEM-W-NOSUCHFILE, no such file  %BACKUP-E-OPENOUT, error opening/ HOME:[000000.JOHN.SUSREST.JOHN.SUSHI.TEST1]TEST  1FILEB.;1 as outputl -RMS-E-DNF, directory not foundc" -SYSTEM-W-NOSUCHFILE, no such file  %BACKUP-E-OPENOUT, error opening/ HOME:[000000.JOHN.SUSREST.JOHN.SUSHI.TEST2]TEST  2FILEA.;1 as output  -RMS-E-DNF, directory not foundh" -SYSTEM-W-NOSUCHFILE, no such file  %BACKUP-E-OPENOUT, error opening/ HOME:[000000.JOHN.SUSREST.JOHN.SUSHI.TEST2]TEST  2FILEB.;1 as outputl -RMS-E-DNF, directory not found-" -SYSTEM-W-NOSUCHFILE, no such file  %BACKUP-E-OPENOUT, error opening/ HOME:[000000.JOHN.SUSREST.JOHN.SUSHI.TEST3]TESTa 3FILEA.;1 as outputg -RMS-E-DNF, directory not found." -SYSTEM-W-NOSUCHFILE, no such file  %BACKUP-E-OPENOUT, error opening/ HOME:[000000.JOHN.SUSREST.JOHN.SUSHI.TEST3]TESTr 3FILEB.;1 as outputn -RMS-E-DNF, directory not found!" -SYSTEM-W-NOSUCHFILE, no such file   ------------------------------  9 Date: Wed, 14 Jun 2000 13:50:33 +0100 (GMT Daylight Time)c From: Andy.Harper@kcl.ac.uka& Subject: Re: Command syntax for BACKUP1 Message-ID: <SIMEON.10006141333.S@odin.kcl.ac.uk>   3 On Wed, 14 Jun 2000 14:08:25 +0200 John Jenniskens  ( <j.jenniskens@HumanInference.com> wrote:   > Hello, > K > I'm an unexpierenced OpenVMS user, and I have a question about the backup 
 > command.H > I want to make a backup of a directory tree, with all the files in it,M > afterwards I want to restore the tree, but in another directory. (The final * > goal is on the customers OpenVMS system)F > In Unix I use the tar command, but since tar is not available on all7 > OpenVMS -systems,  I want to use the BACKUP  command.s > G > Can anyone give me the syntax for the backup and the restore command.  > - > I've tried it, using the folowing commands:y > show defa  >   HOME:[000000.JOHN] > L > I want to backup everything in the [.sushi] directory (so the whole tree), > in the file john.sav > , > backup [.sushi...]*.*;* john.sav /save_set >   <    Suppose you want to save all the files in [sushi...] and !    restore them to [newsushi...]:n  
    Save with:t'      backup [sushi...]    john.sav/save       Restore with:9      backup john.sav/save/select=[sushi...] [newsushi...]i  C    The non-wildcarded leading part of the source directory will be  ?    lexically replaced by the equivalent non-wildcarded leading s    part of the destination.l  A    In this example, it will replace 'sushi' with 'newsushi'; all a>    subsequent parts of the directory pathname remain the same.  C    Remember that if you backup using a relative directory spec, as      in:       backup [.sushi...]  ...   A    The COMPLETE directory path is actually saved in the saveset. q=    So, when you do the restore, the /SELECT qualifier should nA    specify the full absolute path up to the point where you want a    to change it.  B     backup john.sav/save/select=[full.path.sushi...] [newsushi...]   Regards,   Andy Harper0 King's College Londont ----------------------   Andy.Harper@kcl.ac.ukv   ------------------------------  % Date: Wed, 14 Jun 2000 18:11:09 +0100   From: steven.reece@quintiles.com3 Subject: DECnet-Plus and DECnet task to task from Ci> Message-ID: <802568FE.005E812E.00@qedilc01.qedi.quintiles.com>  J A few weeks ago there was a discussion here which included a C program offD DSNlink.  The program used the "classic" task construction in C as :  7 fp=open("node::\"task=taskname\"", O_RDWR,0,"ctx=bin");e" write (fp, buffer_send, BUFFSIZE); read (fp,buffer_read,BUFFSIZE);  close (fp);r  P Although this all works tickety boo when I'm going from my VAXstation 4000-90 toO my AlphaServer 4100s, doing the same thing from the 4100s to the 4100s presents-K problems on some occasions.  The two 4100s are clustered and have a clusterd> alias defined, the alias being the target for the "task=....."  J On some occasions, it appears that the client is "missing" the end of fileP record which is explicitly sent by the server.  The client (the chunk of exampleP code above is from the client) then sits there consuming cpu trying to do a readL which it's never going to be able to since the server has already closed the8 connection, apparently without an error being generated.  M Any ideas how I can either do investigation or stop the cause of the problem?   P The Alphas are presently running VMS 7.1-1H2 with DECnet Plus v7.1 ECO 02 whilst9 the VAXstation is running DECnet Plus v7.1 with VMS v7.1.A   Steve.   ------------------------------  % Date: Wed, 14 Jun 2000 09:19:57 +0100m  From: steven.reece@quintiles.com& Subject: DSM (Was: Re: Fun VMS Facts?)> Message-ID: <802568FE.002DE0E7.00@qedilc01.qedi.quintiles.com>  O I think the DSM subject is a bit like buses - you don't hear of it for ages andF  ages and then two come together.E The other day I came across a letter indicating that although DigitaleN transferred the product in 1995 they were still selling licenses for it at theL time of the Compaq buyout.  From May 1999 they'd no longer sell licenses and: from May 2000 you couldn't get support for it from Compaq.  = Then again, I never used it and probably never will..........    Chris Casey wrote:O >>>If you are talking about the same benchmark that I saw I think that you will H find that another ex-Digital product on the same test run beat Rdb handsH down. This was DSM, now owned by Intersystems, and still producing worldH beating price performance on Alpha systems. In a typical show of DigitalJ double standards and poor marketing decisions the DSM figures were quietlyG forgotten about because Digital was trying to push Rdb at that time.<<<a   ------------------------------  # Date: Wed, 14 Jun 2000 17:11:55 GMT- From: mccrobie@my-deja.com3 Subject: Re: Files-11 ODS-2 Readability for FreeBSDa) Message-ID: <8i8ecb$5f9$1@nnrp1.deja.com>:  - In article <3946BC5F.D7481BF8@earthlink.net>,8:   "David J. Dachtera" <djesys.nospam@earthlink.net> wrote: > mccrobie@my-deja.com wrote:iH > > > > and I don't see any need for writing ODS-2 on FreeBSD to be read by? > > > > OpenVMS.  Why would you recommend adding write support?: > > >nF > > > ...and if you could dual boot *BSD/Linux and OpenVMS? (...and of > > course, F > > > you can!), would you not want to share data between the o.s.-es? > > >s > >  > > ISO9660 isn't good enough? >"( > Can you write ISO-9660 on a hard disk? >  > -- > David J. Dachterad > dba DJE Systemsm$ > http://home.earthlink.net/~djesys/ >a< > Unofficial Affordable OpenVMS Home Page and Message Board:- > http://home.earthlink.net/~djesys/vms/soho/  >     B To a hard disk?  A couple of ideas come to mind.  Use some type ofH container file system available for both Unix and OpenVMS.  OpenVMS withG the LD driver shouldn't be a problem, although I haven't actually tried-H the LD driver with an ISO9660 container file - still, should work.  UnixD would be the issue since "container file systems" seem to be a "new" idea there.-  H Nothing in ISO9660, as far as I remember, prevents it from being written! to a "raw" hard drive.  Consider:   $ UNIX:  mkisofs ... myisofile.iso9660+        dd if=myisofile.iso9660 of=/dev/rda1@  2        where rda1 eventually translates to DKB100:              rda1 on FreeBSD"              rrz0c on Digital Unix  ; OpenVMS:  mount /media=cd /noassist /system DKB100: cdlabelh  C If hard disk isn't your thing, then a CD-RW would do:  just blast aEA ISO9660 file system onto a CD-RW using a CD-writer.  CDRECORD andhF MKISOFS are available for both Unix and OpenVMS.  Use CD-RW so that it can be erased and reused.    Issues include:M#     File names from Unix -> OpenVMSA#     File attributse OpenVMS -> Unixm  - but I would imagine such could be worked out.r    & Sent via Deja.com http://www.deja.com/ Before you buy.d   ------------------------------  # Date: Wed, 14 Jun 2000 17:18:05 GMTn From: mccrobie@my-deja.com3 Subject: Re: Files-11 ODS-2 Readability for FreeBSD ) Message-ID: <8i8enq$5qv$1@nnrp1.deja.com>r  E Well, that is true.  Perhaps an open source version without automaticn> record/file conversion and a payware version with - perhaps...  @ But my original post still stands - there doesn't seem to be any7 interest, open source or payware, not even for testing.i   Chuck McCrobie  - In article <3946BD9F.E6580580@earthlink.net>,y:   "David J. Dachtera" <djesys.nospam@earthlink.net> wrote: > mccrobie@my-deja.com wrote:hD > > I wasn't planning on doing open source.  It seems most companies want@ > > some type of purchasable before committing to use a product. >TE > PKware has had both shareware and payware versions almost since theeB > beginning. Something to think about, with Open Source instead ofG > shareware. Naturally, there should be a subtle difference between theh3 > two versions to make the payware more attractive.  >p > > ThisC > > may/may not include support contracts.  My goal is to provide ae product H > > that someone can use to access their OpenVMS written data on anotherH > > platform (FreeBSD in this case) in a "native file system mode".  HowB > > many in this newsgroup would use open source software in theirF > > production environment without a support contract?  How many would use 6 > > open source software even with a support contract? >tA > If the company does not object, why not? I even know of a place  runningo& > CMU/IP on a mission critical system! >eG > > Yes, I would agree that the only way to make money from open sourcer isG > > to have a good reference for the next job.  I wish to make money bywF > > doing what I know best - writing software.  However, it seems that this1 > > particular product is not the way to do that.  >iH > I dunno. Phil Katz, rest his soul, did pretty well for himself, though) > perhaps he never believed that himself.w >( > -- > David J. Dachterau > dba DJE Systemse$ > http://home.earthlink.net/~djesys/ > < > Unofficial Affordable OpenVMS Home Page and Message Board:- > http://home.earthlink.net/~djesys/vms/soho/B >l    & Sent via Deja.com http://www.deja.com/ Before you buy.    ------------------------------  % Date: Tue, 13 Jun 2000 20:49:47 -0400C* From: David A Froble <davef@tsoft-inc.com> Subject: Re: Fun VMS Facts?t- Message-ID: <3946D6AB.F8518859@tsoft-inc.com>d   Bill Todd wrote: > L > The '18 years without a reboot' story from the Irish National Railway is aJ > good one, though someone else would have to provide the details (kind ofM > makes you wonder about modern systems that seem to accept 'memory leaks' ast > a way of life...). > J > VMS V1.0 would run (and not much else) in 256KB of memory using (IIRC) aL > pair of 30 MB disks.  For that matter, it wouldn't surprise me if VMS V7.2J > were a *lot* smaller (and required considerably less physical memory andE > disk space) than, say, Win2K (again, you'd have to dig up details).v   $ syH OpenVMS V7.2  on node DFE60  13-JUN-2000 20:36:47.26  Uptime  8 08:08:34M   Pid    Process Name    State  Pri      I/O       CPU       Page flts  Pages M 00000041 SWAPPER         HIB     16        0   0 00:00:00.33         0      0mM 00000044 LANACP          HIB     13       44   0 00:00:00.17       380    825sM 00000085 _LTA5003:       LEF      4      181   0 00:00:01.01      1412    3473M 00000046 IPCACP          HIB     10        7   0 00:00:00.05        98    174 M 00000047 ERRFMT          HIB      8     5133   0 00:00:06.22       160    223cM 00000049 OPCOM           HIB      7      832   0 00:00:00.79       414    152aM 0000004A AUDIT_SERVER    HIB     10       52   0 00:00:00.30       531    818uM 0000004B JOB_CONTROL     HIB     10       81   0 00:00:00.23       237    375uM 0000004C QUEUE_MANAGER   HIB      9      211   0 00:00:01.10       792   1141nM 0000004D SECURITY_SERVER HIB     10       54   0 00:00:00.91      1175   1317EM 0000004F NETACP          HIB     10      288   0 00:00:00.51       186    387 P 00000050 EVL             HIB      6       48   0 00:00:00.21       381    431  NM 00000051 REMACP          HIB      8       13   0 00:00:00.04       122     77 M 00000052 LATACP          HIB     14        9   0 00:00:00.07       308    259,M 00000053 TCPIP$INET_ACP  HIB     10      119   0 00:00:00.54       703    738IM 00000054 SYMBIONT_1      HIB      6      124   0 00:00:00.79       451     64 M 000000A2 DFE             CUR      4     5960   0 00:00:21.18      7577    381P $ sh me,@               System Memory Resources on 13-JUN-2000 20:37:21.10  tL Physical Memory Usage (pages):     Total        Free      In Use    ModifiedL   Main Memory (40.00Mb)            81920       58814       21764        1342  @L Virtual I/O Cache Usage (pages):   Total        Free      In Use     MaximumL   Cache Memory                      2994          15        2979       58352   L Slot Usage (slots):                Total        Free    Resident     SwappedL   Process Entry Slots                 40          22          18           0L   Balance Set Slots                   36          20          16           0  lL Dynamic Memory Usage (bytes):      Total        Free      In Use     LargestL   Nonpaged Dynamic Memory        1373184      138752     1234432        6976L   Paged Dynamic Memory            780288      510016      270272      507888  lL Paging File Usage (pages):                      Free  Reservable       TotalL   DISK$VMS072:[SYS0.SYSEXE]SWAPFILE.SYS         1200        1200        1200L   DISK$VMS072:[SYS0.SYSEXE]PAGEFILE.SYS        73696       39667       73696  eO Of the physical pages in use, 11959 pages are permanently allocated to OpenVMS.i  H So, we're looking at about 6 MB permenantly allocated to VMS, and with 2P interactive users logged in, DECnet and TCP/IP running, 11 MB of memory in use. N This is a VAXstation 4000 model 60.  I'd expect an Alpha to be a bit more of a hog.  L Physical Memory Usage (pages):     Total        Free      In Use    ModifiedL   Main Memory (64.00Mb)             8192        2488        4865         839  N In this case, the Alpha is using almost 3/4 of the 64 MB of memory.  Note thatM neither system is set up in any way to save on memory usage, if anything, the4O opposite is true.  The Alpha was running a bit more than the VAX at the time ofn& the snapshot, and more TCP/IP options.  L Not hard to claim that this mature, robust, secure, reliable, etc multi-userM operating system does it's thing using less resources than windoz.  One of myoP people was setting up a server for windoz 2000, and claimed a need for 256 MB of memory.   J > Rdb on VMS held the TPC record back in the mid-'90s (before DEC and then3 > Compaq stopped spending money on VMS benchmarks).  > N > And clustering is now the talk of the town, despite having been pioneered 16K > years ago by VMS.  (Who knows - will record managers be next?  Of course,t- > VMS can't claim to have pioneered those...)n >  > - bill > M > <jbecker@ui.urban.org> wrote in message news:8i5mdp$4eg$1@nnrp1.deja.com...n8 > > I'm putting together an orientation presentation for= > > internal end users who are new to OpenVMS Alpha. For part < > > of the presentation, I'm going to include material along7 > > the lines of "What's OpenVMS" and "What's an Alpha"d: > > because most of these people are completely unfamiliar > > with either. > > : > > Here's where you come in. What fun VMS facts would you; > > include in such an orientation? I'm looking for tidbits : > > that can be stated briefly and effectively to (in this: > > situation) a general but educated audience. I want the1 > > items to show that VMS is a good place to be.- > >09 > > I have several ideas of my own, but I thought I'd ask  > > the VMS community. > >2 > > -- > > Jim Becker/ > > The Urban Institute (http://www.urban.org/)U7 > > DECUS ESILUG (http://eisner.decus.org/lugs/esilug/)  > >  > > * > > Sent via Deja.com http://www.deja.com/ > > Before you buy.    --  4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com6 T-Soft, Inc.  170 Grimplin Road  Vanderbilt, PA  15486   ------------------------------  % Date: Wed, 14 Jun 2000 08:18:51 +0100[5 From: "Chris Casey" <Chris.Casey@HendersonNOSPAM.com>, Subject: Re: Fun VMS Facts?J) Message-ID: <8i7bl4$j7e$1@soap.pipex.net>;  : Bill Todd wrote in message <8i64sl$gsb$1@pyrite.mv.net>...I >Rdb on VMS held the TPC record back in the mid-'90s (before DEC and thenU2 >Compaq stopped spending money on VMS benchmarks).    L If you are talking about the same benchmark that I saw I think that you willH find that another ex-Digital product on the same test run beat Rdb handsH down. This was DSM, now owned by Intersystems, and still producing worldH beating price performance on Alpha systems. In a typical show of DigitalJ double standards and poor marketing decisions the DSM figures were quietlyD forgotten about because Digital was trying to push Rdb at that time.   ------------------------------  % Date: Wed, 14 Jun 2000 11:49:19 +0100gB From: Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com> Subject: Re: Fun VMS Facts?o* Message-ID: <3947632F.E2F036FA@uk.sun.com>   Dan Sugalski wrote:   8 > At 04:10 PM 6/13/00 +0000, jbecker@ui.urban.org wrote:9 > >Here's where you come in. What fun VMS facts would youa: > >include in such an orientation? I'm looking for tidbits9 > >that can be stated briefly and effectively to (in this 9 > >situation) a general but educated audience. I want the 0 > >items to show that VMS is a good place to be. > J > I'd make a quick scan through the CERT archives and pop up the number ofI > CERT security advisories on a per-OS basis. That one's always fun... :)  >a  9 Why is it fun, OpenVMS vunerabilities to a number of CERTn7 advisories have been fixed by Compaq/Digital in patchest4 released for OpenVMS but no response has been posted to CERT.  7 The poster asked for VMS fact's, you don't want fictiono3 in this list and including any CERT/OpenVMS relatede; material in the list is going to make it a work of fiction.o   Regardsp Andrew Harrisonc Enterprise IT Architectd   ------------------------------  # Date: Wed, 14 Jun 2000 13:12:42 GMT / From: "John Nixon" <jorlnixon@worldnet.att.net>  Subject: Re: Fun VMS Facts?,E Message-ID: <exL15.1133$Uw3.78247@bgtnsc04-news.ops.worldnet.att.net>s  H I am not sure if this is what you want, but I thought it was interesting that even whileeJ DEC/Compaq was actively trying to kill VMS, it actually grew its installed baseK thanks to  the users and engineers  that wouldn't let that happen. Now thatT Compaq seemsG willing to allow VMS to grow and improve things should get even better:   - With apologies to Terry Shannon and Acersoft:r1 (http://www.acersoft.com/SKD/00/VMSViewpoint.htm)l  I "Compaq claims an installed base of ~450K OpenVMS systems. According to ayK presentation delivered by OpenVMS executives in 1997, the firm expected the K 1996 installed base of ~435K systems to decline to ~250K systems by CY2001.TD Accordingly, the staying power of OpenVMS has exceeded even Compaq's expectations."      K <jbecker@ui.urban.org> wrote in message news:8i5mdp$4eg$1@nnrp1.deja.com...h6 > I'm putting together an orientation presentation for; > internal end users who are new to OpenVMS Alpha. For part : > of the presentation, I'm going to include material along5 > the lines of "What's OpenVMS" and "What's an Alpha"-8 > because most of these people are completely unfamiliar > with either. >c8 > Here's where you come in. What fun VMS facts would you9 > include in such an orientation? I'm looking for tidbitso8 > that can be stated briefly and effectively to (in this8 > situation) a general but educated audience. I want the/ > items to show that VMS is a good place to be.d >\7 > I have several ideas of my own, but I thought I'd aske > the VMS community. >( > -- > Jim Becker- > The Urban Institute (http://www.urban.org/) 5 > DECUS ESILUG (http://eisner.decus.org/lugs/esilug/)o >y >p( > Sent via Deja.com http://www.deja.com/ > Before you buy.0   ------------------------------  % Date: Wed, 14 Jun 2000 15:59:08 +0200h= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>h Subject: Re: Fun VMS Facts?a) Message-ID: <39478FAC.9088C58E@gtech.com>n  ( Andrew Harrison SUNUK Consultancy wrote:; > Why is it fun, OpenVMS vunerabilities to a number of CERTe9 > advisories have been fixed by Compaq/Digital in patchesy6 > released for OpenVMS but no response has been posted
 > to CERT. > 9 > The poster asked for VMS fact's, you don't want fictionr5 > in this list and including any CERT/OpenVMS relatedn= > material in the list is going to make it a work of fiction.e   ????   I can not see the logic in:n  F VMS engineering does not report back to CERT about fixes => using CERT VMS advisories is fiction-  E (unless you ofcourse mean that VMS will look worse than it really is,r; because problem that have actually been fixed does not haveS
 status fixed)g   Arne   ------------------------------   Date: 14 Jun 2000 14:52:37 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog) Subject: Re: Fun VMS Facts?w, Message-ID: <8i867l$lkp@gap.cco.caltech.edu>  F In article <8i5mdp$4eg$1@nnrp1.deja.com>, jbecker@ui.urban.org writes:7 >Here's where you come in. What fun VMS facts would you 8 >include in such an orientation? I'm looking for tidbits7 >that can be stated briefly and effectively to (in this 7 >situation) a general but educated audience. I want then. >items to show that VMS is a good place to be.   Hmm.  Well, there's always:e  C $! the next program is written in C and writes a couple of thousandt1 $! short records into outfile in stream-lf formatn $ program outfile  $ sort outfile newoutfiles9 $! hey, why did my system just grind to a halt????!!!?!?!s  A Sort goes batshit because the C RTL never puts the correct lrl onnJ the file, and so every record allocates 32k of memory, and Sort now needs . more memory than the system contains.  Whereas  % $ sort/process=tag outfile newoutfile   K works fine and completes nearly instantaneously.  That one has been lots of G "FUN" for me and my users, since that behavior makes no sense at all tonJ people who have no idea what a "record" is, let alone a "record managementL service", and if they TYPE the file it "looks" just like another file which  works correctly.  G Probably not the sort of "fun" you're after though!  I think something  I using the VMS wildcard convention [...]*foo*.*;* would be useful - that'soK a very convenient way of doing something that's often a real pain on other u OS's.  Have a look in:  2   http://seqaxp.bio.caltech.edu/www/vms_sheet.html:   http://seqaxp.bio.caltech.edu/www/vms_beginners_faq.html   for other ideas.   David Mathog mathog@seqaxp.bio.caltech.edua? Manager, sequence analysis facility, biology division, Caltech g   ------------------------------  # Date: Wed, 14 Jun 2000 16:44:51 GMT 9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)o Subject: Re: Fun VMS Facts?s+ Message-ID: <+IyVeOfUTDWh@eisner.decus.org>n  a In article <8i7bl4$j7e$1@soap.pipex.net>, "Chris Casey" <Chris.Casey@HendersonNOSPAM.com> writes:h > < > Bill Todd wrote in message <8i64sl$gsb$1@pyrite.mv.net>...J >>Rdb on VMS held the TPC record back in the mid-'90s (before DEC and then3 >>Compaq stopped spending money on VMS benchmarks).  >  > N > If you are talking about the same benchmark that I saw I think that you willJ > find that another ex-Digital product on the same test run beat Rdb handsJ > down. This was DSM, now owned by Intersystems, and still producing worldJ > beating price performance on Alpha systems. In a typical show of DigitalL > double standards and poor marketing decisions the DSM figures were quietlyF > forgotten about because Digital was trying to push Rdb at that time.  F The product from Intersystems pushed in the latest joint press releaseE from Compaq is called Cache'.  I don't know if it uses any of the old-F DSM code, but they tested it on some big VMS machine and got some sort@ of performance figure that seemed worth issuing a press release.   ------------------------------  % Date: Wed, 14 Jun 2000 15:50:18 +0100uB From: Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com> Subject: Re: Fun VMS Facts?e* Message-ID: <39479BAA.BCBC682F@uk.sun.com>   Arne Vajhj wrote:  * > Andrew Harrison SUNUK Consultancy wrote:= > > Why is it fun, OpenVMS vunerabilities to a number of CERT ; > > advisories have been fixed by Compaq/Digital in patches'8 > > released for OpenVMS but no response has been posted > > to CERT. > > ; > > The poster asked for VMS fact's, you don't want fiction>7 > > in this list and including any CERT/OpenVMS relatedr? > > material in the list is going to make it a work of fiction.i >o > ???? >  > I can not see the logic in:  > H > VMS engineering does not report back to CERT about fixes => using CERT > VMS advisories is fictiona >r  6 So using CERT advisories or the lack of them in a list4 interesting "facts" about OpenVMS is pointless isn't it.l   >hG > (unless you ofcourse mean that VMS will look worse than it really is,e= > because problem that have actually been fixed does not haved > status fixed)t  < No, worse than that. Someone publishes a list of interesting< facts about OpenVMS which includes the "fact" that there are% very few CERT advisories for OpenVMS.l  ? You reading this list of "facts" draw the obvious but incorrectr; conclusion that this means that OpenVMS is more secure thanp< other OS's and even worse that you don't need to worry about> things like out of band data attacks and smurf attacks because: there are no OpenVMS advisories ralating to these attacks.  < The fact that OpenVMS is vunerable to these attacks and that= you require patches for some of them has now been lost to youo4 and you may or may not be vunerable depending on how up-to-date your system is.  > In particular you may also think that you are safe from things3 like SYN FLOOD attacks for which there is no patch..  = Note I am not saying that OpenVMS is any better or worse than ? any other OS, but I am simply making the point that CERt is notc? the place to look for any "facts" relating to OpenVMS security.    regardsi Andrew Harrisonp Enterprise IT Architectg   ------------------------------  % Date: Wed, 14 Jun 2000 11:14:17 -0400D& From: "Mike Duffy" <mdduffy@erols.com> Subject: Re: Fun VMS Facts?-+ Message-ID: <8i87fj$ab7$1@bob.news.rcn.net>   > How about the Fun Fact (tm) that Intel (and others, of course)3 runs its chip manufacturing lines on OpenVMS boxes?i  = (Or at least it did last time I looked - when they opened thet+ new plant in the Dominican Republic (IIRC))    -Mikee    F jbecker@ui.urban.org wrote in message <8i5mdp$4eg$1@nnrp1.deja.com>...5 >I'm putting together an orientation presentation forr: >internal end users who are new to OpenVMS Alpha. For part9 >of the presentation, I'm going to include material alonga4 >the lines of "What's OpenVMS" and "What's an Alpha"7 >because most of these people are completely unfamiliarc
 >with either.s >y7 >Here's where you come in. What fun VMS facts would youy8 >include in such an orientation? I'm looking for tidbits7 >that can be stated briefly and effectively to (in this 7 >situation) a general but educated audience. I want the . >items to show that VMS is a good place to be. > 6 >I have several ideas of my own, but I thought I'd ask >the VMS community.  >1 >--  >Jim Becker0, >The Urban Institute (http://www.urban.org/)4 >DECUS ESILUG (http://eisner.decus.org/lugs/esilug/) >3 >0' >Sent via Deja.com http://www.deja.com/7 >Before you buy.   ------------------------------  % Date: Wed, 14 Jun 2000 11:37:30 -0400.' From: "Bill Todd" <billtodd@foo.mv.com>  Subject: Re: Fun VMS Facts?0( Message-ID: <8i88l4$9kj$1@pyrite.mv.net>  D Larry Kilgallen <Kilgallen@eisner.decus.org.nospam> wrote in message% news:+IyVeOfUTDWh@eisner.decus.org...1   ...:  H > The product from Intersystems pushed in the latest joint press releaseG > from Compaq is called Cache'.  I don't know if it uses any of the old H > DSM code, but they tested it on some big VMS machine and got some sortB > of performance figure that seemed worth issuing a press release.  L Their literature suggests that Cache has avoided the elegance (and resultingK bloat) of relational structure and rediscovered the merits of pointer-based J relationships.  Whether their code is new or DSM-based I have no idea, butI the concept is similar to the kind of RMS-like enhancements I think a newM file system should support.1   - bill   ------------------------------  % Date: Wed, 14 Jun 2000 17:55:02 +02000= From: Arne =?iso-8859-1?Q?Vajh=F8j?= <arne.vajhoej@gtech.com>M Subject: Re: Fun VMS Facts?0) Message-ID: <3947AAD5.CE8DDC8F@gtech.com>g  ( Andrew Harrison SUNUK Consultancy wrote: > Arne Vajhj wrote:, > > Andrew Harrison SUNUK Consultancy wrote:? > > > Why is it fun, OpenVMS vunerabilities to a number of CERT = > > > advisories have been fixed by Compaq/Digital in patches : > > > released for OpenVMS but no response has been posted > > > to CERT. > > > = > > > The poster asked for VMS fact's, you don't want fictionl9 > > > in this list and including any CERT/OpenVMS related A > > > material in the list is going to make it a work of fiction.  > >  > > ???? > >  > > I can not see the logic in:  > > J > > VMS engineering does not report back to CERT about fixes => using CERT > > VMS advisories is fiction  > 8 > So using CERT advisories or the lack of them in a list6 > interesting "facts" about OpenVMS is pointless isn't > it.    ????  @ You claimed that. I argued that your argumentation does not make. any sense. And you think I agree with you ????  I > > (unless you ofcourse mean that VMS will look worse than it really is,p? > > because problem that have actually been fixed does not havee > > status fixed)  > > > No, worse than that. Someone publishes a list of interesting> > facts about OpenVMS which includes the "fact" that there are' > very few CERT advisories for OpenVMS.o > A > You reading this list of "facts" draw the obvious but incorrectm= > conclusion that this means that OpenVMS is more secure thand> > other OS's and even worse that you don't need to worry about@ > things like out of band data attacks and smurf attacks because< > there are no OpenVMS advisories ralating to these attacks. > > > The fact that OpenVMS is vunerable to these attacks and that? > you require patches for some of them has now been lost to youh6 > and you may or may not be vunerable depending on how > up-to-date your system is.  ) Now you have changed your argumentation !t  = First you argued that VMS engineering did not tell CERT, when  a bug was fixed.  8 Now you claim that problems on VMS does not get reported in the first place !  9 That may be true or not true. It does not matter. This isn6 an official list. SUN employees posting to news-groups5 about non-registered VMS vulnerabilities is just FUD. 9 If you have the facts, then report it, or keep your mouth  shot.   = That is just like benchmarks. Tests that are not registederedh* at SPEC or TPC organizations are just FUD.   Arne   ------------------------------  # Date: Wed, 14 Jun 2000 15:57:44 GMT 3 From: Eric Dittman <dittman@narnia.int.dittman.net>r Subject: Re: Fun VMS Facts?a> Message-ID: <YXN15.9894$6d7.525509@news-west.usenetserver.com>  > Wayne Sewell <wayne@tachysoft.xxx.044962.killspam.0138> wrote:M : It contains a section called Ten Ways to Tell if You're an OpenVMS fanatic,iO : contining items like "You know what happened on November 17, 1858"  "You knowuL : why the 'sho system' display had to have the system uptime field increased : beyond 999 days."  And so on.i   Here's an example:  
 $ show sysM VAX/VMS V5.5-2  on node ZEUS  14-JUN-2000 10:03:12.32   Uptime  1312 05:11:50p    That's over 3.5 years of uptime. -- e Eric Dittman dittman@dittman.net    ------------------------------    Date: 14 Jun 2000 12:54:55 +0200* From: eplan@kapsch.net (Peter LANGSTOEGER)6 Subject: Re: how to stop echo on DS90M telnet listener* Message-ID: <3947647f$1@news.kapsch.co.at>  a In article <802568FD.00558227.00@qedilc01.qedi.quintiles.com>, steven.reece@quintiles.com writes: ? >This is the reason I moved to DS700s with DNAS on flash cards. J >- You can telnet onto the console port on the system really quickly afterO >powerup (such as after a power outage), usually quicker than the time taken to:9 >get the host up so you can watch it booting successfullyP  H A DS90M has a flash mem and DNAS, too. If you have the 2MB flash variantN (the 1MB flash variant needs 5min[!!] to decompress the boot image at startup)L the boot of the DS90M is fast enough. It seems, you are thinking of a DS90TLF which has no flash (and no DNAS because it also has not enough RAM)...   >(or not, but then againJ >there is no or not cos we're talking VMS here aren't we boys and girls?);  F Ok, maybe COMP.SYS.DEC fits better, but VMS systems are involved so we could accept it here...n  P >- There's no annoying echo from the DECserver, just the one from the VMS system >sitting behind it.e  G And this is IMHO the point. You will want to disable ECHO on the TELNETrJ connection completely (and then how to do this on the TELNET server in theM DNAS to force every TELNET client to disable echo ?) and let VMS do the echo..  L In the meantime, one could toggle the echo of the TELNET connection with theI appropriate key configured in the TELNET client (usually a ^E) - but onlyrG when the echo in the telnet connection is done locally = client and notl remote = server.  N >There may be other ways, but I got fed up trying so a DS700 with DNAS won the >day.t   You solved the problem ? How ?8 Because DNAS is DNAS, your solution should apply here !?   -- a< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888t< FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------  # Date: Wed, 14 Jun 2000 13:00:31 GMTe= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) 6 Subject: Re: how to stop echo on DS90M telnet listener0 Message-ID: <009EB967.74D5C4B1@SendSpamHere.ORG>  W In article <3947647f$1@news.kapsch.co.at>, eplan@kapsch.net (Peter LANGSTOEGER) writes:Vb >In article <802568FD.00558227.00@qedilc01.qedi.quintiles.com>, steven.reece@quintiles.com writes:@ >>This is the reason I moved to DS700s with DNAS on flash cards.K >>- You can telnet onto the console port on the system really quickly afteriP >>powerup (such as after a power outage), usually quicker than the time taken to: >>get the host up so you can watch it booting successfully >oI >A DS90M has a flash mem and DNAS, too. If you have the 2MB flash variantcO >(the 1MB flash variant needs 5min[!!] to decompress the boot image at startup) M >the boot of the DS90M is fast enough. It seems, you are thinking of a DS90TLdG >which has no flash (and no DNAS because it also has not enough RAM)...s  H Sorry, I posted yesterday and said it was the 90M.  It was, in fact, the- 90TL.  I have no problems when using the 90M.p   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMn   ------------------------------    Date: 14 Jun 2000 17:54:43 +0200* From: eplan@kapsch.net (Peter LANGSTOEGER)6 Subject: Re: how to stop echo on DS90M telnet listener( Message-ID: <3947aac3@news.kapsch.co.at>  p In article <009EB967.74D5C4B1@SendSpamHere.ORG>, system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) writes:X >In article <3947647f$1@news.kapsch.co.at>, eplan@kapsch.net (Peter LANGSTOEGER) writes:c >>In article <802568FD.00558227.00@qedilc01.qedi.quintiles.com>, steven.reece@quintiles.com writes:aA >>>This is the reason I moved to DS700s with DNAS on flash cards.SL >>>- You can telnet onto the console port on the system really quickly afterQ >>>powerup (such as after a power outage), usually quicker than the time taken to:; >>>get the host up so you can watch it booting successfullyu >>J >>A DS90M has a flash mem and DNAS, too. If you have the 2MB flash variantP >>(the 1MB flash variant needs 5min[!!] to decompress the boot image at startup)N >>the boot of the DS90M is fast enough. It seems, you are thinking of a DS90TLH >>which has no flash (and no DNAS because it also has not enough RAM)... >UI >Sorry, I posted yesterday and said it was the 90M.  It was, in fact, thef. >90TL.  I have no problems when using the 90M.   Then use the 90M ;-).e  E I so far didn't see many differences between a 90TL and a 90M in thisoK particular area. "DEFINE PORT x TELNET SERVER ECHO NEGOTIATION INIT|NOINIT"iB is AFAIK the only command available on DNAS and not on the 90TL...   -- w< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888n< FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------  % Date: Wed, 14 Jun 2000 15:41:54 +00007$ From: Steve.Spires@yellowpages.co.uk- Subject: Job logical name table quota problemb/ Message-ID: <002568FE.005654C8.00@quegw01.btyp>u   cc:a bcc:= Contact:   Tel: 3063  -  VSSG, 1st Floor, Bridge Street Plazai  $ Job logical name table quota problem  	 Dear all,n  O I have a Vax4705A running VMS7.1 upon which I need to perform certain functionsa@ which I think probably aren't relevant to the problem, which is;  M When carrying out an operation I am hit with the EXLNMQUOTA error. OriginallyrH both my UAF JTQUOTA and the system PQL_DJTQUOTA were set at 1024. I haveK increased the both of these to 8192. However, after logging out and back intK again to pick up the UAF value (the PQL value is dynamic) I still get a job / logical name table set up with a quota of 1024:    $ show log/job/full   A (LNM$JOB_957EA000)      [kernel]  [shareable]  [Quota=(848,1024)]u  P So, how do I fix this? At present it is a big problem (I have logged a call withK Compaq here in the UK, but I know I'll probably get an answer here quicker)  almost a show-stopper.  2 I get the feeling I'm missing something obvious...   Steve Spires VMS System Manager BT/Yellow Pagese     [Information] -- PostMaster:D This transmission is intended solely for the addressee(s) and may beL confidential. If you are not the named addressee, or if the message has beenP addressed to you in error, you must not read, disclose, reproduce, distribute or use this transmission.  L Delivery of this message to any person other than the named addressee is notH intended in any way to waive confidentiality.  If you have received thisF transmission in error please contact the sender or delete the message.  
 Thank you.   ------------------------------    Date: 14 Jun 2000 16:54:03 +0200* From: eplan@kapsch.net (Peter LANGSTOEGER)1 Subject: Re: Job logical name table quota problemV* Message-ID: <39479c8b$1@news.kapsch.co.at>  V In article <002568FE.005654C8.00@quegw01.btyp>, Steve.Spires@yellowpages.co.uk writes:% >Job logical name table quota problemd  8 How about (temporarily) turning off Job Table Quota with   	UAF>MOD/JT=0 user   at all ?  1 I decided to switch off _my_ JTQUOTA years ago...e   -- o< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888a< FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------  % Date: Wed, 14 Jun 2000 16:19:48 +0000m$ From: Steve.Spires@yellowpages.co.uk1 Subject: Re: Job logical name table quota problem / Message-ID: <002568FE.0059CE4A.00@quegw01.btyp>   = Contact:   Tel: 3063  -  VSSG, 1st Floor, Bridge Street Plazad  O Okay, I did that, logged out and back in again, and still get the table created L with a quota of 1024. I also (for the other response) made sure that I did aC WRITE CURRENT and a WRITE ACTIVE when changing the SYSGEN parameterc
 PQL_DJTQUOTA.F  H Is there any other way I can up the quota on the table after it has been created?   Steve Spires VMS System Manager BT/Yellow PagesO        ; eplan@kapsch.net (Peter LANGSTOEGER) on 14/06/2000 14:54:03f    To:        Info-VAX@Mvb.Saic.Com+ cc:         (bcc: Steve Spires/YellowPages)bH From:      eplan@kapsch.net (Peter LANGSTOEGER), 14 June 2000, 2:54 p.m.  ( Re: Job logical name table quota problem        N In article <002568FE.005654C8.00@quegw01.btyp>, Steve.Spires@yellowpages.co.uk writes:i% >Job logical name table quota problemo  8 How about (temporarily) turning off Job Table Quota with        UAF>MOD/JT=0 user   at all ?  1 I decided to switch off _my_ JTQUOTA years ago...    --< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888S< FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------    Date: 14 Jun 2000 17:43:32 +0200* From: eplan@kapsch.net (Peter LANGSTOEGER)1 Subject: Re: Job logical name table quota problemn* Message-ID: <3947a824$1@news.kapsch.co.at>  V In article <002568FE.0059CE4A.00@quegw01.btyp>, Steve.Spires@yellowpages.co.uk writes:P >Okay, I did that, logged out and back in again, and still get the table createdM >with a quota of 1024. I also (for the other response) made sure that I did anD >WRITE CURRENT and a WRITE ACTIVE when changing the SYSGEN parameter >PQL_DJTQUOTA.  I Assuming that you have PQL_MJTQUOTA at the Default = 0, then it will workeF (at least it did on my systems). Seems you have forgotten to check the5 PQL_MJTQUOTA and kept your eyes only on PQL_DJTQUOTA.i   --  < Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888 < FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------  % Date: Wed, 14 Jun 2000 01:26:01 -0400d% From: mjdowden@panix.com (M J Dowden)t Subject: Re: MVII heat output : Message-ID: <mjdowden-1406000126010001@news.widomaker.com>  F     The "Supermicrosystems Handbook" 1986 gives the system power heatI dissipation as 1978 BTU/k for the floorstand (BA-123), 1177 BTU/k for the 8 pedestal (BA-23) and 5872 BTU/k for the Cabinet (H-9642)   ------------------------------  % Date: Tue, 13 Jun 2000 20:53:05 -0400l* From: David A Froble <davef@tsoft-inc.com>+ Subject: Re: Need VMS media - Austin, Texas , Message-ID: <3946D771.C2BF086@tsoft-inc.com>   Bill Bradford wrote: > 4 > This is probably a long shot, and I apologize, but5 > is anybody on this list / reading this newsgroup inn8 > Austin, Texas and have either the OpenVMS/VAX Hobbyist8 > Media Kit , or a VMS distribution (5.5-2 or better) on9 > CD-ROM that I can borrow for an evening and copy?  I'ves8 > got licenses and my own Media Kit ordered, but its not9 > here yet, and I'd like to fire this system up tonight..: > 2 > Thanks.  If anybody can help, please email me at > mrbill@decvax.org.  M Now that's eagerness!  A good quality.  I seem to remember that the hobby kitcJ came rather quickly, couple of days, from I believe Houston to PA.  Unless; there's a screw-up, you should have the kit rather quickly.    Dave   -- W4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com6 T-Soft, Inc.  170 Grimplin Road  Vanderbilt, PA  15486   ------------------------------  % Date: Wed, 14 Jun 2000 10:52:55 +0200n From: Robin.Goerlach@merck.dee7 Subject: OpenVMS and Windows NT Integration for Dummiesr4 Message-ID: <C12568FE.0030D5D5.00@dedamsg1.merck.de>   Hi,   N I would like to read this book, can anyone tell me the ISBN Nr. or a DEC Order No. ?o  P I didn't manage to find it at amazon or Digital Press (at Butterworth-Heinemann)   any hint is wellcome Robin1   ------------------------------  % Date: Wed, 14 Jun 2000 10:06:49 -04005- From: "Peter Weaver" <peter.weaver@stelco.ca>=; Subject: Re: OpenVMS and Windows NT Integration for Dummiess. Message-ID: <skf4bop0is448@corp.supernews.com>  C You can ask your Sales Rep for a copy, the part number they have tor order is 11N3-0200A-WWEN.t   -- Peter Weaver  , Robin.Goerlach@merck.de wrote in message ... >Hi, >dE >I would like to read this book, can anyone tell me the ISBN Nr. or af	 DEC Orderb
 >No. ? ...   ------------------------------  # Date: Wed, 14 Jun 2000 17:05:21 GMTS9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)c; Subject: Re: OpenVMS and Windows NT Integration for Dummiesi+ Message-ID: <vtHf3bvLM1oN@eisner.decus.org>   B And if you cannot find anyone willing to admit they are your salesF representative, you can try mail to Susan.Skonetski using the InternetH Domain for Compaq.  Please do try the sales representative first though,J because if they haven't heard aboout the book your question will introduce< it to them and get them to send it to their other customers.  ^ In article <skf4bop0is448@corp.supernews.com>, "Peter Weaver" <peter.weaver@stelco.ca> writes:E > You can ask your Sales Rep for a copy, the part number they have toc > order is 11N3-0200A-WWEN.v >  > -- > Peter Weaver > . > Robin.Goerlach@merck.de wrote in message ... >>Hi,s >>F >>I would like to read this book, can anyone tell me the ISBN Nr. or a > DEC Orderi >>No. ? ...e >    ------------------------------  % Date: Wed, 14 Jun 2000 10:12:58 -0700 ! From: Shane.F.Smith@Healthnet.comp; Subject: Re: OpenVMS and Windows NT Integration for DummiesyC Message-ID: <OFD4E161C8.F397A60E-ON882568FE.005E925F@HEALTHNET.COM>a   >Hi, >tI >I would like to read this book, can anyone tell me the ISBN Nr. or a DEC  >Order No. ? >a: >I didn't manage to find it at amazon or Digital Press (at >Butterworth-Heinemann)h >h >any hint is wellcomeh >Robin  A I don't see an ISBN on my copy, but there's a Compaq part number:yF 11N3-0200A-WWEN. There's also a second part number, but it doesn't say< whose. It's probably the Dummies Press number. That's MM034.  G Inside there's a number to call about where to buy books in the series,rE 1-800-434-3422. Alternatively, try e-mailing Sue Skonetski at Compaq:nH Susan.Skonetski@compaq.com. A conversation with her led to me getting onK the mailing list this book was distributed to, she may be able to help you.,  
 Good luck.   Shaneh  H  #####   ---------------------------------------------------------------I #-O-O-# | Shane underbar S on pacbell dot net. Spam to abuse@127.0.0.1  |-H #  L  #  ---------------------------------------------------------------D  #===#   Don't blame HealthNet for anything I say. They're innocent.H   ###    OpenVMS: The operating system God runs the Earth simulation on.   ------------------------------  % Date: Wed, 14 Jun 2000 14:19:32 -0300n1 From: "Boyle, Darren" <boyledj@bankofbermuda.com>t; Subject: RE: OpenVMS and Windows NT Integration for Dummies"K Message-ID: <9F664D538536D411BD3200508B6FF01A024748@bdant027.bda.bobda.com>e  H I just received my copy of this, it was sent from Rich Marcello, you may2 want to contact him at RICHARD.MARCELLO@COMPAQ.COM - Darren   > ---------- > From: K > Kilgallen@eisner.decus.org.nospam[SMTP:Kilgallen@eisner.decus.org.nospam] . > Reply To: 	Kilgallen@eisner.decus.org.nospam) > Sent: 	Wednesday, June 14, 2000 2:05 PMt > To: 	Info-VAX@Mvb.Saic.Com> > Subject: 	Re: OpenVMS and Windows NT Integration for Dummies > D > And if you cannot find anyone willing to admit they are your salesH > representative, you can try mail to Susan.Skonetski using the InternetJ > Domain for Compaq.  Please do try the sales representative first though,L > because if they haven't heard aboout the book your question will introduce> > it to them and get them to send it to their other customers. > ? > In article <skf4bop0is448@corp.supernews.com>, "Peter Weaver" " > <peter.weaver@stelco.ca> writes:G > > You can ask your Sales Rep for a copy, the part number they have to  > > order is 11N3-0200A-WWEN.t > >  > > -- > > Peter Weaver > > 0 > > Robin.Goerlach@merck.de wrote in message ... > >>Hi,D > >>H > >>I would like to read this book, can anyone tell me the ISBN Nr. or a
 > > DEC Order 
 > >>No. ? ...M > >  >     F **********************************************************************C This message and any files transmitted with it are confidential andoJ may be privileged and/or subject to the provisions of privacy legislation.M They are intended solely for the use of the individual or entity to whom theyLL are addressed. If the reader of this message is not the intended recipient, B please notify the sender immediately and then delete this message.I You are notified that reliance on, disclosure of, distribution or copyingc of this message is prohibited.   Bank of BermudahF **********************************************************************   ------------------------------  # Date: Wed, 14 Jun 2000 11:09:43 GMTm% From: hx_101@hotmail.com (Horse Nuts)n# Subject: OpenVMS Training on CD-ROMt, Message-ID: <394767b1.752944767@news.dal.ca>  > Does anyone have any suggestions on a good training CD-ROM for< OpenVMS? Something similar to a UNIX Systems Admin course???   Thanks   ------------------------------   Date: 14 Jun 2000 14:28:08 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)' Subject: Re: OpenVMS Training on CD-ROM06 Message-ID: <8i84po$fp5$3@mailint03.im.hou.compaq.com>  T In article <394767b1.752944767@news.dal.ca>, hx_101@hotmail.com (Horse Nuts) writes:? :Does anyone have any suggestions on a good training CD-ROM for = :OpenVMS? Something similar to a UNIX Systems Admin course???B  ;   Please search for "training" or "TRK" in the OpenVMS FAQ.n  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Wed, 14 Jun 2000 17:45:29 +0000n$ From: Steve.Spires@yellowpages.co.ukF Subject: Problam solved (was Re: Job logical name table quota problem)/ Message-ID: <002568FE.0061A5DB.00@quegw01.btyp>u  = Contact:   Tel: 3063  -  VSSG, 1st Floor, Bridge Street Plazau  P It turns out that (for some reason as yet unknown to me) one of the Oracle DBA'sM had made a copy of SYSUAF.DAT and redefined the SYSUAF system-wide instead ofiO privately for a short time, so I was in fact looking at an incorrect version oftN the UAF when I made my change. Making the change in the correct version solved the problem.  	 Ho-hum...a   Steve Spires VMS System Manager BT/Yellow Pages         1 steven.reece@quintiles.com on 14/06/2000 16:25:25   # To:        Steve Spires/YellowPages  cc:u> From:      steven.reece@quintiles.com, 14 June 2000, 4:25 p.m.  ( Re: Job logical name table quota problem             Steve,P Have you looked at PQL_MJTQUOTA?  That's the maximum value that JTQUOTA can haveO IIRC so it may be that although your process would be allowed to have more fromaB looking at the UAF record the maximum doesn't let it go that high.   Steve.        5 Steve.Spires@yellowpages.co.uk on 14/06/2000 17:19:48I        To:      Info-VAX@Mvb.Saic.Com>  ,  cc:      (bcc: Steven Reece/QRED/Quintiles)      2  Subject: Re: Job logical name table quota problem                  = Contact:   Tel: 3063  -  VSSG, 1st Floor, Bridge Street Plaza0  O Okay, I did that, logged out and back in again, and still get the table createdeL with a quota of 1024. I also (for the other response) made sure that I did aC WRITE CURRENT and a WRITE ACTIVE when changing the SYSGEN parameterb
 PQL_DJTQUOTA.u  H Is there any other way I can up the quota on the table after it has been created?   Steve Spires VMS System Manager BT/Yellow Pages         ; eplan@kapsch.net (Peter LANGSTOEGER) on 14/06/2000 14:54:03t    To:        Info-VAX@Mvb.Saic.Com+ cc:         (bcc: Steve Spires/YellowPages)tH From:      eplan@kapsch.net (Peter LANGSTOEGER), 14 June 2000, 2:54 p.m.  ( Re: Job logical name table quota problem        N In article <002568FE.005654C8.00@quegw01.btyp>, Steve.Spires@yellowpages.co.uk writes:e% >Job logical name table quota problemo  8 How about (temporarily) turning off Job Table Quota with        UAF>MOD/JT=0 user   at all ?  1 I decided to switch off _my_ JTQUOTA years ago...-   --< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888B< FBFV/Information Services           E-mail  eplan@kapsch.netF <<< KAPSCH AG  Wagenseilgasse 1     PSImail PSI%(0232)281001141::EPLANH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"N "VMS is today what Microsoft wants Windows NT V8.0 to be!" Compaq, 22-Sep-1998   ------------------------------  % Date: Wed, 14 Jun 2000 18:00:18 +0100r  From: steven.reece@quintiles.comJ Subject: Re: Problam solved (was Re: Job logical name table quota problem)> Message-ID: <802568FE.005D82A7.00@qedilc01.qedi.quintiles.com>  * Good reason to chop the DBAs quotas off... :-)    Steve Spires wrote:mM >>>It turns out that (for some reason as yet unknown to me) one of the Oracle  DBA's M had made a copy of SYSUAF.DAT and redefined the SYSUAF system-wide instead of O privately for a short time, so I was in fact looking at an incorrect version ofoN the UAF when I made my change. Making the change in the correct version solved the problem.<<<s   ------------------------------  % Date: Wed, 14 Jun 2000 14:21:30 -0300F1 From: "Boyle, Darren" <boyledj@bankofbermuda.com> J Subject: RE: Problam solved (was Re: Job logical name table quota problem)K Message-ID: <9F664D538536D411BD3200508B6FF01A024749@bdant027.bda.bobda.com>o  1 I say give him TMPMBX & NETMBX and lose the rest.v - Darren   > ----------D > From: 	steven.reece@quintiles.com[SMTP:steven.reece@quintiles.com]) > Sent: 	Wednesday, June 14, 2000 2:00 PMl > To: 	Info-VAX@Mvb.Saic.ComD > Subject: 	Re: Problam solved (was Re: Job logical name table quota
 > problem) >  >  > , > Good reason to chop the DBAs quotas off... > :-)s >  > Steve Spires wrote:iH > >>>It turns out that (for some reason as yet unknown to me) one of the > Oracle > DBA's,L > had made a copy of SYSUAF.DAT and redefined the SYSUAF system-wide instead > ofF > privately for a short time, so I was in fact looking at an incorrect > version ofI > the UAF when I made my change. Making the change in the correct versionn > solved > the problem.<<<s >  >  >  >     F **********************************************************************C This message and any files transmitted with it are confidential andjJ may be privileged and/or subject to the provisions of privacy legislation.M They are intended solely for the use of the individual or entity to whom theyyL are addressed. If the reader of this message is not the intended recipient, B please notify the sender immediately and then delete this message.I You are notified that reliance on, disclosure of, distribution or copying  of this message is prohibited.   Bank of Bermuda1F **********************************************************************   ------------------------------   Date: 14 Jun 2000 17:43:30 GMT* From: bleau@umtof.umd.edu (Lawrence Bleau)> Subject: Re: Proxy problem: Why does node:: work but 0:: fail?) Message-ID: <8i8g82$rcg$1@hecate.umd.edu>a  c In article <034901bfd5b7$58523fe0$020a0a0a@xile.realm>, "John E. Malmberg" <wb8tyw@qsl.net> writes:  >eL >I can duplicate your problem if you have a non-default proxy for the SYSTEM, >account in your proxy database to anything. >e >LOCAL:.ULEIS::SYSTEM MUMBLE  H I have such a nondefault proxy.  It maps SYSTEM to HTTPSERVER, and, if IO remember correctly, is needed by some components of the OSU server.  In fact, I M stumbled across this by trying to use helpgate to access the VMS help file onIM this system (see what happens with http://uleis.umd.edu/help).  This isn't anQN OSU or HTTP problem, though; it's an underlying Decnet proxy problem.  Read on before commenting.  J I doubt this is the sole reason, though.  I say this because the problem IM described (0:: fails but ULEIS:: works) exists for my personal, nonprivilegedmO account (BLEAU).  The existence of a nondefault proxy account for SYSTEM shouldr% not matter at all for other accounts.   E >The other thing might be that your local decnet address might not be0( >registered in your local address tower.  ) I already checked this; it is registered.a  M >If you do a SET HOST 0, and you are able to log in but you get a "Failure on @ >the back translate address request", then I would suspect that.  H I did not get this error.  Here is the entry for node ULEIS in the local	 database:6   DECNET_REGISTER> sho node uleisg  " Directory Service: Local name file   Node name:         LOCAL:.uleis  Phase IV synonym:  ULEIS9 Node address:      49::00-06:AA-00-04-00-F7-1A:20 (6.759)L   Number of nodes reported on:  1   - Here are the relevant proxy database entries:    LOCAL:.ULEIS::*t	     * (D)o   LOCAL:.ULEIS::SYSTEM5     SYSTEM (D)                             HTTPSERVERe   I also did the command:04 MCR NCL FLUSH SESSION CONTROL NAMING CACHE ENTRY "*"  N and the problem persists.  I double checked my intrusion database to make sureK it wasn't keeping me out that way - it wasn't - and I double checked on thegI personal account after the above NCL command - it still fails in the samel manner.t   Lawrence Bleau University of Maryland" Physics Dept., Space Physics Group 301-405-6223 bleau@umtof.umd.edua   ------------------------------  % Date: Wed, 14 Jun 2000 11:23:52 -0400g' From: "Bill Todd" <billtodd@foo.mv.com>tB Subject: Re: ramdisk vs. file cache, and the winner is, file cache( Message-ID: <8i87rm$8o5$1@pyrite.mv.net>  H <hein@eps.zko.dec.c*m> wrote in message news:14JUN200000560191@miasys...   ...   $ > NO. RMS sucks CPU cycles big time.  B I've heard this before, and am curious why:  processing RMS-formatK sequential files (with whatever record format) never required all that many L instructions in RMS-11, and I can't see why it should in VMS/RMS.  The majorL difference I'd expect would be the need to probe the user structures on eachJ record access:  is this really that expensive, and if so is there some wayL to pre-verify them in the manner of the recent 'fast I/O' or 'fast path' - I# forget which is which - mechanisms?-   - bill   ------------------------------  % Date: Wed, 14 Jun 2000 08:38:14 -0700b+ From: Sharon Guthrie <sguthrie@stan911.com>p" Subject: Re: Recording CD's on VMS+ Message-ID: <3947A6E6.6C97D8B2@stan911.com>A   "Main, Kerry" wrote:' > The following might be of assistance:- > http://www.tmesis.com/CDROM/  < 	I second the recommendation of this link.   I used CDrecord? with alot of help from Brian, the webmaster at TMESIS.  I put a4
 PlexwriterE burner on a VAXstation 3100, which should work fine.  Turned out that C Brian had to modify CDrecord a bit to work with the VS3100 drivers.nE The version of CDrecord that he modified for me is right there on his / page, with a very helpful list of instructions.t= 	Feel free to email me if you need more info on how I did it.v   -- o4 Sharon Guthrie, Stanislaus County Emergency Dispatch# email:  sguthrie at stan911 dot com % business url:  http://www.stan911.com " *  Check out my personal website, A *  now with "The Guthrie Library" in association with Amazon.Com:  *  http://www.jps.net/redo; ----------------------------------------------------------- % bang head on keyboard to continue....-   ------------------------------  % Date: Wed, 14 Jun 2000 04:36:06 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca>o3 Subject: Removing/Stripping RFC headers from emailsr, Message-ID: <394743F5.B895D69D@videotron.ca>  H I am writing an application will will, as one source of transactions, beT receiving emails either from VMSmail (foreign transport) or Message Router (on VAX).  M Since some (but not all) systems are configured to include the RFC headers of L messages coming from the internet as part of the text of VMSmail messages, IF am wondering if anyone has found a pretty failsafe logic to detect theI presence of an RFC header in text ? (either at the top of bottom, but not C always present (when receiving direct from VMS mail from instance).e  ; Is the "Received:" tag always at the top of an RFC header ?.  K These messages are going to be routed to mobile phones, so I have to removes6 any/all junk in emails to send the text only contents.  K Also, under current SMTP-VMS products, can my software assume that incoming-N emails from the SMTP gateway will have had the quoted-printable converted backN to the latin-1/DEC-MCS characters or do the SMTP products (DEC/Compaq's) leave) the MIME  quoted-printable messages raw ?-  N (I know that DEC's product doesn't handle MIME multipart messages, and I guess* I will eventually have to deal with that).   ------------------------------  9 Date: Wed, 14 Jun 2000 10:29:57 +0100 (GMT Daylight Time)g From: Andy.Harper@kcl.ac.uka3 Subject: Removing/Stripping RFC headers from emailsa1 Message-ID: <SIMEON.10006141057.N@odin.kcl.ac.uk>   , On Wed, 14 Jun 2000 04:36:06 -0400 JF Mezei % <jfmezei.spamnot@videotron.ca> wrote:a  J > I am writing an application will will, as one source of transactions, beV > receiving emails either from VMSmail (foreign transport) or Message Router (on VAX). > O > Since some (but not all) systems are configured to include the RFC headers ofSN > messages coming from the internet as part of the text of VMSmail messages, IH > am wondering if anyone has found a pretty failsafe logic to detect theK > presence of an RFC header in text ? (either at the top of bottom, but notBE > always present (when receiving direct from VMS mail from instance).-  C   You should take a look at the source code of pine for VMS, where l; this is extensive code to handle this (module VMS_MAIL.C). o Available from:g-     http://www.agh.cc.kcl.ac.uk/vms/pine-vms/H    = > Is the "Received:" tag always at the top of an RFC header ?y  A   I believe so, since it's added by the receiving mail server as eC pretty much it's first step. Since it's far easier to add it first  C rather than in the middle before all other received lines, I can't  B imagine any mail server being silly enough to do otherwise. As to B whether it's a requirement that it comes first - don't know - but 4 the RFC should state this if it is (RFC 821/RFC822).    M > These messages are going to be routed to mobile phones, so I have to removel8 > any/all junk in emails to send the text only contents. >o     Regards,   Andy Harper  King's College Londonn   ------------------------------   Date: 14 Jun 2000 11:49:42 GMT* From: helbig@astro.rug.nl (Phillip Helbig)7 Subject: Re: Removing/Stripping RFC headers from emails-. Message-ID: <8i7rgm$2in$1@info.service.rug.nl>  5 In article <394743F5.B895D69D@videotron.ca>, JF Mezei-' <jfmezei.spamnot@videotron.ca> writes: -  N >Since some (but not all) systems are configured to include the RFC headers ofM >messages coming from the internet as part of the text of VMSmail messages, I-G >am wondering if anyone has found a pretty failsafe logic to detect them% >presence of an RFC header in text ? a   SEARCH?j  < >Is the "Received:" tag always at the top of an RFC header ?  ( Return-Path: is always first in my case.  L >These messages are going to be routed to mobile phones, so I have to remove7 >any/all junk in emails to send the text only contents.R   DEFINE MACRO RFKILLr FIND BUFFER RFKILLO insert;change; er -"================== RFC 822 Headers ==================" der p	 FIND LASTl   :-)K  L >Also, under current SMTP-VMS products, can my software assume that incomingO >emails from the SMTP gateway will have had the quoted-printable converted backrO >to the latin-1/DEC-MCS characters or do the SMTP products (DEC/Compaq's) leaves* >the MIME  quoted-printable messages raw ?   They are raw in my case.   ------------------------------  # Date: Wed, 14 Jun 2000 14:21:43 GMTo1 From: schreiber@eisner.decus.org (Jeff Schreiber)o7 Subject: Re: Removing/Stripping RFC headers from emails + Message-ID: <D0bE6P$Y+T9G@eisner.decus.org>n  \ In article <394743F5.B895D69D@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: > O > Since some (but not all) systems are configured to include the RFC headers of N > messages coming from the internet as part of the text of VMSmail messages, IH > am wondering if anyone has found a pretty failsafe logic to detect theK > presence of an RFC header in text ? (either at the top of bottom, but not E > always present (when receiving direct from VMS mail from instance).* > = > Is the "Received:" tag always at the top of an RFC header ?n >   7     No.  For example, MX will put a Return-Path: on it.o  "     Some facts about header lines:  I         1) Headers are seperated from the text by a null line.  So if your7            find a null line, there are no more headers.r  O         2) A header name is printable ascii [ascii(33) -> ascii(126)] excluding 5            the colon, and is terminated with a colon.i  K         3) headers that continue onto a multiple line must have white space*/            preceding each line after the first.*  +     I encourage perusal of RFC 821 and 822.a  4                                             -   Jeff   --1 Jeff Schreiber,            Process Software Corp. 1 schreiber@mx.process.com   http://www.process.coms+      TCPware & MultiNet: Stronger than Evern   ------------------------------  % Date: Wed, 14 Jun 2000 15:34:38 +0100 / From: "Bolton, Lee" <BoltonL@capitagroup.co.uk>e" Subject: Replacing the Quorum Disk@ Message-ID: <BF981E50293ED111BAC2006097BE4934035E9FF4@cbswmbdc1>   Hi,   I I have a 2 node Alpha cluster and have a disk (DKC806) as my quorum disk.cD This disk needs replacing. I am wondering if this would be the right approach to replace the disk.u  # 1)  Take an Image backup of DKC806.  2)  Bring down the Cluster. K 3)  Replace the disk. (as it is a straight replacement of a disk, no system A parameters would need changing, or re-configuring on the HSZ70?) o 4)  Reboot the Cluster.-5 5)  Restore the data from the Image backup of DKC806.n  < Your help or confirmation of this would be much appreciated.   Thanks,e   Lee Bolton.1F **********************************************************************B This email and any files transmitted with it are confidential and G intended solely for the use of the individual or entity to whom they   FF are addressed. If you have received this email in error please notify  the system manager.   F This footnote also confirms that this email message has been swept by 1 MIMEsweeper for the presence of computer viruses.    www.mimesweeper.comgF **********************************************************************   ------------------------------  # Date: Wed, 14 Jun 2000 16:24:11 GMT  From: trdorr@my-deja.com Subject: Seaching SYSUAF.DAT) Message-ID: <8i8bij$33l$1@nnrp1.deja.com>5  E I'm looking for a way to search SYSUAF.DAT for accounts that have notcF been active for greater than 30 days and less than 60 days. Is there aB way to use the LAST Login date to produce a list of users that are	 inactive?m Thanks,l Toma    & Sent via Deja.com http://www.deja.com/ Before you buy.e   ------------------------------  # Date: Wed, 14 Jun 2000 16:38:21 GMTi7 From: "Lyle W. West" <lyle.west@childrenshc.org.nospam>.  Subject: Re: Seaching SYSUAF.DAT5 Message-ID: <39476E74.EF70006@childrenshc.org.nospam>   E You could try Nologin from the V4 Freeware CD. If you don't have the o@ CD, look at the VMS website or http://www.montagar.com/freeware/  ' (as I shamelessly push one of my tools)z   trdorr@my-deja.com wrote:d > G > I'm looking for a way to search SYSUAF.DAT for accounts that have not4H > been active for greater than 30 days and less than 60 days. Is there aD > way to use the LAST Login date to produce a list of users that are > inactive?y	 > Thanks,I > Tom    --  0 My opinions seldom reflect those of my employer.   Lyle W. West   ------------------------------  % Date: Wed, 14 Jun 2000 13:13:36 -0400)[ From: jamese@beast.dtsw.army.mil (Ed James, TCS Inc, 410-295-1919, ed.james@telecomsys.com)Y  Subject: Re: Seaching SYSUAF.DAT0 Message-ID: <00061413133639@beast.dtsw.army.mil>  = trdorr@my-deja.com wrote on Wed, 14 Jun 2000 16:24:11 GMT in e <8i8bij$33l$1@nnrp1.deja.com>:  G > I'm looking for a way to search SYSUAF.DAT for accounts that have nottH > been active for greater than 30 days and less than 60 days. Is there aD > way to use the LAST Login date to produce a list of users that are > inactive?o	 > Thanks,  > TomE  G Attached below my .sig is DORMANT, a C program written by Darrell Blairy and enhanced my me.   : Ed James                           ed.james@telecomsys.com5 TeleCommunications Systems, Inc.   voice 410-295-1919S5 275 West Street, Suite 400         fax   410-280-1094A Annapolis, MD 21401-1740                    sN ============================= dormant.c below ================================N /***************************************************************************** * TITLE: DORMANT.C * AUTHOR: Darrell Blair: * CREATE DATE: March 16, 1990g! * LAST REV. DATE: January 6, 2000s * MODIFIED BY: Darrell Blair7 * MODIFIED BY: V. Edward James  ed.james@telecomsys.comBN ****************************************************************************** * DESCRIPTION:@ *  DORMANT identifies all dormant and near dormant uaf accounts.C *  The dormant and near dormant values can be entered as qualifiers H *  (/WARNING & /WATCH) or (if no qualifiers are present) default values  *  are used.M *****************************************************************************-" **** Version 1.0 - March 20, 1990  ****        (Initial Release)0" **** Version 1.1 - March 22, 1990 5 ****        (Enhanced screen output while WORKING...):E ****        (Increased arrays from 300 to 1000, I like to use memory)3" **** Version 1.2 - March 29, 1990 . ****        (Added date of run to title sheet)" **** Version 1.3 - March 30, 1990 , ****        (Added node name to title sheet)" **** Version 2.0 - March 30, 1990 F ****        (Converted all report file output to DEC Special Graphics)$ **** Version 2.2 - July 26, 1995 vej, ****        (Converted cld to object module)4 ****        (Fixed lib$day bug. Added disacnt check)' **** Version 2.3 - February 3, 1997 vej B ****        (Fixed foreign command to add "DORMANT " before parse)L ****        (Added DECSPEC qualifier.  Changed tabs to spaces, every three.)E ****        (Added SORT qualifier. Added active accounts in array[3])l' **** Version 2.3.2 - March 10, 1997 vejh& ****        (Added /verbose and /help)C ****        (Changed /match default to 90, /warning default to 180) & **** Version 2.3.3 - April 9, 1999 vej, ****        (Added COMMAND to tabbed output)( **** Version 2.3.4 - January 7, 2000 vej> ****        (Fixed lib$day for zero days, made ansi compliant,F ****         removed descript, changed disacnt in displays to disuser)N *****************************************************************************/ #define VERSION "2.3.4"e #include ssdef #include stdlibh #include stdio #include stringe #include descrip #include uaideft #include lnmdefd #include starlet #include lib$routinesf #include climsgdef #define DEFAULT_WATCH 90 #define DEFAULT_WARNING 180* #define MIN_DAYS 1 #define MAX_ACCOUNTS 1000  struct item_struct {%          unsigned short item_length ;G$          unsigned short item_code ;           int *buff_addr ;           int *buff_length ;           } ; struct information {$          int number_of_days_dormant;          short disacnt;a          char username[13];l          } ; struct bin_time { "       unsigned short year_since_0;#       unsigned short month_of_year; "       unsigned short day_of_month;!       unsigned short hour_of_day;:$       unsigned short minute_of_hour;&       unsigned short second_of_minute;*       unsigned short hundredths_of_second;       }; struct dsc$descriptor_s tabnam;a struct dsc$descriptor_s lognam;r struct dsc$descriptor_s usr;! struct information holding_array; * struct information array[4][MAX_ACCOUNTS]; static struct bin_time numtime;T   typedef int BOOL;'      FILE  *outfile, *infile;o<    int   status, warning_subscript = 0, watch_subscript = 0,K          never_subscript = 0, active_subscript = 0, number_of_accounts = 0,UL          watch_value, warning_value, day_zero, last_login_day, days_dormant;A    BOOL  decspec_value, sort_by_user_value, tabbed_output_value,\r*          verbose_value, interactive_value;H    char  extended_month_names[14][10] = {"Error", "January", "February",<          "March", "April", "May", "June", "July", "August", 9          "September", "October", "November", "December"};l    char  equiv_name[256];e6    char recordfile[255] = "DORMANT.RPT", the_date[19];    static char command2[1025];    char *interactive_name;  
 #ifdef __DECC  #pragma extern_model save   #pragma extern_model globalvalue extern DORMANT_TABLE;c #pragma extern_model restore #elseC globalvalue DORMANT_TABLE; #endif  5 #define check(stat) if( (stat&1) != 1) lib$stop(stat)9/ #define max(a,b)        ((a) > (b) ? (a) : (b))e/ #define min(a,b)        ((a) < (b) ? (a) : (b))U  . int cli$present( struct dsc$descriptor_s *id);/ int cli$get_value( struct dsc$descriptor_s *id,e4                    struct dsc$descriptor_s *ret_str,,                    unsigned short *ret_len); int cli$dispatch( void);4 int cli$dcl_parse( struct dsc$descriptor_s *command,                    int table, 3                    unsigned int (*param_routine)(),r4                    unsigned int (*prompt_routine)(),4                    struct dsc$descriptor_s *prompt);+ int compare( const void *a, const void *b);3 void dormant_main( void);- void make_uaf_list( void);" int days_since_zero( int quad[2]);9 int get_dormant_time( struct information *holding_array);R8 void get_logical( char *logical_name, char *table_name); void write_output( void);y! void write_decspec_output( void);* void write_text_output( void);  void write_tabbed_output( void);  * int compare( const void *a, const void *b) {t>    int ascending = FALSE;   /* Set to FALSE for descending. */    if( ascending) {e>       if( (*(struct information *)a).number_of_days_dormant < >           (*(struct information *)b).number_of_days_dormant) {          return(-1);       }e>       if( (*(struct information *)a).number_of_days_dormant ==>           (*(struct information *)b).number_of_days_dormant) {             int ii;n>             ii = strncmp( (*(struct information *)a).username,>                           (*(struct information *)b).username,L                           min( strlen( (*(struct information *)a).username),N                                strlen( (*(struct information *)b).username)));#             if( ii < 0) return(-1); "             if( ii > 0) return(1);             return(0);
             } =       if( (*(struct information *)a).number_of_days_dormant >i>           (*(struct information *)b).number_of_days_dormant) {          return(1);        }i    }	    else {o=       if( (*(struct information *)a).number_of_days_dormant >a>           (*(struct information *)b).number_of_days_dormant) {          return(-1);       }t>       if( (*(struct information *)a).number_of_days_dormant ==>           (*(struct information *)b).number_of_days_dormant) {             int ii;r>             ii = strncmp( (*(struct information *)a).username,>                           (*(struct information *)b).username,L                           min( strlen( (*(struct information *)a).username),N                                strlen( (*(struct information *)b).username)));#             if( ii < 0) return(-1); "             if( ii > 0) return(1);             return(0);
             },=       if( (*(struct information *)a).number_of_days_dormant <i>           (*(struct information *)b).number_of_days_dormant) {          return(1);4       }6    } }r   void dormant_main( void) {c    int quad[2] = { 0, 0, };e    unsigned short ret_len;.    static char cliwarning[255], cliwatch[255];,    $DESCRIPTOR( ptr_cliwarning, cliwarning);(    $DESCRIPTOR( ptr_cliwatch, cliwatch);+    $DESCRIPTOR( ptr_clioutput, recordfile);m'    $DESCRIPTOR( verbose_id, "VERBOSE");a!    $DESCRIPTOR( help_id, "HELP");o%    $DESCRIPTOR( output_id, "OUTPUT"); #    $DESCRIPTOR( watch_id, "WATCH"); '    $DESCRIPTOR( warning_id, "WARNING");u3    $DESCRIPTOR( tabbed_output_id, "TABBED_OUTPUT");S'    $DESCRIPTOR( decspec_id, "DECSPEC");d1    $DESCRIPTOR( sort_by_user_id, "SORT_BY_USER");s/    $DESCRIPTOR( interactive_id, "INTERACTIVE");s  3    if( cli$present( &verbose_id) == CLI$_PRESENT) {t       verbose_value = TRUE;o    }	    else {M       verbose_value= FALSE;a    }0    if( cli$present( &help_id) == CLI$_PRESENT) {K       printf( "DORMANT-I-HELP, DORMANT UAF account finder (version %s).\n",a          VERSION);W       printf( "DORMANT-I-HELP, /WARNING=nn    Days before account reported dormant\n");t^       printf( "DORMANT-I-HELP, /WATCH=nn      Days before account reported almost dormant\n");R       printf( "DORMANT-I-HELP,                Must be less than /WARNING days\n");O       printf( "DORMANT-I-HELP, /OUTPUT=fspec  File spec for output listing\n");pX       printf( "DORMANT-I-HELP, /DECSPEC       Use DEC Special Graphic character set\n");Q       printf( "DORMANT-I-HELP, /SORT_BY_USER  Sort listing by name, not days\n"); b       printf( "DORMANT-I-HELP, /TABBED_OUTPUT Write output suitable for spreadsheet insertion\n");Q       printf( "DORMANT-I-HELP, /VERBOSE       Turn on informational messages\n"); V       printf( "DORMANT-I-HELP, /NOINTERACTIVE Use last non-interactive login date\n");O       printf( "DORMANT-I-HELP, /HELP          Print this message ansd exit\n");h       exit(EXIT_SUCCESS);l    }  $    status = sys$numtim(&numtime, 0);    if( status != SS$_NORMAL) {S       printf("DORMANT-E-NUMTIME, Error converting binary time to numeric time.\n");cF       printf("DORMANT-I-NOREPRT, No report file has been created.\n");       lib$stop(status);o    }{    sprintf(the_date, "%s %d, %d", extended_month_names[numtime.month_of_year], numtime.day_of_month, numtime.year_since_0);"    if( verbose_value) { f       printf("DORMANT-I-WELCOME, Welcome to the DORMANT UAF account finder (version %s).\n", VERSION);    }0    get_logical( "SYS$NODE", "LNM$SYSTEM_TABLE");%    day_zero = days_since_zero( quad);r2    if( cli$present( &output_id) == CLI$_PRESENT) {D       status = cli$get_value( &output_id, &ptr_clioutput, &ret_len);!       recordfile[ret_len] = '\0'; !       if( status != SS$_NORMAL) {i          lib$stop(status);       }e       if( ret_len <= 0) {e,          strcpy( recordfile, "DORMANT.RPT");       }k    }1    if( cli$present( &watch_id) == CLI$_PRESENT) {dB       status = cli$get_value( &watch_id, &ptr_cliwatch, &ret_len);       cliwatch[ret_len] = '\0';4!       if( status != SS$_NORMAL) {           lib$stop(status);       }n#       watch_value = atoi(cliwatch);        if( verbose_value) {S          printf("DORMANT-I-OVRRIDE, Using user supplied WATCH value of %d days.\n",i          watch_value);       } =       if( watch_value < MIN_DAYS || watch_value > day_zero) {hx          printf( "DORMANT-W-OUTLIMT, Your value for /WATCH is outside the limits of %d-%d days.\n", MIN_DAYS, day_zero);          exit(EXIT_FAILURE);       }t    }	    else {c"       watch_value = DEFAULT_WATCH;       if( verbose_value) {[          printf("DORMANT-I-DEFAULT, Using default WATCH value of %d days.\n", watch_value);e       }e    }  3    if( cli$present( &warning_id) == CLI$_PRESENT) {eF       status = cli$get_value( &warning_id, &ptr_cliwarning, &ret_len);!       cliwarning[ret_len] = '\0';e!       if( status != SS$_NORMAL) {R          lib$stop(status);       }r'       warning_value = atoi(cliwarning);r       if( verbose_value) {U          printf("DORMANT-I-OVRRIDE, Using user supplied WARNING value of %d days.\n",d          warning_value);       } A       if( warning_value < MIN_DAYS || warning_value > day_zero) {ey          printf("DORMANT-W-OUTLIMT, Your value for /WARNING is outside the limits of %d-%d days.\n", MIN_DAYS, day_zero);,          exit(EXIT_FAILURE);       }i    }	    else {o&       warning_value = DEFAULT_WARNING;       if( verbose_value) {_          printf("DORMANT-I-DEFAULT, Using default WARNING value of %d days.\n", warning_value);o       }     }%    if( watch_value > warning_value) { `       printf("DORMANT-W-INVALID, The /WATCH value cannot be larger than the /WARNING value.\n");       exit(EXIT_FAILURE);p    }9    if( cli$present( &tabbed_output_id) == CLI$_PRESENT) {h!       tabbed_output_value = TRUE;        if( verbose_value) {N          printf("DORMANT-I-OVRRIDE, Output will be in tabbed text format.\n");       }i    }	    else {V"       tabbed_output_value = FALSE;       if( verbose_value) {I          printf("DORMANT-I-DEFAULT, Output will be in report format.\n");u       }o    }    if( !tabbed_output_value) {6       if( cli$present( &decspec_id) == CLI$_PRESENT) {          decspec_value = TRUE;          if( verbose_value) {eS             printf("DORMANT-I-OVRRIDE, Report includes DEC Special Characters.\n");s
          }       }        else {          decspec_value = FALSE;c          if( verbose_value) {aV             printf("DORMANT-I-DEFAULT, Report includes no DEC Special Characters.\n");
          }       }     }8    if( cli$present( &sort_by_user_id) == CLI$_PRESENT) {        sort_by_user_value = TRUE;       if( verbose_value) {S          printf("DORMANT-I-OVRRIDE, Warning/Watch sections listed by username.\n");        }e    }	    else {n!       sort_by_user_value = FALSE;-       if( verbose_value) {X          printf("DORMANT-I-DEFAULT, Warning/Watch sections sorted by days inactive.\n");       }e    }7    if( cli$present( &interactive_id) == CLI$_NEGATED) {9        interactive_value = FALSE;       if( verbose_value) {V          printf("DORMANT-I-DEFAULT, Non-Interactive last login date will be used.\n");       }e    }	    else {a       interactive_value = TRUE;a       if( verbose_value) {R          printf("DORMANT-I-DEFAULT, Interactive last login date will be used.\n");       }     }    if( interactive_value) {n        if( tabbed_output_value) {*          interactive_name = "INTERACTIVE";       }        else {-          interactive_name = "  Interactive ";n       }     }	    else {m        if( tabbed_output_value) {-          interactive_name = "NONINTERACTIVE";*       }*       else {-          interactive_name = "NonInteractive";d       }a    }    (void) make_uaf_list();    if( !sort_by_user_value) {aO       qsort( array[0], warning_subscript, sizeof(struct information), compare);rM       qsort( array[1], watch_subscript, sizeof(struct information), compare);iN       qsort( array[3], active_subscript, sizeof(struct information), compare);    }    (void) write_output();*    exit(EXIT_SUCCESS); }*   void make_uaf_list( void)* {**    int spawn_status, ii, rough_count = -2;7    static char spawnin[23] =  "SYS$LOGIN:DORMANT.TMP;";a7    static char spawnout[23] = "SYS$LOGIN:DORMANT.OUT;";3     char uafline[132], *work_ptr;&    $DESCRIPTOR( ptr_spawnin, spawnin);(    $DESCRIPTOR( ptr_spawnout, spawnout);  1    if( (outfile = fopen(spawnin, "w")) != NULL) {a       if( verbose_value) {E          printf("DORMANT-I-OPTIMIZ, Optimizing the UAF database.\n");e       }j6       fprintf(outfile, "$ SET DEFAULT SYS$SYSTEM:\n");,       fprintf(outfile, "$ RUN AUTHORIZE\n");)       fprintf(outfile, "LIST/BRIEF *\n");e!       fprintf(outfile, "EXIT\n");j"       fprintf(outfile, "$ EOJ\n");       fclose(outfile);    }	    else {oG       printf("DORMANT-E-NOOPTIM, Cannot optimize the UAF database.\n");e       exit(EXIT_FAILURE);     }\    status = lib$spawn(0, &ptr_spawnin, &ptr_spawnout, 0, 0, 0, spawn_status, 0, 0, 0, 0, 0);    if( status == SS$_NORMAL) {$       lib$delete_file(&ptr_spawnin);%       lib$delete_file(&ptr_spawnout);v    }	    else {nR       printf("DORMANT-E-ERROPTM, Error occured while optimizing UAF database.\n");          lib$stop(status);    }A    if( (infile = fopen("SYS$SYSTEM:SYSUAF.LIS;", "r")) != NULL) {oI       while( (work_ptr = fgets(uafline, sizeof(uafline), infile)) != 0) {e          rough_count++;3       }     }    fclose(infile);A    if( (infile = fopen("SYS$SYSTEM:SYSUAF.LIS;", "r")) != NULL) { 9       work_ptr = fgets(uafline, sizeof(uafline), infile);39       work_ptr = fgets(uafline, sizeof(uafline), infile);1       if( verbose_value) {&          printf("DORMANT-I-WORKING,");       } I       while( (work_ptr = fgets(uafline, sizeof(uafline), infile)) != 0) {o&          for( ii = 0; ii < 12; ii++) {,             if( uafline[ii + 21] == '\40') {                break;o
             }e             else {=                holding_array.username[ii] = uafline[ii + 21];m
             }r
          }+          holding_array.username[ii] = '\0';           if( verbose_value) {s?             if( number_of_accounts % (rough_count / 10) == 0) { J                printf(" %d%%,", (number_of_accounts * 100) / rough_count);
             } 
          }<          last_login_day = get_dormant_time( &holding_array);#          if( last_login_day == 0) {*8             array[2][never_subscript++] = holding_array;
          }          else {aM             holding_array.number_of_days_dormant = day_zero - last_login_day;sH             if( holding_array.number_of_days_dormant >= warning_value) {=                array[0][warning_subscript++] = holding_array;*
             }*K             else if( holding_array.number_of_days_dormant >= watch_value) {r;                array[1][watch_subscript++] = holding_array;n
             }q             else {<                array[3][active_subscript++] = holding_array;
             }.
          }          number_of_accounts++;2          holding_array.number_of_days_dormant = 0;'          holding_array.username[0] = 0;l       }0       fclose(infile);r       if( verbose_value) {          printf(" Done!\n");       }:    }	    else {c^       printf("DORMANT-E-NTFOUND, Optimized UAF database not found. Check your privileges.\n");       exit(EXIT_FAILURE);t    }
    return; }V  ! int days_since_zero( int quad[2])  {     int number_of_days;  &    if( quad[0] == 0 && quad[1] == 0) {/       status = lib$day( &number_of_days, 0, 0);     }	    else {a6       status = lib$day( &number_of_days, &quad[0], 0);    }    if( status != SS$_NORMAL) {       lib$stop(status);b    }    return number_of_days;r }n  8 int get_dormant_time( struct information *holding_array) {d&    static struct item_struct stuff[3];*    int ii, last_login_binary[2], uaiflags;  7    usr.dsc$w_length = strlen( holding_array->username);e/    usr.dsc$a_pointer = holding_array->username;i#    usr.dsc$b_class = DSC$K_CLASS_S; #    usr.dsc$b_dtype = DSC$K_DTYPE_T;*      if( interactive_value) { ,       stuff[0].item_code = UAI$_LASTLOGIN_I;    }	    else { ,       stuff[0].item_code = UAI$_LASTLOGIN_N;    }    stuff[0].item_length = 8;3    stuff[0].buff_addr = (int *) &last_login_binary;a  #    stuff[1].item_code = UAI$_FLAGS;r    stuff[1].item_length = 4;*    stuff[1].buff_addr = (int *) &uaiflags;      stuff[2].item_code = 0;    stuff[2].item_length = 0;  4    status = sys$getuai(0, 0, &usr, &stuff, 0, 0, 0);    if( status != SS$_NORMAL) {       lib$stop(status);p    }"    if( uaiflags & UAI$M_DISACNT) {$       holding_array->disacnt = TRUE;    }	    else {f%       holding_array->disacnt = FALSE;A    }@    if( last_login_binary[0] == 0 && last_login_binary[1] == 0) {
       ii = 0;     }	    else { /       ii = days_since_zero( last_login_binary);m    }
    return ii;t }u  7 void get_logical( char *logical_name, char *table_name)u {i'    static struct item_struct stuff0[2];_
    int ii;  ,    tabnam.dsc$w_length = strlen(table_name);%    tabnam.dsc$a_pointer = table_name; &    tabnam.dsc$b_class = DSC$K_CLASS_S;&    tabnam.dsc$b_dtype = DSC$K_DTYPE_T;  .    lognam.dsc$w_length = strlen(logical_name);'    lognam.dsc$a_pointer = logical_name;t&    lognam.dsc$b_class = DSC$K_CLASS_S;&    lognam.dsc$b_dtype = DSC$K_DTYPE_T;  0    for( ii = 0; ii < sizeof(equiv_name); ii++) {       equiv_name[ii] = '\0';    }%    stuff0[0].item_code = LNM$_STRING;u    stuff0[0].item_length = 256;t,    stuff0[0].buff_addr = (int *) equiv_name;      stuff0[1].item_code = 0;     stuff0[1].item_length = 0;*  8    status = sys$trnlnm(0, &tabnam, &lognam, 0, &stuff0);8    if( status != SS$_NORMAL && status != SS$_NOLOGNAM) {       lib$stop(status);_    }
    return; }    void write_output( void) {d4    if( (outfile = fopen(recordfile, "w")) != NULL) {        if( tabbed_output_value) {&          (void) write_tabbed_output();       }a       else {          if( decspec_value) {4*             (void) write_decspec_output();
          }          else {"'             (void) write_text_output();e
          }       }"       fclose(outfile);    } }    #define BOLD "\033[1m" #define UNDERSCORE "\033[4m" #define NORMAL "\033[0m" #define GRAPH_MODE "\033(0"c #define ASCII_MODE "\033(B"d  void write_decspec_output( void) { 
    int ii;      if( verbose_value) {e\       printf("DORMANT-I-RECORD,  Recording information in reporting file %s\n", recordfile);    }<    fprintf(outfile, "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n");X    fprintf(outfile, "                            %slqqqqqqqqqqqqqqqqqqk\n", GRAPH_MODE);f    fprintf(outfile, "                            x  %sDormant Report%s  x\n", ASCII_MODE, GRAPH_MODE);M    fprintf(outfile, "                            x  %s%s%s  x\n", ASCII_MODE,h$       interactive_name, GRAPH_MODE);J    fprintf(outfile, "                            tqqqqqqqqqqqqqqqqqqu\n");5    fprintf(outfile, "                            x");o;    for( ii = 0; ii < (19 - strlen(equiv_name)) / 2; ii++) {(       fprintf(outfile, " ");    }B    fprintf(outfile, "%s%s%s", ASCII_MODE, equiv_name, GRAPH_MODE);:    for( ii = ii + strlen(equiv_name) + 1; ii < 19; ii++) {       fprintf(outfile, " ");    }    fprintf(outfile, "x\n");r5    fprintf(outfile, "                            x");aG    for( ii = 0; ii < (sizeof(the_date) - strlen(the_date)) / 2; ii++) {c       fprintf(outfile, " ");    }@    fprintf(outfile, "%s%s%s", ASCII_MODE, the_date, GRAPH_MODE);F    for( ii = ii + strlen(the_date) + 1; ii < sizeof(the_date); ii++) {       fprintf(outfile, " ");    }    fprintf(outfile, "x\n"); Z    fprintf(outfile, "                            mqqqqqqqqqqqqqqqqqqj\n\n\n\n\n\n\n\n\n");V    fprintf(outfile, "                  lqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqk\n");    fprintf(outfile, "                  x %sAccount WARNING setting is at %3d days%s x\n", ASCII_MODE, warning_value, GRAPH_MODE);c    fprintf(outfile, "                  x %sAccount WATCH   setting is at %3d days%s x\n", ASCII_MODE, watch_value, GRAPH_MODE); V    fprintf(outfile, "                  tqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqu\n");    fprintf(outfile, "                  x  %sNumber of accounts over WARNING: %4d%s x\n", ASCII_MODE, warning_subscript, GRAPH_MODE);    fprintf(outfile, "                  x  %sNumber of accounts over WATCH  : %4d%s x\n", ASCII_MODE, watch_subscript, GRAPH_MODE);    fprintf(outfile, "                  x  %sNumber of accounts NEVER USED  : %4d%s x\n", ASCII_MODE, never_subscript, GRAPH_MODE);    fprintf(outfile, "                  x  %sTotal accounts on the system   : %4d%s x\n", ASCII_MODE, number_of_accounts, GRAPH_MODE);)V    fprintf(outfile, "                  mqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqqj\n");    fprintf(outfile, "\f\n");I    fprintf(outfile, "                            lqqqqqqqqqqqqqqqqqk\n"); e    fprintf(outfile, "                            x %sWarning Section%s x\n", ASCII_MODE, GRAPH_MODE);aL    fprintf(outfile, "                            x %s%s %s x\n", ASCII_MODE,$       interactive_name, GRAPH_MODE);Y    fprintf(outfile, "                            mqqqqqqqqqqqqqqqqqj%s\n\n", ASCII_MODE); Z    fprintf(outfile, "            (The status of the these accounts should be checked)\n");Z    fprintf(outfile, "                 (D indicates the account is marked DISUSER)\n\n\n");w    fprintf(outfile, "%slqqqqqqqqqqqqqqqqqqqqqqqqqwqqqqqqqqqqqqqqqqqqqqqqqqqwqqqqqqqqqqqqqqqqqqqqqqqqqk\n", GRAPH_MODE);tt    fprintf(outfile, "x%sUSERNAME     D    DORMANT%sx%sUSERNAME     D    DORMANT%sx%sUSERNAME     D    DORMANT%sx\n",N       ASCII_MODE, GRAPH_MODE, ASCII_MODE, GRAPH_MODE, ASCII_MODE, GRAPH_MODE);j    fprintf(outfile, "tqqqqqqqqqqqqqqqqqqqqqqqqqnqqqqqqqqqqqqqqqqqqqqqqqqqnqqqqqqqqqqqqqqqqqqqqqqqqqu\nx");/    for( ii = 0; ii < warning_subscript; ii++) {R       if( ii % 3 == 0 && ii) {!          fprintf(outfile, "\nx");        })!       if( array[0][ii].disacnt) {,?          fprintf(outfile, "%s%-12s D  %4d days%sx", ASCII_MODE,)H             array[0][ii].username,  array[0][ii].number_of_days_dormant,             GRAPH_MODE);       }        else {?          fprintf(outfile, "%s%-12s    %4d days%sx", ASCII_MODE,)G             array[0][ii].username, array[0][ii].number_of_days_dormant,)             GRAPH_MODE);       }r    }    while( ii++ % 3 != 0) {5       fprintf(outfile, "                         x");r    }k    fprintf(outfile, "\nmqqqqqqqqqqqqqqqqqqqqqqqqqvqqqqqqqqqqqqqqqqqqqqqqqqqvqqqqqqqqqqqqqqqqqqqqqqqqqj\n");L    fprintf(outfile, "\f\n");I    fprintf(outfile, "                            lqqqqqqqqqqqqqqqqqk\n");ee    fprintf(outfile, "                            x  %sWatch Section%s  x\n", ASCII_MODE, GRAPH_MODE);hL    fprintf(outfile, "                            x  %s%s%s x\n", ASCII_MODE,$       interactive_name, GRAPH_MODE);Y    fprintf(outfile, "                            mqqqqqqqqqqqqqqqqqj%s\n\n", ASCII_MODE); Z    fprintf(outfile, "            (The status of the these accounts should be checked)\n");Z    fprintf(outfile, "                 (D indicates the account is marked DISUSER)\n\n\n");w    fprintf(outfile, "%slqqqqqqqqqqqqqqqqqqqqqqqqqwqqqqqqqqqqqqqqqqqqqqqqqqqwqqqqqqqqqqqqqqqqqqqqqqqqqk\n", GRAPH_MODE);mt    fprintf(outfile, "x%sUSERNAME     D    DORMANT%sx%sUSERNAME     D    DORMANT%sx%sUSERNAME     D    DORMANT%sx\n",N       ASCII_MODE, GRAPH_MODE, ASCII_MODE, GRAPH_MODE, ASCII_MODE, GRAPH_MODE);j    fprintf(outfile, "tqqqqqqqqqqqqqqqqqqqqqqqqqnqqqqqqqqqqqqqqqqqqqqqqqqqnqqqqqqqqqqqqqqqqqqqqqqqqqu\nx");-    for( ii = 0; ii < watch_subscript; ii++) {i       if( ii % 3 == 0 && ii) {!          fprintf(outfile, "\nx");,       }t!       if( array[1][ii].disacnt) {e?          fprintf(outfile, "%s%-12s D  %4d days%sx", ASCII_MODE,SG             array[1][ii].username, array[1][ii].number_of_days_dormant,t             GRAPH_MODE);       }        else {?          fprintf(outfile, "%s%-12s    %4d days%sx", ASCII_MODE,dG             array[1][ii].username, array[1][ii].number_of_days_dormant,s             GRAPH_MODE);       }e    }    while( ii++ % 3 != 0) {5       fprintf(outfile, "                         x");p    }k    fprintf(outfile, "\nmqqqqqqqqqqqqqqqqqqqqqqqqqvqqqqqqqqqqqqqqqqqqqqqqqqqvqqqqqqqqqqqqqqqqqqqqqqqqqj\n");l    fprintf(outfile, "\f\n");K    fprintf(outfile, "                           lqqqqqqqqqqqqqqqqqqqqk\n");ag    fprintf(outfile, "                           x %sNever Used Section %sx\n", ASCII_MODE, GRAPH_MODE);sN    fprintf(outfile, "                           x   %s%s%s   x\n", ASCII_MODE,$       interactive_name, GRAPH_MODE);[    fprintf(outfile, "                           mqqqqqqqqqqqqqqqqqqqqj%s\n\n", ASCII_MODE);i[    fprintf(outfile, "            (The status of the these accounts should be verified)\n"); Z    fprintf(outfile, "                 (D indicates the account is marked DISUSER)\n\n\n");t    fprintf(outfile, "%slqqqqqqqqqqqqqqwqqqqqqqqqqqqqqwqqqqqqqqqqqqqqwqqqqqqqqqqqqqqwqqqqqqqqqqqqqqk\n", GRAPH_MODE);y    fprintf(outfile, "x%sUSERNAME     D%sx%sUSERNAME     D%sx%sUSERNAME     D%sx%sUSERNAME     D%sx%sUSERNAME     D%sx\n",lZ       ASCII_MODE, GRAPH_MODE, ASCII_MODE, GRAPH_MODE, ASCII_MODE, GRAPH_MODE, ASCII_MODE, *       GRAPH_MODE, ASCII_MODE, GRAPH_MODE);g    fprintf(outfile, "tqqqqqqqqqqqqqqnqqqqqqqqqqqqqqnqqqqqqqqqqqqqqnqqqqqqqqqqqqqqnqqqqqqqqqqqqqqu\nx");"-    for( ii = 0; ii < never_subscript; ii++) {f       if( ii % 5 == 0 && ii) {!          fprintf(outfile, "\nx");        }D!       if( array[2][ii].disacnt) {A5          fprintf(outfile, "%s%-12s D%sx", ASCII_MODE,,/             array[2][ii].username, GRAPH_MODE);        }        else {5          fprintf(outfile, "%s%-12s  %sx", ASCII_MODE,_/             array[2][ii].username, GRAPH_MODE);n       }     }    while( ii++ % 5 != 0) {*       fprintf(outfile, "              x");    }v    fprintf(outfile, "\nmqqqqqqqqqqqqqqvqqqqqqqqqqqqqqvqqqqqqqqqqqqqqvqqqqqqqqqqqqqqvqqqqqqqqqqqqqqj\n%s", ASCII_MODE); }    void write_text_output( void)( {i
    int ii;      if( verbose_value) {E\       printf("DORMANT-I-RECORD,  Recording information in reporting file %s\n", recordfile);    }<    fprintf(outfile, "\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n");K    fprintf(outfile, "                            /------------------\\\n");uJ    fprintf(outfile, "                            |  Dormant Report  |\n");=    fprintf(outfile, "                            |  %s  |\n",        interactive_name);J    fprintf(outfile, "                            |------------------|\n");5    fprintf(outfile, "                            |");n;    for( ii = 0; ii < (19 - strlen(equiv_name)) / 2; ii++) {        fprintf(outfile, " ");    }&    fprintf(outfile, "%s", equiv_name);:    for( ii = ii + strlen(equiv_name) + 1; ii < 19; ii++) {       fprintf(outfile, " ");    }    fprintf(outfile, "|\n");c5    fprintf(outfile, "                            |"); G    for( ii = 0; ii < (sizeof(the_date) - strlen(the_date)) / 2; ii++) {D       fprintf(outfile, " ");    }$    fprintf(outfile, "%s", the_date);F    for( ii = ii + strlen(the_date) + 1; ii < sizeof(the_date); ii++) {       fprintf(outfile, " ");    }    fprintf(outfile, "|\n");eZ    fprintf(outfile,"                            \\------------------/\n\n\n\n\n\n\n\n\n");W    fprintf(outfile, "                  /----------------------------------------\\\n"); e    fprintf(outfile, "                  | Account WARNING setting is at %3d days |\n", warning_value); c    fprintf(outfile, "                  | Account WATCH   setting is at %3d days |\n", watch_value); V    fprintf(outfile, "                  |----------------------------------------|\n");h    fprintf(outfile, "                  |  Number of accounts over WARNING: %4d |\n", warning_subscript);f    fprintf(outfile, "                  |  Number of accounts over WATCH  : %4d |\n", watch_subscript);f    fprintf(outfile, "                  |  Number of accounts NEVER USED  : %4d |\n", never_subscript);i    fprintf(outfile, "                  |  Total accounts on the system   : %4d |\n", number_of_accounts);sV    fprintf(outfile,"                  \\----------------------------------------/\n");    fprintf(outfile,"\f\n");eJ    fprintf(outfile, "                            /-----------------\\\n");I    fprintf(outfile, "                            | Warning Section |\n"); <    fprintf(outfile, "                            | %s  |\n",       interactive_name);K    fprintf(outfile,"                            \\-----------------/\n\n");sZ    fprintf(outfile, "            (The status of the these accounts should be checked)\n");Z    fprintf(outfile, "                 (D indicates the account is marked DISUSER)\n\n\n");j    fprintf(outfile, "/-----------------------------------------------------------------------------\\\n");i    fprintf(outfile, "|USERNAME     D    DORMANT|USERNAME     D    DORMANT|USERNAME     D    DORMANT|\n");ij    fprintf(outfile, "|-------------------------+-------------------------+-------------------------|\n|");/    for( ii = 0; ii < warning_subscript; ii++) {        if( ii % 3 == 0 && ii) {!          fprintf(outfile, "\n|");,       }p!       if( array[0][ii].disacnt) { G          fprintf(outfile, "%-12s D  %4d days|", array[0][ii].username, _=                         array[0][ii].number_of_days_dormant);        }R       else {G          fprintf(outfile, "%-12s    %4d days|", array[0][ii].username,  =                         array[0][ii].number_of_days_dormant);w       }m    }    while( ii++ % 3 != 0) {5       fprintf(outfile, "                         |");Y    }k    fprintf(outfile,"\n\\-----------------------------------------------------------------------------/\n");     fprintf(outfile,"\f\n"); J    fprintf(outfile, "                            /-----------------\\\n");I    fprintf(outfile, "                            |  Watch Section  |\n");l<    fprintf(outfile, "                            |  %s |\n",       interactive_name);K    fprintf(outfile,"                            \\-----------------/\n\n");rZ    fprintf(outfile, "            (The status of the these accounts should be checked)\n");Z    fprintf(outfile, "                 (D indicates the account is marked DISUSER)\n\n\n");j    fprintf(outfile, "/-----------------------------------------------------------------------------\\\n");i    fprintf(outfile, "|USERNAME     D    DORMANT|USERNAME     D    DORMANT|USERNAME     D    DORMANT|\n");oj    fprintf(outfile, "|-------------------------+-------------------------+-------------------------|\n|");-    for( ii = 0; ii < watch_subscript; ii++) {o       if( ii % 3 == 0 && ii) {!          fprintf(outfile, "\n|");i       }0!       if( array[1][ii].disacnt) {bG          fprintf(outfile, "%-12s D  %4d days|", array[1][ii].username, d=                         array[1][ii].number_of_days_dormant);        }        else {G          fprintf(outfile, "%-12s    %4d days|", array[1][ii].username, [=                         array[1][ii].number_of_days_dormant);        }i    }    while( ii++ % 3 != 0) {5       fprintf(outfile, "                         |");_    }k    fprintf(outfile,"\n\\-----------------------------------------------------------------------------/\n");o    fprintf(outfile,"\f\n"); L    fprintf(outfile, "                           /--------------------\\\n");K    fprintf(outfile, "                           | Never Used Section |\n");c>    fprintf(outfile, "                           |   %s   |\n",       interactive_name);M    fprintf(outfile,"                           \\--------------------/\n\n");C[    fprintf(outfile, "            (The status of the these accounts should be verified)\n");eZ    fprintf(outfile, "                 (D indicates the account is marked DISUSER)\n\n\n");g    fprintf(outfile, "/--------------------------------------------------------------------------\\\n");yf    fprintf(outfile, "|USERNAME     D|USERNAME     D|USERNAME     D|USERNAME     D|USERNAME     D|\n");g    fprintf(outfile, "|--------------+--------------+--------------+--------------+--------------|\n|");u-    for( ii = 0; ii < never_subscript; ii++) {u       if( ii % 5 == 0 && ii) {!          fprintf(outfile, "\n|");e       }d!       if( array[2][ii].disacnt) {a=          fprintf(outfile, "%-12s D|", array[2][ii].username);=       }Y       else {=          fprintf(outfile, "%-12s  |", array[2][ii].username);O       }     }    while( ii++ % 5 != 0) {*       fprintf(outfile, "              |");    }i    fprintf(outfile, "\n\\--------------------------------------------------------------------------/\n");  }s   void write_tabbed_output( void)1 {u
    int ii;      if( verbose_value) {f\       printf("DORMANT-I-RECORD,  Recording information in reporting file %s\n", recordfile);    }F    fprintf(outfile, "DORMANT REPORT\t%s\t%s\n", the_date, equiv_name);/    fprintf(outfile, "COMMAND\t%s\n", command2);rO    fprintf(outfile, "%s WARNING VALUE\t%d\n", interactive_name, warning_value); K    fprintf(outfile, "%s WATCH VALUE\t%d\n", interactive_name, watch_value);iS    fprintf(outfile, "%s WARNING FOUND\t%d\n", interactive_name, warning_subscript);rO    fprintf(outfile, "%s WATCH FOUND\t%d\n", interactive_name, watch_subscript); Q    fprintf(outfile, "%s ACTIVE FOUND\t%d\n", interactive_name, active_subscript);eT    fprintf(outfile, "%s NEVER USED FOUND\t%d\n", interactive_name, never_subscript);R    fprintf(outfile, "%s TOTAL FOUND\t%d\n", interactive_name, number_of_accounts);/    for( ii = 0; ii < warning_subscript; ii++) {_!       if( array[0][ii].disacnt) {$N          fprintf(outfile, "%s WARNING USER\t%-12s\t%d\tD\n", interactive_name,/                         array[0][ii].username, 0=                         array[0][ii].number_of_days_dormant);i       }h       else {N          fprintf(outfile, "%s WARNING USER\t%-12s\t%d\t \n", interactive_name,/                         array[0][ii].username, m=                         array[0][ii].number_of_days_dormant);        }     }-    for( ii = 0; ii < watch_subscript; ii++) { !       if( array[1][ii].disacnt) {vL          fprintf(outfile, "%s WATCH USER\t%-12s\t%d\tD\n", interactive_name,/                         array[1][ii].username, o=                         array[1][ii].number_of_days_dormant);4       }        else {L          fprintf(outfile, "%s WATCH USER\t%-12s\t%d\t \n", interactive_name,/                         array[1][ii].username, e=                         array[1][ii].number_of_days_dormant);O       }4    }.    for( ii = 0; ii < active_subscript; ii++) {!       if( array[3][ii].disacnt) { M          fprintf(outfile, "%s ACTIVE USER\t%-12s\t%d\tD\n", interactive_name, .                         array[3][ii].username,=                         array[3][ii].number_of_days_dormant);t       }\       else {M          fprintf(outfile, "%s ACTIVE USER\t%-12s\t%d\t \n", interactive_name,q.                         array[3][ii].username,=                         array[3][ii].number_of_days_dormant);        }E    }-    for( ii = 0; ii < never_subscript; ii++) { !       if( array[2][ii].disacnt) {OJ          fprintf(outfile, "%s NEVER USED\t%-12s\t\tD\n", interactive_name,/                         array[2][ii].username);r       }i       else {J          fprintf(outfile, "%s NEVER USED\t%-12s\t\t \n", interactive_name,/                         array[2][ii].username);        }o    } }%   int main ( void) {i    unsigned int istat, status;    static char command[1025];     char def_cmd[] = "DORMANT ";u'    $DESCRIPTOR( command_desc, command);x)    $DESCRIPTOR( command2_desc, command2); (    $DESCRIPTOR( prompt_id, "DORMANT> ");  $    command_desc.dsc$w_length = 1024;%    command2_desc.dsc$w_length = 1024;;J    istat = lib$get_foreign( &command_desc, 0, &command_desc.dsc$w_length);    check( istat);l-    command[command_desc.dsc$w_length] = '\0';{    strcpy( command2, def_cmd);    strcat( command2, command);<    command2_desc.dsc$w_length = command_desc.dsc$w_length  +       strlen( def_cmd);\9    status = cli$dcl_parse( &command2_desc, DORMANT_TABLE,q2       &lib$get_input, &lib$get_input, &prompt_id);    if( status & 1) {       status = cli$dispatch();    } }yN ============================= dormant.c above ================================N ============================= dormant.hlp below ==============================	 1 DORMANTH  @ 	DORMANT identifies dormant SYSUAF accounts as specified by your 	command line inputs.\   	Format: 		$ DORMANT[/qualifiers...]   A 	You tell DORMANT the criteria for flagging an account as dormantiB 	or almost dormant so you can keep better track of the users that  	are NOT using the system.  
 2 /WARNING  @ 	The /WARNING qualifier indicates the number of days an inactive  	account is reported as dormant.   	Format:" 		$ DORMANT/WARNING=<warning days>  - 		<warning days> can range from 1 - 730 days.e   		DEFAULT:  /WARNING=60o   2 /WATCH  > 	The /WATCH qualifier indicates the number of days an inactive: 	account is reported as being almost dormant.  The accountF 	inactivity must fall between the /WATCH value and the /WARNING value.   	Format: 		$ DORMANT/WATCH=<watch days>  + 		<watch days>	can range from 1 - 730 days.e) 				cannot be larger than <warning days>.S   		DEFAULT:  /WATCH=30,
 2 Examples   		$ DORMANTo  G 	The command above will use the default values for /WATCH and /WARNING.e    ! 		$ DORMANT/WATCH=180/WARNING=210o  C 	The command above will report all accounts that have not been usedME 	interactively for 210 days or more as being dormant.  Also, accountsnH 	which have not been used between 180 days and 210 days will be reported 	as accounts to watch. 	D 2 Output  D 	The final DORMANT report is output to the file DORMANT.RPT in your C 	default directory.  You can save this file for future reference oro! 	print it on a connected printer.N  @ 	NOTE:	The report file contains DEC Special Graphics characters.8 		If you print the file, make sure your printer supports 		these characters.;N ============================= dormant.hlp above ==============================N ========================= dormant_table.cld below ============================ module	dormant_table ident	"v2.3.2" define	verb	dormant\ 	routine	dormant_main % 	qualifier watch, value(type=$number)f' 	qualifier warning, value(type=$number) $ 	qualifier output, value(type=$file) 	qualifier decspec 	qualifier sort_by_usern 	qualifier tabbed_output 	qualifier verbose 	qualifier helpf( 	qualifier interactive,default,negatableN ========================= dormant_table.cld above ============================N ============================= makefile. below ================================ # set up for PRODUCTION   CFLAG	=	/nolist/optimize/nodebug LFLAG	=	/nodebug/notrace  - # uncomment following to override for TESTINGv #CFLAG	=	/list/nooptimizen #LFLAG	=  " # options specific to this program CLOCAL	= LLOCAL	=  0 # options specific to this architecture/compiler CCOMP	=e LCOMP	=   " CFLAGS	=	$(CCOMP)$(CLOCAL)$(CFLAG)$ LINKFLAGS=	$(LCOMP)$(LLOCAL)$(LFLAG)   C	=	.c H	=	.h O	=	.obj E	=	.exe L	=	.lis
 CLD	=	.cld
 RNH	=	.rnh
 HLP	=	.hlp
 HLB	=	.hlb BIN	=	utils:  
 PGM	=	dormanto TBL	=	$(PGM)_table   OBJECTS	=	$(PGM)$(O) $(TBL)$(O)j OBJLIST	=	$(PGM)$(O),$(TBL)$(O)f   MAKEFILE=	makefile.(   compile:	$(PGM)$(E)   " $(PGM)$(E):	$(MAKEFILE) $(OBJECTS)/ 		link /exec=$(PGM)$(E) $(LINKFLAGS) $(OBJLIST)e  " $(PGM)$(O):	$(PGM)$(C) $(MAKEFILE)  $ $(TBL)$(O):	$(TBL)$(CLD) $(MAKEFILE)! 		set command/object $(TBL)$(CLD)w   purge:& 		-purge $(PGM).*,$(TBL).*,$(MAKEFILE)   clean:E 		-if f$$search("$(PGM)$(L)") .nes. "" then delete/nolog $(PGM)$(L);*SE 		-if f$$search("$(PGM)$(O)") .nes. "" then delete/nolog $(PGM)$(O);*EE 		-if f$$search("$(TBL)$(L)") .nes. "" then delete/nolog $(TBL)$(L);*qE 		-if f$$search("$(TBL)$(O)") .nes. "" then delete/nolog $(TBL)$(O);*"   clobber:E 		-if f$$search("$(PGM)$(E)") .nes. "" then delete/nolog $(PGM)$(E);*    install:	compile' 		-copy/log $(PGM)$(E) $(BIN)$(PGM)$(E)[- 		-set file/protection=(g,w) $(BIN)$(PGM)$(E)%   install_purge:	installP 		-if f$$search("$(BIN)$(PGM)$(E)") .nes. "" then purge /keep=2 $(BIN)$(PGM)$(E)N ============================= makefile. above ================================   ------------------------------   Date: 14 Jun 00 13:56:37 EDT From: grant@rigel.cc.wmich.edu  Subject: Re: Seaching SYSUAF.DAT/ Message-ID: <po6wTny1UoJQ@mcduck.acs.wmich.edu>{  G This is pretty trivial if you have Datatrieve and the record definitiono; for SYSUAF.  We do similar ad hoc things fairly frequently.q  D In article <8i8bij$33l$1@nnrp1.deja.com>, trdorr@my-deja.com writes:G > I'm looking for a way to search SYSUAF.DAT for accounts that have not H > been active for greater than 30 days and less than 60 days. Is there aD > way to use the LAST Login date to produce a list of users that are > inactive?s	 > Thanks,, > TomM >  > ( > Sent via Deja.com http://www.deja.com/ > Before you buy.    ------------------------------  % Date: Wed, 14 Jun 2000 18:02:35 +0100n, From: Ted Allwood <support@leva.leeds.ac.uk>! Subject: re: Searching SYSUAF.DAT)3 Message-ID: <009EB9B3.2EE32A6E.12@leva.leeds.ac.uk>s   trdorr@my-deja.com wrote:R > G > I'm looking for a way to search SYSUAF.DAT for accounts that have notqH > been active for greater than 30 days and less than 60 days. Is there aD > way to use the LAST Login date to produce a list of users that are > inactive?A	 > Thanks,\ > Tom   7 I wonder how much scanning sysuaf will really tell you?O  D These days, the majority of accesses to my VMS system are via Samba E (file access and printing), POP3 (mail) and authentication of access qC to Web pages (OSU server).  All of these validate passwords againstiB sysuaf records but none of them change the date fields therein.  I7 would guess that the same applies for many other sites.A  L Username management in a University environment can be especially difficult G (graduate students don't so much leave as appear less frequently 8-) soiD identifying 'inactive' accounts is a problem that I'd like to manageJ in a better way as well - but scanning sysuaf isn't the (complete) answer.     Regards, Ted   K -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-qK Support@leva.leeds.ac.uk                                Tel:  0113 233 2167pG School of Mechanical Engineering,  University of Leeds,  Leeds  LS2 9JTM   ------------------------------  % Date: Wed, 14 Jun 2000 02:13:47 -0400 * From: David A Froble <davef@tsoft-inc.com> Subject: Some HW questions- Message-ID: <3947229B.E1C7209C@tsoft-inc.com>-  M I've recently gotten my hands on a DEC TZ87N.  What tapes does this use?  DLTm III?  O I have a device called a terminal concentrator.  A Digital model PC4XB-DB.  Has J 8 RJ45 jacks, an 8-pin circular jack, and came with a BC13K-10 cable.  Any ideas?   Dave   -- f4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com6 T-Soft, Inc.  170 Grimplin Road  Vanderbilt, PA  15486   ------------------------------  % Date: Mon, 12 Jun 2000 16:07:19 +0200e) From: Etienne Joubert <etienne@citec.net> 
 Subject: test ) Message-ID: <3944EE97.7978E0BA@citec.net>t  > <!doctype html public "-//w3c//dtd html 4.0 transitional//en"> <html>
 &nbsp;</html>i   ------------------------------  # Date: Wed, 14 Jun 2000 16:00:11 GMT(9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)f Subject: Re: VAX on Intel?+ Message-ID: <Up88Xf1EAipi@eisner.decus.org>   g In article <3946F4B3.A83AA58E@earthlink.net>, "David J. Dachtera" <djesys.nospam@earthlink.net> writes:- > Larry Kilgallen wrote: >> rj >> In article <3946BE75.66A86FBA@earthlink.net>, "David J. Dachtera" <djesys.nospam@earthlink.net> writes: >> > Hoff Hoffman wrote: >> >>cm >> >> In article <3945A986.FD0E0421@earthlink.net>, "David J. Dachtera" <djesys.nospam@earthlink.net> writes:-M >> >> :So, someone tell me: under EXACTLY what conditions would expanding thebI >> >> :OpenVMS user base harm Compaq's stock price or its investors???!!!i >> >> M >> >>   Please make your case -- I'd love to see this port.  In the interrum, N >> >>   please see the OpenVMS FAQ, and specifically please see section VMS11. >> >( >> > What further case needs to be made? >>  C >> I think the idea is that you have to make your point (with whichuC >> I happen to disagree) to Compaq management.  It sounds much more-D >> convincing coming from a customer than from employees like Steve. > C > Well, I could be wrong, but I assume they've all been to businessoJ > school. They shouldn't need a tech. nerd like me to come along and teachD > them what they should have known to get their degrees in the first > place.  @ I interpret the phrase "make your case" to mean "convince them".& Presumably you have not done that yet.  K > - I can talk until I'm blue in the face and they still won't understand.    # So another tactic will be required.t   ------------------------------  % Date: Wed, 14 Jun 2000 10:24:19 -0400-' From: "Bill Todd" <billtodd@foo.mv.com>-/ Subject: Re: VMS File Caching Futures and so onE( Message-ID: <8i84bt$5q0$1@pyrite.mv.net>  5 Glenn C. Everhart <Everhart@GCE.com> wrote in message|! news:3946A50F.5D5BEAFD@GCE.com...- > A bit of summary.-? > 1. VMS caching has been a problem for some time, and is beingt= > addressed. The forthcoming 7.3 fieldtest will see the firsttA > public light for the new caching system. There are still things 3 > that can improve but it should be worth the wait.] >e? > 2. On the Veritas homepage are some whitepapers detailing them> > unix caching scheme (and their replacement scheme). Note how; > in both, user data is not written to disk till long after @ > the program is done; there is no mention of even ensuring that- > data is written before closing the program,   J While I haven't checked to make sure it was still there for Glenn to find,L there used to a page indicating that whether files were flushed on close wasF a mount-time option.  So the rest of his comments don't apply - unlessJ that's the behavior desired when the file system was mounted (which is, ofJ course, the expected behavior in the Unix environment, so really shouldn'tK surprise anyone:  people there understand the need to flush files at points + where guaranteed persistence is important).f    so in principle< > a program written for the default situation in unix cannot= > know that its data was written correctly. I have seen casesi? > where treating user data this way, for an app written without-B > being laced with explicit flushes, can lose application metadata+ > and royally corrupt its data as a result.M  L Such cases have a technical term in this industry:  they're called 'bugs' inJ the application, and have been known to occur even in the VMS environment.    Admittedly the Veritas=? > f/s allows one to specify write-through; it appears not to be{ > the default.  L No, it's not the default, any more than writing through RMS sequential filesK is:  in both cases, the only way the user has to ensure that the data is on H disk is to flush it (or, on Unix, 'synch' the file) explicitly.  The oneH difference is that in the Unix environment dirty data can be shared withJ another program without necessarily forcing it to disk - behavior which isH well-defined in that environment and (as with all well-defined behavior)- causes problems only if used inappropriately.-   >-; > 3. When writing to nla0:, every record results in a $qio.-? > When writing to a ramdisk, RMS and language runtime libraries = > can be between your app and the disk. Ditto for hard disks,f@ > though default behavior can change; RMS has lots of variations: > in its defaults depending on device characteristic bits.  K Dear me:  that sounds like at least as much a recipe for user confusion and\. consequent disaster as the Unix behavior does.  @ > It is well to count I/O operations (show dev/full shows these)= > to see whether every record is writing disk or every block.e> > VMS underlying block I/O has generally been faster than unix> > block I/O in tests at low level, but a user app that invokesA > multiple levels of additional interpretation slows things down.|< > VMS engineering is I believe well aware of this. Still one< > should not expect default VMS behavior to ever change to a" > setting that can lose user data.  K Nor would any sensible person advocate such a change.  However, it's always F possible to define new environments (or possibly file attributes) thatE support new default behavior without affecting existing applications.    - bill   > < > 4. The VMS compilers I've seen have had the most wonderful@ > diagnostics I've encountered, pointing out errors which sailed> > by compilers from old RSX, from Microsoft, from Absoft, from@ > Sun. While I must confess (in a joking way here) that it would? > be refreshing for a C dialect to allow itself to be expressedt> > in Macro-11, allowing some of that good old code to migrate,= > C and its descendants (both legitimate and Morganatic) have ; > proven their worth. Much more C has been ported in nativee. > format than has Macro-11 (or pal-11r or...). >,: > 5. Now that QT is ported to VMS, anyone up for compiling > the K office?  >n > Glenn Everhart   ------------------------------  # Date: Wed, 14 Jun 2000 11:48:04 GMT  From: bazzasmith@my-deja.com  Subject: VMS Printing using PCL6) Message-ID: <8i7rdh$m5n$1@nnrp1.deja.com>%  B I have successfully printed to a HP4000TN using PCL6 to format the print.  E What I would like to do is print a bitmap file (ie company logo) alson using PCL via VMS.  
 Any ideas?   Thanks in advance. Barryr    & Sent via Deja.com http://www.deja.com/ Before you buy.e   ------------------------------  % Date: Wed, 14 Jun 2000 08:11:24 -0500 ) From: "John E. Malmberg" <wb8tyw@qsl.net>$$ Subject: Re: VMS Printing using PCL67 Message-ID: <009601bfd602$0bb01130$020a0a0a@xile.realm>   % Barry <bazzasmith@my-deja.com> wrote:   D > I have successfully printed to a HP4000TN using PCL6 to format the > print. > G > What I would like to do is print a bitmap file (ie company logo) also, > using PCL via VMS. >  > Any ideas?  H It can depend on what method you are using to print to the printer.  TheH basic major variations are: DCPS, LPR/LPD, or Normal VMS print symbionts Direct/LAT/TELNETSYM.v  J The specific instructions for DCPS are a bit different from that for otherI print symbionts, as it is documented to allow multiple text libraries per H printer, and the text libraries can be targeted to specific print modes.    H For each model of printer, create a device control library with a set ofH company standard module names.  This will insulate your users / programs# from differences in printer models.4  K For the PCL printers, save the bitmap into a text file of a print image anddI put it in a PCL specific text library with a module name like "corplogo".e  H It is also reasonably easy to do this with a bitmap saved as a postcriptH file.  I have not tried SIXEL, as there is no real need, but that should
 work also.    L Then you define a form such as LOGOPAPER that invokes your "corplogo" module. as either a setup module or page setup module.     -John  wb8tyw@qsl.network   ------------------------------  % Date: Wed, 14 Jun 2000 09:31:08 +0200iE From: "David Andreas Alderud" <aaldv97@student.removethispart.vxu.se>%" Subject: Re: VMS Security features& Message-ID: <8i7brc$1qb$1@news.lth.se>  F > >> > I was not refering to VMS, only to OSs in general, I don't know enough > >aboutL > >> > VMS to make such a statement :). Someone once told me that VMS didn't	 > >suffer K > >> > from buffer overruns the way other OS does, so I'm interested in howh themI > >> > stack is implemented. I'll guess I'll see how it's done when I buyc them > >book. > >>I > >> Buffer overruns and stack implementations really aren't related, but; > >> that's a separate issue.h > >2I > >They are. Mybe not in VMS. But the different implementations I know ofm  > >doesn't give full protection.J > >Just because you can't execute code on the stack doesn't mean you can't< > >overrun something with a return adress to malicious code. > I > You "might" be able to place code and data on the stack to execute some I > "malicious" code.  Chances are that the malefactor will only get as far J > as affecting the process running the program.  VMS maintains a process/-J > job series of bits which describe system level function and/or the scopeK > of these system level functions that the process is permitted to perform..I > The "malicious" code would likely fail if the process is not privied to I > execute the functions needed to do its dirty work.  Can you or the mal-rH > icous code affect these bit?  Not unless the process possesses certain > of these privie bits.m  C Pretty advanced hardware, and a lot of status registers, I imagine.   J > In addition, the stack has guard pages, as well as, PTE protections thatJ > would make a stack overwrite into the areas of a more privileged (inner)I > mode stack cause an ACCess VIOlation (ACCVIO).  Unless the code running H > at usermode attempting to write over the stack knew of the location ofJ > these inner mode stacks and held the appropriate privies and attains theI > appropriate access level to write to these stacks, the malevolence of aeH > stack overrun is likely to be rather benign.  In addition, the "guard"J > pages are normally "NA -- No Access".  In otherwords, even the innermostH > privied access mode -- kernel -- cannot write to these pages without aH > knowledge of the memory management sub-system and how to muck with theI > PTEs mapping these pages and invalidating any hardware TBs (Translationu" > Buffers) related to these pages.  D Ok, but the stack guards will not always be hit on an bufferoverrun.  F > >I don't know how much system cracking you've done, but I know these thingsH > >very well I'm sad to admit, that's why I'm writing this report, to do some > >good. :)  >dJ > Cracking unix is *not* cracking VMS.  You can crack a pistachio nut withJ > your hands and get inside but the same techniques you've employed to getK > at the "kernel" there will not help you if you're working on say a Brazil  > nut.  J As I said, "Mybe not in VMS.", meaning that what I said might not apply to VMS.G There are still several stupid things one can do in VMS, like cloaking.=  F > >Mybe VMS puts some stack-guards in the stack to, sometimes, make it possibleG > >to detect overruns, but it doesn't give full protection for apperentg > >reasons.$ >bI > Guard pages and access mode protection.  Both will thwart your attemptssH > to get at the inner modes to do damage unless you're privied to do so. > K > >> In general, the pages allocated to the stack on VMS aren't executable.a IfK > >> you try and execute code on the stack your process will throw an errore andoD > >> die. Memory a process allocates is also generally not marked asK > >> executable, and attempts to execute randomly allocated data will fail.oH > >> Finally the pages of your process that are executable are generally> > >> marked as read-only, so attempts to alter them will fail. > > G > >But the important thing, how does VMS protect the executable processA memory@ > >space from beeing overwritten with malicious executable data?! > >Is this what the Monitor does?  >	H > You're obviously not going to understand this until you spend a littleI > time understanding VMS and, specifically, memory management and the PTEO  > access mode protection scheme.  $ Feal like donating some hardware? =)  D > >> All the system services that take arbitrary chunks of memory asL > >> parameters, be they strings or buffers or whatnot, use a data structureI > >> called a descriptor to describe the buffer, and one of the pieces ofL thatG > >> descriptor is a buffer length. The system service code (as well asL prettyI > >> much all the languages on VMS except for C) can automagically handle  thesec9 > >> things and also respect the buffer length parameter.	 > >$K > >I know, but isn't C the most common developent language on VMS, or is it	 > >highperformance FORTRAN?n > ) > Gee, it's always been macro for me!  :)l  G Well, personaly I prefer assembler. So the VAX would be perfect for me,n- changing that instructionset and all that. :)a  L > >I recomend Ada95, fast development and well protected programs, a bit big> > >runtime enviorment but for what it does it's well worth it. > >=L > >Thanks for your input, though, I don't agree with everything you said. :) >  > Is it disagreement or denial?u  H I didn't agree with him that stack implementation has nothing to do with buffer overruns, that's all.J I also said "Mybe not in VMS.", meaning that what I was about to say might not apply to VMS.i  I > If you are truly looking to write a paper on some aspect of system sec-cH > utity and you want to include/cover VMS, you'd be best to read some ofH > the "wonderful" VMS documentation available or find a time to sit downI > with a VMS (and I hate this term) guru and learn some of the nuances oft5 > the operating system used to maintain its security.v  E That might be hard, I don't know of anyone running VMS. I know people-( running alphas but they all use OpenBSD.   ------------------------------  % Date: Wed, 14 Jun 2000 11:03:01 +0100a* From: "Richard Brodie" <R.Brodie@rl.ac.uk>" Subject: Re: VMS Security features, Message-ID: <8i7l8j$1430@newton.cc.rl.ac.uk>  P "David Andreas Alderud" <aaldv97@student.removethispart.vxu.se> wrote in message  news:8i7brc$1qb$1@news.lth.se...  K > > The "malicious" code would likely fail if the process is not privied toeK > > execute the functions needed to do its dirty work.  Can you or the mal-rJ > > icous code affect these bit?  Not unless the process possesses certain > > of these privie bits.a >pE > Pretty advanced hardware, and a lot of status registers, I imagine.   M No, this is a software issue. The assignment of the least privilege needed toeR do the job is a fairly fundamental concept in security. At the core, it all reliesL on the system being able to maintain the integrity of kernel data structures via hardware memory protection.d  I chroot is the best example I can think of as a similar mechanism in Unix.a@ The Java online documentation is good background reading on fine grained security.s   ------------------------------  # Date: Wed, 14 Jun 2000 10:34:38 GMT== From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)u" Subject: Re: VMS Security features0 Message-ID: <009EB953.13F20DD5@SendSpamHere.ORG>  n In article <8i7brc$1qb$1@news.lth.se>, "David Andreas Alderud" <aaldv97@student.removethispart.vxu.se> writes:
 >{...snip...}4J >> You "might" be able to place code and data on the stack to execute someJ >> "malicious" code.  Chances are that the malefactor will only get as farK >> as affecting the process running the program.  VMS maintains a process/-dK >> job series of bits which describe system level function and/or the scope5L >> of these system level functions that the process is permitted to perform.J >> The "malicious" code would likely fail if the process is not privied toJ >> execute the functions needed to do its dirty work.  Can you or the mal-I >> icous code affect these bit?  Not unless the process possesses certainb >> of these privie bits. >-D >Pretty advanced hardware, and a lot of status registers, I imagine.  J Not necessarily.  On Alpha, as a performance decision, there are no statusI registers.  Any machine which supports virtual memory will have some verylK similar hardware whether or not the O/S takes advantage of said hardware is, another thread altogether.  K >> In addition, the stack has guard pages, as well as, PTE protections that K >> would make a stack overwrite into the areas of a more privileged (inner)AJ >> mode stack cause an ACCess VIOlation (ACCVIO).  Unless the code runningI >> at usermode attempting to write over the stack knew of the location ofrK >> these inner mode stacks and held the appropriate privies and attains theHJ >> appropriate access level to write to these stacks, the malevolence of aI >> stack overrun is likely to be rather benign.  In addition, the "guard"sK >> pages are normally "NA -- No Access".  In otherwords, even the innermosteI >> privied access mode -- kernel -- cannot write to these pages without a>I >> knowledge of the memory management sub-system and how to muck with the J >> PTEs mapping these pages and invalidating any hardware TBs (Translation# >> Buffers) related to these pages.e >tE >Ok, but the stack guards will not always be hit on an bufferoverrun.n  J Stack regions are carved out of VMS process P1 address space.  This is theJ 32 bit addressable virtual address region which starts at address 7FFFFFFFI hex and grows downward to 40000000 hex.  The act of pushing something on  J the stack involved decrementing the value of the stack pointer used to ad-J dress the stack and then move the value to this newly derived address.  InL VMS, the less privied stacks (sans the user stack) start with the less priv-L ied stack in the higher address locations --  SUPERVISOR, EXECUTIVE and thenK KERNEL.  The access mode protections keep stack overruns in these mods from J corrupting inner mode stacks.  The kernel stack has guard pages at its ex-I tremes because of the nature of kernel mode.  Thus under and over runs ineJ kernel mode will cause an ACCVIO.  The USER mode stack grows downward fromJ the very lowest portion of P1 space (CTL$GL_CTLBASVA) to the virtual maxi-I mum of 40000000 hex.  USER mode stack space is created and destroyed withi each image activation.    K It simply doesn't matter.  IF YOU DO NOT POSSESS THE PRIVILEGE AND YOU HAVEiK NOT ATTAINED THE PROCESSOR ACCESS MODE NEEDED, YOU CAN NOT WRITE TO A PRIV-  IED ACCESS MODE STACK.    K Buffer overruns are also checked in system service routines -- the ultimateaK API in most all of the run-tile libraries and other APIs in VMS.  One post-oJ er pointed out to you that data is passed to these services in the form ofJ a descriptor -- a data structure which describes the data with informationI as to its length and its location.  These services then check the access-tK iblity of the data with access probing instructions (PROBEx).  These probeseJ are done using the "previous" mode of the caller.  Thus, if you've not at-K tained a specified access mode, you can not ask a system service to performtH some task which would corrupt the mechanisms which the particular system, service is designed control or manipulate.    K >> Cracking unix is *not* cracking VMS.  You can crack a pistachio nut witheK >> your hands and get inside but the same techniques you've employed to gettL >> at the "kernel" there will not help you if you're working on say a Brazil >> nut.f >aK >As I said, "Mybe not in VMS.", meaning that what I said might not apply tot >VMS.aH >There are still several stupid things one can do in VMS, like cloaking.  9 Would that entail use of the Romulan Cloaking device?  ;)   K About the only stupid thing one can do in VMS is to employ a stupid system tL manager!  Along with privilege comes liability.  If you've got a stupid sys-L tem manage unknowledgable of the potential problems that granting privilegesK can cause, then your system is vulnerable.  Granting privileges also reliessI on a level of trustworthiness of the individual(s) that are granted theseeH privies.  If you guard your system's security as well as nuclear weaponsG secrets are guarded at Los Alamos labs, you're certainly infor trouble.   I >> You're obviously not going to understand this until you spend a littleeJ >> time understanding VMS and, specifically, memory management and the PTE! >> access mode protection scheme.e >d% >Feal like donating some hardware? =)   G Sure.  I'll send you a VAX 11/780 free of charge.  You only have to payn the shipping charges.  ;)d  * >> Gee, it's always been macro for me!  :) > H >Well, personaly I prefer assembler. So the VAX would be perfect for me,. >changing that instructionset and all that. :)  	 Good boy!i    >> Is it disagreement or denial? >dI >I didn't agree with him that stack implementation has nothing to do with( >buffer overruns, that's all.oK >I also said "Mybe not in VMS.", meaning that what I was about to say mightb >not apply to VMS.  I VMS is well thought out and the engineering group is very concerned with hH the security aspects of the O/S, unlike another popular O/S which was anG assembly of hacks from its genesis.  VMS engineering also doesn't throwuI out the welcome mat to stupid user requested features eventually exploit-lI ed by prepubescent dweebs like those in the Phillipines to perpetrate thew "I Love You" email.o   >eJ >> If you are truly looking to write a paper on some aspect of system sec-I >> utity and you want to include/cover VMS, you'd be best to read some of I >> the "wonderful" VMS documentation available or find a time to sit down J >> with a VMS (and I hate this term) guru and learn some of the nuances of6 >> the operating system used to maintain its security. > F >That might be hard, I don't know of anyone running VMS. I know people) >running alphas but they all use OpenBSD.n  # http://www.openvms.digital.com:8000l   OpenVMS doc set on-line.   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM    ------------------------------  % Date: Wed, 14 Jun 2000 13:31:57 +0200.E From: "David Andreas Alderud" <aaldv97@student.removethispart.vxu.se> " Subject: Re: VMS Security features& Message-ID: <8i7puq$52b$1@news.lth.se>  F > >Pretty advanced hardware, and a lot of status registers, I imagine. >iL > Not necessarily.  On Alpha, as a performance decision, there are no statusK > registers.  Any machine which supports virtual memory will have some verylJ > similar hardware whether or not the O/S takes advantage of said hardware is > another thread altogether.  > Yeah, like NetBSD for example, though it sure is FAST anyways.    G > >Ok, but the stack guards will not always be hit on an bufferoverrun.0 > L > Stack regions are carved out of VMS process P1 address space.  This is theL > 32 bit addressable virtual address region which starts at address 7FFFFFFFJ > hex and grows downward to 40000000 hex.  The act of pushing something onL > the stack involved decrementing the value of the stack pointer used to ad-L > dress the stack and then move the value to this newly derived address.  InH > VMS, the less privied stacks (sans the user stack) start with the less priv-oI > ied stack in the higher address locations --  SUPERVISOR, EXECUTIVE andT thenH > KERNEL.  The access mode protections keep stack overruns in these mods fromL > corrupting inner mode stacks.  The kernel stack has guard pages at its ex-K > tremes because of the nature of kernel mode.  Thus under and over runs in L > kernel mode will cause an ACCVIO.  The USER mode stack grows downward fromL > the very lowest portion of P1 space (CTL$GL_CTLBASVA) to the virtual maxi-K > mum of 40000000 hex.  USER mode stack space is created and destroyed witha > each image activation. >nH > It simply doesn't matter.  IF YOU DO NOT POSSESS THE PRIVILEGE AND YOU HAVEG > NOT ATTAINED THE PROCESSOR ACCESS MODE NEEDED, YOU CAN NOT WRITE TO At PRIV-  > IED ACCESS MODE STACK. >h > D > Buffer overruns are also checked in system service routines -- the ultimateG > API in most all of the run-tile libraries and other APIs in VMS.  Oneo post-sL > er pointed out to you that data is passed to these services in the form ofL > a descriptor -- a data structure which describes the data with informationK > as to its length and its location.  These services then check the access-sF > iblity of the data with access probing instructions (PROBEx).  These probesL > are done using the "previous" mode of the caller.  Thus, if you've not at-E > tained a specified access mode, you can not ask a system service to  performeJ > some task which would corrupt the mechanisms which the particular system, > service is designed control or manipulate.   Interesting.  J > >As I said, "Mybe not in VMS.", meaning that what I said might not apply to > >VMS. J > >There are still several stupid things one can do in VMS, like cloaking. >h; > Would that entail use of the Romulan Cloaking device?  ;)p   Nah, something like this:o   $ save_verify = 'f$verify(0)' 1 $ system = "vax"        !Set to "alpha" for Alpha  $!) $!  File to build Ehud Gavron's INVISIBLEg $! $!  Author:     Hunter Goatley $! $ say := write sys$output   $ on error then goto common_exit# $ on contrl_y then goto common_exit 7 $ say "Extracting $JIBDEF and $PCBDEF from LIB.MLB...."i? $ library/macro/extr=$JIBDEF/out=jibdef.mar sys$library:lib.mlbv? $ library/macro/extr=$PCBDEF/out=pcbdef.mar sys$library:lib.mlbh1 $ say "Converting $*DEF macros to C .H files...."l $ call convert_to_h jibdef.mar $ call convert_to_h pcbdef.mar $ say "Compiling INVISIBLE...."t $ cc invisible $ say "Linking INVISIBLE...."c3 $ link/notrace invisible,invisible.opt_'system'/opts! $ say "INVISIBLE build completed"i $ common_exit:' $       exit f$verify(save_verify).or.1i $ convert_to_h: subroutine! $ name = f$parse(p1,"","","NAME")  $ open/read tmp 'p1' $ create 'name'.H  $ open/append tmph 'name'.Hh $ cvt_loop:m  $    read/error=cvt_fin tmp line9 $    if f$extract(0,4,line).nes."$EQU" then goto cvt_loopn0 $    write tmph "#define ",f$extract(4,255,line) $    goto cvt_loop
 $ cvt_fin: $    close tmp $    close tmpht4 $ write sys$output "C header file ''name'.H created" $ exit $ endsubroutinet  L > About the only stupid thing one can do in VMS is to employ a stupid systemI > manager!  Along with privilege comes liability.  If you've got a stupidt sys-C > tem manage unknowledgable of the potential problems that grantingt
 privilegesF > can cause, then your system is vulnerable.  Granting privileges also reliesK > on a level of trustworthiness of the individual(s) that are granted theselJ > privies.  If you guard your system's security as well as nuclear weaponsI > secrets are guarded at Los Alamos labs, you're certainly infor trouble.r >i' > >Feal like donating some hardware? =)  >vI > Sure.  I'll send you a VAX 11/780 free of charge.  You only have to pay  > the shipping charges.  ;)u   Hehe.oL Well, after I talked to my professor we decided that I should take a look at. VMS and OS/400 when and if I ever do a PhD. ;)J I would like to thank you all for your input, I'd really love to play withH VMS, I will hopefully get the chance one day in a not to distant future.  , > >> Gee, it's always been macro for me!  :) > >dJ > >Well, personaly I prefer assembler. So the VAX would be perfect for me,0 > >changing that instructionset and all that. :) >  > Good boy!e  A Since I'm an old MC68k assembler demo coder I've come to love it.$G Though I think the Intel architecure sucks, not much I can do about it,t atleast they are cheap.n2 Hopefully the IBM POP-PPC boards will be out soon.  " > >> Is it disagreement or denial? > >dK > >I didn't agree with him that stack implementation has nothing to do withn > >buffer overruns, that's all.wG > >I also said "Mybe not in VMS.", meaning that what I was about to sayn mighte > >not apply to VMS. >dJ > VMS is well thought out and the engineering group is very concerned withJ > the security aspects of the O/S, unlike another popular O/S which was anI > assembly of hacks from its genesis.  VMS engineering also doesn't throwgK > out the welcome mat to stupid user requested features eventually exploit-oK > ed by prepubescent dweebs like those in the Phillipines to perpetrate the  > "I Love You" email.y  < There is a difference, it's security on a much higher level.L I bet there will be similar problems with those Linux distros, GNOME and KDEE looks and works very similar to Windows, just a question of time, theA- intelligence of the users seams to match too.k   > >mL > >> If you are truly looking to write a paper on some aspect of system sec-K > >> utity and you want to include/cover VMS, you'd be best to read some ofsK > >> the "wonderful" VMS documentation available or find a time to sit downeL > >> with a VMS (and I hate this term) guru and learn some of the nuances of8 > >> the operating system used to maintain its security. > >eH > >That might be hard, I don't know of anyone running VMS. I know people+ > >running alphas but they all use OpenBSD.e >R% > http://www.openvms.digital.com:8000r >o > OpenVMS doc set on-line.  L Yep, I've read it all. But without hardware I can't really make a subjective report I'm afraid.G I'll try to continue my work and hopefully one day it will include VMS.m   Thanks for the help.   ------------------------------  # Date: Wed, 14 Jun 2000 13:07:45 GMT = From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) " Subject: Re: VMS Security features0 Message-ID: <009EB968.77A4795F@SendSpamHere.ORG>  n In article <8i7puq$52b$1@news.lth.se>, "David Andreas Alderud" <aaldv97@student.removethispart.vxu.se> writes:
 >{...snip...}gK >> >There are still several stupid things one can do in VMS, like cloaking.  >>< >> Would that entail use of the Romulan Cloaking device?  ;) >u >Nah, something like this: >h >$ save_verify = 'f$verify(0)'2 >$ system = "vax"        !Set to "alpha" for Alpha >$! * >$!  File to build Ehud Gavron's INVISIBLE >$!s >$!  Author:     Hunter Goatley  >$!i >$ say := write sys$output! >$ on error then goto common_exito$ >$ on contrl_y then goto common_exit8 >$ say "Extracting $JIBDEF and $PCBDEF from LIB.MLB...."@ >$ library/macro/extr=$JIBDEF/out=jibdef.mar sys$library:lib.mlb@ >$ library/macro/extr=$PCBDEF/out=pcbdef.mar sys$library:lib.mlb2 >$ say "Converting $*DEF macros to C .H files...." >$ call convert_to_h jibdef.maru >$ call convert_to_h pcbdef.mar   >$ say "Compiling INVISIBLE...." >$ cc invisible  >$ say "Linking INVISIBLE...."4 >$ link/notrace invisible,invisible.opt_'system'/opt" >$ say "INVISIBLE build completed" >$ common_exit:r( >$       exit f$verify(save_verify).or.1 >$ convert_to_h: subroutines" >$ name = f$parse(p1,"","","NAME") >$ open/read tmp 'p1'v >$ create 'name'.H >$ open/append tmph 'name'.H >$ cvt_loop:! >$    read/error=cvt_fin tmp linei: >$    if f$extract(0,4,line).nes."$EQU" then goto cvt_loop1 >$    write tmph "#define ",f$extract(4,255,line)i >$    goto cvt_loops >$ cvt_fin:b >$    close tmpe >$    close tmph5 >$ write sys$output "C header file ''name'.H created"  >$ exit  >$ endsubroutine    H Have you attempted to use this code from a process granted basic TMPMBX G and NETMBX privileges?  It doesn't work.  This "INVISIBLE" program will-G require the process to possess CMKRNL privilege.  If you've already at-S2 tained CMKRNL privileges, the game's already over.   >Hehe.M >Well, after I talked to my professor we decided that I should take a look atc/ >VMS and OS/400 when and if I ever do a PhD. ;)eK >I would like to thank you all for your input, I'd really love to play with"I >VMS, I will hopefully get the chance one day in a not to distant future.i   Don't pass it up.     B >Since I'm an old MC68k assembler demo coder I've come to love it.H >Though I think the Intel architecure sucks, not much I can do about it, >atleast they are cheap.3 >Hopefully the IBM POP-PPC boards will be out soon.l  J You ought to find VAX macro quite simple if you're familiar with the 68K's assembly lingo.   = >There is a difference, it's security on a much higher level. M >I bet there will be similar problems with those Linux distros, GNOME and KDEtF >looks and works very similar to Windows, just a question of time, the. >intelligence of the users seams to match too.   Morons get what they deserve.r   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM    ------------------------------    Date: 14 Jun 2000 10:10:15 -0500 From: briggs@eisner.decus.orga" Subject: Re: VMS Security features+ Message-ID: <D76Cg$SZ8LFI@eisner.decus.org>c  n In article <8i23e7$k72$1@news.lth.se>, "David Andreas Alderud" <aaldv97@student.removethispart.vxu.se> writes:M > But the important thing, how does VMS protect the executable process memory ? > space from beeing overwritten with malicious executable data?n  > Is this what the Monitor does?  E Executable process memory is typically allocated in read-only PSECTS.p  J So you can't overwrite it without messing with memory protection/ownership first.  G Don't know what you mean by "Monitor".  But, if you run your code underrL control of the VMS debugger, it will make your executable PSECTS read/write.J This is to allow the debugger itself (which runs with no special privilegeI in user mode) to insert breakpoint traps into the code to implement break  points.e  L But the ability to change memory protection in this manner is not a securityG problem because you have to contrive to get your own code running to dooF it.  And, if you could get your own code running, you wouldn't need to? write over the instruction stream to get your own code running.t  E You can't $ RUN/DEBUG an image linked with /NOTRACEBACK.  And you aretF warned if you try to install a /TRACEBACK-linked image with privilege.  E Fixed back in VMS version 4 was the ability for a user to cause theiroL own shared libraries to be loaded in place of the standard systems librariesK when invoking installed, privileged images (think Unix setuid executables).-  B >> All the system services that take arbitrary chunks of memory asJ >> parameters, be they strings or buffers or whatnot, use a data structureL >> called a descriptor to describe the buffer, and one of the pieces of thatL >> descriptor is a buffer length. The system service code (as well as prettyM >> much all the languages on VMS except for C) can automagically handle theseG7 >> things and also respect the buffer length parameter.S > J > I know, but isn't C the most common developent language on VMS, or is it > highperformance FORTRAN?  H Are you suggesting buffer overrun attacks on portions of VMS that happen to have been implemented in C?  J You're going to have a hard time attacking the kernel that way because, asF was mentioned, the system services tend to use descriptors rather thanJ null-terminated strings.  Even C programmers tend not to use "strcpy" whenD they are not using null terminated strings.  And there's little call for "gets" in the kernel.Y  E A buffer overrun attack on an installed privileged image written in C $ might or might not be more fruitful.  & 	John Briggs			briggs@eisner.decus.org   ------------------------------  # Date: Wed, 14 Jun 2000 17:00:38 GMTi9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)e" Subject: Re: VMS Security features+ Message-ID: <g0kry41e0Exj@eisner.decus.org>c  n In article <8i7puq$52b$1@news.lth.se>, "David Andreas Alderud" <aaldv97@student.removethispart.vxu.se> writes:  K >> >There are still several stupid things one can do in VMS, like cloaking.s >>< >> Would that entail use of the Romulan Cloaking device?  ;) >  > Nah, something like this:n >  > $ save_verify = 'f$verify(0)'p3 > $ system = "vax"        !Set to "alpha" for Alpha. > $!+ > $!  File to build Ehud Gavron's INVISIBLEd > $!  > $!  Author:     Hunter Goatley > $! > $ say := write sys$outputb" > $ on error then goto common_exit% > $ on contrl_y then goto common_exita9 > $ say "Extracting $JIBDEF and $PCBDEF from LIB.MLB...."dA > $ library/macro/extr=$JIBDEF/out=jibdef.mar sys$library:lib.mlbrA > $ library/macro/extr=$PCBDEF/out=pcbdef.mar sys$library:lib.mlb   ? In order to manipulate PCBs and JIBs you need CMKRNL privilege.eG Thus the above is countered by Brian's rule about giving out privilege.f  C So you were correct -- you do need hardware, in order to do a quick A and easy test and determine that the above would not work without 
 privilege.   ------------------------------  # Date: Wed, 14 Jun 2000 16:55:40 GMTr9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)o" Subject: Re: VMS Security features+ Message-ID: <+1keGLIjGCLw@eisner.decus.org>s  p In article <009EB953.13F20DD5@SendSpamHere.ORG>, system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-) writes:p > In article <8i7brc$1qb$1@news.lth.se>, "David Andreas Alderud" <aaldv97@student.removethispart.vxu.se> writes:  L >>As I said, "Mybe not in VMS.", meaning that what I said might not apply to >>VMS.I >>There are still several stupid things one can do in VMS, like cloaking.s > ; > Would that entail use of the Romulan Cloaking device?  ;)n > M > About the only stupid thing one can do in VMS is to employ a stupid system wN > manager!  Along with privilege comes liability.  If you've got a stupid sys-N > tem manage unknowledgable of the potential problems that granting privilegesM > can cause, then your system is vulnerable.  Granting privileges also relieshK > on a level of trustworthiness of the individual(s) that are granted these0
 > privies.  E A slight variation on that stupid thing is to buy from an incompetentcD vendor. A lot of vendor software gets installed with privilege (withB good reason in many cases), but it is critical that the people whoC wrote that privileged code know what they are doing.  Typically yougC do not get source of commercial products, but you can often measurelC their competence by the external security aspects of the product. I?B have been in a situation where a vendor told the customer that theA VMS Default account needed BYPASS privilege because "Cobol cannot @ share files".  I know a vendor who insisted on doing Physical IO( for no good reason (a database product).  @ So far as I know, neither of these vendors has a significant VMS customer base.  , By the way, what do you mean by "cloaking" ?   ------------------------------  % Date: Wed, 14 Jun 2000 13:27:37 +0100lB From: Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com>D Subject: Re: Which cpu does Sun server use: I386, Mip, Ppc or Alpha?* Message-ID: <39477A39.7740A67E@uk.sun.com>   "Main, Kerry" wrote:  	 > Andrew,c >nL > >> One could if one was being cruel make the same point about Compaq, "who9 > knows what Compaq really plan with/for MERCED/ALPHA :-)m > <<<t > K > Yep, one could, except that Compaq has been very clear that Alpha will bemI > its high end chip architecture for OpenVMS, Tru64, Himalayan, and Linux$ > OS's.e >tN > As previous reply stated, to make this very clear, Compaq announced it would$ > not be porting Tru64 UNIX to IA64. >pK > Now, if one was being cruel, one could make the point "What is Sun's high$4 > end chip strategy going to be? SPARC III or IA64?" >s  J Kerry people buy systems they don't buy CPU's in isolation. Compaq/DigitalD have always been long on CPU performance, particulary if measured byI benchmarks that don't stress any other system capability but have for the-H last 2-3 years been very short on systems throughput. It may suprise youF to know that something like Oracle can spend more than 50% of its timeE simply doing loads and stores. With this in mind concentrating on CPU F performance will not deliver the performance benefits people hope for.  O This is illustrated by Compaqs inability to lay claim to performance leadershipt  8 for anything that does stress components other than CPU.  G Now Compaq have just released WildFire and in the rosy glow surroundingeM its launch you seem to have missed a few really sorry aspects of the WildFire J design which tend to suggest that Compaq need to be developing WildFire II" for release as early as next year.   What are the sorry aspects ?  F Well memory latency as a starter, the local on node access in WildFireD is 330ns with remote access being 990ns, this would have been OK 3-4F years ago (its similar to the Origin 2000) but does not look too greatB now. The E10K for example has a flat latency from all CPU's to all> memory of ~450ns and to acheive this as an average on WildFireD you need to be getting 80+% local accesses. This is going to be hardB to do just relying on the OS and this may require people to changeB the way they develop applications. Incedentally the WildFire white< paper published by Compaq where WildFire latency is compared? with E10K latency uses a "made up" E10K latency number of 600nsd? which may have made you think that WildFire is more competitivew than it is.t  ? Just for your information the Origin 2000 a 3-4 year old system 9 had at lanuch 310ns local and 950ns remote as illustratedr6 in the white paper SGI published for the origin intro.  ) http://www.sgi.com/origin/images/isca.pdfw  ? Secondly the bisectional bandwidth of the WildFire is about the : same as the E10K, yes Compaq have published higher STREAMS< numbers but these included local on-node memory access which< assumes that a process never goes off node. This is unlikely< except for a very small subset of applications that WildFire
 will support.t  = Now none of these things would have been an issue if WildFirel9 had been announced and delivered 3-4 years ago, but beingu; marginally faster than the E10K for example is hardly goingd9 to ensure that WildFire is competitive once SuperDome andr7 Suns UltraIII machines and the IBM S80 replacement ships# and none of them are that far away.   5 So more pertinent is what is Compaqs high end systems.5 strategy, as you can see it would be unwise to assumeC  ( that WildFire is the answer. :-) :-) :-)   Regardss Andrew Harrisonl Enterprise IT Architecth   ------------------------------  % Date: Wed, 14 Jun 2000 10:56:39 -0400b' From: "Bill Todd" <billtodd@foo.mv.com>eD Subject: Re: Which cpu does Sun server use: I386, Mip, Ppc or Alpha?( Message-ID: <8i868i$7lf$1@pyrite.mv.net>  E Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com> wrote inu, message news:39477A39.7740A67E@uk.sun.com...   ...d  L > Kerry people buy systems they don't buy CPU's in isolation. Compaq/DigitalF > have always been long on CPU performance, particulary if measured byK > benchmarks that don't stress any other system capability but have for thehJ > last 2-3 years been very short on systems throughput. It may suprise youH > to know that something like Oracle can spend more than 50% of its timeG > simply doing loads and stores. With this in mind concentrating on CPUiH > performance will not deliver the performance benefits people hope for. >eF > This is illustrated by Compaqs inability to lay claim to performance
 leadership >i: > for anything that does stress components other than CPU. > I > Now Compaq have just released WildFire and in the rosy glow surroundingaF > its launch you seem to have missed a few really sorry aspects of the WildFireL > design which tend to suggest that Compaq need to be developing WildFire II$ > for release as early as next year. >o > What are the sorry aspects ? > H > Well memory latency as a starter, the local on node access in WildFireF > is 330ns with remote access being 990ns, this would have been OK 3-4H > years ago (its similar to the Origin 2000) but does not look too greatD > now. The E10K for example has a flat latency from all CPU's to all@ > memory of ~450ns and to acheive this as an average on WildFireF > you need to be getting 80+% local accesses. This is going to be hardD > to do just relying on the OS and this may require people to change$ > the way they develop applications.  L Or not.  VMS already supports locality ('RAD' may be the Wildfire term - I'dB have to check) to help keep accesses local to a single board in anC application-transparent manner.  Since a single board can support 4 G processors (each of which will enjoy the process and generational speedcD improvements of the Alpha road map over time, including extension toI simultaneous multi-threading) and up to 32 GB of on-board memory, this issJ not as significant a restriction as you suggest:  in fact, only very *few*D monolithic applications can really benefit much from more than 4-wayL multi-processing (especially if each processor supports multiple interleavedA threads at the hardware level) and more than 32 GB of memory, and J non-monolithic applications are relatively easy to segregate onto multiple: boards that need to interact only relatively infrequently.  K So on-board latency will indeed likely dominate overall average application?G latency (80+% may well be a very conservative estimate for the on-boardtJ fraction), and IIRC there's a copy of VMS on each board to avoid excessiveE system off-board code (and presumably pure data) access.  Given theseeL considerations, plus Alpha's respectable caching facilities, I'd suggest youI not get too complacent about Sun's next-generation competitiveness - evenlL ignoring the fact that creation of the Wildfire follow-on is also well under way.     Incedentally the WildFire white> > paper published by Compaq where WildFire latency is comparedA > with E10K latency uses a "made up" E10K latency number of 600nsuA > which may have made you think that WildFire is more competitives
 > than it is.   K If Compaq's marketdroids have slung any incorrect FUD at Sun, it's entirelysG appropriate to point this out.  As for competitiveness, let's allow theeK benchmarks to tell the tale:  I'm sure that at least a few apples-to-apples   comparisons will be unavoidable.   >eA > Just for your information the Origin 2000 a 3-4 year old systemc; > had at lanuch 310ns local and 950ns remote as illustratedr8 > in the white paper SGI published for the origin intro. > + > http://www.sgi.com/origin/images/isca.pdf  >rA > Secondly the bisectional bandwidth of the WildFire is about thei< > same as the E10K, yes Compaq have published higher STREAMS> > numbers but these included local on-node memory access which> > assumes that a process never goes off node. This is unlikely> > except for a very small subset of applications that WildFire > will support.t  J As pointed out above, this is in fact likely for the overwhelming majorityG of applications Wildfire will support, and the majority of the rest arepL easily partitionable to meet this criterion.  Leaving aside the fact for forI *really* monstrous applications, neither E10K nor Wildfire is sufficient: I you've got to expand into clustering, and Sun really doesn't want to takee  this competitive argument there.   >n? > Now none of these things would have been an issue if WildFirea; > had been announced and delivered 3-4 years ago, but beinge= > marginally faster than the E10K for example is hardly goingm; > to ensure that WildFire is competitive once SuperDome andk9 > Suns UltraIII machines and the IBM S80 replacement ship % > and none of them are that far away.c > 7 > So more pertinent is what is Compaqs high end systemsu7 > strategy, as you can see it would be unwise to assume  > * > that WildFire is the answer. :-) :-) :-)  L My impression is that for quite a while now your criticisms have been mostlyL constructive (or at least reasonable) ones.  Hope this post doesn't indicate a return to slinging FUD.t   - bill   > 	 > Regardss > Andrew Harrisona > Enterprise IT Architect  >  >c >n   ------------------------------  # Date: Wed, 14 Jun 2000 17:20:10 GMTN9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)sD Subject: Re: Which cpu does Sun server use: I386, Mip, Ppc or Alpha?+ Message-ID: <11IjYRcKc7px@eisner.decus.org>   o In article <39477A39.7740A67E@uk.sun.com>, Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com> writes:s  I > Now Compaq have just released WildFire and in the rosy glow surroundingeO > its launch you seem to have missed a few really sorry aspects of the WildFirerL > design which tend to suggest that Compaq need to be developing WildFire II$ > for release as early as next year.  A Of course "as early as" did not happen with Wildfire, so it mightIA not with a successor.  Under US law, however, the Sparc-3 team isnA unable to bring suit against Compaq for copying their developmentP	 cycle :-)l  H > Well memory latency as a starter, the local on node access in WildFireF > is 330ns with remote access being 990ns, this would have been OK 3-4H > years ago (its similar to the Origin 2000) but does not look too greatD > now. The E10K for example has a flat latency from all CPU's to all@ > memory of ~450ns and to acheive this as an average on WildFireF > you need to be getting 80+% local accesses. This is going to be hardD > to do just relying on the OS and this may require people to change$ > the way they develop applications.  A Compaq also has control of the linker and the compilers.  Some of @ the compilers support a qualifier of /ARCHITECTURE=EV6, and theyA could presumably do something for Wildfire if that were required.g  B There was a notice, by the way, that the VMS SDK will be available? for about $39 US, so you might want to pick up a copy Andrew tot& learn the latest straight from Compaq.  A > Just for your information the Origin 2000 a 3-4 year old system"; > had at lanuch 310ns local and 950ns remote as illustrated 8 > in the white paper SGI published for the origin intro.  . Yes, John Mcalpin mentioned that in comp.arch., How does Origin 2000 compare to your UE10K ?; If Wildfire does better, I suppose it must be the operatingB= system then.  Can we get you to commit to that on the record,: Andrew ?  A > Secondly the bisectional bandwidth of the WildFire is about thea< > same as the E10K, yes Compaq have published higher STREAMS> > numbers but these included local on-node memory access which> > assumes that a process never goes off node. This is unlikely> > except for a very small subset of applications that WildFire > will support.c  @ Kerry has been using VMS so long that he tends to think in terms> of applications being well-written.  That may be unfamiliar to? those from a Unix background (and seems to be, based on some ofa' the code that gets imported to VMS :-)._   ------------------------------  # Date: Wed, 14 Jun 2000 17:56:22 GMT 9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) D Subject: Re: Which cpu does Sun server use: I386, Mip, Ppc or Alpha?+ Message-ID: <OOfcpe4nfRb+@eisner.decus.org>t  R In article <8i868i$7lf$1@pyrite.mv.net>, "Bill Todd" <billtodd@foo.mv.com> writes: > G > Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com> wrote int. > message news:39477A39.7740A67E@uk.sun.com...  I >> Well memory latency as a starter, the local on node access in WildFiresG >> is 330ns with remote access being 990ns, this would have been OK 3-4sI >> years ago (its similar to the Origin 2000) but does not look too greatuE >> now. The E10K for example has a flat latency from all CPU's to all A >> memory of ~450ns and to acheive this as an average on WildFirenG >> you need to be getting 80+% local accesses. This is going to be hardoE >> to do just relying on the OS and this may require people to changev% >> the way they develop applications.g > N > Or not.  VMS already supports locality ('RAD' may be the Wildfire term - I'dD > have to check) to help keep accesses local to a single board in an! > application-transparent manner.K  C My understanding is that the "RAD" term was introduces so as not toxC link the software concept to the hardware implementation.  Wildfire E is known to have a granularity at the "QBB" level, but if some future>E machine were designed to support a different granularity of CPUs thats7 share "local" memory, the "RAD" term would still apply.v  B Perhaps "Alpha - the Cryptographer's Edition" available in varying? RAD granularities of 1, 3, 5, 7, 11, 13, 17, 19, 23, 29,... :-)    ------------------------------  % Date: Wed, 14 Jun 2000 17:28:19 +0100 B From: Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com>D Subject: Re: Which cpu does Sun server use: I386, Mip, Ppc or Alpha?* Message-ID: <3947B2A3.E7D4D027@uk.sun.com>   Bill Todd wrote:  G > Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com> wrote inr. > message news:39477A39.7740A67E@uk.sun.com... >o > ...  >iN > Or not.  VMS already supports locality ('RAD' may be the Wildfire term - I'dD > have to check) to help keep accesses local to a single board in anE > application-transparent manner.  Since a single board can support 4nI > processors (each of which will enjoy the process and generational speed-F > improvements of the Alpha road map over time, including extension toK > simultaneous multi-threading) and up to 32 GB of on-board memory, this iseL > not as significant a restriction as you suggest:  in fact, only very *few*F > monolithic applications can really benefit much from more than 4-wayN > multi-processing (especially if each processor supports multiple interleavedC > threads at the hardware level) and more than 32 GB of memory, andiL > non-monolithic applications are relatively easy to segregate onto multiple< > boards that need to interact only relatively infrequently. >   8 Firstly, other NUMA OS's like Dynix for example have the> same sort of code in the OS to try and encourage locality, theE Sequents also have local hardware assist to cache heavily used pages.I? But even with this support many people using these systems havet? found that it is necessary to modify their applications to helps> the OS. In Sequents case running Oracle Parallel Server in the7 box, partitioning the DBMS across nodes within the samen OS instance has been tried.l  = Even with this level of tuning and with the support of the OSc 80% locality is high.o  9 Secondly large DBMS applications will easily consume moren8 than 4 CPU's worth of resource making it necessary to go8 of node with things like the SGA being a chunk of memory1 which all nodes may need to have equal access to.u  5 Thirdly the latencies quoted by Compaq in their paperl4 for the WildFire are "idle" latencies, I assume that3 these are for an unloaded system and if that is theM6 case then the latency of a node under load or off node7 access under load is likely to be worse than the valuesd that have been quoted.     >eM > So on-board latency will indeed likely dominate overall average applicationsI > latency (80+% may well be a very conservative estimate for the on-board"L > fraction), and IIRC there's a copy of VMS on each board to avoid excessiveG > system off-board code (and presumably pure data) access.  Given thesejN > considerations, plus Alpha's respectable caching facilities, I'd suggest youK > not get too complacent about Sun's next-generation competitiveness - evenON > ignoring the fact that creation of the Wildfire follow-on is also well under > way. >   I I can't really comment on Sun's next generation servers but consider thist? example. The HP N4000 which occupies the 1-8 CPU space which isdD also occupied by the GS80 has an Idle latency of 130ns, true it does> not have the same throughput as WildFire but it does have muchE lower latency. There is very little actual technical detail availablei> on HP's SuperDome but I would suggest that they will be aimingC for sub 200ns local idle access because that is what they currentlyr# are able to support with the N4000.P  H This is the sort of memory latency you should be looking at for WildFireE otherwise you will have huge issues delivering the levels of ILP thate? the CPU's are capable and moving forward Alpha will support SMT / which will only increase the demand for memory.h  C > > Just for your information the Origin 2000 a 3-4 year old systemi= > > had at launch 310ns local and 950ns remote as illustratede: > > in the white paper SGI published for the origin intro. > >o- > > http://www.sgi.com/origin/images/isca.pdft > > C > > Secondly the bisectional bandwidth of the WildFire is about thep> > > same as the E10K, yes Compaq have published higher STREAMS@ > > numbers but these included local on-node memory access which@ > > assumes that a process never goes off node. This is unlikely@ > > except for a very small subset of applications that WildFire > > will support.e >yL > As pointed out above, this is in fact likely for the overwhelming majorityI > of applications Wildfire will support, and the majority of the rest aretN > easily partitionable to meet this criterion.  Leaving aside the fact for forK > *really* monstrous applications, neither E10K nor Wildfire is sufficient:'K > you've got to expand into clustering, and Sun really doesn't want to take " > this competitive argument there. >   < Sure you can partition applictions and people using OPS in a: single Sequent box are examples of this. Would you do this< however if the alternative was a similarly sized system with8 a more uniform memory architecture which did not require; you to make these changes. I think on the basis of Sequentsm/ experience in this are the answer has to be no.    regardst Andrew Harrison  Enterprise IT Architect0   ------------------------------  % Date: Wed, 14 Jun 2000 17:49:17 +0100uB From: Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com>D Subject: Re: Which cpu does Sun server use: I386, Mip, Ppc or Alpha?* Message-ID: <3947B78D.ADE2F037@uk.sun.com>   Larry Kilgallen wrote:  q > In article <39477A39.7740A67E@uk.sun.com>, Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com> writes:n >rJ > > Well memory latency as a starter, the local on node access in WildFireH > > is 330ns with remote access being 990ns, this would have been OK 3-4J > > years ago (its similar to the Origin 2000) but does not look too greatF > > now. The E10K for example has a flat latency from all CPU's to allB > > memory of ~450ns and to acheive this as an average on WildFireH > > you need to be getting 80+% local accesses. This is going to be hardF > > to do just relying on the OS and this may require people to change& > > the way they develop applications. >sC > Compaq also has control of the linker and the compilers.  Some of B > the compilers support a qualifier of /ARCHITECTURE=EV6, and theyC > could presumably do something for Wildfire if that were required.  >l  A Well that may help but you are more likely to have to profile the : application to get much help because the behaviour you are= looking at improving will be determined by a lot of variables>7 that you are unlikely to be able model at compile time.   0 > Yes, John Mcalpin mentioned that in comp.arch.. > How does Origin 2000 compare to your UE10K ?  5 The Origin 2000 was introduced at about the same timey= as the E10K but the both have very different characteristics.t  8 The Origin like the Wildfire is a NUMA system with nodes5 with local memory connected together via a high speedg
 interconnect.n   The E10K is a UMA system.i  8 Both the Origin and the WildFire have lower local memory7 latencys than the E10Ks uniform access. 310 or 330ns vso5 440ns. But 440ns is for any CPU on the E10K accessinga9 any block of memory while both the Origin ad the WildFirer& have a remote memory access of 900+ns.  = > If Wildfire does better, I suppose it must be the operating ? > system then.  Can we get you to commit to that on the record, 
 > Andrew ?  D The point is that WildFire has to do better and a lot better. Compaq? have marketed it against machines like the E10K but the realitysE is that not long into the WildFires product life it will be competinglA not against Sun's E10K but the follow on, not agains the HP V2500 ( but against the follow on ditto for IBM.  > At the moment Compaq have published a set of benchmark results@ that do not establish that WildFire has a performance leadership= over any of its major competitors and so thats why I ask whend  the WildFire replacement is due.     >eC > > Secondly the bisectional bandwidth of the WildFire is about thea> > > same as the E10K, yes Compaq have published higher STREAMS@ > > numbers but these included local on-node memory access which@ > > assumes that a process never goes off node. This is unlikely@ > > except for a very small subset of applications that WildFire > > will support.n >aB > Kerry has been using VMS so long that he tends to think in terms@ > of applications being well-written.  That may be unfamiliar toA > those from a Unix background (and seems to be, based on some ofm) > the code that gets imported to VMS :-).   A I doubt that any of the applications currently running on OpenVMSn8 have been designed to cope with a system that has a NUMA architecture. :-)m     Regardsy Andrew Harrisong Enterprise IT Architecte   ------------------------------  % Date: Wed, 14 Jun 2000 13:17:15 -0400 ' From: "Bill Todd" <billtodd@foo.mv.com>cD Subject: Re: Which cpu does Sun server use: I386, Mip, Ppc or Alpha?( Message-ID: <8i8eg5$gej$1@pyrite.mv.net>  E Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com> wrote ing, message news:3947B78D.ADE2F037@uk.sun.com...   ...e  C > I doubt that any of the applications currently running on OpenVMSo: > have been designed to cope with a system that has a NUMA > architecture. :-)e  J I suggest that you check deja for my answer to this argument (if it didn'tG make it to your server):  VMS has that well under control, and very fewe6 applications will have any reason to be aware of NUMA.   - bill   >W >f	 > Regardso > Andrew Harrisonr > Enterprise IT Architect  >d >e >i   ------------------------------  % Date: Wed, 14 Jun 2000 08:02:08 -0700E7 From: David D Miller <ddmiller@notes.west.raytheon.com>u& Subject: ZLXp-E1/E2/E3 Switch SettingsF Message-ID: <OFBCAAA74F.F218A025-ON072568FE.0050DF4F@rsc.raytheon.com>   Folksn  C I'm reposting this message because my mailer scrambled the table sou badly.  E From the manual (EK-T2323-OG), this is how the switch settings on theu& ZLXp-E1/E2/E3 video board are defined:    . SW1 SW2 SW3 SW4   Pixel    Resolution  Refresh/                  Freq(MHz)             Rate(Hz)s, D    D   D   D     130     1280x1024      72, U    D   D   D     119     1280x1024      66, D    U   D   D     108     1280x1024      60, U    U   D   D     104     1152x900       72, D    D   U   D      93     1152x900       66, U    D   U   D      75     1024x768       70, D    U   U   D      74     1024x768       72, U    U   U   D      69     1024x864       60, D    D   D   U      65     1024x768       60, U    D   D   U      50      800x600       72, D    U   D   U      40      800x600       60, R    U   D   U      32      642x480       72, D    D   U   U      25      640x480       60, U    D   U   U     135     1280x1024      75, D    U   U   U     110     1280x1024      60$ U    U   U   U              Reserved  2 1. Up and Down as viewed from the back of the card$    with the stereo jack on the left.5 2. In this configuration, SW1 is the leftmost switch,e    nearest the stereo jack.h   I hope this is useful.   dave.n   ------------------------------   End of INFO-VAX 2000.331 ************************    