1 INFO-VAX	Thu, 22 Apr 2004	Volume 2004 : Issue 223       Contents:2 Re: =?ISO-8859-1?Q?T=E9tris_assembleur_VAX_+_SMG?= Re: Advanced Server Roadmap  Connecting Ethernet Printers  Re: Connecting Ethernet Printers  Re: Connecting Ethernet Printers  Re: Connecting Ethernet Printers  Re: Connecting Ethernet Printers  Re: Connecting Ethernet Printers  Re: Connecting Ethernet Printers  Re: Connecting Ethernet Printers  Re: Connecting Ethernet Printers1 gcc/g++ compiler (for coverage analysis) for vms? > Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> RE: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...- Re: ISM Version 6.4-F14 03 and Open VMS 7.3-2 - Re: ISM Version 6.4-F14 03 and Open VMS 7.3-2 - Re: ISM Version 6.4-F14 03 and Open VMS 7.3-2 , Java & OpenVMS : Problem with Thread.sleep()0 Re: Java & OpenVMS : Problem with Thread.sleep()4 Re: Linux on its way out - unless you are a geek ...4 Re: Linux on its way out - unless you are a geek ...4 Re: Linux on its way out - unless you are a geek ...! Re: LYNX, character sets, DECterm ! Re: LYNX, character sets, DECterm ! Re: LYNX, character sets, DECterm ! Re: LYNX, character sets, DECterm P OpenVMS PearlHewlett-Packard and the Politecnico di Torino - The OpenVMS I-64 Op' PCSI installation on a application disk # Ping: Alan Fay - Netbackup Question 	 Re: RWINS  RE: Support Dates for VMS  Re: Support Dates for VMS  Re: Support Dates for VMS  Re: Trying to run an EXE/ Re: Using Windows XP hyperterminal with OpenVMS " Re: VAX Macro language and file..., Re: [OT] An interesting view of Linux's cost  F ----------------------------------------------------------------------    Date: 22 Apr 2004 07:36:40 -0700 From: ohm62@hotmail.com (OHM) ; Subject: Re: =?ISO-8859-1?Q?T=E9tris_assembleur_VAX_+_SMG?= < Message-ID: <9d337b47.0404220636.5f6f467@posting.google.com>   >    Babelfish came up with: >  >       "Hello, A >        Continuation a council d'une nobody, I m'inscris on this C >        newsgroup to ask you l'aide. I am with the research of the @ >        codes d'un titris out of assembler VAX with SMG (Screen@ >        Management) in graphic interface. J'ai found on the Net= >        (http://decwarch.free.fr/index.html) a titris out of > >        assembler VAX but utlisant like graphic interface X11B >        called also DECWindows or DECWindows/Motif. It m'est also@ >        impossible of l'exicuter (via a waiter VAX "polarhome")< >        because I cannot read the codes. By thanking you by: >        advance, I request d'accepter, l'expression of my' >        respectful feelings from you."    LOL!!!  3 This thing is useless, but maybe for good laughs...   2 Here's a somewhat better inter-linear translation:  H ------------------------------------------------------------------------ Bonjour, Hi,   > Suite un conseil d'une personne, je m'inscris sur ce groupe de
 discussion4 Upon someone's advice, I subscribe to this newsgroup   pour vous demander de l'aide.  to ask for help.  C Je suis  la recherche des codes d'un ttris en assembleur VAX avec  SMG @ I am looking for VAX Assembly language sources for a Tetris game
 utilizing SMG   + (Screen Management) en interface graphique.  for a graphical interface.  E J'ai trouv sur le net (http://decwarch.free.fr/index.html) un ttris  en; I already found on the Internet (...) a Tetris game written   : assembleur VAX mais utlisant comme interface graphique X11< in VAX Assembly language, but featuring a X11 graphical user
 interface,  F appel aussi DECWindows ou DECWindows/Motif. Il m'est aussi impossible de= also known as DECwindows or DECwindows/MOTIF.  I can't even     C l'excuter (via un serveur VAX "polarhome") car je ne peux lire les  codes.E build and run it (through a VAX server "polarhome") since I am unable  to understand the sources.  D En vous remerciant par avance, je vous prie d'accepter, l'expression de mes sentiments respectueux. $ Thank you in advance.  Best Regards.   ------------------------------  % Date: Thu, 22 Apr 2004 07:54:47 -0400 $ From: "PEN" <paul.nuneznosp@mhp.com>$ Subject: Re: Advanced Server Roadmap, Message-ID: <c68bq9$e1t$1@hplms2.hpl.hp.com>   Hi,   F Yes, current plans are for Advanced Server for OpenVMS to be ported to OpenVMS Itanium.  = The OpenVMS roadmap was recently updated and is available at:   > http://h71000.www7.hp.com/openvms/roadmap/openvms_roadmaps.htm  G Slide 36 covers Advanced Server for OpenVMS, PATHWORKS for OpenVMS, and  PATHWORKS32.  - More info on OpenVMS Itanium can be found at:   J http://www.hp.com/products1/evolution/alpha_retaintrust/openvms/index.html   HTH,   Paul  0 "John" <hayesRMV1966@yahoo.com> wrote in message% news:8kyhc.3895$pg.1163@okepread05... H > I know that HP has ported OpenVMS to the new Intel chips, what are the plans L > for the layered products? Currently Advanced Server 7.3 emulates an NT 4.0$ > server, any plans to upgrade this? >  > Thanks >  >    ------------------------------  % Date: Thu, 22 Apr 2004 12:13:07 +0100  From: "Graham Forward" <a@b.c>% Subject: Connecting Ethernet Printers + Message-ID: <cNNhc.1$pm2.6594@news.ecrc.de>   I Up until now I have run the printers of my DS10 by connecting through the J printers' parallel ports to an Emulex print server (P3000/P4000)  and thenI creating a LAT port.  I would drive the printers either by addressing the J port directly, or by using a queue to address the port using /PROC=LATSYM.  J I now have a printer with an Ethernet port of its own.  How can I tell the system to use that instead?    TIA    ------------------------------    Date: 22 Apr 2004 13:41:25 +0200' From: huber@mppmu.mpg.de (Joseph Huber) ) Subject: Re: Connecting Ethernet Printers + Message-ID: <WNDPZ+SIWukz@vms.mppmu.mpg.de>   L In article <cNNhc.1$pm2.6594@news.ecrc.de>, "Graham Forward" <a@b.c> writes:K > Up until now I have run the printers of my DS10 by connecting through the L > printers' parallel ports to an Emulex print server (P3000/P4000)  and thenK > creating a LAT port.  I would drive the printers either by addressing the L > port directly, or by using a queue to address the port using /PROC=LATSYM. > L > I now have a printer with an Ethernet port of its own.  How can I tell the > system to use that instead?  >   I It depends of the printers network I/O capabilities. Most running LPD and I Telnet protocols. In this case You need to setup either the LPD client in G Your TCPIP package. For telnet protocols there is a telnet symbiont in  ; VMS TCPIP services (in Multinet and TCPware also I assume).   D Times some HP printer-network cards had LAT capabilities, but that's probably gone since long.     --  >    Joseph "Sepp" Huber, Muenchen   http://www.huber-joseph.de/   ------------------------------  % Date: Thu, 22 Apr 2004 13:42:40 +0100  From: "Graham Forward" <a@b.c>) Subject: Re: Connecting Ethernet Printers + Message-ID: <f5Phc.2$pm2.7239@news.ecrc.de>   4 "Joseph Huber" <huber@mppmu.mpg.de> wrote in message% news:WNDPZ+SIWukz@vms.mppmu.mpg.de... F > In article <cNNhc.1$pm2.6594@news.ecrc.de>, "Graham Forward" <a@b.c> writes: I > > Up until now I have run the printers of my DS10 by connecting through  the I > > printers' parallel ports to an Emulex print server (P3000/P4000)  and  thenI > > creating a LAT port.  I would drive the printers either by addressing  the @ > > port directly, or by using a queue to address the port using
 /PROC=LATSYM.  > > J > > I now have a printer with an Ethernet port of its own.  How can I tell the  > > system to use that instead?  > >  > K > It depends of the printers network I/O capabilities. Most running LPD and K > Telnet protocols. In this case You need to setup either the LPD client in H > Your TCPIP package. For telnet protocols there is a telnet symbiont in= > VMS TCPIP services (in Multinet and TCPware also I assume).  > F > Times some HP printer-network cards had LAT capabilities, but that's > probably gone since long.  >  > --  @ >    Joseph "Sepp" Huber, Muenchen   http://www.huber-joseph.de/  I The printer is a Xerox N40.  It can support a wide range of protocols - I K can print to it over the network from a PC.  My question really is how do I J set things up on the VMS system to be able to access the printer directly,H rather than through an ancient LAT print server?  I'm afraid I need more% detail on HOW to set up the protocol.    Thanks,    Graham   ------------------------------    Date: 22 Apr 2004 15:37:31 +0200' From: huber@mppmu.mpg.de (Joseph Huber) ) Subject: Re: Connecting Ethernet Printers + Message-ID: <n2qC235TDRUy@vms.mppmu.mpg.de>   L In article <f5Phc.2$pm2.7239@news.ecrc.de>, "Graham Forward" <a@b.c> writes: >  > K > The printer is a Xerox N40.  It can support a wide range of protocols - I M > can print to it over the network from a PC.  My question really is how do I L > set things up on the VMS system to be able to access the printer directly,J > rather than through an ancient LAT print server?  I'm afraid I need more' > detail on HOW to set up the protocol.   9 For the LPD setup, which TCPIP stack do You have on VMS ? ; For DEC/HP TCPIP services , use (under the system account)  
 @tcpip$config 8 In the menu , use both server and client submenu for LPD7 (if I remember well, setup the LPD client also reuires  ; LPD server, although in Your case only LPD client is used.) 3 Then afterwards the LPD "printcap" has to be setup. @ If Yours is TCPIP services, I will give You examples of printcap and the queu definition...      --  >    Joseph "Sepp" Huber, Muenchen   http://www.huber-joseph.de/   ------------------------------    Date: 22 Apr 2004 16:08:22 +0200' From: huber@mppmu.mpg.de (Joseph Huber) ) Subject: Re: Connecting Ethernet Printers + Message-ID: <Tl71RL$LOE$t@vms.mppmu.mpg.de>   U In article <1qc3W$PPLV3B@vms.mppmu.mpg.de>, huber@mppmu.mpg.de (Joseph Huber) writes:  > ; > After tcpip$config for LPD, the TCPIP$LPD_PRINTCAP.DAT in + > sys$specific:[tcpip$lpd] or sys$manager:     Correction: $  the file name is TCPIP$PRINTCAP.DAT    --  >    Joseph "Sepp" Huber, Muenchen   http://www.huber-joseph.de/   ------------------------------    Date: 22 Apr 2004 16:05:46 +0200' From: huber@mppmu.mpg.de (Joseph Huber) ) Subject: Re: Connecting Ethernet Printers + Message-ID: <1qc3W$PPLV3B@vms.mppmu.mpg.de>   L In article <f5Phc.2$pm2.7239@news.ecrc.de>, "Graham Forward" <a@b.c> writes: >  > K > The printer is a Xerox N40.  It can support a wide range of protocols - I M > can print to it over the network from a PC.  My question really is how do I L > set things up on the VMS system to be able to access the printer directly,J > rather than through an ancient LAT print server?  I'm afraid I need more' > detail on HOW to set up the protocol.  >   = For tcpip services, one of our printer setup looks like this:   9 After tcpip$config for LPD, the TCPIP$LPD_PRINTCAP.DAT in ) sys$specific:[tcpip$lpd] or sys$manager:    ; -----------------------------------------------------------  TCPIP$LPD_QUEUE:\          :lp=TCPIP$LPD_QUEUE:\          :sd=TCPIP$LPD_SPOOL: # ( # postscript, LJ6 in room 209 (Oberlack) PRPS106|prps106|pr106:\ 1         :lf=/SYS$SPECIFIC/TCPIP$LPD/PRPS106.LOG:\          :lp=PRPS106:\          :rm=iws132a:\          :rp=prps106:\ ,         :sd=/SYS$SPECIFIC/TCPIP$LPD/PRPS106:; -----------------------------------------------------------     $ Where "prps106" is the printer name,@ "iws132a" is the print-server IP node (our central printserver),? maybe the same as the printers IP nodename, if You go directly   to the printer. 9 Create a subdirectory as specified in the :sd= statement.   @ There is a utility to produce printcap interactively, just don't; remeber its name, I always edit the printcap file directly.   ; ----------------------------------------------------------- 3 $! define the VMS queue to point to the LPD server:  $ init/queue/start- )   /descrip="HP LJ6 in room 209 Oberlack"-                  /owner=[1,4]-                  /retain=error--                 /DEFAULT=(FEED,FORM=DEFAULT)-                  /device=server- '                 /process=tcpip$lpd_smb- "                 /LIBRARY=PSDEVCTL-5                 /protection=(s:e,g:r,w:w,o:d) PRPS106 ; -----------------------------------------------------------   = Library DEVCTL is a local library of postscript setup modules = for postscript printers (to select single/double sided, paper ? source etc.). Leave off the /LIBRARY line if You don't have one / for the Xerox (only default setup in use then).     --  >    Joseph "Sepp" Huber, Muenchen   http://www.huber-joseph.de/   ------------------------------  # Date: Thu, 22 Apr 2004 14:57:15 GMT * From: Paul Anderson <paul.anderson@hp.com>) Subject: Re: Connecting Ethernet Printers 5 Message-ID: <220420041038227840%paul.anderson@hp.com>   B In article <f5Phc.2$pm2.7239@news.ecrc.de>, Graham Forward <a@b.c> wrote:  G > The printer is a Xerox N40.  It can support a wide range of protocols E > - I can print to it over the network from a PC.  My question really F > is how do I set things up on the VMS system to be able to access theD > printer directly, rather than through an ancient LAT print server?  D You certainly can use a generic LPD or Telnet queue, but you can getG more features by using DCPS.  If you are not familiar with DCPS, it's a C layered product that requires no license that provides full support E over IP to your Xerox N40 printer.  The current version is DCPS V2.3.   D After installing DCPS, you would change the DCPS$STARTUP.COM file toF create a queue for the printer.  You would use your desired queue nameF in the P1 parameter and the IP_RawTCP protocol and the name or address$ of your printer in the P2 parameter.   Paul   --    Paul Anderson   OpenVMS Engineering    Hewlett-Packard Company    ------------------------------  % Date: Thu, 22 Apr 2004 15:42:17 +0100  From: "Graham Forward" <a@b.c>) Subject: Re: Connecting Ethernet Printers + Message-ID: <lRQhc.3$pm2.7819@news.ecrc.de>   4 "Joseph Huber" <huber@mppmu.mpg.de> wrote in message% news:1qc3W$PPLV3B@vms.mppmu.mpg.de... F > In article <f5Phc.2$pm2.7239@news.ecrc.de>, "Graham Forward" <a@b.c> writes:  > >  > > K > > The printer is a Xerox N40.  It can support a wide range of protocols -  I J > > can print to it over the network from a PC.  My question really is how do ID > > set things up on the VMS system to be able to access the printer	 directly, L > > rather than through an ancient LAT print server?  I'm afraid I need more) > > detail on HOW to set up the protocol.  > >  > ? > For tcpip services, one of our printer setup looks like this:  > ; > After tcpip$config for LPD, the TCPIP$LPD_PRINTCAP.DAT in * > sys$specific:[tcpip$lpd] or sys$manager: > = > -----------------------------------------------------------  > TCPIP$LPD_QUEUE:\  >         :lp=TCPIP$LPD_QUEUE:\  >         :sd=TCPIP$LPD_SPOOL: > # * > # postscript, LJ6 in room 209 (Oberlack) > PRPS106|prps106|pr106:\ 3 >         :lf=/SYS$SPECIFIC/TCPIP$LPD/PRPS106.LOG:\  >         :lp=PRPS106:\  >         :rm=iws132a:\  >         :rp=prps106:\ . >         :sd=/SYS$SPECIFIC/TCPIP$LPD/PRPS106:= > -----------------------------------------------------------  >  > & > Where "prps106" is the printer name,B > "iws132a" is the print-server IP node (our central printserver),@ > maybe the same as the printers IP nodename, if You go directly > to the printer. ; > Create a subdirectory as specified in the :sd= statement.  > B > There is a utility to produce printcap interactively, just don't= > remeber its name, I always edit the printcap file directly.  > = > ----------------------------------------------------------- 5 > $! define the VMS queue to point to the LPD server:  > $ init/queue/start- + >   /descrip="HP LJ6 in room 209 Oberlack"-  >                 /owner=[1,4]-   >                 /retain=error-/ >                 /DEFAULT=(FEED,FORM=DEFAULT)- ! >                 /device=server- ) >                 /process=tcpip$lpd_smb- $ >                 /LIBRARY=PSDEVCTL-7 >                 /protection=(s:e,g:r,w:w,o:d) PRPS106 = > -----------------------------------------------------------  > ? > Library DEVCTL is a local library of postscript setup modules ? > for postscript printers (to select single/double sided, paper A > source etc.). Leave off the /LIBRARY line if You don't have one 1 > for the Xerox (only default setup in use then).  >  > --  @ >    Joseph "Sepp" Huber, Muenchen   http://www.huber-joseph.de/  I The example you have given of the queue initialisation looks exactly like J the sort of thing I'm after.  I am using TCPware rather than the Compaq/HPF stack.  However I have just read the relevant (LPD) bit of the TCPwareJ manual and it seems to me that this relates to allowing access to users onL remote hosts.   The manuals are pretty poor - the indexing particularly, butF I just want to print to a local printer.  Is LPD the right thing to be* looking at?  I am extremely ignorant here!   ------------------------------    Date: 22 Apr 2004 17:21:22 +0200' From: huber@mppmu.mpg.de (Joseph Huber) ) Subject: Re: Connecting Ethernet Printers + Message-ID: <HQOahRqq428X@vms.mppmu.mpg.de>   L In article <lRQhc.3$pm2.7819@news.ecrc.de>, "Graham Forward" <a@b.c> writes: > K > The example you have given of the queue initialisation looks exactly like L > the sort of thing I'm after.  I am using TCPware rather than the Compaq/HPH > stack.  However I have just read the relevant (LPD) bit of the TCPwareL > manual and it seems to me that this relates to allowing access to users onN > remote hosts.   The manuals are pretty poor - the indexing particularly, butH > I just want to print to a local printer.  Is LPD the right thing to be, > looking at?  I am extremely ignorant here!  H DCPS (see P.Andersons response) is independent of the underlaying TCPIP  stack.E If DCPS is not option an for You, then yes, LPD is the right thing.   @ I just can't help You with the TCPware setup, somebody else may.   --  >    Joseph "Sepp" Huber, Muenchen   http://www.huber-joseph.de/   ------------------------------    Date: 22 Apr 2004 02:47:00 -07002 From: newgroup@jcaps.com (Morten Jagd Christensen): Subject: gcc/g++ compiler (for coverage analysis) for vms?= Message-ID: <6728d13d.0404220147.6f40f497@posting.google.com>    Hi all  6 I am developing an automated testframework for module 7 testing on a vms based platform. As part of this effort   I want to collect coverage data.  A I have tried the PCA profiler but it has a few shortcomings. For  3 example I want a report per source file where I can > see the lines covered and the lines not covered, but PCA gives& this kind of report per function only!  7 Using the -ftest-coverage and -fprofile-arcs options to 5 gcc I can get this information per file which is more  useful for my purpose.  > So I am looking for a port of gcc (and gcov) so I can use gcov instead.  E Any hints (links, emails etc.) on how to do this will be appreciated.    Cheers   /Morten    ------------------------------  % Date: Thu, 22 Apr 2004 07:50:27 +0200 * From: "Karsten Nyblad" <nospam@nospam.com>G Subject: Re: How to turn linux into VMS - memory refresher for Dave ... + Message-ID: <c67mf7$l0$1@news.cybercity.dk>   5 "Bob Ceculski" <bob@instantwhip.com> wrote in message 7 news:d7791aa1.0404211441.14cd912e@posting.google.com... F > Andrew Harrison <andrew_._remove_harrison@su_n.com> wrote in message, news:<c66bdh$rua$1@new-usenet.uk.sun.com>...: > > Well he may not but Solaris uses PAM and it would also# > > appear that HP-UX does as well.  > >  > > O http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?objectID=tis07674  > > > > > I cannot be bothered to look and see if AIX does but sinceA > > Solaris and HP-UX are the 1st and 2and largest UNIX platforms A > > its rather safe to say that the majority of UNIX platforms do 8 > > not fit the description you applied to them earlier. > @ > one description they fit is they are both convuluted, unsecure > unreliable garbage ...  H In <c5qav1$c7d$1@news.cybercity.dk> I asked you which CERT reports couldL have been avoided if Unix did not execute user programs and command languageH interpreter in separate.  Why don't you answer that question in stead of! trolling?  Don't you have answer?    Karsten Nyblad ibpit1202 at sneakemail dot com    ------------------------------  % Date: Thu, 22 Apr 2004 08:27:50 +0200 * From: "Karsten Nyblad" <nospam@nospam.com>G Subject: Re: How to turn linux into VMS - memory refresher for Dave ... , Message-ID: <c67olc$3m1$1@news.cybercity.dk>  8 In <mOtQB4dyFEBG@eisner.encompasserve.org> "Bob Koehler", koehler@eisner.aspm.encompasserve.org wrote:; >Please remember that you don't speak for all UNIX vendors. G I don't speak for any of the Unix vendors.  I am no and I have not been L employed by any Unix vendor.  An I have been using VMS a lot more than Unix.! Are you running out of arguments?   E >This reminds me of a key difference between DECnet (on ANY platform) F >and IP (on ANY platform):  DECnet will NOT open a connection until itD >knows who you are and verifies that you have access to the machine.C >This requires one of: a) an explicit username/password (anyone can C >do that), b) a proxy (only a privileged user can do that), or c) a E >default account for a well known service (only a privileged user can F >do that).  Unprivileged users cannot add faulty network services that >skip authentication.   J What has DECnet to do with security on the Internet?  DECnet might be usedE on Intranets  (I doubt that many firms actually do it), but DECnet is L virtually non existing on today's Internet.  The important issue is securityI in TCP/IP and related protocols on the Internet.  That is what we all use K for connecting to others and that is where the bad guys are.  That was what < we were discussing before you tried to divert our attention.  J Besides, DECnet might be secure, but I also remember it as needing lots ofL resources, e.g., it was extremely slow to start a file transfer because someJ fool had got the idea that peoples login.com should be executed before theI actual file transfer started.  It would have been slow anyway.  We always J exited login.com at the start when it was executed from a network process, and it was still slow.   Karsten Nyblad ibpit1202 at sneakemail dot com    ------------------------------    Date: 22 Apr 2004 09:21:51 +0200' From: huber@mppmu.mpg.de (Joseph Huber) G Subject: Re: How to turn linux into VMS - memory refresher for Dave ... + Message-ID: <2p7rsSuF00QW@vms.mppmu.mpg.de>   , In article <c67olc$3m1$1@news.cybercity.dk>,-  "Karsten Nyblad" <nospam@nospam.com> writes:  > L > Besides, DECnet might be secure, but I also remember it as needing lots ofN > resources, e.g., it was extremely slow to start a file transfer because someL > fool had got the idea that peoples login.com should be executed before theK > actual file transfer started.  It would have been slow anyway.  We always L > exited login.com at the start when it was executed from a network process, > and it was still slow. >   1 I don't want to go into the Linux-VMS discussion, 9 just for DECnet file access , the equivalent in *x is NOT = NFS (this would be either access to cluster-mounted disks or  + DECnet DFS), but it compares to Unix "rcp": : There is no real difference, in both cases a network login= is executed on the remote system, and in both cases some kind . of login script/profile/*rc file is executed. 4 And it is the recommended and documented way to exit9 (sy)login.com early for network logins, the same is true  8 for e.g. .bashrc executed for rsh/rcp on a Linux system.  ; DECnet file-access was/is convenient because of zero admin  7 work, but no replacement for a distributed file-system.     --  >    Joseph "Sepp" Huber, Muenchen   http://www.huber-joseph.de/   ------------------------------  % Date: Thu, 22 Apr 2004 12:18:13 +0200 * From: "Karsten Nyblad" <nospam@nospam.com>G Subject: Re: How to turn linux into VMS - memory refresher for Dave ... , Message-ID: <c68659$n41$1@news.cybercity.dk>  4 "Joseph Huber" <huber@mppmu.mpg.de> wrote in message% news:2p7rsSuF00QW@vms.mppmu.mpg.de... . > In article <c67olc$3m1$1@news.cybercity.dk>,/ >  "Karsten Nyblad" <nospam@nospam.com> writes: 3 > I don't want to go into the Linux-VMS discussion,   G Point well taken.  I am simply trying to find out if security in VMS is F superior by design when compared to *x.  Until now what I have seen isK rather missy and VMS may not be capable of claiming more than been at least G as secure as modern Unixes.  Maybe that TCP/IP is implemented in a more ? secure way on VMS than on *x.  But TCP/IP is not a part of VMS.   ; > just for DECnet file access , the equivalent in *x is NOT > > NFS (this would be either access to cluster-mounted disks or- > DECnet DFS), but it compares to Unix "rcp": < > There is no real difference, in both cases a network login? > is executed on the remote system, and in both cases some kind / > of login script/profile/*rc file is executed. 6 > And it is the recommended and documented way to exit: > (sy)login.com early for network logins, the same is true: > for e.g. .bashrc executed for rsh/rcp on a Linux system.  C Yes, but I think not that many are using those old protocols on the I Internet.  Today people will require encryption on the Internet.  I would J like to keep the discussion to which operating system is best in a hostile environment.   Karsten Nyblad ibpit1202 at sneakemail dot com    ------------------------------  + Date: Thu, 22 Apr 2004 11:08:43 +0000 (UTC)  From: david20@alpha2.mdx.ac.ukG Subject: Re: How to turn linux into VMS - memory refresher for Dave ... ) Message-ID: <c6893r$3q9$1@news.mdx.ac.uk>   l In article <c66bdh$rua$1@new-usenet.uk.sun.com>, Andrew Harrison <andrew_._remove_harrison@su_n.com> writes: >Bob Koehler wrote:  > ] >> In article <c660mk$10cg$1@news.cybercity.dk>, "Karsten Nyblad" <nospam@nospam.com> writes:  >>  K >>>"Bob Koehler" <koehler@eisner.nospam.encompasserve.org> wrote in message 0 >>>news:YrHGDADLmuOz@eisner.encompasserve.org... >>> D >>>>In article <c65htd$jbd$1@new-usenet.uk.sun.com>, Andrew Harrison >>> . >>><andrew_._remove_harrison@su_n.com> writes: >>> ) >>>>> From the Solaris ftdp manual pages.  >>>>1 >>>>   Somehow I don't think Solaris is all UNIX.  >>>>F >>>>   _I_ (me, myself, first person), have fixed ftpd that were usingI >>>>   different code for authentication than login was.  Any single UNIX 0 >>>>   not doing so does not change the example. >>>> >>> K >>>Or that the Unix people have realized that they too needed to centralize O >>>authentication and authorization.  Please remember that many of the APIs for < >>>authentication and authorization in VMS are not that old. >>   >>  @ >>    Please remember that you don't speak for all UNIX vendors. >>   > 7 >Well he may not but Solaris uses PAM and it would also   >appear that HP-UX does as well. > P >http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?objectID=tis07674 > ; >I cannot be bothered to look and see if AIX does but since > >Solaris and HP-UX are the 1st and 2and largest UNIX platforms> >its rather safe to say that the majority of UNIX platforms do5 >not fit the description you applied to them earlier.  >   J PAM is relatively new. Hence even if all the major vendors latest versionsM support PAM that would still not invalidate the statement which was about the E users personal experience of Unix systems which undoubtedly covered a E historical period rather than being restricted to newer Unix systems.     
 David Webb VMs and Unix team leader CCSS Middlesex University >Regards >Andrew HarrisonL >>    This reminmds me of a key difference between DECnet (on ANY platform) K >>    and IP (on ANY platform):  DECnet will NOT open a connection until it I >>    knows who you are and verifies that you have access to the machine. H >>    This requires one of: a) an explicit username/password (anyone canH >>    do that), b) a proxy (only a privileged user can do that), or c) aJ >>    default account for a well known service (only a privileged user canL >>    do that).  Unprivileged users cannot add faulty network services that  >>    skip authentication. >>  > >>    IP will let anyone to connect to a socket.  If you writeK >>    authentication into your server, then you get what you write.  If you . >>    don't bother, then you get what you get. >>   >    ------------------------------    Date: 22 Apr 2004 05:34:09 -0700( From: bob@instantwhip.com (Bob Ceculski)G Subject: Re: How to turn linux into VMS - memory refresher for Dave ... = Message-ID: <d7791aa1.0404220434.12707ee2@posting.google.com>   ^ "Karsten Nyblad" <nospam@nospam.com> wrote in message news:<c67olc$3m1$1@news.cybercity.dk>...: > In <mOtQB4dyFEBG@eisner.encompasserve.org> "Bob Koehler". > koehler@eisner.aspm.encompasserve.org wrote:= > >Please remember that you don't speak for all UNIX vendors. I > I don't speak for any of the Unix vendors.  I am no and I have not been N > employed by any Unix vendor.  An I have been using VMS a lot more than Unix.# > Are you running out of arguments?  > G > >This reminds me of a key difference between DECnet (on ANY platform) H > >and IP (on ANY platform):  DECnet will NOT open a connection until itF > >knows who you are and verifies that you have access to the machine.E > >This requires one of: a) an explicit username/password (anyone can E > >do that), b) a proxy (only a privileged user can do that), or c) a G > >default account for a well known service (only a privileged user can H > >do that).  Unprivileged users cannot add faulty network services that > >skip authentication.  > L > What has DECnet to do with security on the Internet?  DECnet might be usedG > on Intranets  (I doubt that many firms actually do it), but DECnet is N > virtually non existing on today's Internet.  The important issue is securityK > in TCP/IP and related protocols on the Internet.  That is what we all use M > for connecting to others and that is where the bad guys are.  That was what > > we were discussing before you tried to divert our attention. > L > Besides, DECnet might be secure, but I also remember it as needing lots ofN > resources, e.g., it was extremely slow to start a file transfer because someL > fool had got the idea that peoples login.com should be executed before theK > actual file transfer started.  It would have been slow anyway.  We always L > exited login.com at the start when it was executed from a network process, > and it was still slow. >  > Karsten Nyblad! > ibpit1202 at sneakemail dot com   ? ever hear of Decnet over IP?  And I can assure you it is hardly * slow, and a lot more reliable than FTP ...   ------------------------------    Date: 22 Apr 2004 05:35:14 -0700( From: bob@instantwhip.com (Bob Ceculski)G Subject: Re: How to turn linux into VMS - memory refresher for Dave ... = Message-ID: <d7791aa1.0404220435.46ba4cc8@posting.google.com>   ] "Karsten Nyblad" <nospam@nospam.com> wrote in message news:<c67mf7$l0$1@news.cybercity.dk>... 7 > "Bob Ceculski" <bob@instantwhip.com> wrote in message29 > news:d7791aa1.0404211441.14cd912e@posting.google.com...iH > > Andrew Harrison <andrew_._remove_harrison@su_n.com> wrote in message/ >  news:<c66bdh$rua$1@new-usenet.uk.sun.com>... < > > > Well he may not but Solaris uses PAM and it would also% > > > appear that HP-UX does as well.  > > >i > > >rR >  http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?objectID=tis07674 > > >c@ > > > I cannot be bothered to look and see if AIX does but sinceC > > > Solaris and HP-UX are the 1st and 2and largest UNIX platforms.C > > > its rather safe to say that the majority of UNIX platforms doe: > > > not fit the description you applied to them earlier. > > B > > one description they fit is they are both convuluted, unsecure > > unreliable garbage ... > J > In <c5qav1$c7d$1@news.cybercity.dk> I asked you which CERT reports couldN > have been avoided if Unix did not execute user programs and command languageJ > interpreter in separate.  Why don't you answer that question in stead of# > trolling?  Don't you have answer?  >  > Karsten Nyblad! > ibpit1202 at sneakemail dot come  ; and while you are checking out the cert counts for the laste< ten years, check out this link below to TCPware, an IP stack8 for VMS ... notice the few certs listed had no or little= effect (access violation error) on VMS ... how did unix fare?   , http://www.process.com/techsupport/cert.html   ------------------------------    Date: 22 Apr 2004 07:22:10 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)VG Subject: Re: How to turn linux into VMS - memory refresher for Dave ...r3 Message-ID: <p4j6Mci3hkr5@eisner.encompasserve.org>V  l In article <c66ar5$rnt$1@new-usenet.uk.sun.com>, Andrew Harrison <andrew_._remove_harrison@su_n.com> writes: > 8 > What was it about my previous posts that you disagreed > with ?  I    You claim I was technically in error, but everything technical I wrotemH    was true.  You then invented a "new systems" aspect to the discussion    which I never claimed.    > The correction > The Sun development of PAM2 > or the Sun having the largest UNIX market share.  B    Your implied claim that since Solaris now uses PAM the multipleE    implementations of authentication routines do not exist in UNIX.  )I    Obviously you've only actually shown that they don't exist in current t    versions of Solaris..  G    Others have posted that there are UNIX other than Solaris also usingwF    PAM.  Good.  But it doesn't make all UNIX systems in use today freeF    of multiple authentication routines.  IT doesn't make UNIX that way:    by design.  It only makes some UNIX systems that way by    implementation.   ------------------------------    Date: 22 Apr 2004 07:23:25 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)EG Subject: Re: How to turn linux into VMS - memory refresher for Dave ...o3 Message-ID: <cWDgoJIMc9D3@eisner.encompasserve.org>P  n In article <c66bpm$9cb$1@charm.magnus.acs.ohio-state.edu>, JONESD@er6.eng.ohio-state.edu (David Jones) writes: > Q > You left out d) privileged process can declare itself a known object and accept2L > connections to that object.  In any event, I think you are overstating theO > case for DECnet security.  In TCP/IP, only privileged processes can listen oniK > port numbers below 1024, and well-known services use ports in that range.   G    What on earth makes you think that security holes require well knowno    services?   ------------------------------    Date: 22 Apr 2004 07:25:10 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)rG Subject: Re: How to turn linux into VMS - memory refresher for Dave .../3 Message-ID: <vGxE92MX+U4g@eisner.encompasserve.org>b  l In article <c66bdh$rua$1@new-usenet.uk.sun.com>, Andrew Harrison <andrew_._remove_harrison@su_n.com> writes: > < > I cannot be bothered to look and see if AIX does but since? > Solaris and HP-UX are the 1st and 2and largest UNIX platforms ? > its rather safe to say that the majority of UNIX platforms doe6 > not fit the description you applied to them earlier.  D    Just because they are the top 2 shipping today does not make them    the top 2 in use.  D    But I never claimed all UNIX systems were faulty.  I only claimedF    the solution was not designed-in, and that the fault exists on some    UNIX systems.   ------------------------------    Date: 22 Apr 2004 07:29:52 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)eG Subject: Re: How to turn linux into VMS - memory refresher for Dave ...r3 Message-ID: <+uz+Tp8XVD6$@eisner.encompasserve.org>l  X In article <c67mf7$l0$1@news.cybercity.dk>, "Karsten Nyblad" <nospam@nospam.com> writes: > 7 > "Bob Ceculski" <bob@instantwhip.com> wrote in messaget  J > In <c5qav1$c7d$1@news.cybercity.dk> I asked you which CERT reports couldN > have been avoided if Unix did not execute user programs and command languageJ > interpreter in separate.  Why don't you answer that question in stead of# > trolling?  Don't you have answer?   G    While I don't know of any specific CERT reports related to this, and F    I don't know if Bob Ceculksi does, I do know UNIX users who hacked 6    thier own system by using copies of shell programs.  @    Ever try to "find" a file who's name starts with a NULL byte?  F    You can copy DCL.EXE all you want, but you can't make the copy intoF    a CLI unless you've already got privilege, nor can you hide it from#    routine system management scans.p   ------------------------------  % Date: Thu, 22 Apr 2004 08:59:02 -0400A# From: "Dan Allen" <dallen@nist.gov>cG Subject: RE: How to turn linux into VMS - memory refresher for Dave ...n: Message-ID: <JFEPKAPBPMDFDBOIANGDIEJAFBAA.dallen@nist.gov>  O PAM is optional even in Solaris. Lots of older subsystems even on Solaris still7F use older interfaces directly into the password/shadow files includingF roll-your-own. Been there seen that more than once and quite recently.   DanO   > -----Original Message-----D > From: Bob Koehler [mailto:koehler@eisner.nospam.encompasserve.org]( > Sent: Thursday, April 22, 2004 9:22 AM > To: Info-VAX@Mvb.Saic.ComsI > Subject: Re: How to turn linux into VMS - memory refresher for Dave ...p >M >oB > In article <c66ar5$rnt$1@new-usenet.uk.sun.com>, Andrew Harrison- > <andrew_._remove_harrison@su_n.com> writes:T > > : > > What was it about my previous posts that you disagreed
 > > with ? >-K >    You claim I was technically in error, but everything technical I wrotewJ >    was true.  You then invented a "new systems" aspect to the discussion >    which I never claimed.u >l > > The correction > > The Sun development of PAM4 > > or the Sun having the largest UNIX market share. >nD >    Your implied claim that since Solaris now uses PAM the multipleE >    implementations of authentication routines do not exist in UNIX.eJ >    Obviously you've only actually shown that they don't exist in current >    versions of Solaris.o >aI >    Others have posted that there are UNIX other than Solaris also using-H >    PAM.  Good.  But it doesn't make all UNIX systems in use today freeH >    of multiple authentication routines.  IT doesn't make UNIX that way< >    by design.  It only makes some UNIX systems that way by >    implementation. >w >" >@ >>   ------------------------------    Date: 22 Apr 2004 07:34:51 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)nG Subject: Re: How to turn linux into VMS - memory refresher for Dave ...h3 Message-ID: <ibcP2tXkltQa@eisner.encompasserve.org>,  Y In article <c67olc$3m1$1@news.cybercity.dk>, "Karsten Nyblad" <nospam@nospam.com> writes:  > 6 > What has DECnet to do with security on the Internet?  5    It only serves to show that there are better ways.t  L > Besides, DECnet might be secure, but I also remember it as needing lots ofN > resources, e.g., it was extremely slow to start a file transfer because someL > fool had got the idea that peoples login.com should be executed before the > actual file transfer started.H  ?    The same fool which made it more secure, maybe?  The VAX/VMSgH    implementation of DECnet was known for process startup issues, but it@    is not generic to DECnet.  The TOPS-20, RSX, Ultirx, and AIX H    implementations of DECnet which I used all serve as counter-examples.B    I suspect many of the implmentations I didn't try also serve as    counter-examples.  C    I'll give up a little bit of performance for a heap of security.p2    I end up getting a lot more work done that way.   ------------------------------    Date: 22 Apr 2004 07:37:47 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)eG Subject: Re: How to turn linux into VMS - memory refresher for Dave ... 3 Message-ID: <TOPgew6jESju@eisner.encompasserve.org>t  U In article <2p7rsSuF00QW@vms.mppmu.mpg.de>, huber@mppmu.mpg.de (Joseph Huber) writes:g > = > DECnet file-access was/is convenient because of zero admin o9 > work, but no replacement for a distributed file-system.t  E    We didn't need a distributed file system when we used DECnet sincee@    programs could open, read, and write files over the network. F    Something which DECnet and ISO/OSI have in their file access levels@    but IP doesn't.  For IP networks you have to have a an entire:    network file system just to get that little capability.   ------------------------------    Date: 22 Apr 2004 08:14:12 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) G Subject: Re: How to turn linux into VMS - memory refresher for Dave ...r3 Message-ID: <KsPWa8jYe286@eisner.encompasserve.org>r  Y In article <c68659$n41$1@news.cybercity.dk>, "Karsten Nyblad" <nospam@nospam.com> writes:b > I > Point well taken.  I am simply trying to find out if security in VMS ist) > superior by design when compared to *x.   @    By design, yes.  By implentation, yes.  If you want technicalD    details then let's stick to them and stay away from flame bating.  H    There are lots of good references to support my claim.  If you simplyH    pull the technical data out of the "UNIX Hater's Handbook" and ignore0    the snide remarks you'll get about 80% of it.  E    Most of the difference is in the implementation, not the design.  lB    Most security problems are buffer-overflow based, and that's an@    implementation issue.  History shows lots of vendors have had?    implementation flaws opening buffer overflows.  Most of UNIXLE    was implemented im C, where buffer overflows are trivial to setup, C    though possible to avoid.  Most of VMS was not implemented in C, E    it was in languages where buffer overflows are difficult to setup.pB    Much of the latest work in VMS is also done in C but so far has    avoided buffer oeverflows.L  C    C enthusiats will tell you that the buffer overflow problem is aiC    compiler problem because the compilers simply use addresses for  C    pointeres, which C does not require.  The fact remains that the  C    compilers do it and C doesn't require any behaviour which is not     implementable that way.  D    An excellent discussion of the many problems with pointers can be@    found in Sun's original white paper on the invention of Java,C    where the authors discuss why they did not put pointers in Java.n  E    Many security problems can be traced to UNIX security design, suchrF    as having only one real privilege level (either you are root or youG    aren't), and accessing it via suid/sgid bits in the file protection.fA    History shows lots of vendors have tripped over this, shippinguC    suid root programs with flaws that let them do things they were -@    never intended to do.  It's not that the security model isn't>    workable, it's just that it's hard to get right.  Many UNIXB    vendors are moving away from this, creating role-based securityE    concepts not tied to the file systems, and therefor more like VMS.e  E    UNIX and VMS have ways to log activity and avoid attacks, but I'verE    not yet seen any UNIX which implements a security audit trail with H    all the features of the VMS security audit log, such as logging everyD    login failure or every file access failure, or the VMS feature ofG    logging every media mount (historically logged in a different log). cE    Nor have I seen VMS like rich set of password protection features.rE    I have never seen a UNIX which will evade password guessing to theaE    point of refusing a valid password, without creating an avenue for E    a denial of service attack.   Many UNIX still use minimum password$?    lifetimes, which I guess is to prevent users from imediately|?    restoring expired passwords, leaving open an opportunity for:@    vulnerability, instead of using password histories to control    password reuse.  B    So OK, UNIX is getting better, changing some designs and fixingF    implementation bugs.  Meanwhile I've had all this in VMS since 19845    even though UNIX had a 10 year starting advantage.e  G    And this is just a peek at the differences.  You should get onto theoC    VMS web page and study the VMS Guide to Security to see just howe"    rich its security features are.   ------------------------------  % Date: Thu, 22 Apr 2004 15:41:03 +0200r* From: "Karsten Nyblad" <nospam@nospam.com>G Subject: Re: How to turn linux into VMS - memory refresher for Dave ...i- Message-ID: <c68i1k$17r3$1@news.cybercity.dk>   H "Bob Koehler" <koehler@eisner.nospam.encompasserve.org> wrote in message- news:+uz+Tp8XVD6$@eisner.encompasserve.org...D$ >    I do know UNIX users who hacked8 >    thier own system by using copies of shell programs. >sB >    Ever try to "find" a file who's name starts with a NULL byte?  I VMS does not allow the users to change which CLI they are using, but Unix-H does.  (VMS users will have to ask their system manager for changing theG CLI.)  Thus such a bug is ulikely on VMS.  The bug you where describingrI seems to be due to pour implementation and Unix having a feature that VMS H does not have.  If this bug is related to pour design, then it has to beK running shells with privilliges.  That should not be necessary.  Yes, there J have been pourly designed Unixes, but hopefully they are on their way out.  H >    You can copy DCL.EXE all you want, but you can't make the copy intoH >    a CLI unless you've already got privilege, nor can you hide it from% >    routine system management scans.n > I It is difficult to write a new CLI for VMS without access to the sources. J If you copy from the sources of VMS, then you cannot distribute the sourceF code.  That makes life hard to hacker, but it is also security throughI obscurity.  Security experts usually frown on security through obscurity.    ------------------------------  % Date: Thu, 22 Apr 2004 07:07:58 -0700 + From: "Barry Treahy, Jr." <Treahy@MMaz.com>aG Subject: Re: How to turn linux into VMS - memory refresher for Dave ...I' Message-ID: <4087D1BE.7070905@MMaz.com>    Karsten Nyblad wrote:   I >"Bob Koehler" <koehler@eisner.nospam.encompasserve.org> wrote in messagea. >news:+uz+Tp8XVD6$@eisner.encompasserve.org... >  u > $ >>   I do know UNIX users who hacked8 >>   thier own system by using copies of shell programs. >>B >>   Ever try to "find" a file who's name starts with a NULL byte? >>     >> >aJ >VMS does not allow the users to change which CLI they are using, but UnixI >does.  (VMS users will have to ask their system manager for changing thee >CLI.) >r Not 100% true:        LOGIN1  e   /CLI   )         /CLI=command-language-interpretert  nD      Specifies the name of an alternate command language interpreterA      (CLI) to override the default CLI listed in the UAF. The CLI E      you specify must be located in SYS$SYSTEM and have the file typer
      .EXE.   B      If you do not specify a command interpreter by using the /CLIC      qualifier and you do not have a default CLI listed in the UAF,s;      the system supplies the qualifier /CLI=DCL by default.L   -- t  > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                            ------------------------------  % Date: Thu, 22 Apr 2004 15:53:28 +0200-* From: "Karsten Nyblad" <nospam@nospam.com>G Subject: Re: How to turn linux into VMS - memory refresher for Dave ...P- Message-ID: <c68ip0$191u$1@news.cybercity.dk>e  5 "Bob Ceculski" <bob@instantwhip.com> wrote in messageE7 news:d7791aa1.0404220435.46ba4cc8@posting.google.com... = > and while you are checking out the cert counts for the lasts> > ten years, check out this link below to TCPware, an IP stack: > for VMS ... notice the few certs listed had no or little? > effect (access violation error) on VMS ... how did unix fare?o >c. > http://www.process.com/techsupport/cert.html  K Bob, you are trying to change subject.  I will not get into a discussion ontE why there are more CERT reports on *x than on VMS.  You have had thatrJ discussion with Andrew a number of times, and I think there is nothing newJ to add.  We are discussing whether or not VMS is more secure than Unix dueJ to how the operating systems are designed, and it was you who started that discussion, remember?    Karsten Nyblad ibpit1202 at sneakemail dot com    ------------------------------  % Date: Thu, 22 Apr 2004 16:45:24 +0200 * From: "Karsten Nyblad" <nospam@nospam.com>G Subject: Re: How to turn linux into VMS - memory refresher for Dave ... - Message-ID: <c68lq8$1dai$1@news.cybercity.dk>   H "Bob Koehler" <koehler@eisner.nospam.encompasserve.org> wrote in message- news:ibcP2tXkltQa@eisner.encompasserve.org...r? > In article <c67olc$3m1$1@news.cybercity.dk>, "Karsten Nyblad"o <nospam@nospam.com> writes:s > >/K > > Besides, DECnet might be secure, but I also remember it as needing lotst ofK > > resources, e.g., it was extremely slow to start a file transfer becauseJ someJ > > fool had got the idea that peoples login.com should be executed before then! > > actual file transfer started.e >tE >    I'll give up a little bit of performance for a heap of security.t4 >    I end up getting a lot more work done that way. >nH The idea of staring a user process is fine, but much overhead could haveI been avoided by not starting the CLI and not executing the LOGIN.COM.  In H stead file transfers could be started right after checking user name andG password.  I have always thought starting the CLI as an attempt to make % Digitals customers buy more machines.e   Karsten Nyblad ibpit1202 at sneakemail dot com.   ------------------------------  % Date: Thu, 22 Apr 2004 16:32:54 +0100 > From: Andrew Harrison <andrew.d.harrison_remove_the_d@sun.com>G Subject: Re: How to turn linux into VMS - memory refresher for Dave ...a0 Message-ID: <c68pm6$nke$1@new-usenet.uk.sun.com>   Bob Koehler wrote:n > In article <c66ar5$rnt$1@new-usenet.uk.sun.com>, Andrew Harrison <andrew_._remove_harrison@su_n.com> writes: > 8 >>What was it about my previous posts that you disagreed >>with ? >  > K >    You claim I was technically in error, but everything technical I wrotenJ >    was true.  You then invented a "new systems" aspect to the discussion >    which I never claimed.k >    Rubbishx  = You have claimed in the past to have used Solaris and if I amp very much mistaken HP-UX.C   This is what you posted.  > "Every UNIX I have ever used has logon implemented by at least@ login and ftpd.  I've had to fix some of those ftpd to do things login could do."  = Now either your earlier claims to have used Solaris and HP-UX-> are untrue or your most current claim is untrue. Which is it ?  ? so which UNIX's are you refering to when you say all the UNIX'sr, you have ever used suffer from the feature ?     >  >>The correction >>The Sun development of PAM2 >>or the Sun having the largest UNIX market share. >  > D >    Your implied claim that since Solaris now uses PAM the multipleG >    implementations of authentication routines do not exist in UNIX.  oK >    Obviously you've only actually shown that they don't exist in current h >    versions of Solaris.e >   F I didn't say anything about Solaris NOW using PAM, because PAM isn't aH recent addition to Solaris. I also pointed out that HP-UX also uses PAM.  I >    Others have posted that there are UNIX other than Solaris also usingoH >    PAM.  Good.  But it doesn't make all UNIX systems in use today freeH >    of multiple authentication routines.  IT doesn't make UNIX that way< >    by design.  It only makes some UNIX systems that way by >    implementation. >   = It does however severely limit the list of UNIX OS's that youx= could produce in your list of all the UNIX OS's you have ever$ used.w   Regards. Andrew Harrisonl >    ------------------------------  % Date: Thu, 22 Apr 2004 16:37:53 +0100@> From: Andrew Harrison <andrew.d.harrison_remove_the_d@sun.com>G Subject: Re: How to turn linux into VMS - memory refresher for Dave ....0 Message-ID: <c68pvh$nln$1@new-usenet.uk.sun.com>   Dan Allen wrote:Q > PAM is optional even in Solaris. Lots of older subsystems even on Solaris stillcH > use older interfaces directly into the password/shadow files includingH > roll-your-own. Been there seen that more than once and quite recently. >   ? Well of course you can roll your own, however in.ftpd which Bobo was refering to uses PAM.    Regards  Andrew Harrisono   ------------------------------  % Date: Thu, 22 Apr 2004 16:07:21 +0100 > From: Andrew Harrison <andrew.d.harrison_remove_the_d@sun.com>G Subject: Re: How to turn linux into VMS - memory refresher for Dave ...i0 Message-ID: <c68o69$n3k$1@new-usenet.uk.sun.com>   Bob Ceculski wrote:e_ > "Karsten Nyblad" <nospam@nospam.com> wrote in message news:<c67mf7$l0$1@news.cybercity.dk>...  > 7 >>"Bob Ceculski" <bob@instantwhip.com> wrote in message-9 >>news:d7791aa1.0404211441.14cd912e@posting.google.com...  >>G >>>Andrew Harrison <andrew_._remove_harrison@su_n.com> wrote in messagel >>/ >> news:<c66bdh$rua$1@new-usenet.uk.sun.com>...g >>: >>>>Well he may not but Solaris uses PAM and it would also# >>>>appear that HP-UX does as well.o >>>> >>>> >>R >> http://h20000.www2.hp.com/bizsupport/TechSupport/Document.jsp?objectID=tis07674 >>> >>>>I cannot be bothered to look and see if AIX does but sinceA >>>>Solaris and HP-UX are the 1st and 2and largest UNIX platformsaA >>>>its rather safe to say that the majority of UNIX platforms doe8 >>>>not fit the description you applied to them earlier. >>> A >>>one description they fit is they are both convuluted, unsecure  >>>unreliable garbage ...v >>J >>In <c5qav1$c7d$1@news.cybercity.dk> I asked you which CERT reports couldN >>have been avoided if Unix did not execute user programs and command languageJ >>interpreter in separate.  Why don't you answer that question in stead of# >>trolling?  Don't you have answer?r >> >>Karsten Nyblad! >>ibpit1202 at sneakemail dot coms >  > = > and while you are checking out the cert counts for the lastr> > ten years, check out this link below to TCPware, an IP stack: > for VMS ... notice the few certs listed had no or little? > effect (access violation error) on VMS ... how did unix fare?n >   3 And while you are at it Bob why not reflect on thats: fact that you have never managed to justify the comparison8 of OpenVMS vs any other OS using CERT counts because you= have never managed to explain why OpenVMS security advisoriest; do not get posted to CERT reliably or even to explain wherea; you ought to look for them as CERT is the last place to go.h  ; I have asked you to justify your claims or quit making theme8 on more occasions than is worth recalling you have never9 managed and all you are doing now is the standard openVMSi CERT troll.d  ; so I will ask you again, justify your claims or quit makingl them.i     Regardsk ANdrew Harrison    ------------------------------  % Date: Thu, 22 Apr 2004 16:27:28 +0100u> From: Andrew Harrison <andrew.d.harrison_remove_the_d@sun.com>G Subject: Re: How to turn linux into VMS - memory refresher for Dave ....0 Message-ID: <c68pbv$ngh$1@new-usenet.uk.sun.com>   Bob Koehler wrote:n > In article <c66bdh$rua$1@new-usenet.uk.sun.com>, Andrew Harrison <andrew_._remove_harrison@su_n.com> writes: > < >>I cannot be bothered to look and see if AIX does but since? >>Solaris and HP-UX are the 1st and 2and largest UNIX platformsn? >>its rather safe to say that the majority of UNIX platforms dod6 >>not fit the description you applied to them earlier. >  > F >    Just because they are the top 2 shipping today does not make them >    the top 2 in use. > / Fine spin and not what you origionally claimed.t  F >    But I never claimed all UNIX systems were faulty.  I only claimedH >    the solution was not designed-in, and that the fault exists on some >    UNIX systems. > 7 This is actually what you posted so that you can recap.e  > "Every UNIX I have ever used has logon implemented by at least@ login and ftpd.  I've had to fix some of those ftpd to do things login could do."  @ You also have said that you have used Solaris and if I recollectD correctly HP-UX, I will leave you to work out what your posting says! in the light of this information.u  C Remember your point is incorrect if the UNIX OS is Solaris or HP-UXt  @ So which UNIX OS's were you refering to when you said every UNIX$ you have ever used (except Solaris).   25.    Regards, Andrew Harrison8   ------------------------------    Date: 22 Apr 2004 10:33:11 -0700( From: bob@instantwhip.com (Bob Ceculski)G Subject: Re: How to turn linux into VMS - memory refresher for Dave ... = Message-ID: <d7791aa1.0404220933.465a7c6f@posting.google.com>@  ^ "Karsten Nyblad" <nospam@nospam.com> wrote in message news:<c68659$n41$1@news.cybercity.dk>... > E > Yes, but I think not that many are using those old protocols on the K > Internet.  Today people will require encryption on the Internet.  I woulduL > like to keep the discussion to which operating system is best in a hostile > environment. >  > Karsten Nyblad! > ibpit1202 at sneakemail dot coms    / the best?  ok, first unix vs VMS clustering ...n  < http://h71000.www7.hp.com/openvms/whitepapers/Illuminata.pdf    ( then how about unix/linux TCO vs VMS ...  K http://h71000.www7.hp.com/openvms/whitepapers/tco_clusters/TCO_WP_Feb04.pdfe     looks like VMS wins ... :)   ------------------------------  % Date: Thu, 22 Apr 2004 16:34:47 +0200c* From: "Karsten Nyblad" <nospam@nospam.com>G Subject: Re: How to turn linux into VMS - memory refresher for Dave ... - Message-ID: <c68l6d$1cdk$1@news.cybercity.dk>   H "Bob Koehler" <koehler@eisner.nospam.encompasserve.org> wrote in message- news:KsPWa8jYe286@eisner.encompasserve.org.../ > >nE >    Most of the difference is in the implementation, not the design.pD >    Most security problems are buffer-overflow based, and that's anB >    implementation issue.  History shows lots of vendors have hadA >    implementation flaws opening buffer overflows.  Most of UNIXiG >    was implemented im C, where buffer overflows are trivial to setup,hE >    though possible to avoid.  Most of VMS was not implemented in C,uG >    it was in languages where buffer overflows are difficult to setup.eD >    Much of the latest work in VMS is also done in C but so far has >    avoided buffer oeverflows.r >nE >    C enthusiats will tell you that the buffer overflow problem is a D >    compiler problem because the compilers simply use addresses forD >    pointeres, which C does not require.  The fact remains that theE >    compilers do it and C doesn't require any behaviour which is noto >    implementable that way. >tF >    An excellent discussion of the many problems with pointers can beB >    found in Sun's original white paper on the invention of Java,E >    where the authors discuss why they did not put pointers in Java.l  K I think we have already agreed that the zero terminated strings where a badeL idea.  Even worse was the idea of system routines not checking the length ofI buffers used for returning values.  However, code writting to profesionalh$ standards do not use those routines.  H I cannot se that BLISS is a better language for implementing secure codeG than C.  You can write pour code in BLISS too.  The differences between F Unix/C and VMS/BLISS is that the designers of VMS set up better codingL standards from the start.  Please note that the descriptors on VMS are not aH 100% certain way of avoiding buffer overflows.  If you want to be nearlyJ 100% sure that your code does not have any buffer overflows, then you need4 to use other programming languages than BLISS and C.  G >    Many security problems can be traced to UNIX security design, suchtH >    as having only one real privilege level (either you are root or youI >    aren't), and accessing it via suid/sgid bits in the file protection.eC >    History shows lots of vendors have tripped over this, shippingiD >    suid root programs with flaws that let them do things they wereB >    never intended to do.  It's not that the security model isn't@ >    workable, it's just that it's hard to get right.  Many UNIXD >    vendors are moving away from this, creating role-based securityG >    concepts not tied to the file systems, and therefor more like VMS.D  K Most of the priviliges on VMS are in the ALL category meaning that a hacker G with sufficient skills can use then to get full control of the machine.e  D >    So OK, UNIX is getting better, changing some designs and fixingH >    implementation bugs.  Meanwhile I've had all this in VMS since 19847 >    even though UNIX had a 10 year starting advantage.e  I Not true.  The security auditing logs are later.  But it seems to me thatsK Unix has been held back by compatibility issues, e.g., one vendor could not I implement a better password encryption algorithm because then the vendorsh: system would not work together with other vendors systems.   Karsten Nyblad ibpit1202 at sneakemail dot come   ------------------------------   Date: 22 Apr 04 05:29:29 PST From: mckinneyj@cpva.saic.comt6 Subject: Re: ISM Version 6.4-F14 03 and Open VMS 7.3-2( Message-ID: <D1$tcBXRiJJj@cpva.saic.com>  9 In article <c66plv$8cne8$1@ID-143435.news.uni-berlin.de>,c/  "Hans Vlems" <hvlems.dotweg@zonnet.nl> writes:e > 8 > "Bill Schmidt" <bill@wcschmidt.com> schreef in bericht9 > news:ed08a912.0404211304.46a15125@posting.google.com...eI >> I verified that I can upgrade VMS from 7.3 to 7.3-2. I dislike talkingp >> to Intersystems.o >>* >> I have a problem after upgrading my are >> disappearing. >>H >> I have a very short window to do the upgrade and was wondering if any@ >> knows which  parameters or user quotas need to be increased ? >>	 >> Thanks  > = > Would you care to explain the term "jobbed-off processes" ?N >   9 	Essentially, a call to $creprc to create a detached job.s   ------------------------------    Date: 22 Apr 2004 07:16:59 -0700' From: bill@wcschmidt.com (Bill Schmidt)e6 Subject: Re: ISM Version 6.4-F14 03 and Open VMS 7.3-2= Message-ID: <ed08a912.0404220616.42e3ca71@posting.google.com>a  m "Hans Vlems" <hvlems.dotweg@zonnet.nl> wrote in message news:<c66plv$8cne8$1@ID-143435.news.uni-berlin.de>...i8 > "Bill Schmidt" <bill@wcschmidt.com> schreef in bericht9 > news:ed08a912.0404211304.46a15125@posting.google.com... J > > I verified that I can upgrade VMS from 7.3 to 7.3-2. I dislike talking > > to Intersystems. > >o+ > > I have a problem after upgrading my arei > > disappearing.  > >eI > > I have a very short window to do the upgrade and was wondering if any A > > knows which  parameters or user quotas need to be increased ?r > >c
 > > Thanks > = > Would you care to explain the term "jobbed-off processes" ?e >  > Hans   Sorry about that s  1 Jobbed-off is the ISM term for detached processes    Thankt   ------------------------------  % Date: Thu, 22 Apr 2004 18:42:01 +0200n, From: "Hans Vlems" <hvlems.dotweg@zonnet.nl>6 Subject: Re: ISM Version 6.4-F14 03 and Open VMS 7.3-29 Message-ID: <c68soq$96tlu$1@ID-143435.news.uni-berlin.de>c  6 "Bill Schmidt" <bill@wcschmidt.com> schreef in bericht7 news:ed08a912.0404220616.42e3ca71@posting.google.com... 9 > "Hans Vlems" <hvlems.dotweg@zonnet.nl> wrote in messager5 news:<c66plv$8cne8$1@ID-143435.news.uni-berlin.de>...-: > > "Bill Schmidt" <bill@wcschmidt.com> schreef in bericht; > > news:ed08a912.0404211304.46a15125@posting.google.com...wL > > > I verified that I can upgrade VMS from 7.3 to 7.3-2. I dislike talking > > > to Intersystems. > > >o- > > > I have a problem after upgrading my areo > > > disappearing.e > > >mK > > > I have a very short window to do the upgrade and was wondering if anyiC > > > knows which  parameters or user quotas need to be increased ?i > > >n > > > Thanks > >e? > > Would you care to explain the term "jobbed-off processes" ?u > >e > > Hans >u > Sorry about that >h3 > Jobbed-off is the ISM term for detached processes  >p > Thank   8 OK, now that you mention it I should have guessed it :-)L I'd start with Authorize and check whether the usercode that starts the jobsI has the proper values set for MAXJOBS (MAXDETACH perhaps as well), PRCLM, 5 and has authorization to run processes in Batch mode. G Other than that, the output of MC SYSGEN SHO /PQL might be of interest.o   Hans   ------------------------------  % Date: Thu, 22 Apr 2004 15:16:38 +0200f& From: "Walter Kuhn" <w.kuhn@ksg.co.at>5 Subject: Java & OpenVMS : Problem with Thread.sleep()wG Message-ID: <4087c5b9$0$19638$91cee783@newsreader02.highway.telekom.at>    Hello Group,  E We encountered a problem with the Thread.sleep() method in Java underh OpenVMS.  F Sun's description of this method says: "Causes the currently executingK thread to sleep (cease execution) for the specified number of milliseconds. 4 The thread does not lose ownership of any monitors."  J This will not work under OpenVMS, if the system clock is changed (e.g. dueA to an external time synchronization) while the tread is sleeping.V  I The attached Java program demonstrates this effect (endless loop: sleep 5 E seconds & write current time to stdout), which we do not recognize inm Windows or Linux environments.  . We're currently using JRE 1.4.2 and VMS 7.3-2.  # Doen anynone know how to fix this ?D   OUTPUT ON WINDOWS (OK):M   OS: Windows XP   Thu Apr 22 13:23:22 CEST 20046 Thu Apr 22 13:23:28 CEST 2004m Thu Apr 22 13:23:33 CEST 2004m Thu Apr 22 13:23:38 CEST 2004 @ -- time was set 2 minutes ahead; thread does sleep for 5 seconds Thu Apr 22 13:25:38 CEST 2004n Thu Apr 22 13:25:43 CEST 2004 ? -- time was set 2 minutes back, thread does sleep for 5 secondss Thu Apr 22 13:23:48 CEST 2004w   OUTPUT ON OpenVMS (NOT OK):    $ java "TestSleep"   OS: OpenVMSP   Thu Apr 22 13:51:35 MEST 2004s Thu Apr 22 13:51:40 MEST 2004  Thu Apr 22 13:51:45 MEST 2004s Thu Apr 22 13:51:50 MEST 2004i Thu Apr 22 13:51:55 MEST 2004n Thu Apr 22 13:52:00 MEST 2004 G -- time was set 2 minutes ahead; Thread was only sleeping for 1 secondseA (i.e. the time we needed to press ENTER for the SET TIME command)a Thu Apr 22 13:54:01 MEST 2004  Thu Apr 22 13:54:06 MEST 2004d Thu Apr 22 13:54:11 MEST 2004  Thu Apr 22 13:54:16 MEST 2004a Thu Apr 22 13:54:21 MEST 2004s Thu Apr 22 13:54:26 MEST 2004eG -- time was set 2 minutes back; Thread was sleeping for 2 minutes and 5 
 seconds !! Thu Apr 22 13:54:31 MEST 2004A Thu Apr 22 13:54:36 MEST 2004o   Java Source:   import java.util.Date;   /**r>  * Test class that starts an endless loop, in which the threadH  * sleeps for 5 seconds, prints the current time to stdout, sleeps for 5 seconds ...m  *  */a public class TestSleep {  )  public static void main(String[] args) {e=   System.err.println("OS: " + System.getProperty("os.name"));e     while (true) {"    System.out.println(new Date());    try {     Thread.sleep(5000);u%    } catch (InterruptedException e) {      e.printStackTrace();    }   }   } }e     Thank you for help.n   regardsw   Walter Kuhn-> KSG Kommunikations- und Sicherheitssysteme Gesellschaft m.b.H.% Computerstrasse 6, 1101 Wien, AustriaeH Tel.: +43 (1) 601 01 2486; Fax: +43 (1) 603 88 62; Mob.: +43 676 87 44 1 2486 w dot kuhn at ksg dot co dot ati   ------------------------------    Date: 22 Apr 2004 10:17:18 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)/9 Subject: Re: Java & OpenVMS : Problem with Thread.sleep()n3 Message-ID: <aXL7YXXM$zzo@eisner.encompasserve.org>   p In article <4087c5b9$0$19638$91cee783@newsreader02.highway.telekom.at>, "Walter Kuhn" <w.kuhn@ksg.co.at> writes: > Hello Group, > G > We encountered a problem with the Thread.sleep() method in Java undery
 > OpenVMS. > H > Sun's description of this method says: "Causes the currently executingM > thread to sleep (cease execution) for the specified number of milliseconds. 6 > The thread does not lose ownership of any monitors." > L > This will not work under OpenVMS, if the system clock is changed (e.g. dueC > to an external time synchronization) while the tread is sleeping.- > K > The attached Java program demonstrates this effect (endless loop: sleep 5sG > seconds & write current time to stdout), which we do not recognize iny  > Windows or Linux environments. > 0 > We're currently using JRE 1.4.2 and VMS 7.3-2. > % > Doen anynone know how to fix this ?t  B Similar problems in DECnet were fixed by switching from specifying= absolute timer queue entries to relative timer queue entries.c  < I imagine those who maintain Java for VMS could do the same,< but when importing Unix-centric features there is a tendencyB to use C RTL features rather than going back to the specification.   ------------------------------    Date: 22 Apr 2004 07:14:40 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)t= Subject: Re: Linux on its way out - unless you are a geek ... 3 Message-ID: <Y2Py54TBqxo3@eisner.encompasserve.org>   l In article <c66af7$rjp$1@new-usenet.uk.sun.com>, Andrew Harrison <andrew_._remove_harrison@su_n.com> writes: > ; > Of course not but since we are talking about implimentingp/ > new systems its interesting but not relevant.s  ?    Who's talking about implementing new systems?  I was talkinghA    about VMS built-in capabilities vs. add on QFS's capabilities.c   ------------------------------  % Date: Thu, 22 Apr 2004 15:06:17 +0200 * From: "Karsten Nyblad" <nospam@nospam.com>= Subject: Re: Linux on its way out - unless you are a geek ...r- Message-ID: <c68g0d$14u7$1@news.cybercity.dk>n  H "Bob Koehler" <koehler@eisner.nospam.encompasserve.org> wrote in message- news:Y2Py54TBqxo3@eisner.encompasserve.org...cB > In article <c66af7$rjp$1@new-usenet.uk.sun.com>, Andrew Harrison+ <andrew_._remove_harrison@su_n.com> writes:r > > = > > Of course not but since we are talking about implimenting.1 > > new systems its interesting but not relevant.e >yA >    Who's talking about implementing new systems?  I was talkingeC >    about VMS built-in capabilities vs. add on QFS's capabilities.m >r  L Did VMS have clusters before V4.0?  When did the APIs for authentication andA authorization get into OpenVMS?  You are spinning when you try to 5 distinguish between built-in and add on capabilities.    Karsten Nyblad ibpit1202 at sneakemail dot comI   ------------------------------  % Date: Thu, 22 Apr 2004 15:58:36 +0100 > From: Andrew Harrison <andrew.d.harrison_remove_the_d@sun.com>= Subject: Re: Linux on its way out - unless you are a geek ... 0 Message-ID: <c68nls$mtn$1@new-usenet.uk.sun.com>   Bob Koehler wrote:n > In article <c66af7$rjp$1@new-usenet.uk.sun.com>, Andrew Harrison <andrew_._remove_harrison@su_n.com> writes: > ; >>Of course not but since we are talking about implimentingn/ >>new systems its interesting but not relevant.X >  > A >    Who's talking about implementing new systems?  I was talkingoC >    about VMS built-in capabilities vs. add on QFS's capabilities.s >   - Oh dear another added on vs bundled argument.>  7 Does it alter the basic facts which are that other OS's3; do support multi-writer, multi-reader clustered filesystemsb; and that the assumption that they don't so prevelant in ther= OpenVMS community and well illustrated in this thread is alsoa
 incorrect.  9 And as you should also know your assumption that QFS onlyl9 supports FC-SAN based storage was also incorrect, it alsod% supports dual ported storage as well.I  @ It was fun watching you struggle but since your basic assumptionA was incorrect the discussion about MASSBUS etc was fun but hardlye	 relevant.g   Regardsx Andrew Harrisont   ------------------------------  + Date: Thu, 22 Apr 2004 05:12:12 +0000 (UTC)sP From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)* Subject: Re: LYNX, character sets, DECterm$ Message-ID: <c67k7b$t4s$1@online.de>  D In article <070e80lu6eetllc28bajetm8hfdaqfcup2@4ax.com>, John Sauter <J_Sauter@Empire.Net> writes:     > Is EDT still being maintained? >  > John Sauter responded: > 4 > I certainly hope so, it's still being distributed.  H My understanding was that while it is still being distributed, there is G no further development.  I might be wrong, though.  Was it EDT or TECO r! which is a VESTed image on ALPHA?-   ------------------------------  % Date: Thu, 22 Apr 2004 06:51:06 -0400s' From: John Sauter <J_Sauter@Empire.Net>j* Subject: Re: LYNX, character sets, DECterm8 Message-ID: <8j8f80d3s73v9rr2qb82p0qbqicsfu40s6@4ax.com>   Phillip Helbig wrote:n  E My understanding was that while it is still being distributed, there rE is no further development.  I might be wrong, though.  Was it EDT or a& TECO which is a VESTed image on ALPHA?   John Sauter responded:  : There is a big difference between "no further development"7 and "unsupported."  HP has an obligation to support EDTe: as long as they distribute it with OpenVMS.  In January of7 2000 I found a Y2K bug in an obscure corner of the DTSS 9 software.  They fixed it even though we were probably they only customer still using it.p  6 I don't know about TECO, but I believe EDT was Vested.7 It wouldn't have taken much effort to do a native port.s! After all, EDT ran on the PDP-11.R%     John Sauter (J_Sauter@Empire.Net)r   ------------------------------    Date: 22 Apr 2004 08:19:05 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)e* Subject: Re: LYNX, character sets, DECterm3 Message-ID: <fHn+CsWaweI0@eisner.encompasserve.org>f  w In article <c66g6n$8kj$1@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:tF > In article <cpu98056oq5bcndmiiqtde6ov791p34hq6@4ax.com>, John Sauter  > <J_Sauter@Empire.Net> writes:  > @ >> If you can get this to happen reliably, please submit an SPR. >> This is definitely a bug. >   > Is EDT still being maintained?  %    The SPD says it's still supported.M   ------------------------------    Date: 22 Apr 2004 08:23:37 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) * Subject: Re: LYNX, character sets, DECterm3 Message-ID: <APa91zhH$MuJ@eisner.encompasserve.org>n  w In article <c67k7b$t4s$1@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:s > J > My understanding was that while it is still being distributed, there is I > no further development.  I might be wrong, though.  Was it EDT or TECO i# > which is a VESTed image on ALPHA?s  C    TECO is linked to TIE images, so it was most likely VESTed.  EDTt0    is not.  IIRC TECO was quite a test for VEST.  C    TECO is also not available under 8.1.  The VEST-like utility forcC    Alpha to I64 is also not yet available, it is supposed to handlet*    Alpha images that were VESTed from VAX.   ------------------------------    Date: 22 Apr 2004 07:50:41 -07001 From: susan_skonetski@hotmail.com (Sue Skonetski)uY Subject: OpenVMS PearlHewlett-Packard and the Politecnico di Torino - The OpenVMS I-64 Opi= Message-ID: <857e9e41.0404220650.2e1aff7d@posting.google.com>e   Dear Folks,   = Many thanks to OpenVMS Ambassador Massimo Belloni (Italy) forl6 submitting this excellent article as an OpenVMS Pearl.  
 Warm Regards,b Suen. ______________________________________________E Hewlett-Packard and the Politecnico di Torino - The OpenVMS I-64 Openf- Source Porting Program at the starting blocksv  E HP recently started a new initiative called "OpenVMS I-64 Open Sourcei> Porting Program". Promoted by the HP Corporate Engineering, itF provides Itanium hardware to educational institutions that include theD porting of Open Source packages to OpenVMS in their classes. SeveralF universities have been involved in the program, among them the ItalianE Politecnico di Torino, which is the first one to fully implement thisa new initiative.s  > The program has been started by HP on the basis of the renewedF customer interest for the OpenVMS operating system, an industry leaderE solutions providing high stability, reliability, clustering, industryl= standard features and low TCO, very appreciated in e-business E applications. The next release of OpenVMS for Integrity Servers, wellu= amalgamated in the HP Adaptive Enterprise concept, is alreadynD generating a strong interest. This initiative will provide customers@ with further applications, tools and suites coming from the open% source world to the OpenVMS platform.a  E The "OpenVMS I-64 Open Source Porting Program" is very appreciated by ? Professor Marco Mezzalama, pro-rettore (vice-chancellor) of therF Politecnico di Torino (http://www.polito.it/index.en.html). As part ofF the initiative, four systems have been assigned to the Turin athenaeumC (two RX2600 Itanium servers and two Alpha XP1000 workstations) with F the current last release of OpenVMS installed, together with tools andC compilers. The RX2600s run OpenVMS V8.1, today's Itanium developers-E release. The students will be able to perform the open source porting B with support directly provided by HP's Corporate Engineering, thusD living the experience of working with HP's development laboratories,C supported directly by Corporate OpenVMS Engineering  and locally bysE OpenVMS Ambassadors from C&I. Very few universities can offer such an.@ experience to their students, and this will help them to an evenB better preparation for their future in the computer industry, with- particular reference to the OpenVMS platform.s   ------------------------------    Date: 22 Apr 2004 07:17:05 -0700+ From: andreas.kanon@home.se (Andreas Kanon) 0 Subject: PCSI installation on a application disk< Message-ID: <dbbc6dc6.0404220617.ade37dd@posting.google.com>   Hi  C I am in the process of trying to incorporate PCSI as a standard for. our company developed software.dD The problem is that we are currently using VMS Install and it allowsD you to do basically anything since its dcl commandfiles that specify what you will do.eD Our application is never installed on the system disk , but rather a dedicated application disk.   F Now the PCSI$DESTINATION logical is used to tell PCSI where the 'root'B will be so that it can use its relative path inside the package to7 create directory structures and move files in to place.p  E The problem i have is that i want that to appear as a question insideTB the pcsi package rather then as a switch when you issue the instal5 command from the prompt with the /destination switch.u3 I would preffer to do this in the preinstall phase.a@ And i haven't found a really convenient way of checking that theD target disk is the desired one , but rather the only way is to check+ that the target is anything BUT sys$common.0  E I find this way rather crude and would like to find a way around thisn@ since i really like the way pcsi would enable us to uninstal our application easily..  D Any way i can define a target disk from within the packet as well asA verifying that its the same disk as a previous installation (thisd& should be available in the database?)?   ------------------------------  % Date: Thu, 22 Apr 2004 12:10:21 -0400y< From: "Peter Weaver" <WeaverConsultingServices@sympatico.ca>, Subject: Ping: Alan Fay - Netbackup Question9 Message-ID: <c68qpe$9eped$1@ID-141708.news.uni-berlin.de>C   Alan,d  H If we have NetBackup V3.4GA installed and plan on going to V4.5MP6 do weH need to wait for the 4.5 CD to arrive? Or is the MP6 download a complete kit?   --   Peter Weaver Weaver Consulting Services Inc.M Canadian VAR for CHARON-VAX  www.weaverconsulting.cao   ------------------------------    Date: 22 Apr 2004 03:16:03 -0700. From: alexdaniels@themail.co.uk (Alex Daniels) Subject: Re: RWINS= Message-ID: <9f7f13a8.0404220216.4aba01c7@posting.google.com>   ~ "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> wrote in message news:<4085DDA7.269DAE67@NeOaSrPtAhMlNiOnWk.net>... > Alex Daniels wrote:  > > l > > rules@telia.com (Esa Ryh?nen) wrote in message news:<101c337d.0404190015.661dca84@posting.google.com>...A > > > I have a process (running a Java app) stuck in RWINS state.cK > > > I can't stop the process with STOP PROC/ID=xxxx and need to reboot toD > > > get rid of the process..> > > > Anybody know how to stop a process stuck in RWINS state? > > >7
 > > > Regards5 > >  > > This may be worth a try... > >  > > STOP > >  > >   /IMAGE > > 7 > >         /IMAGE [/IDENTIFICATION=pid] [process-name]t > > I > >      Calls the $FORCEX system service to stop the image of the targeteD > >      process specified in the process id or process name that is@ > >      currently executing. The target process is not deleted. > > H > >      If you omit the /IDENTIFICATION qualifier and the process name,A > >      the STOP/IMAGE command is identical to the STOP command.w > J > This still depends on the process to exit the wait state and service the > image exit request.o > C > Remember: neither $DELPRC nor $FORCEX are something you do *TO* agH > process, they are requests you make of a process, and the process must$ > volunteer to honor those requests. > D > A process in a resource wait (RWxxx) state remains there until the. > resource being waited for becomes available.  E I have certainlly managed to kill processes in RW states with $FORCEX A where $DELPRC has not worked. Hence why I posted that it is stille worth trying it.   ------------------------------  % Date: Thu, 22 Apr 2004 05:15:59 -0700e# From: "Tom Linden" <tom@kednos.com>n" Subject: RE: Support Dates for VMS9 Message-ID: <NDEMLKKEBOIFBMJLCECICEAGDCAA.tom@kednos.com>i  E I think you mean 8.2, due Oct 1 on Alpha and Itanium and a few months,
 later on VAX.x     -----Original Message-----6   From: Bhushan Narkhede [mailto:bhushann@hotmail.com]*   Sent: Wednesday, April 21, 2004 10:58 PM   To: Info-VAX@Mvb.Saic.Coms    Subject: Support Dates for VMS         Hi,e   @        Can any one let me know till what date is the OpenVMS 7.2,   supported by HP. Is it declared as of now.      Rgds,i	   Bhushan       ---a(   Incoming mail is certified Virus Free.<   Checked by AVG anti-virus system (http://www.grisoft.com).A   Version: 6.0.655 / Virus Database: 420 - Release Date: 4/8/2004d    --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).? Version: 6.0.655 / Virus Database: 420 - Release Date: 4/8/2004-   ------------------------------    Date: 22 Apr 2004 05:44:24 -07002 From: williamwebb@openvms-rocks.com (William Webb)" Subject: Re: Support Dates for VMS= Message-ID: <bf98c417.0404220444.7d1de1f2@posting.google.com>@  p bhushann@hotmail.com (Bhushan Narkhede) wrote in message news:<8a3b834.0404212157.217df49@posting.google.com>... > Hi,r > @ >      Can any one let me know till what date is the OpenVMS 7.2, > supported by HP. Is it declared as of now. >  > Rgds,I	 > Bhushanc    . The answer to your question can be found here:  3 ftp://ftp.compaq.com/pub/services/software/ovms.pdft   WWWebb   ------------------------------    Date: 22 Apr 2004 08:14:35 -0700' From: bill@wcschmidt.com (Bill Schmidt)f" Subject: Re: Support Dates for VMS= Message-ID: <ed08a912.0404220714.4f781d68@posting.google.com>l  p bhushann@hotmail.com (Bhushan Narkhede) wrote in message news:<8a3b834.0404212157.217df49@posting.google.com>... > Hi,h > @ >      Can any one let me know till what date is the OpenVMS 7.2, > supported by HP. Is it declared as of now. >  > Rgds,h	 > Bhushan   E according to HP, offical end of support was 12/31/02. I still know ofrE sites running 6.x and HP is still supporting them. The only issues isME if you need to patch the operatoring system, they may charge for them    Bill Schmidt' OpenVMS Operating System Support Chart e   OpenVMS VAX Version   Version Release Date   Standard Support   	 End Date O)  Entry Date to Prior Version Support -SE e*  Ending Date for Prior Version Support-SE   l 5.5-2 
  Nov 1991   Sep 30, 1995 
  Oct 1, 1996 O  W/24 mo notice   P 5.5-2H4 
  Jul 1993   Sep 30, 1995 
  Oct 1, 1996 d  W/24 mo notice   u 6.2 
  May 1995   Mar 31, 1998 
  Apr 1, 1998 e  W/24 mo notice   r 7.0 
  Jan 1996   Mar 31, 1998   Not Available k  l 7.1 
  Jan 1997   Jun 30, 2000 
  Jul 1, 2000 e  Dec 31, 2001   t 7.2 
  Feb 1999   Dec 31, 2002 
  Jan 1, 2003 i  June 30, 2004 s  m 7.3 
  May 2001   TBD y  Yes - TBD e  t OpenVMS    Alpha    Version   Version Release Date   Standard Support End Date d)  Entry Date to Prior Version Support -SE a*  Ending Date for Prior Version Support-SE     6.2 
  May 1995   Mar 31, 1998 
  Apr 1, 1998 ,  W/24 mo notice     6.2-1HX   Feb/May 1996   Mar 31, 1998 
  Apr 1, 1998 h  W/24 mo notice   : 7.0 
  Jan 1996   Mar 31, 1998   Not Available f  D 7.1 
  Jan 1997 
  Jul 1, 2000 $  Not available u  o 7.1-1H2 
  Apr 1998 
  Jul 1, 2000 o  Not Available i  L 7.1-2 
  Dec 1998   Jun 30, 2000 
  Jul 1, 2000 s  Dec 31, 2001   d 7.2 
  Feb 1999   Jun 30, 2002   Not available    l 7.2-1 
  Aug 1999   June 30,2002   Not Available i  r 7.2-1H1 
  Jun 2000   Jun 30, 2002   Not available o  e 7.2-2 
  Sep 2001   Dec 31, 2002 
  Jan 1, 2003    June 30, 2004 H  i 7.3 
  May 2001   June 30, 2004 -  Not Available 2  p 7.3-1 
  Aug 2002   Dec 31, 2004   Not Available .  a 7.3-2 
  Dec 2003   TBD >  YES - TBD :  W/24 mo noticex   ------------------------------  # Date: Thu, 22 Apr 2004 04:34:43 GMT-, From: Wesley Dunnahoo <wdunnahoo@mndspng.cm>! Subject: Re: Trying to run an EXE>D Message-ID: <DXHhc.18247$l75.16108@newsread2.news.atl.earthlink.net>    This may be what you need to do.   $ cc agvctl ) $ link agvctl,file1/library,file2/library. $ run agvctl  5 The first line takes agvctl.c and creates agvctl.obj.o  I The second line takes the main program code in agvctl.obj and tells LINK >F to look in file1.olb and file2.olb for any .obj code modules/routines 9 that are referenced in agvctl.obj but wasn't in agvctl.c.0  H The "/library" tells LINK that file1 is file1.olb which is a collection E of object (.obj) files in a library (sort of like a ZIP file but for  I storing related ".obj" files as a collection).  There are other types of  E library files.  If you want to store a collection of text files in a t? library, they have a ".tlb" extension (text library file type).i  F I hope this helps explain how the source, object and executable files F relate to each other in VMS.  It's also handy to use object libraries G when compiling the same program on more than one CPU type (VAX, Alpha,  F Itamium, whatever).  I won't get into the specifics of that right now.  H The RUN statement above actually runs the executable file "AGVCTL.EXE". I   Notice that each command looks for a certain file type/extension so it t( doesn't have to be typed out completely.  : The lines above typed out completely would look like this:  
 $ cc agvctl.cw5 $ link agvctl.obj,file1.olb/library,file2.olb/libraryi $ run agvctl.exe  L Saves quite a bit of typing not having to specify the file types/extensions.  ? If you want to create or update an object library, you can get O< information on object, text or whatever libraries by typing:   $ help library   Aron Hoekstra wrote:G > Wow, thanks everyone for your help!  I am really suprised at how manyaE > people are willing to give me a hand, and i really appreciate it.. e  > Now I have one more problem... > G > I tried using the run command, as suggested... and now i have anotherr	 > error..n >    >  $ $run agvctl' > Error 2424 reported by process SYSTEM 9 > in file $DISK1:[GCSBIII.AGVCTL]AGVCTL.C;192, at line 81l/ > %SYSTEM-W-NOSUCHSEC, no such (global) section-/ > %SYSTEM-W-NOSUCHSEC, no such (global) section4 >  $ >  0F > I have some library files (OLB) that came with the executables..  isG > there anything special I need to do to get those library's registeredeA > in the system?  Is there any way I could tell where that global:( > section is supposed to be initialized? > 5 > the line in the code which generates that error is:o > 1 > init_error_exit(process,__FILE__,__LINE__,err);i >  > thanks >  > -arona   ------------------------------    Date: 22 Apr 2004 07:12:36 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 8 Subject: Re: Using Windows XP hyperterminal with OpenVMS3 Message-ID: <bVhkMPPoLd1F@eisner.encompasserve.org>o  h In article <n4bd80hohmkikmsn2spgc6d3qp041i5lnu@4ax.com>, John Laird <nospam@laird-towers.org.uk> writes: > L > Bob, if you are using a laptop it's going to be very hard indeed to find aJ > "single keystroke" for all 48 Fn, PFn, KPn, and En keys that exist on anI > LK4xx keyboard, because laptops just don't have that number of keys :-)   &    My point exactly!  I'll keep my VT.   ------------------------------  % Date: Fri, 23 Apr 2004 02:11:04 +1000o' From: nospam <x@wedontwantyourspam.com> + Subject: Re: VAX Macro language and file...r5 Message-ID: <BCAE2BB8.2629D%x@wedontwantyourspam.com>t  E in article c63p5c$7hfg8$1@ID-143435.news.uni-berlin.de, Hans Vlems atc2 hvlems.dotweg@zonnet.nl wrote on 21/04/2004 04:09:   > ; > "Crepitus" <le_dieu_crepitus@yahoo.fr> schreef in berichte9 > news:b4c72b33.0404200708.56cc051f@posting.google.com...r >> HelloG >> I would like to create a small program to see how to use file in VAX  >> Macro Language.G >> I'd like to open a file and read it line by line, print each line onr1 >> screen and then close the file for the moment.5A >> Could you give me some documentation to create such a program?w	 >> Thankss >  > ;+G > ; Most of this program originates from the example on page 4-6 of thea. > ; "Introduction to RMS" manual (SIMPLE.MAR).J > ; My extension is simply to count the copied lines and when equal to theI > ; size of a screen to ask for a <cr>. One dirty trick: the line counteriM > ; also serves as the one byte terminal buffer and must therefore be clearedtL > ; after the $GET macro has been executed.                    hv, 6-jul-84.9 > ; V2 now calls $TRNLNM and removes the need to use DCL.CK > ; V2.1 adds support for one parameter: the filename.            21-jan-93O > ;- >       .TITLE  TERM_LISTS > ;               =========- >       .IDENT  /2.1/e! >       .PSECT  DATA,RD,WRT,NOEXE  > ;               ---- >       $SSDEF >       $LNMDEFH >       $NAMDEFi >       $FABDEFr8 >       HCSIZE = 58             ;SIZE OF A HARDCOPY PAGEH > BUFFER: .BLKB   256             ;MAXIMUM RECORDSIZE OF THE SOURCE FILEG > BUFF_DESC:                      ;DESCRIPTOR FOR THE TT: PRINT SERVICE  >       .LONG   0e >       .LONG   BUFFERK ^^^^^^^^^^^^^^^^^^^^^^ Should be .address BUFFER though both work but later," on  its done correctly as .address   ------------------------------  % Date: Thu, 22 Apr 2004 12:44:33 +0200p* From: Paul Sture <nospam@sture.homeip.net>5 Subject: Re: [OT] An interesting view of Linux's costa9 Message-ID: <c687mi$8pn7e$1@ID-132135.news.uni-berlin.de>a   Andrew Harrison wrote: > Hans Vlems wrote:g > < >> "Paul Sture" <nospam@sture.homeip.net> schreef in bericht6 >> news:c60svh$6k0a8$1@ID-132135.news.uni-berlin.de... >> >>> John Smith wrote:o >>> K >>>> http://update.informationweek.com/cgi-bin4/DM/y/egYw0GMMYy0G6p0CUFe0AD  >>>>	 >>>> YMMVe >>>> >>>hG >>> Interesting given my experience with Linux. Back with SuSE 7.n, theqJ >>> networking was not only a pain to configure, but when I added an extraK >>> NIC (identical), it wouldn't see both. Solution: get a couple of new ellH >>> cheapo NICs, wnd they coexisted happily. I was quite happy with that5 >>> distribution once I'd got it configured properly.  >>>eL >>> I finally upgraded to SuSE 9.0 last year, and that was a mistake. On oneL >>> PC, the graphics came out so tiny they were unreadable, and didn't offerL >>> the means to change the resolution. On the PC with twin NICs, I couldn'tJ >>> get either to talk, so I was off to buy another cheap NIC. On that PC,H >>> it would often "lose" both sound and randomly start interpreting the/ >>> keyboard as German instead of Swiss German.p >>>oD >>> I also found the SuSE 9.0 installation somewhat too dumbed down.J >>> Whenever I tried to get into "expert mode" it really tried to dissuade >>> me from doing so.i >>> H >>> OK, I could no doubt have coaxed them both into life eventually, butF >>> once I'd got OpenOffice running on my Mac, I simply lost interest. >>>, >>> YMMV etc >>>u >>E >> On a similar note: I'd been running RedHat 7.1 quite wel on a PC, l >> standardaD >> configuration. Even Windows2000 installs without a problem on it.E >> The Audigy soundcard was not supported but that did not bother me.vA >> RedHat 8 did support the soundcard but messed up the graphics. F >> RedHat 9 offered proper graphics (KDE) but the soundcard no longer 
 >> worked.. >> After that I gave the diskspace to Windows. >> > E > Strange, I am running JDS (based on SuSE enterprise Desktop) on twoaC > PC's a Dell Lattitude and a Evesham box. In both cases I have hadoB > no issues at all with either sound or graphics though I did need1 > to download the NVIDIA drivers for the Evesham.o > D > The Evesham is less than a year old so its relatively new the Dell > is older.  >   & A couple of differences straight away.  G My PCs started out life as nonames running Windows which I upgraded to  B what I thought were industry compatible NICs, graphics, sound, as H required, but the mix wouldn't have been as mainstream as your Dell and  Evesham.  F Also, I would expect that both JDS and the underlying SuSE Enterprise C desktop have had a good deal more work put into them than SuSE Pro.   : > Both the systems NIC's also work as do USB FireWire etc.  G My problem with NICs and graphics was that they worked perfectly under  A previous versions of SuSE. A new version of the Linux kernel was -E responsible for one change of NICs. The published fix worked for one i NIC, but not two.:  
 > The onlyB > thing that doesn't work currently is the Dells Winmodem which is@ > unlikely to be fixed and I cannot change the resolution of the@ > Dells display using the hot keys which is a feature of the ATI > driver for xFree.c >   F I gave up on Winmodems long ago. I still have an aversion to internal G modems unless they have a reset button, as you can power external ones a; off without shutting the entire system if they get "stuck".p  ? > Everything else works remarkably well, DVD reader, CD writer, % > USB flash cards, Palm USB sync etc.a >   E I believe you, but the point Hans and I are making is that there are TH still pitfalls, upgrades can break stuff, and if we as IT professionals ? find it hard work, what chance do Joe and Jane Home User stand?s   ------------------------------   End of INFO-VAX 2004.223 ************************dt.com (Bill Schmidt)e6 Subject: Re: ISM Version 6.4-F14 03 and Open VMS 7.3-2= Message-ID: <ed08a912.0404220616.42e3ca71@posting.google.com>a  m "Hans Vlems" <hvlems.dotweg@
> Remember, Sunbeam is here to cure excessive religious confusion.

    It will suffice, I hope, merely to reduce the suffering from it....as
for a cure...well, one can only hope to reduce the confusion, not cure it.


--
Then Pallas breath'd in Tydeus' sonne --to render whom
  supreame
To all the Greekes at all his parts she cast a hoter beame
On his high mind, his body fild with much superiour might
And made his compleate armor cast a farre more complete light.



^~00001466:0000040927:034694:From: Guillaume de Noers <libertas@bellsouth.net>
Newsgroups: neworleans.general
Subject: Re: Kumbaya moment of the day.
Date: Fri, 31 Jan 2003 20:18:48 -0600
Message-ID: <3ebm3vcsquek0a3cqu5qn8r82rg1hdii5m@4ax.com>
References: <3e3b7f69.3401060@news.direcway.com> <qn0l3vkrjcbqccfki5fjv1abt3ihmo2410@4ax.com> <3e3aa278.1256702@news.direcway.com> <ut9l3vsstbc15i8t27dapi8975j3h2f9da@4ax.com> <3e3b02d5.25928905@news.direcway.com>
X-Newsreader: Forte Agent 1.8/32.548
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
NNTP-Posting-Host: host-209-214-195-172.mem.bellsouth.net
X-Trace: news.nb.nu 1044065856 host-209-214-195-172.mem.bellsouth.net (1 Feb 2003 03:17:36 +0100 (GMT))
Lines: 18
X-Complaints-To: abuse@nb.nu
X-Visit: http://www.nb.nu/serv04.htm to gain access
Path: mvb.saic.com!news-out.cwix.com!newsfeed.cwix.com!iad-peer.news.verio.net!news.verio.net!newspeer1.nwr.nac.net!newsfeed.freenet.de!amsnews01.chello.com!news.nb.nu!not-for-mail
Xref: mvb.