0 INFO-VAX	Fri, 09 Jan 2004	Volume 2004 : Issue 17      Contents: 7.3-1 to 7.3-2 Re: 7.3-1 to 7.3-2- Re: Alpha macro (64) and the OTS$... routines  Anonymous Read-Only  Change System Node Identifier ! Re: Change System Node Identifier ! Re: Could OpenVMS-Athlon ever be? ! RE: Could OpenVMS-Athlon ever be? ! Re: Could OpenVMS-Athlon ever be? ! Re: Could OpenVMS-Athlon ever be? ! Re: Could OpenVMS-Athlon ever be? ! Re: Could OpenVMS-Athlon ever be? ! Re: Could OpenVMS-Athlon ever be? ! Re: Could OpenVMS-Athlon ever be?  CSWS V2 and ODS5 Re: CSWS V2 and ODS54 Re: DCPS and LPD and Xerox Document Centre printers?4 Re: DCPS and LPD and Xerox Document Centre printers?4 RE: DCPS and LPD and Xerox Document Centre printers?4 Re: DCPS and LPD and Xerox Document Centre printers? Re: ds10l blowout  Re: ds10l blowout  Re: ds10l blowout P Hewlett-Packard plans to EOL the HP AlphaServer Tru64. Find out how Sun can help0 Hints to upgrade DS10L ? (Was Re: ds10l blowout)0 Hints to upgrade DS10L ? (Was Re: ds10l blowout)4 Re: Hints to upgrade DS10L ? (Was Re: ds10l blowout)' Re: how to flush a file opened by DCL ?   Re: HP TCPIP SMTP Queues (again)  Re: HP TCPIP SMTP Queues (again)  Re: HP TCPIP SMTP Queues (again)% Re: I just scored an AlphaServer 800!  Multnet Documentation V4.0?  Re: MySQL 4.0.17 for OpenVMS3 Re: Non-WS Process Quotas not in Performance Manual , RE: number of ethernet adapters in a cluster Re: OpenVMS Certifications Re: OpenVMS CertificationsP Re: Oracle Rdb on GS1280 with 7.3-2 exceeds 1 million transactions per minute us Re: Python 2.3.3A Re: Searchlist behaviour differences between DCL/RMS and CRTL/EXE 8 Re: Secure Web Server Version 2.0 available for download8 Re: Secure Web Server Version 2.0 available for download8 Re: Secure Web Server Version 2.0 available for download8 Re: Secure Web Server Version 2.0 available for download8 Re: Secure Web Server Version 2.0 available for download< supported connection methods for $ @ SYS$SYSTEM:SHUTDOWN.COM@ Re: The Register: OpenVMS among most-secure of operating systems@ Re: The Register: OpenVMS among most-secure of operating systems@ Re: The Register: OpenVMS among most-secure of operating systems Re: Vaxstation some question...  Re: Vaxstation some question...  Re: Vaxstation some question...  RE: Vaxstation some question... , Re: VMS naming conventions for disk volumes?, Re: VMS naming conventions for disk volumes?, Re: VMS naming conventions for disk volumes?, RE: VMS naming conventions for disk volumes?, Re: VMS naming conventions for disk volumes?, RE: VMS naming conventions for disk volumes?, Re: VMS naming conventions for disk volumes?$ Re: VMS to VMS copying between Nodes$ Re: VMS to VMS copying between Nodes$ Re: VMS to VMS copying between Nodes ZLIB 1.2.1 form OpenVMS 9 Re: [Survey] Your best choice for an X Client on Lostdows 9 Re: [Survey] Your best choice for an X Client on Lostdows 9 Re: [Survey] Your best choice for an X Client on Lostdows   F ----------------------------------------------------------------------   Date: 9 Jan 2004 00:45:18 -0800 - From: bhushann@hotmail.com (Bhushan Narkhede)  Subject: 7.3-1 to 7.3-2 < Message-ID: <8a3b834.0401090045.47417059@posting.google.com>   Hi Guys,  B   I am looking at the efforts that might be involved in moving our0 system applications from OpenVMS 7.3-1 to 7.3-2.  - can anyone guide on how to proceed on this??? F What all changes have gone in from 7.3-1 to 7.3-2 ??? or point to some  kind of documentation on it ????E I am trying to get the release notes for 7.3-2, but looks like the HP F site is undergoing some kind of work or due to some reason not able to get any thing from their site.   Rgds,  Bhushan    ------------------------------   Date: 9 Jan 2004 07:18:35 -0600 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: 7.3-1 to 7.3-2 3 Message-ID: <wYB39ndHKktX@eisner.encompasserve.org>   l In article <8a3b834.0401090045.47417059@posting.google.com>, bhushann@hotmail.com (Bhushan Narkhede) writes:
 > Hi Guys, > D >   I am looking at the efforts that might be involved in moving our2 > system applications from OpenVMS 7.3-1 to 7.3-2.  ?    Most likely, nothing.  You'ld have to know your applications D    quite well to know if there's some small nit that you'ld actually
    trip over.    ------------------------------   Date: 9 Jan 2004 04:01:37 -0800 ( From: don.braffitt@hp.com (Don Braffitt)6 Subject: Re: Alpha macro (64) and the OTS$... routines< Message-ID: <e5c311b8.0401090401.f62b379@posting.google.com>  n denny.rich@swagelok.com (Denny Rich) wrote in message news:<d28306e.0401081303.28e963d4@posting.google.com>...@ > We are researching a COBOL problem. There is a difference in aD > comparision (IF statement) between a program compiled on VAX COBOL; > (V7.1, Cobol T5.4), and Alpha Cobol (V7.3-1, Cobol V2.7).  > E > While researching this, we looked at the machine code.  On the VAX, @ > the compiler sets up a CMPC5 instruction, followed by a BLEQU,F > comparing the data to ^x39 (digit 9). On the Alpha, the code sets upD > what looks like a comparison to a ^d32 (space) but it does this by > calling some OTS$ routines.  > E > We can make the Alpha work like the VAX if we compile on Alpha with  > "/convert=leading_blanks". > F > All well and good, but we are nervous about not understanding WHY it > works. > G > In the Alpha machine code, we see calls to one of two routines to set ; > up the compare: OTS$FETCH_U_OPN0_1 (with /noconvert), and = > OTS$FETCH_U_OPN0_1_B (with /convert=leading). Then it calls : > OTS$CMP_LE_DEC_1 in both cases to do the actual compare. > F > Does anybody have any documentation on what are the Inputs, Outputs,) > and Functions of these 3 OTS$ routines?  >   > Grateful for any assistance... >  > denny    Denny,  # There are two issues involved here.   D On Alpha and I64, various routines in LIBOTS2 were written to handleE decimal numeric and string operations where VAX instructions had been B used but no equivalent or similar instructions were present in the@ Alpha and I64 architectures. The code generator for COBOL on VAXE generates instructions like CMPC5 whereas the code generator on Alpha < and I64 generates calls to OTS$ routines. There is not a 1-1= correspondence between VAX instructions and OTS$ routines. In ; particular, the decimal numeric routines in LIBOTS2 such as D OTS$FETCH_U_OPN0_1_B are strictly used for communication between the> COBOL and BASIC compilers on Alpha and I64 and the LIBOTS2.EXED run-time support. These routines are not externally callable outside2 of this context and are not externally documented.  E Also, blanks in decimal numeric data are invalid decimal data on VAX, ? Alpha, and I64, but invalid data decimal detection with the VAX F hardware is not 100% compatible with invalid data decimal detection inF RTL routines on Alpha and I64. Some other hardware architectures (e.g.F PDP-11) treat blanks in decimal numeric data as 0. To ease conversionsB to Alpha and I64, we added an option to COBOL so that the compiler@ will generate code to handle at run-time conversion of blanks in" decimal display numeric data to 0.  C You should also compile on Alpha or I64 with /CHECK=DECIMAL to help B detect situations where you still might have invalid decimal data.? Results with invalid decimal data are undefined and may produce > different results on VAX, Alpha, and I64. This is probably the> difference you are seeing when you don't compile on Alpha with /CONVERT=LEADING_BLANKS.   $ help cobol/convert COBOL   
   /CONVERT      /CONVERT[=(option[,...])]    /NOCONVERT        (D)  ;    Instructs the compiler to perform certain conversions to 
    your data.   '    You can select the following option:   ?    [NO]LEADING_BLANKS      Controls whether or not the compiler 5                            changes blanks to zeros in 1                            numeric display items.      E    Specifying /CONVERT=LEADING_BLANKS instructs the compiler to check ;    for and change blanks to zeros in numeric display items.   F    Specify /CONVERT=LEADING_BLANKS, if you are using a non-VAX system,E    to convert your existing COBOL programs to run on an OpenVMS Alpha E    or I64 system by changing blanks in the data to zeros at run time.   D    Specifying /CONVERT=LEADING_BLANKS produces extra instructions toF    perform these data conversions, which may result in slightly largerA    images and slightly longer execution times than the /NOCONVERT 
    qualifier.       The default is /NOCONVERT.    - Don Braffitt5   project leader, HP COBOL for OpenVMS and Tru64 UNIX    ------------------------------  $ Date: Fri, 9 Jan 2004 09:49:26 -07008 From: "Michael D. Ober" <obermd-.@.-alum-mit-edu-nospam> Subject: Anonymous Read-Only0 Message-ID: <rYALb.24$WK6.12510@news.uswest.net>  L Does anyone know how to get NFS (GuardianOS Linux kernal) to allow unlimitedI read-only access to an NFS share.  All the access will be from one client H (VMS 7.3) , but the user accounts are highly variable, so keeping a user& list on the NFS server isn't feasible.   Thanks, 
 Mike Ober.   ------------------------------   Date: 9 Jan 2004 04:05:49 -0800 ) From: thelockmaster@hotmail.com (Nigel H) & Subject: Change System Node Identifier= Message-ID: <71650955.0401090405.27f3b796@posting.google.com>    Hi,   D We have a new 2 node alpha cluster to which we are going to port ourF existing VAX application. The new cluster was delivered with VMS 7.3-2 etc. installed.   < We have a problem with the system node identifiers, i.e. theD SYS$NODE_<cluster_node1> and ...node2 identifiers, which are defined@ with values of %X80010001 and 3. These values conflict with some= application defined identifiers used on the VAX system we are E migrating - these are used in ACLs and granted on our RDB database to  give the required access.   C As a result of this when we restore files/databases from the VAX to A the alpha the security is wrong as the value of the identifier is F maintained and not the name. I know it is possible to write scripts toC ammend the ACLs and database protection, however this would leave a B difference betweem the VAX and alpha systems which we would ratherF avoid (not least because every time we copy anything across we have to mod. the security).   < Is there a safe way of changing the value of the system nodeC identifiers so that we can define our existing identifiers with the @ same values? For example define our identifiers with the correct< valeus (i.e. replacing the system node identifiers) and then re-install DECnet?   Any help much appreciated!   Thanks Nigel    ------------------------------  % Date: Fri, 09 Jan 2004 13:17:14 +0000 * From: Nic Clews <sendspamhere@[127.0.0.1]>* Subject: Re: Change System Node Identifier' Message-ID: <btm9kk$8hq$1@lore.csc.com>    Nigel H wrote: > > > We have a problem with the system node identifiers, i.e. theF > SYS$NODE_<cluster_node1> and ...node2 identifiers, which are definedB > with values of %X80010001 and 3. These values conflict with some? > application defined identifiers used on the VAX system we are G > migrating - these are used in ACLs and granted on our RDB database to  > give the required access.  > > > Is there a safe way of changing the value of the system nodeE > identifiers so that we can define our existing identifiers with the B > same values? For example define our identifiers with the correct> > valeus (i.e. replacing the system node identifiers) and then > re-install DECnet?  2 This problem may be simpler than you think to fix.  H If you rename the identifier in question (using RENAME/IDE in AUTHORIZE)E the now "missing" identifier of the SYS$NODE_... will be added to the C rightslist database at the next reboot. It is honestly that simple.   G The values and even the names are somewhat arbitrary, the rightslist is H effectively a lookup table of the numbers, and in some cases is just theE next sequential number when a new identifier is added. If you want to D keep some parity with your VAX systems, then add all the identifiers= with values you'd like, see the HELP in authorize to do this.   9 There is absolutely no need to remove or reinstall DECnet    Hope this helps  --  ? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences  nclews at csc dot com    ------------------------------   Date: 9 Jan 2004 07:55:32 GMT 2 From: "Dave Weatherall" <djw-nothere@nospam.nohow>* Subject: Re: Could OpenVMS-Athlon ever be?5 Message-ID: <DTiotGxQ0bj6-pn2-dYyn5EuNIK43@localhost>   B On Thu, 8 Jan 2004 20:28:44 UTC, John Reagan <john.reagan@hp.com>  wrote:  3 > I really don't know where to start on this one...  > I > - Porting Macro-32 to 8086 would be very difficult.  The biggest issue  J > the lack of sufficient registers on 8086.  Think of Macro32 source code I > has having several 'predeclared variables' named R0, R1, etc. that are  E > global to all routines.  On Alpha and Itanium, there are sufficent  K > hardware registers for all of that to work although the Calling Standard  K > on OpenVMS I64 makes it a little interesting.  On 8086, I really have no  I > clue how to efficiently generate code for Macro32 routines that pass 8  D > 'registers' between each other (or between Macro32 and BLISS or C J > routines) for instance.  Macro32's use of GEM is unique compared to the K > other GEM-based compilers.  We have our own code generation and just use  H > GEM for object file writing, peepholing, instruction scheduling, etc. K > The fact that there exists a current 8086 GEM doesn't help Macro32 much.  B >   For instance, for the port to Itanium, we had to gut the code D > generation modules from the OpenVMS Alpha compiler and completely A > reimplement them for Itanium at a cost of several person-years.  > J > - Not all of the compiler people are at Intel.  I'm not.  We still have K > engineers to support GEM and our front-ends on all their current targets  K > as needed.  Intel isn't porting any compilers to OpenVMS.  We are taking  H > some compilers from Intel (C++ right now, C in the future, Fortran is G > somewhat unique) and WE are porting/adapting them to OpenVMS.  Also,  G > OpenVMS customers require more than just C, C++, Fortran,  There are  @ > COBOL, Pascal, and BASIC customers as well as BLISS customers A > internally.  I know that even the current 8086 GEM wouldn't be  I > sufficient for those compilers on traditional 8086 targets.  More work   > would be needed. > H > - Even if Intel produced some sort of mythical 64-bit 8086, again, is I > Intel going to give us a BLISS or any other OpenVMS compiler?  I think  H > not.  For our compilers that are GEM-based, Intel isn't going to do a B > 64-bit 8086 GEM target.  That would all be work we'd have to do.  F The response is appeciated John and the argument understood. The lack F of registers has always been a reason why I was never keen on x86. I'm6 sure the argument was even more powerful 10 years ago.   That said :-  E Any x86 MACRO-32 compiler would assume that it was targetting an x86  ; version of VMS. If this was done with a virtual machine or  A memory-based register stacks then the problem of mapping the VAX  C registers would/could have been resolved. I do remember reading in  E some book, way back when I was still playing with x86 machines, that  F the x86 architecture was not so IIRC 'reliant' on registers because it@ viewed all available memory as if it were a set of registers. I F confess to not having been convinced - we were barely into 286/386 at C the time. Since then, however, Intel, Cyrix and AMD have made this  F statement more of a reality (and quietly added more registers onto the silicon)  E How that would relate to GEM I'm not sure as I know nothing about it   really.   9 Did the NT BLISS compiler use GEM or was that a complete  / re-implemenation?  Does VMS BLISS use GEM even?   B One area that would definitely be a problem is DEC Floating Point : support but that has nowt to do with programming language.   --   Cheers - Dave.   ------------------------------  $ Date: Fri, 9 Jan 2004 04:46:56 -0800# From: "Tom Linden" <tom@kednos.com> * Subject: RE: Could OpenVMS-Athlon ever be?9 Message-ID: <NDEMLKKEBOIFBMJLCECIEEAPCKAA.tom@kednos.com>      -----Original Message-----9   From: Dave Weatherall [mailto:djw-nothere@nospam.nohow] +   Sent: Thursday, January 08, 2004 11:56 PM    To: Info-VAX@Mvb.Saic.Com ,   Subject: Re: Could OpenVMS-Athlon ever be?    C   On Thu, 8 Jan 2004 20:28:44 UTC, John Reagan <john.reagan@hp.com>    wrote:  5   > I really don't know where to start on this one...    > J   > - Porting Macro-32 to 8086 would be very difficult.  The biggest issueK   > the lack of sufficient registers on 8086.  Think of Macro32 source code J   > has having several 'predeclared variables' named R0, R1, etc. that areF   > global to all routines.  On Alpha and Itanium, there are sufficentC   > hardware registers for all of that to work although the Calling 
   Standard=   > on OpenVMS I64 makes it a little interesting.  On 8086, I    really have noJ   > clue how to efficiently generate code for Macro32 routines that pass 8E   > 'registers' between each other (or between Macro32 and BLISS or C K   > routines) for instance.  Macro32's use of GEM is unique compared to the C   > other GEM-based compilers.  We have our own code generation and 
   just useI   > GEM for object file writing, peepholing, instruction scheduling, etc. >   > The fact that there exists a current 8086 GEM doesn't help   Macro32 much. C   >   For instance, for the port to Itanium, we had to gut the code E   > generation modules from the OpenVMS Alpha compiler and completely C   > reimplement them for Itanium at a cost of several person-years.    > K   > - Not all of the compiler people are at Intel.  I'm not.  We still have <   > engineers to support GEM and our front-ends on all their   current targets A   > as needed.  Intel isn't porting any compilers to OpenVMS.  We    are takingI   > some compilers from Intel (C++ right now, C in the future, Fortran is H   > somewhat unique) and WE are porting/adapting them to OpenVMS.  Also,H   > OpenVMS customers require more than just C, C++, Fortran,  There areA   > COBOL, Pascal, and BASIC customers as well as BLISS customers B   > internally.  I know that even the current 8086 GEM wouldn't beJ   > sufficient for those compilers on traditional 8086 targets.  More work   > would be needed.   > I   > - Even if Intel produced some sort of mythical 64-bit 8086, again, is J   > Intel going to give us a BLISS or any other OpenVMS compiler?  I thinkI   > not.  For our compilers that are GEM-based, Intel isn't going to do a D   > 64-bit 8086 GEM target.  That would all be work we'd have to do.  G   The response is appeciated John and the argument understood. The lack H   of registers has always been a reason why I was never keen on x86. I'm8   sure the argument was even more powerful 10 years ago.     That said :-  F   Any x86 MACRO-32 compiler would assume that it was targetting an x86<   version of VMS. If this was done with a virtual machine orB   memory-based register stacks then the problem of mapping the VAXD   registers would/could have been resolved. I do remember reading inF   some book, way back when I was still playing with x86 machines, thatH   the x86 architecture was not so IIRC 'reliant' on registers because itA   viewed all available memory as if it were a set of registers. I G   confess to not having been convinced - we were barely into 286/386 at D   the time. Since then, however, Intel, Cyrix and AMD have made thisH   statement more of a reality (and quietly added more registers onto the
   silicon)  K It was the TI 990 that view all of memory as a set of registers, the x86 is E an accululator architecture, which was obsolete when it was designed,  largely J by engineers with little or no computer background.  The core of x86 theseL days is an 801 architecure effectively emulating this horrible architecture.  F   How that would relate to GEM I'm not sure as I know nothing about it	   really.   :   Did the NT BLISS compiler use GEM or was that a complete1   re-implemenation?  Does VMS BLISS use GEM even?   C   One area that would definitely be a problem is DEC Floating Point <   support but that has nowt to do with programming language.     --   Cheers - Dave.     --- (   Incoming mail is certified Virus Free.<   Checked by AVG anti-virus system (http://www.grisoft.com).A   Version: 6.0.560 / Virus Database: 352 - Release Date: 1/8/2004    --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).? Version: 6.0.560 / Virus Database: 352 - Release Date: 1/8/2004    ------------------------------  $ Date: Fri, 9 Jan 2004 12:23:24 -0000$ From: "L. Blunt" <someone@out.there>* Subject: Re: Could OpenVMS-Athlon ever be?+ Message-ID: <btm6cm01jt9@news1.randori.com>   H > - Porting Macro-32 to 8086 would be very difficult.  The biggest issueI > the lack of sufficient registers on 8086.  Think of Macro32 source code H > has having several 'predeclared variables' named R0, R1, etc. that areD > global to all routines.  On Alpha and Itanium, there are sufficentJ > hardware registers for all of that to work although the Calling StandardJ > on OpenVMS I64 makes it a little interesting.  On 8086, I really have noH > clue how to efficiently generate code for Macro32 routines that pass 8C > 'registers' between each other (or between Macro32 and BLISS or C  > routines) for instance.   H Although I can see your point about all current Intel x86 chips, but theI original link was about the new AMD 64 chips. When running in 64bit mode, H these now have 16 registers (all 64bit), and 16 floating point registersK (also 64bit). Now I'm no compiler writer, but that could have made the task 
 much simpler.   L From a personal point of view; I really like the idea of a native VMS-AMD64.K You would actually get all that was claimed for Itanium (Merced,IA64), as a 1 low cost very high performance hardware platform.   K You would also get, simple(cheap) and very fast, 1-4 (maybe 8) SMP systems, I using AMDs technology, which very similar to EV7. Bigger systems could be = built around multiple 4 cpu nodes (as is being done by Cray).   G Unlike Itanic which is so-so on performance (unless you need a floating L point co-processor :) and is likely to always be a high cost product, due toL low sales. The low sales prediction is because I cannot see Intel not havingL to create a x86-64 chip to stay competitive in anything other than the lowerH end of the PC market, at which point IA64 becomes a very expensive niche product.   ------------------------------  # Date: Fri, 09 Jan 2004 15:05:00 GMT & From: John Reagan <john.reagan@hp.com>* Subject: Re: Could OpenVMS-Athlon ever be?3 Message-ID: <wqzLb.11989$7c4.1686@news.cpqcorp.net>    Dave Weatherall wrote:   > H > The response is appeciated John and the argument understood. The lack H > of registers has always been a reason why I was never keen on x86. I'm8 > sure the argument was even more powerful 10 years ago. >  > That said :- > G > Any x86 MACRO-32 compiler would assume that it was targetting an x86  = > version of VMS. If this was done with a virtual machine or  C > memory-based register stacks then the problem of mapping the VAX  E > registers would/could have been resolved. I do remember reading in  G > some book, way back when I was still playing with x86 machines, that  H > the x86 architecture was not so IIRC 'reliant' on registers because itB > viewed all available memory as if it were a set of registers. I H > confess to not having been convinced - we were barely into 286/386 at E > the time. Since then, however, Intel, Cyrix and AMD have made this  H > statement more of a reality (and quietly added more registers onto the
 > silicon)  E That would be OK, if Macro32 just called other Macro32 routines.  We  I could put the "registers" into memory just fine.  However, you then have  G to coordinate with the other compilers used in the OpenVMS source pool  H that pass things in "registers" between languages (namely BLISS and C). E   You'd have to teach those languages and their code generator about  H passing "registers" parameters which are now in static memory locations.  F It gets even more complicated with saving register state at exception H entry, AST entry, interrupts, etc.  Given the amount of Macro32 code in C OpenVMS, the "registers in memory" concept would spread across the   entire operating system.  ; > Did the NT BLISS compiler use GEM or was that a complete  1 > re-implemenation?  Does VMS BLISS use GEM even?   = All BLISS compilers other than VAX BLISS use the GEM backend.      --   John Reagan ' Compaq Pascal/{A|I}MACRO Project Leader  Hewlett-Packard Company    ------------------------------  # Date: Fri, 09 Jan 2004 15:09:58 GMT & From: John Reagan <john.reagan@hp.com>* Subject: Re: Could OpenVMS-Athlon ever be?4 Message-ID: <avzLb.11990$qe4.11108@news.cpqcorp.net>   L. Blunt wrote:    > J > Although I can see your point about all current Intel x86 chips, but theK > original link was about the new AMD 64 chips. When running in 64bit mode, J > these now have 16 registers (all 64bit), and 16 floating point registersM > (also 64bit). Now I'm no compiler writer, but that could have made the task  > much simpler.   G Actually, I downloaded and did a quick read o the AMD64 files over the  I holidays.  The 16 registers help, but it would still be very hard to do.  E    The OpenVMS Alpha and I64 code base has evolved over the years to  G take advantage of more than just the VAX registers of R2-R11.  Macro32  H provides syntax to talk about all 32 Alpha registers and we providing a F mapping for all of them on Itanium.  This allowed much of the code to @ just recompile without modification.  On a machine with just 16 C registers, all/most of that code would have to be revisited or the  E Macro32 compiler would have to invent some solution and remap behind  1 your back.  I'm not sure how that would all work.      --   John Reagan ' Compaq Pascal/{A|I}MACRO Project Leader  Hewlett-Packard Company    ------------------------------  $ Date: Fri, 9 Jan 2004 16:52:34 -0000$ From: "L. Blunt" <someone@out.there>* Subject: Re: Could OpenVMS-Athlon ever be?+ Message-ID: <btmm5c02nti@news1.randori.com>   H Ah, that's the thing, I thought Macro32 was the VAX assembler, which wasL just compiled on Alphas. Not that it had been extended for that architecture0 as well (that's what I thought Macro64 was for).  3 "John Reagan" <john.reagan@hp.com> wrote in message . news:avzLb.11990$qe4.11108@news.cpqcorp.net... > L. Blunt wrote:  >  > > L > > Although I can see your point about all current Intel x86 chips, but theG > > original link was about the new AMD 64 chips. When running in 64bit  mode, L > > these now have 16 registers (all 64bit), and 16 floating point registersJ > > (also 64bit). Now I'm no compiler writer, but that could have made the task > > much simpler.  > H > Actually, I downloaded and did a quick read o the AMD64 files over theJ > holidays.  The 16 registers help, but it would still be very hard to do.F >    The OpenVMS Alpha and I64 code base has evolved over the years toH > take advantage of more than just the VAX registers of R2-R11.  Macro32I > provides syntax to talk about all 32 Alpha registers and we providing a G > mapping for all of them on Itanium.  This allowed much of the code to A > just recompile without modification.  On a machine with just 16 D > registers, all/most of that code would have to be revisited or theF > Macro32 compiler would have to invent some solution and remap behind3 > your back.  I'm not sure how that would all work.  >  >  > --
 > John Reagan ) > Compaq Pascal/{A|I}MACRO Project Leader  > Hewlett-Packard Company  >    ------------------------------  # Date: Fri, 09 Jan 2004 18:00:14 GMT & From: John Reagan <john.reagan@hp.com>* Subject: Re: Could OpenVMS-Athlon ever be?3 Message-ID: <O_BLb.12013$Jh4.4911@news.cpqcorp.net>    L. Blunt wrote: J > Ah, that's the thing, I thought Macro32 was the VAX assembler, which wasN > just compiled on Alphas. Not that it had been extended for that architecture2 > as well (that's what I thought Macro64 was for).  E Macro32 on Alpha allows you to say any register R0 thru R31 in a VAX  G instruction (ie, ADDL2 (R22)[R23], R24).  There are also various EVAX_  I builtins and now IA64_ builtins as well.  However, those builtins aren't   the 'difficult' problem.   --   John Reagan ' Compaq Pascal/{A|I}MACRO Project Leader  Hewlett-Packard Company    ------------------------------  # Date: Fri, 09 Jan 2004 18:13:36 GMT 9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> * Subject: Re: Could OpenVMS-Athlon ever be?3 Message-ID: <kbCLb.12020$Rt4.6069@news.cpqcorp.net>   / "L. Blunt" <someone@out.there> wrote in message % news:btmm5c02nti@news1.randori.com... J > Ah, that's the thing, I thought Macro32 was the VAX assembler, which wasA > just compiled on Alphas. Not that it had been extended for that  architecture2 > as well (that's what I thought Macro64 was for). >   H The Alpha and IA64 assemblers are there for things you just can't do anyK other way.  But there is a large amount of existing code that originated on I VAX and is Macro-32.  *Very* often it was the case that it was simpler to J make Alpha-specific changes to the Macro-32 code, than to rewrite the code
 in, say C.    5 > "John Reagan" <john.reagan@hp.com> wrote in message 0 > news:avzLb.11990$qe4.11108@news.cpqcorp.net... > > L. Blunt wrote:  > >  > > > J > > > Although I can see your point about all current Intel x86 chips, but the I > > > original link was about the new AMD 64 chips. When running in 64bit  > mode, D > > > these now have 16 registers (all 64bit), and 16 floating point	 registers L > > > (also 64bit). Now I'm no compiler writer, but that could have made the > task > > > much simpler.  > > J > > Actually, I downloaded and did a quick read o the AMD64 files over theL > > holidays.  The 16 registers help, but it would still be very hard to do.H > >    The OpenVMS Alpha and I64 code base has evolved over the years toJ > > take advantage of more than just the VAX registers of R2-R11.  Macro32K > > provides syntax to talk about all 32 Alpha registers and we providing a I > > mapping for all of them on Itanium.  This allowed much of the code to C > > just recompile without modification.  On a machine with just 16 F > > registers, all/most of that code would have to be revisited or theH > > Macro32 compiler would have to invent some solution and remap behind5 > > your back.  I'm not sure how that would all work.  > >  > >  > > -- > > John Reagan + > > Compaq Pascal/{A|I}MACRO Project Leader  > > Hewlett-Packard Company  > >  >  >    ------------------------------  $ Date: Fri, 9 Jan 2004 08:43:43 +0100( From: "Rudolf Wingert" <win@fom.fgan.de> Subject: CSWS V2 and ODS5 : Message-ID: <MCELKPMOKPMNDNKJNIONKEKNCIAA.win@fom.fgan.de>   Hello,  L did Im read right, that you need ODS5 for CSWS V2? We dont like to use ODS5P disks. We love ODS2 with its lovely filename restriction. Esspacialy that all isC uppercase. If I do not have ODS5, am I not able to install CSWS V2?    Best regards R. Wingert    ------------------------------  % Date: Fri, 09 Jan 2004 09:15:22 +0100 : From: Karl Rohwedder <extern.karl.rohwedder@volkswagen.de> Subject: Re: CSWS V2 and ODS5 0 Message-ID: <btlnps$dfq11@doiweb4.volkswagen.de>   Rudolf Wingert wrote:  > Hello, > N > did Im read right, that you need ODS5 for CSWS V2? We dont like to use ODS5R > disks. We love ODS2 with its lovely filename restriction. Esspacialy that all isE > uppercase. If I do not have ODS5, am I not able to install CSWS V2?  >  > Best regards R. Wingert  > B The installation guides explicitly states: you need an ODS-5 disk!   --    + mit freundlichen Gren | with best regards   3 Karl Rohwedder          | it-ingteam(at)t-online.de A                          | extern.karl.rohwedder(at)volkswagen.de    ------------------------------   Date: 9 Jan 2004 04:34:15 -0800 . From: fabiopenvms@yahoo.com.br (Fabio Cardoso)= Subject: Re: DCPS and LPD and Xerox Document Centre printers? = Message-ID: <f30679fb.0401090434.6d6e19cc@posting.google.com>   g Paul Anderson <paul.anderson@hp.com> wrote in message news:<080120041542584263%paul.anderson@hp.com>... ; > In article <btka5u$uli$1@news01.intel.com>, Ken Fairfield ! > <My.Full.Name@intel.com> wrote:  > D > > Paul, do you have a rough estimate when V2.4 might be far enough > > along to field test? > E > We have scheduled release for July, so if we stay on track, a field64 > test might be available between now and then.  :-) >  > April?  May? > @ > We're working on completing the Itanium version of V2.3 first. >  > Paul    / Any feature to print / generate PDF files ? :-)-   Regards    FC   ------------------------------  # Date: Fri, 09 Jan 2004 17:04:06 GMTe* From: Paul Anderson <paul.anderson@hp.com>= Subject: Re: DCPS and LPD and Xerox Document Centre printers?S5 Message-ID: <090120041204452684%paul.anderson@hp.com>a  C In article <f30679fb.0401090434.6d6e19cc@posting.google.com>, Fabio ) Cardoso <fabiopenvms@yahoo.com.br> wrote:x  1 > Any feature to print / generate PDF files ? :-)x  D A PDF-to-PostScript translator is on the wish list, but not for DCPS? V2.4.  It's more likely we will be able to send PDF files to HPr% printers that have PDF functionality.o   Paul   -- ?  Paul Anderson   OpenVMS Engineeringa   Hewlett-Packard Companyl   ------------------------------  $ Date: Fri, 9 Jan 2004 09:33:20 -0800# From: "Tom Linden" <tom@kednos.com>T= Subject: RE: DCPS and LPD and Xerox Document Centre printers? 9 Message-ID: <NDEMLKKEBOIFBMJLCECIOEBGCKAA.tom@kednos.com>v  F Why reinvent the wheel, make it possible to use Ghostscript as a print# filter, one less thing to maintain.n     -----Original Message-----3   From: Paul Anderson [mailto:paul.anderson@hp.com] (   Sent: Friday, January 09, 2004 9:04 AM   To: Info-VAX@Mvb.Saic.Com ?   Subject: Re: DCPS and LPD and Xerox Document Centre printers?h      E   In article <f30679fb.0401090434.6d6e19cc@posting.google.com>, Fabio +   Cardoso <fabiopenvms@yahoo.com.br> wrote:e   3   > Any feature to print / generate PDF files ? :-)e   F   A PDF-to-PostScript translator is on the wish list, but not for DCPSA   V2.4.  It's more likely we will be able to send PDF files to HPi'   printers that have PDF functionality.m      Paul      -- E    Paul Anderson     OpenVMS Engineeringm     Hewlett-Packard Companys      --- (   Incoming mail is certified Virus Free.<   Checked by AVG anti-virus system (http://www.grisoft.com).A   Version: 6.0.560 / Virus Database: 352 - Release Date: 1/8/2004a    --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).? Version: 6.0.560 / Virus Database: 352 - Release Date: 1/8/2004o   ------------------------------  # Date: Fri, 09 Jan 2004 18:27:11 GMTe* From: Paul Anderson <paul.anderson@hp.com>= Subject: Re: DCPS and LPD and Xerox Document Centre printers? 5 Message-ID: <090120041327511848%paul.anderson@hp.com>   D In article <NDEMLKKEBOIFBMJLCECIOEBGCKAA.tom@kednos.com>, Tom Linden <tom@kednos.com> wrote:g  B > Why reinvent the wheel, make it possible to use Ghostscript as a+ > print filter, one less thing to maintain.Y  B You're right.  We wouldn't necessarily write a new translator, but0 might incorporate one already written into DCPS.   Paul   -- -  Paul Anderson   OpenVMS Engineering-   Hewlett-Packard Company5   ------------------------------  # Date: Fri, 09 Jan 2004 13:24:16 GMT12 From: "Ken Farmer" <KFarmer@NOSPAM.SpyderByte.com> Subject: Re: ds10l blowout? Message-ID: <4YxLb.227883$dl.11373901@twister.southeast.rr.com>o  > "Keith Parris" <keithparris_NOSPAM@yahoo.com> wrote in message6 news:cf15391e.0401081626.b50fd16@posting.google.com...3 > jordan@ccs4vms.com (Rich Jordan) wrote in message 9 news:<cc5619f2.0401081152.7f6bef7e@posting.google.com>...tD > > If only DS10Ls had two PCI slots... one just won't cut it.  NeedJ > > decent graphics _and_ either KZPCA or KZPEA SCSI for it to replace any) > > existing EV5x box we have in service.- >-E > There's a Storage/Network/Video Combo card in the Supported Optionsr' > List for the DS10L.  Would this work: H > 3X-DEPVZ-AA  PCI TO 10/100MBPS EHTERNET, GRAPHICS,LVD SCSI COMBO ADPTR >oB > Graphics capability described as "2D Video with 4 MB of graphicsG > memory (based on 3Dlabs Permedia2V video ASIC), similar to the Compaqm@ > SN-PBXGK-BB".  Does this qualify as "decent graphics" for your > purposes?i  7 I have that card in 6 DS10L's.  They work like a charm..   -- Kenneth Farmer  <><a    EnterpriseUnix.org  |  Tru64.org OpenVMS.org  |  dcl.OpenVMS.orgu$ EnterpriseLinux.org  |  LinuxHPC.org   ------------------------------  $ Date: Fri, 9 Jan 2004 14:51:57 +0100- From: "Martin Vorlaender" <mv@pdv-systeme.de>a Subject: Re: ds10l blowout8 Message-ID: <btmbm0$8hoa0$1@ID-56200.news.uni-berlin.de>   Keith Parris wrote:i, > jordan@ccs4vms.com (Rich Jordan) wrote... C >> If only DS10Ls had two PCI slots... one just won't cut it.  NeedsE >> decent graphics _and_ either KZPCA or KZPEA SCSI for it to replace , >> any existing EV5x box we have in service. > E > There's a Storage/Network/Video Combo card in the Supported Optionsh' > List for the DS10L.  Would this work:cH > 3X-DEPVZ-AA  PCI TO 10/100MBPS EHTERNET, GRAPHICS,LVD SCSI COMBO ADPTR  @ If I'm informed correctly, the list price of that card is higher+ than the price David wants for the DS10L...p   cu,t   Martin -- )F   OpenVMS:                | Martin Vorlaender  |  VMS & WNT programmer3    The operating system   | work: mv@pdv-systeme.detF    God runs the           |   http://www.pdv-systeme.de/users/martinv/:    earth simulation on.   | home: martin@radiogaga.harz.de   ------------------------------   Date: 9 Jan 2004 05:55:34 -0800p1 From: matthew.finbow@btinternet.com (Matt Finbow)  Subject: Re: ds10l blowout= Message-ID: <ec25d2bf.0401090555.71b3f9e5@posting.google.com>   _ Wayne Sewell <wayne@tachysoft.com> wrote in message news:<00A2B998.D5F63C48.1@tachysoft.com>...u) > >From: jordan@ccs4vms.com (Rich Jordan)S > >X-Newsgroups: comp.os.vms > >Subject: Re: ds10l blowoutn" > >Date: 8 Jan 2004 11:52:46 -0800 > >lC > >If only DS10Ls had two PCI slots... one just won't cut it.  NeedtI > >decent graphics _and_ either KZPCA or KZPEA SCSI for it to replace anyh( > >existing EV5x box we have in service. > Q > Just once I'd like to find a DS10L (on ebay or wherever) with a scsi controller " > and disk instead of dumbass IDE.Q > ===============================================================================oP > Wayne Sewell, Tachyon Software Consulting  (281)812-0738   wayne@tachysoft.com= > http://www.tachysoft.com/www/tachyon.html and wayne.html   uQ > =============================================================================== D > Jed Clampett, checking into hotel: "This place got a cement pond?". > 	Ellie May: "And do yuh let critters in it?"  B SCSI controllers can be bought cheap second-hand - a plain Adaptec< 39160 card should be recognised as KZPEA (dual channel U160)  ; One caution - there at at least three variants of the 39160  KZPEA (HP) - Works (obviously) Adaptec 39160 - Should workaE Compaq OEM 39160 - Doesnt work (at least the two I've got won't work)i   Matt Finbow.   ------------------------------   Date: 9 Jan 2004 04:50:45 -0800s. From: fabiopenvms@yahoo.com.br (Fabio Cardoso)Y Subject: Hewlett-Packard plans to EOL the HP AlphaServer Tru64. Find out how Sun can helpi< Message-ID: <f30679fb.0401090450.f7e915a@posting.google.com>  C Hewlett-Packard plans to EOL the HP AlphaServer Tru64. Find out howt* Sun can help you migrate to a new platform    2 http://www.sun.com/solutions/infrastructure/tru64/  : Helping Organizations Move to an Industry-Leading Platform  A "Sun worked with us every step of the way to deliver a successful E migration. They helped with information and support, and were here as   friends as well as consultants."  
 Ken Gordon4 Strategic Operations Division Manager, IT Department City of Oakland   D Many enterprises committed to the Tru64/AlphaServer system are facedE with the reality that they will need to migrate to another platform.  C While some companies promise alternatives in the future, Sun offersn> migration solutions for companies who want to reduce costs and0 increase the quality of business services-today.  < The Sun Infrastructure Solution for Tru64 Migration offers aB comprehensive, integrated package of lifecycle services, developer@ migration programs, hardware, software, financial offerings, andC partner skills to ease the transition to Sun systems.  With the Sun C Infrastructure Solution for Tru64 Migration, Sun can provide a new,o? secure IT infrastructure that can accelerate development of new C applications to help meet evolving business needs, while helping toVB maintain current business functionality and protect investments in business applications.   ------------------------------  * Date: Fri, 9 Jan 2004 14:15:54 +0000 (UTC)* From: Osmo Kujala <kujala@tukki.cc.jyu.fi>9 Subject: Hints to upgrade DS10L ? (Was Re: ds10l blowout)y, Message-ID: <btmd2q$pu6$1@mordred.cc.jyu.fi>  ' Rich Jordan <jordan@ccs4vms.com> wrote:tB > If only DS10Ls had two PCI slots... one just won't cut it.  NeedH > decent graphics _and_ either KZPCA or KZPEA SCSI for it to replace any' > existing EV5x box we have in service.l  F Been there. I tried to use Adaptec 39160 SCSI card (should be equal toH KZPEA). For some reason SRM won't show the drives. I can run the bios ofI 39160 and it sees disks all right. SRM shows both channels of 39160 fine,ND but no disks. I don't care much of graphics but when I couldn't bootD from 39160, another SCSI-card would have been handy (OK something to use IDE for then).  J I thought I could use temporarily higher PCI-riser. But do you anyone knowI if the PCI-riser of DS10 is only possibility? I don't happen to own such.   F Another "problem" is the memory. Is there any source of 200-pin SDRAMsK at PC prices. I know that Island has good low-profile model(can be used in i0 both banks), but even that is way too expensive.  J By the way, anyone can exlpain why there is a resistor in the cdrom/floppy. combo set, which doesn't do anything but heat?   Osmo   ------------------------------  $ Date: Fri, 9 Jan 2004 09:14:17 -0700, From: David D Miller <ddmiller@Raytheon.com>9 Subject: Hints to upgrade DS10L ? (Was Re: ds10l blowout)eD Message-ID: <OF070C8BE6.12E8C12E-ON07256E16.00590EEA@mck.us.ray.com>  H Have you tried to get memory from www.kahlon.com?  I've had good luck in	 the past.e   dave.m      ' Rich Jordan <jordan@ccs4vms.com> wrote:sB > If only DS10Ls had two PCI slots... one just won't cut it.  NeedH > decent graphics _and_ either KZPCA or KZPEA SCSI for it to replace any' > existing EV5x box we have in service.d  F Been there. I tried to use Adaptec 39160 SCSI card (should be equal toH KZPEA). For some reason SRM won't show the drives. I can run the bios ofI 39160 and it sees disks all right. SRM shows both channels of 39160 fine,1D but no disks. I don't care much of graphics but when I couldn't bootD from 39160, another SCSI-card would have been handy (OK something to use IDE for then).  J I thought I could use temporarily higher PCI-riser. But do you anyone knowI if the PCI-riser of DS10 is only possibility? I don't happen to own such.f  F Another "problem" is the memory. Is there any source of 200-pin SDRAMsJ at PC prices. I know that Island has good low-profile model(can be used in0 both banks), but even that is way too expensive.  J By the way, anyone can exlpain why there is a resistor in the cdrom/floppy. combo set, which doesn't do anything but heat?   Osmo   ------------------------------  % Date: Fri, 09 Jan 2004 09:22:07 -0700r% From: Dan O'Reilly <dano@process.com>a= Subject: Re: Hints to upgrade DS10L ? (Was Re: ds10l blowout)?B Message-ID: <6.0.0.22.2.20040109092144.020d9760@raptor.psccos.com>  + Me, too.  They're good people to deal with.E  + At 09:14 AM 1/9/2004, David D Miller wrote:   I >Have you tried to get memory from www.kahlon.com?  I've had good luck int
 >the past. >y >dave. >e( >Rich Jordan <jordan@ccs4vms.com> wrote:D > > If only DS10Ls had two PCI slots... one just won't cut it.  NeedJ > > decent graphics _and_ either KZPCA or KZPEA SCSI for it to replace any) > > existing EV5x box we have in service.s > G >Been there. I tried to use Adaptec 39160 SCSI card (should be equal tooI >KZPEA). For some reason SRM won't show the drives. I can run the bios ofnJ >39160 and it sees disks all right. SRM shows both channels of 39160 fine,E >but no disks. I don't care much of graphics but when I couldn't bootoE >from 39160, another SCSI-card would have been handy (OK something to  >use IDE for then).l >dK >I thought I could use temporarily higher PCI-riser. But do you anyone know-J >if the PCI-riser of DS10 is only possibility? I don't happen to own such. >TG >Another "problem" is the memory. Is there any source of 200-pin SDRAMsrK >at PC prices. I know that Island has good low-profile model(can be used inf1 >both banks), but even that is way too expensive.e >rK >By the way, anyone can exlpain why there is a resistor in the cdrom/floppyc/ >combo set, which doesn't do anything but heat?  >b >Osmol   ------J +-------------------------------+----------------------------------------+J | Dan O'Reilly                  |  "There are 10 types of people in this |J | Principal Engineer            |   world: those who understand binary   |J | Process Software              |   and those who don't."                |J | http://www.process.com        |                                        |J +-------------------------------+----------------------------------------+   ------------------------------  * Date: Fri, 9 Jan 2004 09:45:29 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)0 Subject: Re: how to flush a file opened by DCL ?$ Message-ID: <btlt7p$218$3@online.de>  1 > any idea of how to flush a file opened by DCL ?M   From another process:l  1    $ OPEN/READ/SHARE=WRITE logical-name file-namee    $ CLOSE logical-name    ------------------------------  # Date: Fri, 09 Jan 2004 06:52:35 GMT ( From: Alder <PGDEHMKOKIMD@spammotel.com>) Subject: Re: HP TCPIP SMTP Queues (again)n( Message-ID: <TcsLb.2557$zk.518@clgrps12>   JF Mezei wrote::  ' >>Relay                         : FALSEr' >>Local-Alias-Only              : FALSE  > ? > You may want the relay to be true as well as local_alias-only  > ' >>Nosey                         : FALSEC > K > You need to set the right logical for this. This will show you details ofS > messages coming in.l  G The docs suggest NOSEY will merely cause the the full Subject: line to tB be printed in the log file, but I'll try it, along with the RELAY I option, next time I restart SMTP.  Before I do that though, I'll wait to  I see if the log files you requested shed any more light on the issue (see   below).    >  >>Number of MX entries: 0a > - > I don't think that this is an error message8  ; I guess I was expecting it to count the MX entry I added...    >  >>#smtp_symbiont\917:out of ASTu >  > M > What we really need to see is the log entries for one specific message thato > you know will hang.g  C Not sure which log you are referring to here, but here is the hung A! message as displayed by VMS Mail:t   MAIL> 1t)      #1           8-JAN-2004 17:12:29.82 o8       From:   SMTP%"Terrence.Branscombe@gems8.gov.bc.ca"C To:     "'tbransco@erebus.homeip.net'" <tbransco@erebus.homeip.net>b CC:e Subj:   Sent at 5:11  0 Return-Path: Terrence.Branscombe@gems8.gov.bc.ca3 Received: from aspen.itsd.gov.bc.ca (142.32.11.114)e?      by szeged.erebus.homeip.net (V5.4-15, OpenVMS V7.3 Alpha);a)      Thu, 8 Jan 2004 17:12:03 -0800 (PST)eB Received: from pipe.gov.bc.ca (pipe.bcsc.gov.bc.ca [142.32.11.55])B      by aspen.itsd.gov.bc.ca (<snip>) with ESMTP id i091Bv5W017831E      for <tbransco@erebus.homeip.net>; Thu, 8 Jan 2004 17:11:58 -0800t: Received: by pipe.bcsc.gov.bc.ca with <snip> (5.5.2657.72)2      id <CJ3WKLWX>; Thu, 8 Jan 2004 17:11:57 -0800D Message-ID: <4A122BDCB6F1D34B912C800546F10F8A4BCA76@royal.gov.bc.ca>H From: "Branscombe, Terrence EM:EX" <Terrence.Branscombe@gems8.gov.bc.ca>? To: "'tbransco@erebus.homeip.net'" <tbransco@erebus.homeip.net>  Subject: Sent at 5:11g$ Date: Thu, 8 Jan 2004 17:11:53 -0800 MIME-Version: 1.0-- X-Mailer: Internet Mail Service (5.5.2657.72)m Content-Type: text/plain;c          charset="utf-8"   < snip message content > MAIL>    > M > And you will also want to set the receiver logging and provide the relevantmU > contents of the TCPIP$SMTP_RECV_RUN.LOG (one is created for each message received).4  4 Here's the receiver log trace for the above message:   $ TYPE TCPIP$SMTP_RECV_RUN.LOG
 $ Set NoOn/ $ VERIFY = F$VERIFY(F$TRNLNM("SYLOGIN_VERIFY"))@9    TCPIP$SMTP   job terminated at  8-JAN-2004 17:12:29.28t      Accounting information:F    Buffered I/O count:          241      Peak working set size:   8528F    Direct I/O count:             98      Peak virtual size:     188496F    Page faults:                 709      Mounted volumes:            0F    Charged CPU time:  0 00:00:00.95      Elapsed time:   0 00:00:30.09 $e  6 ...and the execution queue stuck on this same message:   $ TCPIP SHOW MAILn  1    SMTP Mail Queue Entry:   971  User: TCPIP$SMTPg      File:  F _SZEGED$DKA0:[SYS0.TCPIP$SMTP]04010817120394_TCPIP$SMTP-1ABF.TCPIP_SZE GED;1n      Status: Processings   $   F As expected, the associated SMTP control/message files are sitting in  the default SMTP directory:E  3 $ DIR/DATE SYS$SPECIFIC:[TCPIP$SMTP]*.TCPIP_SZEGED*l  # Directory SYS$SPECIFIC:[TCPIP$SMTP]n  - 04010817120394_TCPIP$SMTP-1ABF.TCPIP_SZEGED;1p-                        8-JAN-2004 17:12:03.94h2 04010817120394_TCPIP$SMTP-1ABF.TCPIP_SZEGED_TEXT;1-                        8-JAN-2004 17:12:28.62    Total of 2 files.w     Regards,   Alder    ------------------------------  % Date: Fri, 09 Jan 2004 04:02:11 -0500 * From: JF Mezei <jfmezei.spamnot@istop.com>) Subject: Re: HP TCPIP SMTP Queues (again)f) Message-ID: <3FFE6DE0.AFC2303E@istop.com>.   Alder wrote:D > Not sure which log you are referring to here, but here is the hung# > message as displayed by VMS Mail:s  K When a message is received, it generates 2 log "activity". The first one isoK the TCPIP$SMTP_RECV_RUN.LOG which usually contains nothing, unless you haved! the proper logical names enabled.e- ---------------------------------------------e $!   TCPIP$SMTP_RECV_TRACEI $!        Logs all messages received from and transmitted to remote SMTP rJ $!        clients. Used to trace the SMTP application layer protocol. The I $!        same conventions for logging nonprinting characters or control  H $!        characters are used. The logical name UCX$SMTP_PROTO_TRACE is  $!        obsolete.i $!' $!DEFINE/SYSTEM TCPIP$SMTP_RECV_TRACE 1  $! $!---------------- $! $!   TCPIP$SMTP_RECV_DEBUGG $!        Logs full diagnostics, similar to the TCPIP$SMTP_LOG_LEVEL 5  J $!        logical name. Obsoletes the logical name TCPIP$SMTP_PROTO_DEBUG. $!' $!DEFINE/SYSTEM TCPIP$SMTP_RECV_DEBUG 1c- ---------------------------------------------     6 > Here's the receiver log trace for the above message:  L Because you didn'T have the proper logging for the receiver process, nothingL was logged. (the received is poorly designed with one log file per message).  A Also, the really important portion to show is the contents of theeJ TCPIP$SMTP_LOGFILE.LOG that are relevant to that message. (or confirmationJ that the file contains absolutely no trace of that message being processed even with all logging enabled).e   ------------------------------  $ Date: Fri, 9 Jan 2004 09:59:06 -0800* From: "Alder" <MUNDDGNTDYTV@spammotel.com>) Subject: Re: HP TCPIP SMTP Queues (again)w+ Message-ID: <3ffeebea$1@obsidian.gov.bc.ca>i  < OK, RECV_DEBUG is now set.  Thanks for clearing that up, JF.  , Here's the logicals I now have set for SMTP:   $ SHOW LOGICAL/SYSTEM *SMTP*   (LNM$SYSTEM_TABLE)  -   "MAIL$PROTOCOL_SMTP" = "TCPIP$SMTP_MAILSHR"6"   "TCPIP$SMTP_8BITMIME_HACK" = "1"3   "TCPIP$SMTP_COMMON" = "SYS$SPECIFIC:[TCPIP$SMTP]"y   "TCPIP$SMTP_ENABLE" = ".1.."   "TCPIP$SMTP_LOG_LEVEL" = "5"   "TCPIP$SMTP_NOSEY" = "1".   "TCPIP$SMTP_POSTMASTER_ALIAS" = "Postmaster"   "TCPIP$SMTP_RECV_DEBUG" = "1"l+   "TCPIP$SMTP_RECV_SNAPSHOT_BLOCKS" = "200"u   "TCPIP$SMTP_RECV_TRACE" = "1"B&   "TCPIP$SMTP_SFF_REQUIRES_PRIV" = "1"+   "TCPIP$SMTP_SYMB_SNAPSHOT_BLOCKS" = "200"    "TCPIP$SMTP_SYMB_TRACE" = "1"l  F After clearing the mail queue, deleting all control and log files, andI restarting SMTP to use the new RECV_DEBUG logical, I sent myself an email I from an external client, as before.  The RECV log still shows nothing and(L the SYMB log is pretty much identical to the one I posted before.  That "out8 of AST" statement is looking more and more suspicious...  G Here's the logs as they appeared after the restart and the new incoming6 email arrived:   $ TYPE TCPIP$SMTP_RECV_RUN.LOG
 $ Set NoOn/ $ VERIFY = F$VERIFY(F$TRNLNM("SYLOGIN_VERIFY"))08   TCPIP$SMTP   job terminated at  9-JAN-2004 08:51:20.33     Accounting information:oI   Buffered I/O count:          241      Peak working set size:       8768rI   Direct I/O count:            100      Peak virtual size:         188496fI   Page faults:                 745      Mounted volumes:                0rI   Charged CPU time:  0 00:00:00.97      Elapsed time:       0 00:00:30.08t   $ type TCPIP$SMTP_LOGFILE.LOGe  I Dump of log snapshot buffer follows (smtp$symb_condh dumping log snapshotl buffer)a Snapshot buffer is empty End Dump of log snapshot buffer6  D %%%%%%%%%%%%                    9-JAN-2004 08:49:28.13  %%%%%%%%%%%%$ %TCPIP-I-SMTP_LOGSUC, using log file/ SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_LOGFILE.LOG   I Dump of log snapshot buffer follows (smtp$symb_condh dumping log snapshotr buffer) " #smtp_symbiont\465:SMB_DEBUG=65535   TCPIP SMTP configuration data:& Server-Nodes                  : SZEGED4 Queue-Name                    : TCPIP$SMTP_SZEGED_00 Alternate-Gateway             :m General-Gateway               :  Substitute-Domain             :  Zone                          :r* Postmaster-Alias              : Postmaster& Postmaster-Forwards-To        : SYSTEM Foreign-Transport-Synonyms    : - Initial-Interval              : 0 00:30:00.00t- Retry-Interval                : 0 01:00:00.00 - Retry-Maximum                 : 3 00:00:00.00M! Receive-Timeout               : 5o" Retry-Address                 : 16" Hop-Count                     : 16! Symbiont-Snapshot-Blocks      : 0a! Receiver-Snapshot-Blocks      : 0u! Utilities-Snapshot-Blocks     : 0o! Send-Timeout-Init             : 5y! Send-Timeout-Mail             : 5o! Send-Timeout-Rcpt             : 5s! Send-Timeout-Data             : 3i" Send-Timeout-Term             : 10! Log-Level                     : 5t! Receiver-Debug                : 0w! Receiver-Trace                : 0n! Symbiont-Debug                : 0-! Symbiont-Trace                : 1 ! Utilities-Debug               : 0R! Utilities-Trace               : 0:! EF-Debug-Level                : 0 ! Channel-Debug-Level           : 0e# Header-Placement              : TOPr% Eightbit                      : FALSEd% Relay                         : FALSEl% Altgate-Always                : FALSEr% Mx-If-Noaltgate               : FALSE % No-Mx                         : FALSEf% No-Subs-Domain-Inbound        : FALSE $ Smtp-Jacket-Local             : TRUE$ Cent-Sign-Hack                : TRUE$ Nosey                         : TRUE% Log-Line-Numbers              : FALSEd% Memory-Debug                  : FALSEp% Mail$Protocol-Debug           : FALSE % CF-Debug                      : FALSEn% Parse-Debug                   : FALSEs% Deliver-VMS-Def-To            : FALSE % Deliver-NoXVMS                : FALSEC% MTS-From-Hack                 : FALSEi% Rewrite-MTS-From              : FALSE % Local-Alias-Only              : FALSEt% Relay-Based-On-Mx             : FALSE % Reject-Unbacktranslatable-IP  : FALSEn% Accept-Unqualified-Senders    : FALSEd% Accept-Unresolvable-Domains   : FALSE $ SFF-Requires-Priv             : TRUE$ 8BitMIME-Hack                 : TRUE% Suppress-Version-Info         : FALSEu  " Other TCPIP SMTP environment data:* SMTP Software Username        : TCPIP$SMTP9 SMTP Software Default Director: SYS$SPECIFIC:[TCPIP$SMTP]O Symbiont Log File             :h/ SYS$SPECIFIC:[TCPIP$SMTP]TCPIP$SMTP_LOGFILE.LOGS   List of (5) local aliases: EREBUS.HOMEIP.NET 	 localhosto localhost.erebus.homeip.net  SZEGED szeged.erebus.homeip.net  C Ordered list of gateways for relaying mail to alternate gateway ():v Number of MX entries: 0e- Expiration Date:      17-NOV-1858 00:00:00.00e= #smtp_symbiont\503:SYS$SCRATCH (On entry to symbiont)  st:444m5 #smtp_symbiont\1479 :*******Process Statistics*******g SYS$OUTPUT = _NLA0: 9 SYS$INPUT = _NLA0:QMAN00000055_00000062.SYS$QUEUE_MANAGERp SYS$ERROR = _NLA0:+ accnt <start>   astcnt 10240    astlm 10240  authpri 4       biolm 600e) filcnt 193      fillm 200       jobtype 0p/ owner 0 uaf 0   uic 65540       username SYSTEMk: curpriv=203015E081 imagpriv=0000000000 procpriv=203015E081 authpriv=2000004001 < #smtp_symbiont\1510 :*******End of Process Statistics*******8 #smtp_symbiont\587:SYS$SCRATCH SYS$SPECIFIC:[TCPIP$SMTP]. #smtp_symbiont\591:DBG$OUTPUT SYS$OUTPUT: st:1, #smtp_symbiont\596:DBG$INPUT SYS$INPUT: st:1$ #smtp_symbiont\601:SYS$LOGIN  st:444F BUILD8$:[TCPIP_V54_BL15.SRC.SMTP]SMTP_SYMBIONT.C;1: Symbiont activatedD #smtp_symbiont\623:Waiting for the job controller to start the queue #smtp_symbiont\747:ast_handler$ #smtp_symbiont\753:read_message st:1  #smtp_symbiont\781: start_stream! #smtp_symbiont\967:unpack_messageq* #smtp_symbiont\975:item 9 status 1 size 20+ #smtp_symbiont\975:item 12 status 1 size 20i- SMBMSG$K_EXECUTOR_QUEUE: TCPIP$SMTP_SZEGED_01a+ #smtp_symbiont\975:item 29 status 1 size 25o' #smtp_symbiont\1363:smtp$send_to_jobctlV+ #smtp_symbiont\1386:smb$send_to_jobctl st:1s$ #smtp_symbiont\1187:smtp$init_stream# #smtp_symbiont\1207:sys$setprn st:1  #smtp_symbiont\917:out of ASTl+ #smtp_symbiont\629: Symbiont is initializedtC smtp$symb condh entered with condition code. %TCPIP-I-SMTP_SYMBRUN,f! symbiont is running the queue !ASe End Dump of log snapshot bufferd  D %%%%%%%%%%%%                    9-JAN-2004 08:49:39.85  %%%%%%%%%%%%I %TCPIP-I-SMTP_SYMBRUN, symbiont is running the queue TCPIP$SMTP_SZEGED_01C   ------------------------------  % Date: Sat, 10 Jan 2004 00:55:19 +1300-1 From: Tux Wonder-Dog <wes.parish@paradise.net.nz>L. Subject: Re: I just scored an AlphaServer 800!4 Message-ID: <mIwLb.16506$ws.1897741@news02.tsnz.net>   Bob Koehler wrote:  ? > In article <ee336853.0401080543.3bfb950a@posting.google.com>,w, > stevekulpa@yahoo.com (steve kulpa) writes: >> xA >> Now the sad part - I need software.  I remember looking at the4H >> hobbyist program a while back and when I checked into it again, I nowE >> see that you have to join DECUS (or Encompass, whatever).  I was a H >> DECUS member for years in the past and never remember having to pay aI >> fee.  I see where Encompass requires a $90/year fee, plus the hobbyist  >> license fees on top of that.e > G >    See the FAQ.  You can be an associate member, it's free.  There is'K >    no hobbyist license fee.  You can, if you need it, purchase a hobbyist I >    CD which has the OS and popular layered products, or you can get theeJ >    real distribution elsewhere.  Like many hobbyists, I borrow mine from >    someone else.  K I've been thinking much the same sort of thing.  I'm a registered encompasstD associate in my own region, but unfortunately I haven't been given aH registration number, which is what the Hobbyist License requires - or it did the last time I looked.e  L Do you think that the American encompass outfit would take me on?  And would( that be suitable for a Hobbyist License?  
 Wesley Parisho -- eG "Good, late in to more rewarding well."  "Well, you tonight.  And I was K lookintelligent woman of Ming home.  I trust you with a tender silence."  I C get a word into my hands, a different and unbelike, probably - 'shel= fortunate fat woman', wrong word.  I think to me, I justupid.aG Let not emacs meta-X dissociate-press write your romantic dialogs...!!!>   ------------------------------  % Date: Fri, 09 Jan 2004 09:06:19 -06004( From: brandon@dalsemi.com (John Brandon)$ Subject: Multnet Documentation V4.0?1 Message-ID: <04010909061946@dscis6-0.dalsemi.com>i   Anyone using Multinet V4.0?b  ) And where could I find the documentation?   / Is the current version (V4.4 or V4.3) similiar?    TIA!     J*o*h*n B*r*a*n*d*o*ns VMS Systems Administrator * firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  % Date: Fri, 09 Jan 2004 10:07:54 +0100e From: jf.pieronne@laposte.neth% Subject: Re: MySQL 4.0.17 for OpenVMSa2 Message-ID: <btls50$g9i$2@news-reader1.wanadoo.fr>   Rich Jordan wrote:    > jf.pieronne@laposte.net wrote: > 3 >> MySQL 4.0.17 is the latest MySQL stable version.r >>5 >> The kit also include the client shareable library.O >>H >> Download from http://www.pi-net.dyndns.org/anonymous/kits/ or any of  >> the two mirrors.@ >> >> Jean-Franois >  > E > Thank you kindly!  Only I haven't had time to install 4.0.16 yet...h > ( > I think you may be porting too fast ;) >    :-)e  J I have a nearly completely automatic porting process for MySQL, Python,...K So it is, generally, very easy to port a new version of one of these tools.   
 > Rich Jordann >     
 Jean-Franoisy   ------------------------------  % Date: Fri, 09 Jan 2004 09:22:26 +0100o9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> < Subject: Re: Non-WS Process Quotas not in Performance Manual' Message-ID: <3FFE64C1.D1CA07AD@aaa.com>   u > spamsink2001@yahoo.com (Alan E. Feldman) wrote in message news:<b096a4ee.0401081323.2bbf489b@posting.google.com>...BA > > Why aren't process quotas like DIOlm, BIOlm, ASTlm, etc., notu( > > discussed in the performance manual? > >   ; Perhaps becuse that are not *performance* specific quotas ?49 They are rather process *limits*, either there are enought; of them, or there isn't. Other quotas (WS and others) has a 3 more "soft" limit and are more performance related.   	 Jan-Erik.    ------------------------------  * Date: Fri, 9 Jan 2004 09:57:36 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)5 Subject: RE: number of ethernet adapters in a clustere$ Message-ID: <btltuf$218$4@online.de>  
 In articleG <FD827B33AB0D9C4E92EACEEFEE2BA2FB1E1321@tayexc19.americas.cpqcorp.net>,a* "Main, Kerry" <kerry.main@hp.com> writes:   = > > Again, this may or may not be an option for Phillip whoseo4 > > hardware is quite old, e.g., VAXstation 4000/60!  H > Tony, My apologies - I thought he was talking about a work environment > ..  B Just to make things clear, while I do have a VMS "day job" (at theC German stock exchange (lots and lots of VMS there, fortunately)), ItB don't do any hardware stuff there and relatively little VMS systemH management etc (I'm mainly concerned with Rdb databases and applicationsG which interact with them).  My questions in this forum are thus almost 3F always to do with my hobbyist cluster, which as Tony mentions runs on C rather old hardware on a shoestring budget, even if I don't always e mention that explicitly.  C (After I left academia for greener pasture$, and before I had a DSLoF connection at home (and thus no way to conveniently and cheaply accessH an NNTP server from my cluster at home, which at the time had a dial-outB and dial-in on demand ISDN connection), I accessed comp.os.vms viaH Info-VAX from my email address at work.  Since I prefer NNTP access and D now with flat-rate DSL neither cost nor speed is an issue, I am now 5 happily using NEWSRDR to follow the group from home.)    ------------------------------  $ Date: Fri, 9 Jan 2004 11:53:42 -0500< From: "Peter Weaver" <WeaverConsultingServices@sympatico.ca># Subject: Re: OpenVMS Certificationss9 Message-ID: <btmman$91ksi$1@ID-141708.news.uni-berlin.de>a   Keith Parris wrote:t >...> > DECUS.  The certificates look nice on my cubicle wall, and I also: > enjoyed getting the CONDIST/CONOLD for free.  Most folks
 here could >...  7 That's the second person in this thread to say they get = CONDIST through their certifications. All Chris and I get are-7 "Compaq Support Software CD" for Desktops, Workstation, = Notebooks and Handhelds. And I have not even received that in: the last year or so.   -- o Peter Weaver Weaver Consulting Services Inc.. Canadian VAR for CHARON-VAXo www.weaverconsulting.cao   ------------------------------  % Date: Fri, 09 Jan 2004 12:35:57 -0500e& From: David M Smith <dsmit115@csc.com># Subject: Re: OpenVMS Certifications 8 Message-ID: <adptvvg8un4tduvddek4g9rkr4loffp036@4ax.com>  1 On Fri, 9 Jan 2004 11:53:42 -0500, "Peter Weaver" . <WeaverConsultingServices@sympatico.ca> wrote:  8 >That's the second person in this thread to say they get> >CONDIST through their certifications. All Chris and I get are8 >"Compaq Support Software CD" for Desktops, Workstation,> >Notebooks and Handhelds. And I have not even received that in >the last year or so.a  P I received my certifications in January, 2003. I've received 3 sets of SPL / ODLN for OpenVMS Alpha (only, no VAX): Q1, Q2 and Q3 2003. These (at least the onesP I've received) have a different part number and name (e.g., OVMS ALPHA LP ASE Q3P 2003) than the ones I receive at work with my company, and are shipped about 1-2O months after the ones to work (so, I'll look for the Q4 SPL / ODL in a month ore so).  M I also received a couple sets of "Compaq PC" CDs but haven't even looked them 6 over to see what they are, since I was not interested.I ------------------------------------------------------------------------- I David M. Smith 302.391.8533                       dsmit115 at csc dot com-I Computer Sciences Corporation     (Opinions are those of the writer only)eI -------------------------------------------------------------------------    ------------------------------   Date: 9 Jan 2004 05:06:59 -0800n. From: fabiopenvms@yahoo.com.br (Fabio Cardoso)Y Subject: Re: Oracle Rdb on GS1280 with 7.3-2 exceeds 1 million transactions per minute usm= Message-ID: <f30679fb.0401090506.5766cac2@posting.google.com>   v keithparris_NOSPAM@yahoo.com (Keith Parris) wrote in message news:<cf15391e.0401081508.2b02c5ac@posting.google.com>...G > HP OpenVMS operating system and Oracle Rdb software break performancea	 > barrierd > by Marc Courchesne > F > Recent performance tests of Oracle Rdb version 7.1.2 and HP OpenVMS? > version 7.3-2 running on an HP AlphaServer GS1280 system have @ > demonstrated breakthrough transaction performance for a singleF > 32-processor symmetric multiprocessing (SMP) machine configured withE > 256 gigabytes of main memory. The AlphaServer system, together withoA > Oracle Rdb software, achieved sustained throughput of 1,010,160cD > database transactions per minute. In this test, five tables with aF > total of one billion (109) rows were created and their contents wereH > mapped in main memory using Oracle Rdb's Row Cache feature. This test,B > designed to explore the limits of the Row Cache feature, was notH > highly optimized for performance, but additional performance tuning is= > likely to result in even higher transaction rates. This newjD > implementation of Row Cache, first available in Oracle Rdb versionB > 7.1.2, allows customers to fully utilize all the physical memoryG > present in the largest AlphaServer systems for caching databases thatgG > retain complete transaction integrity and full recovery from process, E > disk, or system failure. The OpenVMS version 7.3-2 operating systemtH > provides a number of key features for improved scheduling and resourceC > utilization that enhance scaling on systems with many processors.n2 > These features were fully utilized in this test. > + > For more information, visit HP's site at M7 > http://h71000.www7.hp.com/solutions/oracle/index.htmlu1 > and Oracle's site at http://www.oracle.com/rdb/$ > ---3B > Note: This test does not reflect tpm-C or tpm-H results. --Keith    @ May be its time to Re-bundle Oracle RDB under OVMS ! Even as an@ Oracle-HP agreement.... may be with some kind of activation key.> As RDB is exclusive for OVMS, there is no reason to mantain it? unbundled...Would be good for Sybase / Ingres customers just toe> port their old database/applications to ORDB. And ORacle could justify RDB investments...   Regardse   FC   ------------------------------  % Date: Fri, 09 Jan 2004 11:03:24 +0100r" From: labadie <labadie_g@decus.fr> Subject: Re: Python 2.3.3P2 Message-ID: <btlul9$3mh$1@news-reader3.wanadoo.fr>   jf.pieronne@laposte.net wrote:   > & > Two very good introduction to PythonH > http://www.ferg.org/python_presentations/python_intro_for_managers.ppt. > http://www.ferg.org/python_slides/index.html<   May be I should be a manager, I even understood the first , presentation, which is very good indeed  :-)   Gerard   ------------------------------   Date: 9 Jan 2004 07:08:11 -0600M; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)-J Subject: Re: Searchlist behaviour differences between DCL/RMS and CRTL/EXE3 Message-ID: <ccltCcihcMuR@eisner.encompasserve.org>o  k In article <8emLb.24982$Tz1.16834@news.chello.at>, peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) writes:h > L > Which parsing does the program ? It is opening a file entered as parameterM > and it is opening it in the wrong directory. How do I change ORACLE SQL.EXE F > to open its outputfile in the correct directory of the search list ?  +    You beg racle to fix it.  Then you wait.h   ------------------------------   Date: 9 Jan 2004 07:55:29 GMT02 From: "Dave Weatherall" <djw-nothere@nospam.nohow>A Subject: Re: Secure Web Server Version 2.0 available for downloadh5 Message-ID: <DTiotGxQ0bj6-pn2-dx1O9Lolb9Lc@localhost>a  E On Thu, 8 Jan 2004 15:46:16 UTC, "Rick Barry" <richard.barry@hp.com> r wrote:  
 > Hi Mark, > N > As a few posters already guessed, the stream-lf requirement for served filesM > was to minimize the portability (and maintenance) effort and to address thea@ > problem of quickly determining the number of data bytes in theM > Content-Length calculation (without having to read every byte in the file).d > L > This requirement applies mainly to textual content, such as .HTML, .SHTML,M > and .TXT files, that are typically represented on VMS using variable-lengthhI > record format. Binary data files, such as executables, are not an issue:8 > because fixed-length record format is also acceptable. >  > Rick Barry > OpenVMS System Software Group  > Hewlett-Packard Companyo > Nashua, NH   Hm.   ? I'm at home and can't check my RMS manual but I don't remember  C anything about Stream_LF files that make it easier to determine or iC estimate byte-wise file-length over and above normal fixed length. n What am I missing?  F Whatever happened to VMS dedication to backwards compatibility? Do youD really think it a good idea to expect customers to suddenly have to > convert all the ascii/text files on their sites to be able to B 'upgrade' to CSWS V2? Not to mention changing any file-generation D procedures thay have in place, pluse any testing/qualification this  might involve?   1 Was this a decision really made in the VMS group?    --   Cheers - Dave.   ------------------------------  % Date: Fri, 09 Jan 2004 04:15:04 -05006* From: JF Mezei <jfmezei.spamnot@istop.com>A Subject: Re: Secure Web Server Version 2.0 available for downloadh) Message-ID: <3FFE70E5.E996A3F8@istop.com>d   Dave Weatherall wrote:@ > I'm at home and can't check my RMS manual but I don't rememberD > anything about Stream_LF files that make it easier to determine orD > estimate byte-wise file-length over and above normal fixed length.  ? ((number of blocks - 1) * 512) + offset of EOF into last block.i  H Stream-LF is really a "binary" file. The only difference is that the RMS1 attributes allow an application to see "records".l  L With variable length records, an application does not see the same number of  bytes as are stored in the file.  J Consider that a lot of modern applications require file length, and randomK access to a specific location into a file.  Even older applicatiosn such ast Xmodem, Ymodem require this.  N It is time for VMS to truly support the storage of file length into a file. ItM costs a lot less to "pay" for this when you write it once, that when you ready, the file thousands of times during the day .  N Perhaps for variable length files, RMS could maintain the number of data butesM stored as well as the number of records. Then, an application could calculate N the file size by adding to the total data bytes, the number of bytes that willL be used by the rcord delimiter (number of records * number of bytes for each record delimiter).   ------------------------------  % Date: Fri, 09 Jan 2004 10:05:16 +0100o From: jf.pieronne@laposte.net A Subject: Re: Secure Web Server Version 2.0 available for downloadS2 Message-ID: <btls4v$g9i$1@news-reader1.wanadoo.fr>   JF Mezei wrote:, > Mark Berryman wrote: > J >>with the V1.3 version of the server.  Are you telling me that, with thisA >>latest "upgrade", I am no longer going to be able to use the HPcA >>supported web-server just so the port could be a little easier?a >  > O > You could use the WASD or OSU Web servers. The later was written specificallyT@ > for VMS. Or just stick to your old version of VMS0Apache untilL > Digital/Compaq"HP/whatever realise the mistake in dropping support for VMS > native files.J  D No it's the opposite, WASD was written specifically for VMS, it use / extensively AST and many specific VMS features. B OSU is thread based and if I remember correctly run also on Linux.   ------------------------------  # Date: Fri, 09 Jan 2004 15:49:03 GMTS) From: "Rick Barry" <richard.barry@hp.com>oA Subject: Re: Secure Web Server Version 2.0 available for downloadn3 Message-ID: <P3ALb.11994$cj4.2526@news.cpqcorp.net>   K It's the record formats with metadata, such as variable-length format, thatoK cause problems determining actual data size of a file. Since Content-Length J is recommended by RFC 2616 for all content types, not just binary, we madeH the decision to limit support in this version to stream and fixed record formats.  > We understood that this would affect customers who already hadK variable-length files, which is why we provided the convert utility to help( in some of those situations.  L Thanks to all for taking the time to provide feedback. We'll consider addingK support for variable-length format files in the future, if we discover thathJ the conversion utility is does not solve the problem in many cases. But we  need more time to evaluate this.  
 Rick Barry OpenVMS System Software Groupr Hewlett-Packard Companyl
 Nashua, NH  7 "JF Mezei" <jfmezei.spamnot@istop.com> wrote in message.# news:3FFE70E5.E996A3F8@istop.com...o > Dave Weatherall wrote:B > > I'm at home and can't check my RMS manual but I don't rememberF > > anything about Stream_LF files that make it easier to determine orF > > estimate byte-wise file-length over and above normal fixed length. >iA > ((number of blocks - 1) * 512) + offset of EOF into last block.t >eJ > Stream-LF is really a "binary" file. The only difference is that the RMS3 > attributes allow an application to see "records".o >fK > With variable length records, an application does not see the same number  of" > bytes as are stored in the file. > L > Consider that a lot of modern applications require file length, and randomJ > access to a specific location into a file.  Even older applicatiosn such as > Xmodem, Ymodem require this. > G > It is time for VMS to truly support the storage of file length into af file. ItJ > costs a lot less to "pay" for this when you write it once, that when you read. > the file thousands of times during the day . >uJ > Perhaps for variable length files, RMS could maintain the number of data buteseE > stored as well as the number of records. Then, an application coulde	 calculatetK > the file size by adding to the total data bytes, the number of bytes that  willI > be used by the rcord delimiter (number of records * number of bytes fort each > record delimiter).   ------------------------------   Date: 9 JAN 2004 16:02:13 GMTt4 From: karcher@thuria.waisman.wisc.edu (Carl Karcher)A Subject: Re: Secure Web Server Version 2.0 available for downloado5 Message-ID: <9JAN04.16021354@thuria.waisman.wisc.edu>o  J In a previous article, "Dave Weatherall" <djw-nothere@nospam.nohow> wrote:  H ->Whatever happened to VMS dedication to backwards compatibility? Do youF ->really think it a good idea to expect customers to suddenly have to @ ->convert all the ascii/text files on their sites to be able to D ->'upgrade' to CSWS V2? Not to mention changing any file-generation E ->procedures that have in place, plus any testing/qualification this a ->might involve?  G I concur that this is a huge mistake and an unreasonable expectation ofrG customer effort. I was almost ready to ditch the OSU webserver and jumpa- into CSWS V2.0. Now this. This will require: e  G  1. Converting EVERY file in EVERY www directory on the system. This isBF     not a simple file header change either - every record will need to     be processed and rewritten.A  <  2. Changing every application/procedure that produces html.  H  3. Changing the default for EVERY Advanced Server (aka pathworks) shareF     on the system. STREAM (not STREAM_LF) is STILL the default even in/     the most recent version/eco (from October).   D Certainly, there needs to be a way to restore the previous behavior.   --G -- Carl Karcher, Waisman Computing Services, Waisman Center, UW-Madison-: --                   karcher.nomorespxm@waisman.wisc.edu     ------------------------------  $ Date: Fri, 9 Jan 2004 08:34:42 +0100( From: "Rudolf Wingert" <win@fom.fgan.de>E Subject: supported connection methods for $ @ SYS$SYSTEM:SHUTDOWN.COM : Message-ID: <MCELKPMOKPMNDNKJNIONEEKNCIAA.win@fom.fgan.de>   Hello,  O if you would like to shutdown a remote OpenVMS node from an OpenVMS node SYSMAN- is a supported way to do so.  
 	$ MCR SYSMANn) 	SYSMAN> SET ENVIRONMENT/NODE=remote_nodet 		....( 		Password: your system account password" 	SYSMAN> SHUTDOWN NODE /qualifiers  L Use the help capability of SYSMAN for all possible qualifiers. It works very well. I have done it often.d   Best regards R. Wingertp   ------------------------------   Date: 9 Jan 2004 07:10:31 -0600C; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) I Subject: Re: The Register: OpenVMS among most-secure of operating systemsu3 Message-ID: <UAinCE7DgRq1@eisner.encompasserve.org>   q In article <cf15391e.0401081721.3e9462d2@posting.google.com>, keithparris_NOSPAM@yahoo.com (Keith Parris) writes:  > G > You're trying to convince us that because you claim CERT, etc. reportiG > counts are not accurate, therefore the actual (unknown to you) number A > of VMS security problems discovered must be large (or, for youreB > purposes, at least larger than the reported number for Solaris).  H    He's also assuming that CERT is the only info we have.  Oh how I wishG    I could cite my other sources.  Unfortunately those sources restricth    thier info.   ------------------------------   Date: 9 Jan 2004 06:11:21 -0800s( From: bob@instantwhip.com (Bob Ceculski)I Subject: Re: The Register: OpenVMS among most-secure of operating systems = Message-ID: <d7791aa1.0401090611.2cc480e0@posting.google.com>s   Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<btk2br$a43$1@new-usenet.uk.sun.com>...g > Keith Parris wrote:y9 > > Software vulnerabilities still dog operating systems s- > > http://www.theinquirer.net/?article=13420  > > 2 > > "Proprietary systems are the least vulnerable  > > F > > The operating systems with fewest vulnerabilities in 2003 are HP's( > > OpenVMS, IBM's OS/400 and IBM's zOS. > > F > > These three are all proprietary and they all have security that isJ > > fully integrated, not applied as some kind of after-thought. CertainlyJ > > they come with a decent price-tag but they can be well worth the moneyI > > when the result is fewer security problems, less unscheduled downtimes$ > > and less downtime for patching." > > ...vI > > "the most secure operating systems continue to be certain proprietary0H > > systems from HP and IBM. Some may refer to these more secure systemsF > > as legacy systems but if legacy means secure and reliable it seems0 > > that legacy should be the preferred option." > A > Unfortunately its a pointless piece of research because OpenVMS'C > security advisories do not get reliably reported to CERT, BugtraqSA > etc so counting the ones that do only catches the excpetions to. > the rule.  > @ > You know this, I know this, its well documented so why did you= > bother posting a reference to the The Registers article its 4 > not very responsible behaviour on your part is it. > 	 > Regardsr > Andrew Harrisons  A and I have been on VMS for 19 years now and security mup kits arei> rare ... only 2 that I can remember ... that last security bug> found was decwindows, and how many people use decwindows?  UCXA has had some issues, with TCPware having even less, but as I have8A shown you, VMS squashes many tcpip bugs cold in their tracks with/B that nasty little "ACCESS VIOLATION ERROR" ... you sound desparate@ Andrew ... are your slowaris customer tired of being in the CERT of the week club?    ------------------------------   Date: 9 Jan 2004 06:16:00 -0800 ( From: bob@instantwhip.com (Bob Ceculski)I Subject: Re: The Register: OpenVMS among most-secure of operating systemsC= Message-ID: <d7791aa1.0401090616.7936c388@posting.google.com>S   Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<btk2br$a43$1@new-usenet.uk.sun.com>...p  @ "The other significant feature of these operating systems is theE language in which they are written. The two from IBM are both writtenID in assembler and OpenVMS uses a range of about ten languages, one of which is C.F  = C and similar languages that use pass-by-value techniques are C exceptionally prone to buffer overflow and the consequent potential D for unauthorized users to execute either their own malicious code orF other programs which run with enhanced access privileges. Avoiding theE use of these languages at the most vulnerable points, namely user I/OCE and network I/O, would appear to be wise. Linux, Unix and Windows are > almost entirely written in C, and most of their middleware andA application software is also in these vulnerable languages, so itmB should come as no surprise that they are less secure than OpenVMS, OS/400 and zOS.R  ? The other operating system that had very few vulnerabilities isrD Apple's OS 9, with the Secunia database showing just one in 2003 andB none in 2002. Again this is a proprietary operating system and theF decisions and integration of security rest with one organisation whichF does not have to concern itself with compatibility with other vendors.  D Apple recently moved to a Unix-based operating system, OS X, and the@ 24 vulnerabilities reported for it by Secunia in 2003 are a very telling comment."I   ------------------------------   Date: 9 Jan 2004 13:56:48 GMT ( From: bill@cs.uofs.edu (Bill Gunshannon)( Subject: Re: Vaxstation some question...9 Message-ID: <btmbv0$8mm01$1@ID-135708.news.uni-berlin.de>   9 In article <btkl30$8cim7$1@id-207001.news.uni-berlin.de>,R, 	Tony Arnold <tony.arnold@man.ac.uk> writes: > K > sysadmin on VMS is very different to that on Unix. Most sysadmin on Unix aH > is done by editing config files somewhere. On VMS, there is usually a I > utility with a command line interface for doing much of the qeuivalent .H > sysadmin work. For example, on Unix one edits /etc/passwd to create a E > user account (plus some other stuff). On VMS you run the AUTHORIZE n+ > utility and execute the ADD user command.o >   @ You know, considering how little people here actually know aboutB Unix you really should stop trying to compare it to VMS.  You justC come off looking silly.  While the possibility of editing the filesP@ (in this case, /etc/passwd) exists, command line utilities to doB this in a more reasonable way have pretty much always existed.  AsB far back as Ultrix-11 (which pre-dates VMS) on the PDP-11 this was" not only possible but recommended:   -----L   NAME.      nu - setup and manage user login accounts   SYNTAX      /etc/nu   DESCRIPTIONSA      The nu command is a program to help a system manager create,S?      modify, and delete accounts on that machine.  While every-O=      thing accomplished by nu can be done manually by editingl@      files and issuing shell commands, nu will guide you through@      getting all the details right, worrying about file locking,#      checking for typos, and so on.    ----  C And the same is true for most other tasks.  Unix offers tools to dooF all the usual stuff but usually also offers the ability to bypass themE if that is what the user/admin really wants.  Kind of like doing whatRF the user wants instead of dictating to the user how he will do things.  ? If you're going to continue to criticize Unix at least take theg- time to learn at least a little bit about it.    bill   -- dJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  # Date: Fri, 09 Jan 2004 14:23:45 GMTr! From: "DaviS*" <davis@spamsux.it>0( Subject: Re: Vaxstation some question...1 Message-ID: <RPyLb.81064$_P.3046277@news4.tin.it>u    Thanks...Thanks for your reply!! Good NG....fast and precise..p  I So, i know that VMS is different from Unix, and is the reason for installo OpenVMS and remove BSD!    Thanks again     Davis.   ------------------------------   Date: 9 Jan 2004 07:58:04 -0800h1 From: keithparris_NOSPAM@yahoo.com (Keith Parris)-( Subject: Re: Vaxstation some question...= Message-ID: <cf15391e.0401090758.4df7e1bd@posting.google.com>_  Z "DaviS*" <davis@spamsux.it> wrote in message news:<hZYKb.54382$VW.2385148@news3.tin.it>...$ > - Can i install OpenVMS from net??D >   I haven't external Cdrom, but i can "get" this unit from another > workstation.  F I'm not aware of any way to install OpenVMS from the Internet, but you? can buy a CD with binary media for VMS and many popular layered : products for US$30 through the OpenVMS Hobbyist Program atE http://www.openvmshobbyist.org/, which is where you'll also go to gete your free Hobbyist licenses.  E The OpenVMS webpage at HP is at http://www.hp.com/go/openvms/ and the C OpenVMS FAQ is at http://h71000.www7.hp.com/wizard/openvms_faq.html    ------------------------------  $ Date: Fri, 9 Jan 2004 08:01:45 -0800# From: "Tom Linden" <tom@kednos.com>b( Subject: RE: Vaxstation some question...9 Message-ID: <NDEMLKKEBOIFBMJLCECIAEBFCKAA.tom@kednos.com>o  D BTW, a zipped iso image is only about 230MB, wouldn't that be easier! and better for everyone involved?s     -----Original Message-----:   From: Keith Parris [mailto:keithparris_NOSPAM@yahoo.com](   Sent: Friday, January 09, 2004 7:58 AM   To: Info-VAX@Mvb.Saic.Com *   Subject: Re: Vaxstation some question...      /   "DaviS*" <davis@spamsux.it> wrote in message e/   news:<hZYKb.54382$VW.2385148@news3.tin.it>... &   > - Can i install OpenVMS from net??F   >   I haven't external Cdrom, but i can "get" this unit from another   > workstation.   H   I'm not aware of any way to install OpenVMS from the Internet, but youA   can buy a CD with binary media for VMS and many popular layeredS<   products for US$30 through the OpenVMS Hobbyist Program atG   http://www.openvmshobbyist.org/, which is where you'll also go to getk   your free Hobbyist licenses.   G   The OpenVMS webpage at HP is at http://www.hp.com/go/openvms/ and theDE   OpenVMS FAQ is at http://h71000.www7.hp.com/wizard/openvms_faq.html       --- (   Incoming mail is certified Virus Free.<   Checked by AVG anti-virus system (http://www.grisoft.com).A   Version: 6.0.560 / Virus Database: 352 - Release Date: 1/8/2004     ---A& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).? Version: 6.0.560 / Virus Database: 352 - Release Date: 1/8/2004    ------------------------------  $ Date: Fri, 9 Jan 2004 07:52:03 -0000$ From: "L. Blunt" <someone@out.there>5 Subject: Re: VMS naming conventions for disk volumes?r* Message-ID: <btlmfs07mr@news1.randori.com>  2 "Brian Chase" <bdc@world.std.com> wrote in message" news:btl4c2$s1u$1@pcls4.std.com...L > Are there any commonly adopted naming conventions used for organizing diskH > based storage under VMS?  For example, in the land of Unix one usuallyL > places user account directories under "/home", often on a separate disk orJ > disk partition.  It's also common to find software specific to the localH > installation under "/usr/local", and at one time it wasn't uncommon to/ > find GNU software installed under "/usr/gnu".t >tH > I was wondering what's commonly done out there on VMS systems so I canJ > setup mine in such a way that they don't end up looking weird to someoneJ > familiar with VMS.  I can certainly imagine logical device names playingH > a useful role here, so underlying disk volumes can be shuffled around,D > without having to change how everything references the moved data. >i > Thanks for any advice. >0	 > -brian.v > --  L The usual method on VMS is to create directories for suitable groups of user, accounts, typically with the same group UIC.  E For typical user accounts this would location would be something likeIL $DISK1:[USERS], then under there would be the directories for the individual user accounts.   eg:    $DISK1:[USER.J_SMITH]  I The most important thing is the creation of rooted logical names for easysL management of these areas. This not only makes moving accounts between disksE easy, and allows the directories to be spread over multiple disks. ItoI removes the need for the user to know where their account is when writingi command files and such.:  = eg:  Put something like the following into systartup_vms.com:2:         $ define/system/translate=(conc,term)    user_root $disk1:[user.],$disk2:[users.]  : Then all SYSUAF accounts are defined using the location asK USER_ROOT:[username], and this is also what the user will see as their home  (SYS$LOGIN) directory.  K Similarly you create other rooted logical names for specific uses, such as: K PROJ_ROOT for development, UTILS_ROOT, etc... Using logical names like this L is one of the reason why VMS systems are so easy to manage and maintain, andJ therefore require far fewer administrators/operators than just about everyL other OS. Being able to shift data/directories/applications/operating systemK locations, without having to edit/alter vast numbers or scripts, databases,m. and uaf entries. Just alter the logical names.  E What is shown above is very similarly to what VMS itself used for the.K operating system, using the logical names like sys$system, which translatesr= to sys$sysroot: and sys$common:, these then translate down torE sys$sysdevice:[sys0.], and so on until a file on a specific device is  located.   ------------------------------  * Date: Fri, 9 Jan 2004 09:37:54 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)5 Subject: Re: VMS naming conventions for disk volumes?n$ Message-ID: <btlspi$218$1@online.de>  H In article <btl4c2$s1u$1@pcls4.std.com>, bdc@world.std.com (Brian Chase) writes:   L > Are there any commonly adopted naming conventions used for organizing diskH > based storage under VMS?  For example, in the land of Unix one usuallyL > places user account directories under "/home", often on a separate disk orJ > disk partition.  It's also common to find software specific to the localH > installation under "/usr/local", and at one time it wasn't uncommon to/ > find GNU software installed under "/usr/gnu".r  - At the moment, on my hobbyist cluster I have:u   DSA122:     OVMSVAXSYS_2 DSA133:     ALPHASYS_3 DSA144:     OVMSVAXSYS_4 DSA510:     USER         DSA520:     SOFT         $22$DKA400: OVMSDOC071   $33$DKA0:   DATA         $44$DKA200: SCRATCH       D In a cluster, each disk must have a unique label; this explains the C suffixes for the system disks.  OVMSDOC071 is a documentation CD.  yI Otherwise, I have USER, SOFT, DATA and SCRATCH.  The top-level directory eF structure on USER, SCRATCH and DATA is similar.  The idea is that the H USER disk is for "user files" and the SCRATCH disk for really temporary  stuff.  In SYS$SYLOGIN I have   ' $  IF F$GETDVI("DISK$SCRATCH","EXISTS")V $  THENsM $    IF F$GETDVI("DISK$SCRATCH","AVL") .AND. F$GETDVI("DISK$SCRATCH","MNT") - ;       THEN DEFINE/JOB SYS$SCRATCH DISK$SCRATCH:['USERNAME']t $  ENDIF  H I use the DATA disk for temporary files which take up a lot of space and> are available somewhere else (e.g. on tape).  The idea is thatD SYS$SCRATCH might be periodically purged, all files deleted etc for D performance reasons, whereas the stuff on the DATA disk would stick A around (but wouldn't be backed up).  (At the moment, I also have  G secondary swap and page files for the machine it is directly connected t to on this disk.)r  H SOFT has software which a) is not from DEC/Compaq/HP and b) is designed I to be used by more than one user.  At the moment, the following stuff is e there:  	 B64DECODE> BINHEX CATDOC CHARM  COMPRESS DVIPSy GS GV LATEX' LYNX METAFONT MFTU MLSEARCH NETSCAPE NEWSRDR  NGOLD  PACK PCXt	 READ_DOCS  SDIR SOFT TARa TEXn
 UNCOMPRESS UNPACK UNZIPp UUDECODE UUENCODE VMSTAR XDVI ZIPo ZIPCLOAK ZIPINFOv ZIPNOTEt ZIPSPLIT FTNCHEKn  > A lot of this is from the Freeware-CD and/or Hunter's archive:  0    http://vms.process.com/fileserv-software.html  E I really think one should use separate physical disks for the system aH disk, user disk, and third-party software.  I also have the SCRATCH and E DATA disks as separate disks, mainly to improve performance and make lF housekeeping easier.  Note that backup frequency, strategy etc should H probably be different in each case.  (Now that (almost) all of my disks D are shadow sets, backup is only for recovering accidentally deleted . files etc, not to guard against a disk crash.)  E Getting back to your original question, naming conventions aren't as e uniform as in unix.  With   =    $  DEFINE/SYS/EXEC/TRAN=(CONC,TERM) logical-name disk-namem  B you can set up basically any naming convention you want, which is E transparent to almost everything.  Thus, I think one should put more v' thought into the physical organisation.   ( By the way, USERNAME mentioned above is   4 $  USERNAME = F$EDIT(F$GETJPI("","USERNAME"),"TRIM")  : which I also use in the following contexts in SYS$SYLOGIN:   $PROCESS_NAME_LOOP:> $  I = I + 1< $  SET MESSAGE/NOFACILITY/NOIDENTIFICATION/NOSEVERITY/NOTEXT' $  ON ERROR THEN GOTO PROCESS_NAME_LOOP & $  SET PROCESS/NAME="''USERNAME'-''I'"4 $  SET MESSAGE/FACILITY/IDENTIFICATION/SEVERITY/TEXT    $  STRING = F$TRNLNM("SYS$NODE") $  END = F$LOCATE(":",STRING)w7 $  STRING = "''F$EXTRACT(0,END,STRING)'_''USERNAME'_> "a $  SET PROMPT = "''STRING'"r  D (make sure you do these two things only for interactive processes!).   ------------------------------  * Date: Fri, 9 Jan 2004 09:42:58 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)5 Subject: Re: VMS naming conventions for disk volumes?y$ Message-ID: <btlt31$218$2@online.de>  < In article <00A2B9BC.1BBCE8DA.7@tachysoft.com>, Wayne Sewell <wayne@tachysoft.com> writes:   H > As far as user accounts are concerned, I have seen two strategies thatH > are commonly used.  One is similar to the eunuchs approach in that the4 > user directories are all under a common root, i.e.H > [users.joe],[users.bob], etc.  In the other case, each user is given aG > top level directory on a disk, i.e. [joe], [bob].  In either case thefE > users can be spread out over multiple disks. (Except for the systemnG > disk.  Most sites prefer for that to contain just the system files.) -H > Even if the [users] approach is used, the users can still be placed onJ > more than one disk by using multiple top level [users] directories, such< > as disk1:[users.joe],disk1:[users.bob], disk2:[users.sam],F > disk1:[users.mike].  However the user directories are organized, theG > entry in the user authorization file points directly to the directoryf > for the user.   ' One can use the first approach and haver  :    $  DEF/SYS/EXEC/TRAN=CONC DISK$USER DISK$BLABLA:[USER.]  D and have it look like they are on DISK$USER.  Even if one used many 2 disks, one could have a search list for DISK$USER.   ------------------------------  $ Date: Fri, 9 Jan 2004 06:07:24 -0500' From: "Main, Kerry" <kerry.main@hp.com>r5 Subject: RE: VMS naming conventions for disk volumes?-R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB1E1375@tayexc19.americas.cpqcorp.net>   > -----Original Message-----6 > From: JF Mezei [mailto:jfmezei.spamnot@istop.com]=20 > Sent: January 9, 2004 1:25 AM  > To: Info-VAX@Mvb.Saic.Come7 > Subject: Re: VMS naming conventions for disk volumes?t >=20 > Wayne Sewell wrote:oJ > > system files.)  Even if the [users] approach is used, the users can=20H > > still be placed on more than one disk by using multiple top level=20H > > [users] directories, such as disk1:[users.joe],disk1:[users.bob],=20? > > disk2:[users.sam], disk1:[users.mike].  However the user=200 > directories=20> > > are organized, the entry in the user authorization file=200 > points directly to the directory for the user. >=20< > You forgot the rooted logicals. Define USER to point to=20C > $disk2:[users.] with the /translation=3Dconceile,terminal) and=20d4 > in the SYSUAF, you use "USER:" as the device name. >=20> > This allows you to move user files to different locations=205 > without having to change all the entries in SYSAUF.  >=20? > If you wish to spread users on different drives/locations,=20-< > you could define USER1, USER2, USER3, USER4 to point to=20< > different locations. Randomly assign users to USERx, to=20 > spread the load. >=20   JF -=20r  F Something else to rconsider wrt to those with current SAN technologiesG and the latest V7.3-2 version of OpenVMS - you can now create a dynamic-E volume using virtual storage of multiple disks in a group. Hence, youoH could keep users all on one "volume" (actually multiple disks in a groupG on SAN) and simply add disks if required by additional users are quired5* to be added. Could be HBVS volume as well.  @ Hence, in your example, USER: would not have to change - even as2 additional storage was added to handle more users.   Regardsi  
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477e Email: kerryDOTmainAThpDOTcome. (remove the DOT's and AT for email address)=20   ------------------------------   Date: 9 Jan 2004 05:29:53 -0600b- From: Kilgallen@SpamCop.net (Larry Kilgallen)e5 Subject: Re: VMS naming conventions for disk volumes?e3 Message-ID: <uBokwgB3o6Zm@eisner.encompasserve.org>   P In article <btl4c2$s1u$1@pcls4.std.com>, bdc@world.std.com (Brian Chase) writes:L > Are there any commonly adopted naming conventions used for organizing diskH > based storage under VMS?  For example, in the land of Unix one usuallyL > places user account directories under "/home", often on a separate disk orJ > disk partition.  It's also common to find software specific to the localH > installation under "/usr/local", and at one time it wasn't uncommon to/ > find GNU software installed under "/usr/gnu".. > H > I was wondering what's commonly done out there on VMS systems so I canJ > setup mine in such a way that they don't end up looking weird to someone > familiar with VMS.  D I have been using VMS for 25 years at many different sites and there< are a variety of methods people use without confusing me :-)  7 > I can certainly imagine logical device names playing  H > a useful role here, so underlying disk volumes can be shuffled around,D > without having to change how everything references the moved data.  A Oh !  Yes, at that level you should _never_ embed a physical diskf@ name into any command procedure, SYSUAF field or similar, except/ for the command procedure that mounts the disk.t   ------------------------------  $ Date: Fri, 9 Jan 2004 07:42:53 -05005 From: "Bochnik, William J" <William_Bochnik@acml.com>b5 Subject: RE: VMS naming conventions for disk volumes?eQ Message-ID: <2D75787AAF09C64481BDFD89113BE6D502D06D66@ac2kama0102.ac.lp.acml.com>h  I to add my 2 cents - look into what a system wide logical is and give each G group one and use it as the device name in case you need to move groupsc7 around from disk to disk - it will cut down on the work  for example-   an accounting user may be,   ACC_DSK:[SMITH]2   while a helpdesk person may be   HLP_DSK:[JONES]:   etc.         -----Original Message-----0 From: Wayne Sewell [mailto:wayne@tachysoft.com] ) Sent: Thursday, January 08, 2004 11:06 PMi To: Info-VAX@Mvb.Saic.Comf5 Subject: Re: VMS naming conventions for disk volumes?p    & >From: bdc@world.std.com (Brian Chase) >X-Newsgroups: comp.os.vms2 >Subject: VMS naming conventions for disk volumes?+ >Date: Fri, 9 Jan 2004 02:41:06 +0000 (UTC)    >-G >Are there any commonly adopted naming conventions used for organizing hE >disk based storage under VMS?  For example, in the land of Unix one lC >usually places user account directories under "/home", often on a cE >separate disk or disk partition.  It's also common to find software 6H >specific to the local installation under "/usr/local", and at one time D >it wasn't uncommon to find GNU software installed under "/usr/gnu". > H >I was wondering what's commonly done out there on VMS systems so I can B >setup mine in such a way that they don't end up looking weird to D >someone familiar with VMS.  I can certainly imagine logical device E >names playing a useful role here, so underlying disk volumes can be rI >shuffled around, without having to change how everything references the e >moved data. >a    J As far as user accounts are concerned, I have seen two strategies that areG commonly used.  One is similar to the eunuchs approach in that the user K directories are all under a common root, i.e. [users.joe],[users.bob], etc.eK In the other case, each user is given a top level directory on a disk, i.e.eG [joe], [bob].  In either case the users can be spread out over multiple  disks. yL (Except for the system disk.  Most sites prefer for that to contain just theL system files.)  Even if the [users] approach is used, the users can still be@ placed on more than one disk by using multiple top level [users]L directories, such as disk1:[users.joe],disk1:[users.bob], disk2:[users.sam],J disk1:[users.mike].  However the user directories are organized, the entryG in the user authorization file points directly to the directory for the- user.   J Another option is to break up the users into groups, such as by department# or function or some other criteria.S  ? For instance: [software.joe],[software.sam],[accounting.larry],-- [communications.george],[communications.dave]nL ============================================================================ ===a8 Wayne Sewell, Tachyon Software Consulting  (281)812-0738 wayne@tachysoft.coma; http://www.tachysoft.com/www/tachyon.html and wayne.html    L ============================================================================ ===dB Jed Clampett, checking into hotel: "This place got a cement pond?", 	Ellie May: "And do yuh let critters in it?"  ) -----------------------------------------  The information contained in this transmission may contain privileged and confidential information and is intended only for the use of the person(s) named above. If you are not the intended recipient, or an employee or agent responsible for delivering this message to the intended recipient, any review, dissemination, distribution or duplication of this communication is strictly prohibited. If you are not the intended recipient, please contact the sender immediately by reply e-mail and destroy all copies of  the original message. Please note that we do not accept account orders and/or instructions by e-mail, and therefore will not be responsible for carrying out such orders and/or instructions.n   ------------------------------   Date: 9 Jan 2004 07:15:21 -0600 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)h5 Subject: Re: VMS naming conventions for disk volumes?.3 Message-ID: <ZVPOcsyKNEt9@eisner.encompasserve.org>q  P In article <btl4c2$s1u$1@pcls4.std.com>, bdc@world.std.com (Brian Chase) writes:L > Are there any commonly adopted naming conventions used for organizing diskH > based storage under VMS?  For example, in the land of Unix one usuallyL > places user account directories under "/home", often on a separate disk orJ > disk partition.  It's also common to find software specific to the localH > installation under "/usr/local", and at one time it wasn't uncommon to/ > find GNU software installed under "/usr/gnu".   H    Generally the OS is on one disk (aka the system disk) and the user's I    directories are at the top level of other disks.  It's not unusual to  H    have a rooted directory to put the user's directories in, instead of     the top.4  H    The only thing I've seen that I would consider unusual, or unwize, isH    user directories on the system disk when there is more than one disk.5    No point in letting users fill up the system disk.a  A    I generally start by puting user's directories on the first ofaD    several non-system disks, and then add directories to other disksC    as different groups come along or certain users need more space.r   ------------------------------  $ Date: Fri, 9 Jan 2004 08:29:02 +0100( From: "Rudolf Wingert" <win@fom.fgan.de>- Subject: Re: VMS to VMS copying between Nodesw: Message-ID: <MCELKPMOKPMNDNKJNIONOEKMCIAA.win@fom.fgan.de>   Hello,  O sorry, why does nobody say, you have to define the hosts database. Unknown host.N means, that there ist no local host database or DNS server which can translateH the hostname to IP address. Did you try to use the IP address instead ofN hostname? If you have proxies (TCP/IP proxies) enabled, the best way is to use" RCP as transfer tool (RemoteCoPy).   Best regards R. Wingertl   ------------------------------   Date: 9 Jan 2004 03:13:54 -0800f' From: simon.adaway@ukaea.org.uk (Simon)e- Subject: Re: VMS to VMS copying between Nodest= Message-ID: <b2d9b89e.0401090313.51d92374@posting.google.com>   m bob@instantwhip.com (Bob Ceculski) wrote in message news:<d7791aa1.0401080547.55e09114@posting.google.com>...ln > simon.adaway@ukaea.org.uk (Simon) wrote in message news:<b2d9b89e.0401070723.2a440932@posting.google.com>...
 > > Hi Folks,e > > J > > With Decnet running I could do Node to Node VMS copying easily (did itI > > use LAT?) with the copy command and specifying a remote node - it wassI > > neat. With only TCP/IP this doesn't work for me, it just says 'remoteiI > > node unknown'. Does anyone know how to make it work with TCP/IP? Am I   > > trying the wrong thing here? > >  > > Thanks in advancee > > 	 > > Simon7 > 9 > use TCPware IP stack so you can do phase IV over IP ...t  , Thanks guys there's some useful stuff there,  E But....Isn't it one heck of a load of trouble installing Decnet V (orm> IV) just to do a proper copy between nodes. FTP doesn't handleF sub-directories at all well. Would it be possible to 'share' a disk atE a level and then 'mount' the share on another OpenVMS node...that wayfC I get a true files-11 structure and full copying of sub-directorieso etc?   Simoni   ------------------------------   Date: 9 Jan 2004 13:15:09 +0100e' From: huber@mppmu.mpg.de (Joseph Huber).- Subject: Re: VMS to VMS copying between Nodeso+ Message-ID: <mJaZxV4eZR3W@vms.mppmu.mpg.de>t  g In article <b2d9b89e.0401090313.51d92374@posting.google.com>, simon.adaway@ukaea.org.uk (Simon) writes: o > bob@instantwhip.com (Bob Ceculski) wrote in message news:<d7791aa1.0401080547.55e09114@posting.google.com>... o >> simon.adaway@ukaea.org.uk (Simon) wrote in message news:<b2d9b89e.0401070723.2a440932@posting.google.com>...o >> > Hi Folks, >> > ,K >> > With Decnet running I could do Node to Node VMS copying easily (did itrJ >> > use LAT?) with the copy command and specifying a remote node - it wasJ >> > neat. With only TCP/IP this doesn't work for me, it just says 'remoteJ >> > node unknown'. Does anyone know how to make it work with TCP/IP? Am I! >> > trying the wrong thing here?- >> > -   > G > But....Isn't it one heck of a load of trouble installing Decnet V (orrE > IV) just to do a proper copy between nodes. FTP doesn't handle sub-w > directories at all well. 6G DECnet V installation is not complicated at all, as long as You have a lF simple configuration without routers, name-servers etc. Just configure( to use only LOCAL and DOMAIN namespaces.  + > Would it be possible to 'share' a disk atrG > a level and then 'mount' the share on another OpenVMS node...that way E > I get a true files-11 structure and full copying of sub-directoriese > etc?  B Yes of course, but this has nothing to do with DECnet IV vs. V  in Your initial question:    5 With TCPIP only:  NFS    (part of the TCPIP software)e  / With DECnet:      DECdfs  (VMS layered product)d   -- t>    Joseph "Sepp" Huber, Muenchen   http://www.huber-joseph.de/   ------------------------------  % Date: Fri, 09 Jan 2004 10:19:59 +0100y From: jf.pieronne@laposte.nete  Subject: ZLIB 1.2.1 form OpenVMS2 Message-ID: <btls51$g9i$3@news-reader1.wanadoo.fr>  . Version of ZLIB 1.2.1 for OpenVMS is released.  ; inflate is about 20% faster and minimizes memory allocation  crc32 is about 50% fastere etc...  C The shareables images 1.2.1 are compatible whith the 1.1.4 version.h  M Download from http://www.pi-net.dyndns.org/anonymous/kits/ or any of the two t mirrors.  
 Jean-Franoisv   ------------------------------  # Date: Fri, 09 Jan 2004 15:15:03 GMTU4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk>B Subject: Re: [Survey] Your best choice for an X Client on Lostdows0 Message-ID: <3FFEC271.4E6B37BD@blueyonder.co.uk>   David M Smith wrote: > H > On Thu, 08 Jan 2004 15:02:20 +0100, RvS_Internals <no@spam.com> wrote: > R > >What would be your best choice for a cheap professional X-Windows Client on XP? > >o > R > The responses so far run to eXceed and eXcursions. I happen to have eXceed on myP > company-owned PC, but can't (won't) spend that kind of money for my home PC to > access my hobbyist system. > > > I'd like to know if anyone has experience with CygWin/X, see  1 First, I presume you mean X Server, not X client?   Q I did have a good bash at getting it working for me about 1 year ago. It did haveaQ the annoying protperty of a rather frequent memory leak or somesuch which gobbled W memory on Lostdows, as you call it, and eventually caused a crash (timescale of hours).   G I moved to using VNC which has been adequate for my modest needs since..  C I see there is a new version of cygwin, which I am downloading now.s   regards,   4 ----------------------------------------------------   -- t tim.llewellyn@blueyonder.co.uk   ------------------------------  # Date: Fri, 09 Jan 2004 15:11:43 GMTr4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk>B Subject: Re: [Survey] Your best choice for an X Client on Lostdows0 Message-ID: <3FFEC1A9.A7C0F4F0@blueyonder.co.uk>   "Keith A. Lewis" wrote:u >  > koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes in article <57sqX4ZqGYOt@eisner.encompasserve.org> dated 8 Jan 2004 12:14:24 -0600:` > >In article <btjrdu$1af$1@newslocal.mitre.org>, lewis@mazda.mitre.org (Keith A. Lewis) writes:M > >> Or if you have an Intel/Linux box available, Xvnc.  Then your VMS box ismN > >> talking X to the virtual display server on Linux, and your Windows systemN > >> talks VNC (not X) to the Linux server.  This setup has the advantage that> > >> your VMS X-windows session will survive a Windows reboot. > > J > >   Not necessarily a Linux box.  VNC runs on most boxes.  Unfortunately> > >   IIRC the server doesn't run on VMS yet, just the client. > I > I've used the VMS<-->Xvnc setup on several platforms, and the one whichOM > stands out as ultra-reliable is Intel/Linux.  SPARC/Solaris, PPC/Linux, andsD > Alpha/Linux all work, but the server process tends to crash, whichC > effectively ends your X session.  That's why I was specific in myr > recommendation.h  J I have seen the Linux VNC server crash, on the time scale of months rather that hours, though.c   --   tim.llewellyn@blueyonder.co.uk   ------------------------------  % Date: Fri, 09 Jan 2004 16:42:53 +0100n2 From: martin@radiogaga.harz.de (Martin Vorlaender)B Subject: Re: [Survey] Your best choice for an X Client on Lostdows; Message-ID: <3ffecbfd.524144494f47414741@radiogaga.harz.de>n  9 David McKenzie (david.mckenzie@paradigm-shift.biz) wrote:' > David M Smith wrote:& > > RvS_Internals <no@spam.com> wrote:M > >> What would be your best choice for a cheap professional X-Windows Clienth > >> on XP?  > > N > > The responses so far run to eXceed and eXcursions. I happen to have eXceedM > > on my company-owned PC, but can't (won't) spend that kind of money for my ) > > home PC to access my hobbyist system.H > > @ > > I'd like to know if anyone has experience with CygWin/X, see > >   > > http://www.cygwin.com/xfree/ > M > I use Cygwin as an Xserver regularly to talk to my Linux boxes and have no nN > problem. I think there are lots of other goodies in their like a full blown N > development environment but have not really played with them. Unfortunately L > I do not have VMS client to test cygwin against. But for me so far Cygwin  > is great.e  D What I don't like about Cygwin (and gcc) is that they put a Unix capG over whatever the OS is. If I want to work under Unix, I boot up one of H my Linux machines. If I work/program under VMS or MicroCrap, I'd like to have the OS without a *ix cap.   cu,.   Martin -- cB                         | Martin Vorlaender | VMS & WNT programmer1  OpenVMS: Where do you  | work: mv@pdv-systeme.denD  want to BE today?      |   http://www.pdv-systeme.de/users/martinv/8                         | home: martin@radiogaga.harz.de   ------------------------------   End of INFO-VAX 2004.017 ************************