1 INFO-VAX	Sat, 16 Apr 2005	Volume 2005 : Issue 211       Contents: 'touch' command under VMS? Re: 'touch' command under VMS? Re: 'touch' command under VMS? Re: 'touch' command under VMS? Re: 'touch' command under VMS? Re: FA: VMS SW kits. Re: Itanium ovms 8.2 migration* Re: Microsoft Joins the Battle for Itanium* Re: Microsoft Joins the Battle for Itanium* Re: Microsoft Joins the Battle for Itanium* Re: Microsoft Joins the Battle for Itanium* Re: Microsoft Joins the Battle for Itanium* Re: Microsoft Joins the Battle for Itanium* Re: Microsoft Joins the Battle for Itanium* Re: Microsoft Joins the Battle for Itanium$ Re: Operating System Security Survey$ Re: Operating System Security Survey$ Re: Operating System Security Survey$ Re: Operating System Security Survey$ Re: Operating System Security Survey$ Re: Operating System Security Survey$ Re: Operating System Security Survey$ Re: Operating System Security Survey$ Re: Operating System Security Survey$ Re: Operating System Security Survey$ Re: Operating System Security Survey$ Re: Operating System Security Survey$ Re: Operating System Security Survey$ Re: Operating System Security Survey$ Re: Operating System Security Survey OT: Requiem for a Feline Secure wireless network + Re: Unix - VMS compatible encryption method + Re: Unix - VMS compatible encryption method + Re: Unix - VMS compatible encryption method 6 Re: Updated VMS information please keep for future use1 Re: What do you expect from DIR/SIZ=(ALLO,USED) ? ; Re: [OpenVMS] What do you expect from DIR/SIZ=(ALLO,USED) ?  Re: [PCSI] Feature request Re: [PCSI] Feature requestB Re: [PCSI] PACKAGE vs COPY, was How to compress .PCSI to .PCSI$COMB Re: [PCSI] PACKAGE vs COPY, was How to compress .PCSI to .PCSI$COMB Re: [PCSI] PACKAGE vs COPY, was How to compress .PCSI to .PCSI$COM  F ----------------------------------------------------------------------  % Date: Fri, 15 Apr 2005 13:14:27 -0500 % From: Neil Cherry <njc@wolfgang.uucp> # Subject: 'touch' command under VMS? . Message-ID: <slrnd60143.r5d.njc@wolfgang.uucp>  A Is there an equivalent command to Unix's touch command under VMS? E I need this for tftp where we need to touch the file before we put it 
 there. Thanks    --  C Linux Home Automation         Neil Cherry       ncherry@comcast.net ; http://home.comcast.net/~ncherry/               (Text only) 8 http://hcs.sourceforge.net/                     (HCS II): http://linuxha.blogspot.com/                    My HA Blog   ------------------------------  % Date: Fri, 15 Apr 2005 14:28:04 -0400 - From: "John E. Malmberg" <wb8tyw@qsl.network> ' Subject: Re: 'touch' command under VMS? 1 Message-ID: <r5idnZr14PMrmv3fRVn-hg@adelphia.com>    Neil Cherry wrote:C > Is there an equivalent command to Unix's touch command under VMS? G > I need this for tftp where we need to touch the file before we put it  > there. Thanks   B The CREATE command will create an empty file and populate it from E SYS$INPUT until and end-of-file is received.  On OpenVMS, the end of  2 file character from terminal input is a Control-Z.  H If you use the CREATE command from a DCL command file and the next line 4 starts with a "$" then it will create an empty file.  G One issue with using the CREATE command is that it may create the file  * in a different organization than intended.  J The use of an FDL description file on the CREATE command can control this.  : If you have the GNV kit installed, it has a touch command.   -John  wb8tyw@qsl.network Personal Opinion Only    ------------------------------    Date: 15 Apr 2005 16:19:07 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) ' Subject: Re: 'touch' command under VMS? 3 Message-ID: <WZCz3wXjqBj9@eisner.encompasserve.org>   V In article <slrnd60143.r5d.njc@wolfgang.uucp>, Neil Cherry <njc@wolfgang.uucp> writes:C > Is there an equivalent command to Unix's touch command under VMS? G > I need this for tftp where we need to touch the file before we put it  > there. Thanks          $touch == "copy nl: "   D    Will do that, but will not do other "touch" functions.  For those	    I use    6       $touch == "set file/expiration_date=17-nov-1858"  F    You could create a command file which uses f$search to decide whichB    to do.  And beware of sites actually using the expiration date.   ------------------------------  % Date: Fri, 15 Apr 2005 20:33:22 -0400 - From: Sandra Golas <Sandra.Golas@comcast.net> ' Subject: Re: 'touch' command under VMS? 0 Message-ID: <_8CdnSWLJs_SwP3fRVn-pw@comcast.com>   Neil Cherry wrote:C > Is there an equivalent command to Unix's touch command under VMS? G > I need this for tftp where we need to touch the file before we put it  > there. Thanks  > I Define a global symbol:  touch == "Append nl:". Then touch a file and do   dir/dat=mod.   ------------------------------  % Date: Fri, 15 Apr 2005 17:54:41 -0700 ( From: Jeff Cameron <roktsci@comcast.net>' Subject: Re: 'touch' command under VMS? / Message-ID: <BE85B061.C02B%roktsci@comcast.net>   G On 4/15/05 11:14 AM, in article slrnd60143.r5d.njc@wolfgang.uucp, "Neil " Cherry" <njc@wolfgang.uucp> wrote:  C > Is there an equivalent command to Unix's touch command under VMS? G > I need this for tftp where we need to touch the file before we put it  > there. Thanks   % I like : $SET FILE/TRUNCATE file-spec G It resets the modification date and either does nothing to the file, or K releases unused allocation clusters from the file for files that have them.    ------------------------------  # Date: Fri, 15 Apr 2005 19:44:22 GMT % From: "Bob Lail" <Robert.Lail@hp.com>  Subject: Re: FA: VMS SW kits. 2 Message-ID: <qQU7e.3992$d%4.1586@news.cpqcorp.net>  = "Rob Brooks" <brooks@cuebid.zko.dec.nospam> wrote in message  ' news:C3rUG0INq7nQ@cuebid.zko.dec.com... , > bill@cs.uofs.edu (Bill Gunshannon) writes:I >> (Still hoping someone from HP would jump in here with the real story!)  > K > Official statements would need to come from product management; the other J > engineers who frequent this joint may not know the official position (I 	 > don't).  >  > --   > 2 > Rob Brooks    VMS Engineering -- I/O Exec Group  > brooks!cuebid.zko.dec.com      Bill and Others   L The standard software license terms and conditions are available for ALL to # read on the HP website at this URL:   / http://licensing.hp.com/swl/view.slm?page=index   0 this page talks to the question being discussed.  / http://licensing.hp.com/swl/view.slm?page=stdtm   G My read, and I am NOT a lawyer either, is no, it is a violation of the  K standard terms an conditions to sell the software (media) without explicit   permission from HP.   	 \Bob Lail      --  
 Robert G Lail  Pre-Sales Solution Consultant  Corporate Accounts Hewlett-Packard Company 
 Merrimack, NH  TEL: 603.424.6272  CEL: 603.315.0556    ------------------------------    Date: 15 Apr 2005 16:13:12 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) ' Subject: Re: Itanium ovms 8.2 migration 3 Message-ID: <ivHEHIieVKXc@eisner.encompasserve.org>   j In article <1113574696.425304.268200@o13g2000cwo.googlegroups.com>, "ugex" <edgar_ulloa@yahoo.com> writes:
 > Hi Guys: > F > I need to migrate all aplications in vax vms 7.1 to Itanium ovms 8.2 > (all builded in cobol).  >  > The questions are..  > H > 1.-- Can I do it direcly .. vax - vms  Itanium..? or is necesary Vax - > alpha -itanium..?   F    Well, I once ported Fortran from IBM FORTRAN EXTENDED ENHANCED H toE    VAX-11 Fortran 3.0.  I think porting Cobol from VAX VMS to I64 VMS     would be much easier.  I    (300000 lines of code; took 1 person half time 3 months, during which  H     I learned EDT, DCL, Macro-32, reading 360 Assembly, ANSI/ASCII tape J     labels, IBM SL (EBCDIC) tape labels, System Services, LIB$ RTL calls, K     and some other stuff.  Darn good think I had VMS to make things easy!).   D > 2.- Some one knows what itanium licences I need for start ovms 8.2 > itanium..?      See www.hp.com/go/vms   ------------------------------  % Date: Fri, 15 Apr 2005 15:58:52 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 3 Subject: Re: Microsoft Joins the Battle for Itanium B Message-ID: <1113595135.263ce7941bd6059fde046d912420cb97@teranews>   Keith Parris wrote: E > Microsoft's next server operating system, code-named "Longhorn," is 8 > expected to ship in 2007, and with an Itanium version.  F I have read an artile on CNET not long ago with a different spin.  TheB IA64 version will be released later than for the real version, butE released nevertheless. It was seen as damned if you do, damned if you A don't.  The 64 bit version for 8086 will be released much sooner.   H Also, there will still be a very limited set of applications availabl on. IA64-Windows. No different than Alpha-Windows.  C Also, remember that in 2007, the real future of IA64 will likely be F publicly known, so Longhorn on IA64 may be moot by then and never make it out.    ------------------------------  % Date: Fri, 15 Apr 2005 20:19:04 -0400 ( From: Bill Todd <billtodd@metrocast.net>3 Subject: Re: Microsoft Joins the Battle for Itanium = Message-ID: <v_WdnYhiUpJhxP3fRVn-2g@metrocastcablevision.com>    Keith Parris wrote:   B His usual fluff, this time courtesy of Interex - that notoriously H independent and unbiased adjunct to HP marketing.  Let's take a look at  some:    ...      Microsoft isB > already trumpeting the benefits of Longhorn on Itanium in a new D > marketing drive, starting with a multi-city tour called "Route64."  C Alan beat me to the observation that this marketing drive seems at  L *least* as much aimed at x86-64 as at Itanic.  For example, these statements  I "the combination of faster 64-bit processors and Windows 64-bit provides  3 the best performance, even for 32-bit applications"    and   ? "32-bit Compatibility:  even when moving to 64-bit, you retain  C compatibility with 32-bit. Windows 64-bit can run high-performance   32-bit applications"  D are true *only* for x86-64 environments:  Itanic 32-bit application H performance still really sucks by comparison, just not *quite* as badly  as it once did.   H And even two of the three special *HP* system offerings associated with L the marketing drive (and used in the training classes) are Xeons.  Whoops...   ...      HPI > executives are taking the long view, pointing out that HP's plan spans   > 10 years or more  L Hard to argue with that, since they're already over 16 years into said plan.  7 , and that the benefits of performance, scalability and J > cost savings will give HP a huge advantage against competitors like IBM 	 > and Sun   C Chortle.  Like the over 3:1 performance/scalability *deficit* that  I Itanic has given them in TPC-C at 64 processors, perhaps?  On the bright  G side, though, HP charges 60% more *per unit of work* on their 1M TPC-C  F Madison system than IBM charges on their 3.2M TPC-C POWER5 system, so H customers can feel doubly good about their decision to take a voyage on  the Itanic.   @ Looks like HP Itanics are 0 for 3 (in terms of 'the benefits of E performance, scalability, and cost savings' referred to above) at 64  2 processors, at least as reflected in TPC-C scores.  @ Large systems not your bag?  Then HP will cheerfully sell you a H 16-processor Itanic which will handle only 37% of the TPC-C workload of H a 16-core IBM system - and while there, at least, HP can save you a bit C of cash per unit of work if your needs don't exceed their system's  I relatively limited capacity, the 8% difference isn't exactly anything to   write home about.   E So HP is only 1-for-3 at 16 processors for the metrics it claims are  3 important (and misrepresents itself as leading in).   ? Still too big for you?  IBM beats HP's performance even at the  F 4-processor level, but not by as much, and HP may actually be able to H save you some noticeable money there per unit of work.  But in terms of E the three metrics chosen by HP's own spokespeople, HP still leads in  @ only one of them at the 4-processor system size - and Itanic is D perilously close to being overtaken in performance there by its own G sibling Xeon, which has recently posted a 150K score (vs. 161K for the  I newest Itanics) using IBM's new x3 chipset (which is designed to support  G Xeon configurations up to 32 processors, so it may offer a significant  * challenge in larger systems soon as well).  E Perhaps SAP SD 2-tier better reflects your workload, though - or you  G just (quite prudently) wouldn't want to use only a single benchmark to   evaluate such systems.  H Unfortunately, HP doesn't give you as many options to evaluate Itanic's E 'scalability' there, because the largest system it has submitted has  C only 16 processors (the Big Kahunas of SAP SD 2-tier are - gasp! -  H Fujitsu's 128-processor SPARC64 system at 21K users, with IBM's 64-core / POWER5 system right on its heels at 20K users).   B But we can only work with what HP sees fit to benchmark, so let's I compare its 16-processor score of 2880 to IBM's 16-core score of - hmmm,  F 5056.  Looks as if POWER5 has only a 75% performance advantage there, F but that's still not exactly chopped liver.  And while Fujitsu hasn't I benchmarked recent systems this small, one can observe that even if HP's  H Itanic box scaled *completely linearly* up to 128 processors (which one > might consider a bit optimistic) it would only beat Fujitsu's  128-processor score by 10%.   D And at the low end, both Opteron and Xeon beat Itanic's 4-processor > score (though not by nearly as large a margin as POWER5 does).  E No, SAP SD 2-tier doesn't appear to be the kind of workload those HP  L execs touting the purported 'advantages' of Itanic were thinking of, either.  G Well, there's always jbb2000 - a popular Java server benchmark, and HP  G has submitted configurations up to 64 processors (hitting a high score  E of just a smidge over 1M) which we can evaluate for 'scalability' as  D well as for per-processor performance.  Unfortunately, IBM's POWER5 F 64-core system hits 2.2M and the Fujitsu 128-processor SPARC64 system I does a hair better than 2.2M (and an 'HPC' version of that system scores  ! 2.43M using only 120 processors).   C If you were thinking things might get better for Itanic at smaller  G system sizes - they don't.  IBM's POWER5 continues to devastate Itanic  E by roughly 2:1 margins at the 32-, 16-, and 8-core system sizes (and  G even at just 4 cores leads it by 46%; even previous-generation POWER4+  F systems hold their own pretty well against Itanic up to their 32-core H limit), Fujitsu's SPARC64 continues to lead Itanic at the 64-, 32-, and > 16-processor system sizes (I didn't notice any smaller recent G submissions from Fujitsu), and HP's own PA-RISC 8800 matches Itanic at  F 32 cores (in normal configurations; a 16-socket, two-chips-per-socket D mx2 Itanic configuration beats PA-RISC 8800 by about 4%, due to the A additional large, off-chip mx2 caches) and trounces Itanic quite   thoroughly at 8 cores.  D And when you get down to 4-processor systems, both Xeon and Opteron H soundly thrash Itanic (the former being especially effective when mated I with the new IBM x3 series chipset that I mentioned above - so watch out  - here as well for larger configurations soon).   I Thus in jbb2000 Itanic, rather than enjoying any performance/scalability  F *advantage* (as those HP spokespeople would have us believe), is tied E with PA-RISC 8800 for 3rd place (behind POWER5 and SPARC64) in large  5 systems - and things just get worse in small systems.    Tough luck again, guys.   G Just what kind of useful server workload do you suppose those HP execs  F (or might they have been more like marketing flacks?) *were* thinking  about, anyway?   - bill   ------------------------------  # Date: Sat, 16 Apr 2005 00:31:09 GMT & From: Rick Jones <foo@bar.baz.invalid>3 Subject: Re: Microsoft Joins the Battle for Itanium 2 Message-ID: <h1Z7e.4018$fa5.2101@news.cpqcorp.net>  ) Bill Todd <billtodd@metrocast.net> wrote: K > "the combination of faster 64-bit processors and Windows 64-bit provides  5 > the best performance, even for 32-bit applications"    > and   A > "32-bit Compatibility:  even when moving to 64-bit, you retain  E > compatibility with 32-bit. Windows 64-bit can run high-performance   > 32-bit applications"  F > are true *only* for x86-64 environments:  Itanic 32-bit application J > performance still really sucks by comparison, just not *quite* as badly  > as it once did.   H 32-bit _X86_ applications, not 32-bit applications in general.  On thoseD OSes which support it (HP-UX, not sure about the others, I know that6 Linux does not) native 32-bit IPF binaries can be run.  
 rick jones --  B firebug n, the idiot who tosses a lit cigarette out his car windowF these opinions are mine, all mine; HP might not want them anyway... :)A feel free to post, OR email to raj in cup.hp.com  but NOT BOTH...    ------------------------------  % Date: Fri, 15 Apr 2005 21:59:40 -0400 ( From: Bill Todd <billtodd@metrocast.net>3 Subject: Re: Microsoft Joins the Battle for Itanium = Message-ID: <0Nedne9FKOQT7P3fRVn-hg@metrocastcablevision.com>    Rick Jones wrote: + > Bill Todd <billtodd@metrocast.net> wrote:  > K >>"the combination of faster 64-bit processors and Windows 64-bit provides  5 >>the best performance, even for 32-bit applications"  >  >  >>and  >  > A >>"32-bit Compatibility:  even when moving to 64-bit, you retain  E >>compatibility with 32-bit. Windows 64-bit can run high-performance   >>32-bit applications" >  > F >>are true *only* for x86-64 environments:  Itanic 32-bit application J >>performance still really sucks by comparison, just not *quite* as badly  >>as it once did.  >  > @ > 32-bit _X86_ applications, not 32-bit applications in general.  H I think you'll agree that 32-bit x86 applications are beyond any shadow I of a doubt the context of the above statements about Windows performance  ?   (and therefore the context of my comment about them as well).    - bill   ------------------------------  % Date: Fri, 15 Apr 2005 21:52:54 -0400 # From: "John Smith" <a@nonymous.com> 3 Subject: Re: Microsoft Joins the Battle for Itanium , Message-ID: <CdydnQQeh7ko7P3fRVn-qw@igs.net>   Keith Parris wrote: : > From Interex's HP World News newsletter, April 15, 2005: > --- ( > Microsoft Joins the Battle for Itanium > C > Suddenly, Microsoft is joining Hewlett-Packard and Intel in their E > quest to aggressively drive 64-bit Itanium into enterprises. Here's 8 > what Microsoft is doing to promote Itanium -- and why. > G > Go here for more: http://www.interex.org/hpwnews/content/4.14.05.html  > --- 
 > which says:  > --- ) > "Microsoft Joins the Battle For Itanium  > ? > Suddenly, Microsoft is joining HP and Intel in their quest to 5 > aggressively drive 64-bit Itanium into enterprises.  > E > Microsoft's next server operating system, code-named "Longhorn," is E > expected to ship in 2007, and with an Itanium version. Microsoft is A > already trumpeting the benefits of Longhorn on Itanium in a new D > marketing drive, starting with a multi-city tour called "Route64." > D > Microsoft also plans to buy 1,600 Itanium servers from HP, many of9 > them destined for Itanium 2-based Longhorn lab testing.  > F > Microsoft has already developed or is working on Itanium versions of> > SQL Server 2005, .Net Framework 2005 and Visual Studio 2005. > E > It's an interesting battle with much at stake. ... Microsoft, Intel F > and HP have to varying degrees bet their respective companies on the > platform.  > B > Microsoft hopes Longhorn will compete with Linux as the high-endC > replacement for UNIX systems in enterprises. Intel is counting on C > Itanium to help it continue to dominate the server microprocessor ? > market. And HP has committed nearly all its server lines to a  > migration to Itanium.  > F > HP has been both criticized and applauded for its aggressive supportG > of the Itanium platform, which it co-developed with Intel. ... But HP H > executives are taking the long view, pointing out that HP's plan spansE > 10 years or more, and that the benefits of performance, scalability D > and cost savings will give HP a huge advantage against competitorsE > like IBM and Sun. ... Part of the marketing effort by Intel, HP and E > now Microsoft will be designed to drive momentum behind Itanium ...  > D > And it's more than marketing. Intel plans to ship a motherboard inA > 2007 that will enable companies to yank the Xeon chips on those F > motherboards and replace them with more powerful Itanium processors. > G > ... With the full backing of Intel, HP and now Microsoft, Itanium has ' > a real shot of owning the enterprise.  > 0 > Stay tuned. This is about to get interesting."    , There was a version of NT for the Alpha too.  4 Those who forget history are condemned to repeat it.   ------------------------------  % Date: Fri, 15 Apr 2005 22:00:07 -0400 # From: "John Smith" <a@nonymous.com> 3 Subject: Re: Microsoft Joins the Battle for Itanium , Message-ID: <CdydnQceh7kr7P3fRVn-qw@igs.net>   Bill Todd wrote: > Keith Parris wrote:  > C > His usual fluff, this time courtesy of Interex - that notoriously F > independent and unbiased adjunct to HP marketing.  Let's take a look
 > at some: >  > ...  >  >   Microsoft isB >> already trumpeting the benefits of Longhorn on Itanium in a newE >> marketing drive, starting with a multi-city tour called "Route64."  > D > Alan beat me to the observation that this marketing drive seems atC > *least* as much aimed at x86-64 as at Itanic.  For example, these  > statements > A > "the combination of faster 64-bit processors and Windows 64-bit > > provides the best performance, even for 32-bit applications" >  > and  > @ > "32-bit Compatibility:  even when moving to 64-bit, you retainD > compatibility with 32-bit. Windows 64-bit can run high-performance > 32-bit applications" > E > are true *only* for x86-64 environments:  Itanic 32-bit application C > performance still really sucks by comparison, just not *quite* as  > badly  > as it once did.  > D > And even two of the three special *HP* system offerings associated > withC > the marketing drive (and used in the training classes) are Xeons.  > Whoops...     2 I'd rather have an Opteron-based system than Xeon.     >   HPC >> executives are taking the long view, pointing out that HP's plan  >> spans 10 years or more  > C > Hard to argue with that, since they're already over 16 years into  > said plan. > 9 > , and that the benefits of performance, scalability and F >> cost savings will give HP a huge advantage against competitors like >> IBM and Sun > D > Chortle.  Like the over 3:1 performance/scalability *deficit* thatC > Itanic has given them in TPC-C at 64 processors, perhaps?  On the F > bright side, though, HP charges 60% more *per unit of work* on theirE > 1M TPC-C Madison system than IBM charges on their 3.2M TPC-C POWER5 C > system, so customers can feel doubly good about their decision to  > take a voyage on
 > the Itanic.  > A > Looks like HP Itanics are 0 for 3 (in terms of 'the benefits of F > performance, scalability, and cost savings' referred to above) at 644 > processors, at least as reflected in TPC-C scores. > A > Large systems not your bag?  Then HP will cheerfully sell you a F > 16-processor Itanic which will handle only 37% of the TPC-C workload > ofE > a 16-core IBM system - and while there, at least, HP can save you a  > bit D > of cash per unit of work if your needs don't exceed their system'sG > relatively limited capacity, the 8% difference isn't exactly anything  > to write home about. > F > So HP is only 1-for-3 at 16 processors for the metrics it claims are5 > important (and misrepresents itself as leading in).  > @ > Still too big for you?  IBM beats HP's performance even at theG > 4-processor level, but not by as much, and HP may actually be able to F > save you some noticeable money there per unit of work.  But in terms > ofF > the three metrics chosen by HP's own spokespeople, HP still leads inA > only one of them at the 4-processor system size - and Itanic is E > perilously close to being overtaken in performance there by its own H > sibling Xeon, which has recently posted a 150K score (vs. 161K for theB > newest Itanics) using IBM's new x3 chipset (which is designed toD > support Xeon configurations up to 32 processors, so it may offer a8 > significant challenge in larger systems soon as well). > F > Perhaps SAP SD 2-tier better reflects your workload, though - or youH > just (quite prudently) wouldn't want to use only a single benchmark to > evaluate such systems. > @ > Unfortunately, HP doesn't give you as many options to evaluateA > Itanic's 'scalability' there, because the largest system it has  > submitted has D > only 16 processors (the Big Kahunas of SAP SD 2-tier are - gasp! -A > Fujitsu's 128-processor SPARC64 system at 21K users, with IBM's 9 > 64-core POWER5 system right on its heels at 20K users).  > C > But we can only work with what HP sees fit to benchmark, so let's D > compare its 16-processor score of 2880 to IBM's 16-core score of -F > hmmm, 5056.  Looks as if POWER5 has only a 75% performance advantage > there,G > but that's still not exactly chopped liver.  And while Fujitsu hasn't E > benchmarked recent systems this small, one can observe that even if C > HP's Itanic box scaled *completely linearly* up to 128 processors @ > (which one might consider a bit optimistic) it would only beat' > Fujitsu's 128-processor score by 10%.  > E > And at the low end, both Opteron and Xeon beat Itanic's 4-processor @ > score (though not by nearly as large a margin as POWER5 does). > F > No, SAP SD 2-tier doesn't appear to be the kind of workload those HPF > execs touting the purported 'advantages' of Itanic were thinking of,	 > either.  > H > Well, there's always jbb2000 - a popular Java server benchmark, and HPH > has submitted configurations up to 64 processors (hitting a high scoreF > of just a smidge over 1M) which we can evaluate for 'scalability' asE > well as for per-processor performance.  Unfortunately, IBM's POWER5 G > 64-core system hits 2.2M and the Fujitsu 128-processor SPARC64 system C > does a hair better than 2.2M (and an 'HPC' version of that system  > scores# > 2.43M using only 120 processors).  > D > If you were thinking things might get better for Itanic at smallerH > system sizes - they don't.  IBM's POWER5 continues to devastate ItanicF > by roughly 2:1 margins at the 32-, 16-, and 8-core system sizes (andH > even at just 4 cores leads it by 46%; even previous-generation POWER4+G > systems hold their own pretty well against Itanic up to their 32-core E > limit), Fujitsu's SPARC64 continues to lead Itanic at the 64-, 32-, C > and 16-processor system sizes (I didn't notice any smaller recent H > submissions from Fujitsu), and HP's own PA-RISC 8800 matches Itanic atG > 32 cores (in normal configurations; a 16-socket, two-chips-per-socket E > mx2 Itanic configuration beats PA-RISC 8800 by about 4%, due to the B > additional large, off-chip mx2 caches) and trounces Itanic quite > thoroughly at 8 cores. > E > And when you get down to 4-processor systems, both Xeon and Opteron C > soundly thrash Itanic (the former being especially effective when F > mated with the new IBM x3 series chipset that I mentioned above - so9 > watch out here as well for larger configurations soon).  > 2 > Thus in jbb2000 Itanic, rather than enjoying anyE > performance/scalability *advantage* (as those HP spokespeople would  > have us believe), is tied F > with PA-RISC 8800 for 3rd place (behind POWER5 and SPARC64) in large7 > systems - and things just get worse in small systems.  >  > Tough luck again, guys.  > H > Just what kind of useful server workload do you suppose those HP execsG > (or might they have been more like marketing flacks?) *were* thinking  > about, anyway?    L I had dinner with IBM's Marketing Manager for Global Small & Medium BusinessH on Wednesday night. They have massive programs in place for ISV's to putI software on Power5/AIX - far better than anything I've seen out of HP for  anything on Itanic.   J We got to talking about a wide variety of things, and unsolicited by me, IH was told by the IBM'er that the biggest mistake HP did in the enterpriseH space in the Compaq deal made was not ditching Itanic in favor of Alpha.   ------------------------------  % Date: Fri, 15 Apr 2005 22:46:04 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 3 Subject: Re: Microsoft Joins the Battle for Itanium B Message-ID: <1113619572.1da9eda01024b381a5741b6457c97539@teranews>  H The one thing that Windows-IA64 will have that Windows-8086 won't is the ability to scale to big-iron.   G In other words, Microsoft may be able to showcase some very large scale D systems running on IA64 which would be too big to run on 8086 (yet).    F A bit line Intel/HP giving away large scale IA64 based systens to someG visible research labs for markleting purposes (being able to brag about % IA64 running some very large system).   K I don't think anyone expects Windows on IA64 to be commercially succesfull.    ------------------------------  % Date: Sat, 16 Apr 2005 01:09:58 -0400 ' From: Dave Froble <davef@tsoft-inc.com> 3 Subject: Re: Microsoft Joins the Battle for Itanium 0 Message-ID: <11617hc18kjm49b@corp.supernews.com>   John Smith wrote:   N > I had dinner with IBM's Marketing Manager for Global Small & Medium BusinessJ > on Wednesday night. They have massive programs in place for ISV's to putK > software on Power5/AIX - far better than anything I've seen out of HP for  > anything on Itanic.  > L > We got to talking about a wide variety of things, and unsolicited by me, IJ > was told by the IBM'er that the biggest mistake HP did in the enterpriseJ > space in the Compaq deal made was not ditching Itanic in favor of Alpha. >  >   D What else would you expect.  For some time the performance race was G Power vs Alpha, with everyone else far behind.  EV8 showing up on time  F and showing anticipated strength would likely have relegated Power to ) second place in large SMP configurations.   5 IBM is the largest beneficiary of the death of Alpha.    Dave   ------------------------------  % Date: Fri, 15 Apr 2005 14:43:16 -0400 ' From: Dave Froble <davef@tsoft-inc.com> - Subject: Re: Operating System Security Survey 0 Message-ID: <11602q9740d7318@corp.supernews.com>   JF Mezei wrote: C >>>rabid fanatics have to resort to ballot box stuffing in order to E >>>skew the results of a stupid web poll in order to try and convince  >>>people it isn't?  >  > 3 > We don't know that there was ballot box stuffing.  > H > If there were a total of 100 votes, and 43 people on comp.os.vms votes5 > for VMS, would that be called ballot box stuffing ?  > G > Now, if one or more VMS-fanatics truly did write scripts to vote many F > times, then yeah, that is bad and it does invalidate the survey. But@ > frankly, I wouldn't expect this from people in this newsgroup.  + Uhhh... Didier did admit to multiple votes.   G As far as what you may expect, it's what you don't expect that usually  	 gets you.    ------------------------------  % Date: Fri, 15 Apr 2005 14:38:44 -0400 ' From: Dave Froble <davef@tsoft-inc.com> - Subject: Re: Operating System Security Survey 0 Message-ID: <11602hpcvbe3481@corp.supernews.com>   Mitch Wagner wrote: 8 > On Wed, 13 Apr 2005 20:27:47 -0400, Dave Froble wrote: >  >  >>Mitch Wagner wrote:  >>+ >>>On 13 Apr 2005 09:43:36 GMT, Doc. wrote:  >>>  >>>  >>> M >>>>%NEWS-I-NEWMSG, Didier MORANDI wrote in news:425c355a$1_3@news.bluewin.ch  >>>> >>>> >>>>6 >>>>>(I added 4 votes with ie/NS/Opera and Firefox :-) >>>>! >>>>Vote early.   Vote often. :-)  >>>  >>>  >>> ( >>>Ooooo, you guys are SOOOOOOOO busted. >>>  >> >>But, are we correct? >  > G > I'm not sufficiently familiar with VMS to be entitled to an opinion.   > K > I'm inclined to vote with the majority -- or the majority before YOU GUYS H > CAME IN AND MESSED THINGS UP!!!!!!!!!!!!!!!! -- that the skills of the > sysadmin are matters most.   >  > Mitch Wagner  F If I could have checked two items, I would have selected both VMS and H sysadmin skills.  Reality is that both are required.  Anybody running a I VMS system can allow full access from the Internet, no firewall, and set  C the password of the SYSTEM user account to MANAGER, just as it was  ( distributed many years and versions ago.  F Your poll is flawed in several ways.  The question of skills vs OS is C not really valid.  Well, Ok, there is the possibility of an entity  F thinking that they can have a poor sysadmin as long as they use an OS D known for good security.  See above.  However, to look at security, I there are really two issues.  Knowledgable system set-up and maintenance n> is one issue.  The tools used (Operating System) is the other.  F A workman is only as good as his tools, and, the tools respond to the  skill of the workman.s   Dave   ------------------------------  % Date: Fri, 15 Apr 2005 14:57:11 -0400i' From: Dave Froble <davef@tsoft-inc.com> - Subject: Re: Operating System Security Surveyr/ Message-ID: <11603kerffns24@corp.supernews.com>    Mitch Wagner wrote:a8 > On Thu, 14 Apr 2005 04:15:06 -0400, Dave Froble wrote: >  >  >>Mitch Wagner wrote:r >>+ >>>On 13 Apr 2005 09:43:36 GMT, Doc. wrote:m >>>r >>>u >>> M >>>>%NEWS-I-NEWMSG, Didier MORANDI wrote in news:425c355a$1_3@news.bluewin.ch  >>>> >>>> >>>>6 >>>>>(I added 4 votes with ie/NS/Opera and Firefox :-) >>>>! >>>>Vote early.   Vote often. :-)p >>>n >>>t >>>c( >>>Ooooo, you guys are SOOOOOOOO busted. >>>v >> >>Ok, I'll confess to curosity.v >>J >>If you're not too familiar with VMS, then what was the occasion for you G >>to read some of these posts on c.o.v concerning your popularity poll?r >  > H > I cover this in this week's edition of the newsletter which <shameless+ > promotion> you can still sign up for herea > 2 > http://www.securitypipeline.com/newsletter.jhtml >  > </shameless promotion> > M > The poll has been up two weeks. Last week, VMS was getting 3 percent of theoJ > vote. This week, allofasudden it's getting more than half (IIRC). And weG > got more responses in the second week of the poll than in the first.   > L > That told me the poll was being fixed by VMS advocates. This is hardly theG > first time this has happened. Or the second. Or the 50th. It's always-7 > pretty obvious from voting patterns what's going on. a  H I'd guess that 'the fix' would be that notice of the poll was posted on I c.o.v.  This newsgroup has many participants, and I'm guessing most have fI strong feelings about VMS.  With the neglect by several owners of the OS VG only those with strong opinions remain.  Let such a group know about a iE chance to express their feelings, and you get a large response.  I'm  5 guessing that such could happen in other venues also.r  C Care to share the actual count for the poll?  If you got less than  H several hundred VMS votes, I'd be a bit surprised, given the desire for   exposure of the OS by many here.  L > So I searched on Google, then Google Groups, for "VMS security" (and maybeI > Security Pipeline -- I forgot which exact search strings I used) -- andr > oho, here y'all were.c >  > Mitch Wagner  # Yep, just where we want to be.  :-)d   ------------------------------  % Date: Fri, 15 Apr 2005 15:01:46 -0400t' From: Dave Froble <davef@tsoft-inc.com>r- Subject: Re: Operating System Security Survey 0 Message-ID: <11603sumft655c4@corp.supernews.com>   JF Mezei wrote:r > Mitch Wagner wrote:h > K >>As of now, we have 2,303 responses. I have no idea how many actual peopleCM >>this corresponds to; looks like several people in this newsgroup voted moren" >>than once. A lot more than once. >  > 2 > Ok, 921 responses for VMS is perhaps a bit high.  E I think you underestimate the participants of this newsgroup.  Those DH still here are the hard core VMS people.  Others left long ago.  A high D percentage of c.o.v responding to such a poll would be what I would  expect, not an aberation.l   ------------------------------  % Date: Fri, 15 Apr 2005 15:16:24 -0400r' From: Dave Froble <davef@tsoft-inc.com>r- Subject: Re: Operating System Security Surveyu0 Message-ID: <11604offbscsce4@corp.supernews.com>   Bill Gunshannon wrote:D > In article <1113511910.b18c1da8cd2d613b33880387f7d503f5@teranews>,2 > 	JF Mezei <jfmezei.spamnot@teksavvy.com> writes: >  >>Bill Gunshannon wrote: >>E >>>If I voted more than once, I have no reason to believe that othershE >>>haven't also voted more than once, so I would be a fool to put anylC >>>faith inthe resulting ballot, so yes, it pretty much proves thatu >>>the result is invalid.d >>J >>After I voted, I was presented with the voting results, and returning to- >>the site automatically pesents the results.a >>$ >>here are the cookies they sent me: >>F >>www.securitypipeline.com	FALSE		FALSE	1144947324	160500005	160500005F >>www.securitypipeline.com	FALSE		FALSE	1144971750	162401003	162401003 >>F >>I assume that each cookie is for one survey I filled (I answered the >>second one too). >>G >>Today, I returned to the site and saw that it remembers I had alreadyw6 >>voted for the stuff on the spam legislation/spyware. >>E >>So those who hacked votes for VMS would have had to disable cookiesaG >>and/or delete their cookies to vote again. I expect windows aor linuxr7 >>weenies to do such things. But not VMS professionals.c >  >  > JF, H >   Go back and read the whole thread.  Not only did at least one personE > start off by voting from three different browsers.  Other suggestedwF > voting multiple times and someone specifically pointed out that theyA > were using a cookie to track voting and that it could be easilyt > defeated.w > C >   I don't make this stuff up.  I said from the beginning that theiA > poll was usless and meaningless  That has since been borne out.  >  > bill >   B I for one keep a close watch on cookies.  When I noticed the ones D similar to what JF posted again, I considered their purpose.  Logic E would suggest that cookies could be used to track prior voting, so I cE deleted the cookies and voted a second time to confirm the suspicion.n  H My post complaining about how easy it was to defeat the poll was partly G to show that the results were meaningless, but even more, a popularity lH rather than technical contest.  I doubt that many respondents voted for  other than the OS they use.   H I didn't set up any type of script or whatever to stuff the ballot box. =   I also doubt anyone else did either.  My reasoning is that oH participation in c.o.v is rather large, and those participants would be 4 eager to take actions that reflect favorably on VMS.   ------------------------------  % Date: Fri, 15 Apr 2005 12:08:01 -0700a& From: Mitch Wagner <mitch@wagmail.com>- Subject: Re: Operating System Security Survey/+ Message-ID: <13cxl2mzy5dpk.dlg@wagmail.com>e  6 On Fri, 15 Apr 2005 14:38:44 -0400, Dave Froble wrote:   > Mitch Wagner wrote:y9 >> On Wed, 13 Apr 2005 20:27:47 -0400, Dave Froble wrote:o >> t >>   >>>Mitch Wagner wrote: >>>t, >>>>On 13 Apr 2005 09:43:36 GMT, Doc. wrote: >>>> >>>> >>>>N >>>>>%NEWS-I-NEWMSG, Didier MORANDI wrote in news:425c355a$1_3@news.bluewin.ch >>>>>  >>>>>s >>>>>t7 >>>>>>(I added 4 votes with ie/NS/Opera and Firefox :-)E >>>>>3" >>>>>Vote early.   Vote often. :-) >>>> >>>> >>>>) >>>>Ooooo, you guys are SOOOOOOOO busted.n >>>> >>>n >>>But, are we correct?l >>   >> cH >> I'm not sufficiently familiar with VMS to be entitled to an opinion.  >> rL >> I'm inclined to vote with the majority -- or the majority before YOU GUYSI >> CAME IN AND MESSED THINGS UP!!!!!!!!!!!!!!!! -- that the skills of thet >> sysadmin are matters most.  >> t >> Mitch Wagner  > H > If I could have checked two items, I would have selected both VMS and J > sysadmin skills.  Reality is that both are required.  Anybody running a K > VMS system can allow full access from the Internet, no firewall, and set VE > the password of the SYSTEM user account to MANAGER, just as it was w* > distributed many years and versions ago. > H > Your poll is flawed in several ways.  The question of skills vs OS is E > not really valid.  Well, Ok, there is the possibility of an entity (H > thinking that they can have a poor sysadmin as long as they use an OS F > known for good security.  See above.  However, to look at security, K > there are really two issues.  Knowledgable system set-up and maintenance  @ > is one issue.  The tools used (Operating System) is the other. > H > A workman is only as good as his tools, and, the tools respond to the  > skill of the workman.t  H Heh-heh.... I find it ironic that you say the "poll is flawed in several; was" and then say "a workman is only as good as his tools."   G Because a tool must be judged for the purpose to which it's meant to bepJ put. A screwdriver makes a poor, but sometimes functional hammer. However,* a screwdriver is great for driving screws.  H Likewise, our Voting Booth polls are designed to get people thinking and@ generate discussion. Which is what this one has certainly done.    Mitch Wagner   ------------------------------  % Date: Fri, 15 Apr 2005 15:53:10 -0400e- From: JF Mezei <jfmezei.spamnot@teksavvy.com>4- Subject: Re: Operating System Security SurveylB Message-ID: <1113594882.f1b92da3e13b93d1220702179f8bc3dc@teranews>   Ken Robinson wrote:hC > The real problem that I saw was that you had to "vote" to see the-E > results. I voted once from work. When I got home, I want to see thei8 > results, but I had to "vote" again in order to do so.     G It seems to be that a few people voting twice wouldn't have made such abF hige "ballot stuffing". This would be expected behaviour for any voter@ and thus distributed evenly amongts all possible voting options.   ------------------------------  % Date: Fri, 15 Apr 2005 16:20:49 -0400 4 From: "Peter Weaver" <newsgroup@weaverconsulting.ca>- Subject: Re: Operating System Security Surveyi, Message-ID: <3cam12F6n85qvU1@individual.net>   Mitch Wagner wrote:t >...F > Likewise, our Voting Booth polls are designed to get people thinkingE > and generate discussion. Which is what this one has certainly done.6 >i > Mitch Wagner  c True, and I like the fact that you seem to like putting jokes in your polls. When you asked "What'spd your favourite way to stay informed on IT topics?" you had "Calling the Psychic Friends Network" andd "I'd rather not read about IT topics" and admitted that the people who answered these "makes me takeb the whole poll with a grain of salt." But then the best joke was allowing people to vote "Windows"6 for the "Which operating system is more secure?" poll.   --   Peter Weaver Weaver Consulting Services Inc.  Canadian VAR for CHARON-VAXX www.weaverconsulting.ca    ------------------------------  % Date: Fri, 15 Apr 2005 16:14:15 -0400s- From: JF Mezei <jfmezei.spamnot@teksavvy.com>h- Subject: Re: Operating System Security Survey B Message-ID: <1113596052.105ac17fe2501085c4b7044a75bad311@teranews>   Dave Froble wrote:D >> I for one keep a close watch on cookies.  When I noticed the onesE > similar to what JF posted again, I considered their purpose.  Logic F > would suggest that cookies could be used to track prior voting, so IG > deleted the cookies and voted a second time to confirm the suspicion.C  D Do now, you're blaming your double voting on me ???  :-) :-) :-) :-)  H Anyone who understands HTTP knows that it is really a stateless protocolF and that keeping tracks of a "session" in a 100% foolproof way is justG impossible since the client can play with the browser's information and) muck about.c  D In effect, you used a real live vote as a playground for you to test  your HTTP and browser skills :-) (and you were not alone).   @ Elections Canada considered allowing internet voting for federalC elections, but decided against it for similar reasons. The issue is D about conserving your vote's anonymity, while ensuring you vote onlyF once and of course that you cannot vote on the internet and then go to/ the local physical voting office to vote again.c   ------------------------------  % Date: Fri, 15 Apr 2005 16:40:22 -0400 4 From: "Peter Weaver" <newsgroup@weaverconsulting.ca>- Subject: Re: Operating System Security Surveye, Message-ID: <3can5nF6md6s0U1@individual.net>   Dave Froble wrote: > JF Mezei wrote:t >> Mitch Wagner wrote: >>F >>> As of now, we have 2,303 responses. I have no idea how many actualA >>> people this corresponds to; looks like several people in this 9 >>> newsgroup voted more than once. A lot more than once.n >> >>3 >> Ok, 921 responses for VMS is perhaps a bit high.  >5F > I think you underestimate the participants of this newsgroup.  ThoseD > still here are the hard core VMS people.  Others left long ago.  AD > high percentage of c.o.v responding to such a poll would be what I! > would expect, not an aberation.   ^ Don't forget Sue sent the address out to her distribution list which would include a number ofc people who do not frequent COV. Less than an hour ago I was talking to someone who mentioned seeingsI the poll in Sue's note but had not been to COV to see any of this thread.h   -- i Peter Weaver Weaver Consulting Services Inc.  Canadian VAR for CHARON-VAX- www.weaverconsulting.cao   ------------------------------  % Date: Fri, 15 Apr 2005 17:10:41 -0400i' From: Dave Froble <davef@tsoft-inc.com>l- Subject: Re: Operating System Security Surveye0 Message-ID: <1160bepbe8282c1@corp.supernews.com>   JF Mezei wrote:f > Dave Froble wrote: > D >>>I for one keep a close watch on cookies.  When I noticed the ones >>E >>similar to what JF posted again, I considered their purpose.  LogiceF >>would suggest that cookies could be used to track prior voting, so IG >>deleted the cookies and voted a second time to confirm the suspicion.d >  > F > Do now, you're blaming your double voting on me ???  :-) :-) :-) :-)  D Not at all.  I mentioned that what you posted was similar to what I ; observed, but reading your post came long after my actions.s  J > Anyone who understands HTTP knows that it is really a stateless protocolH > and that keeping tracks of a "session" in a 100% foolproof way is justI > impossible since the client can play with the browser's information andP
 > muck about.e > F > In effect, you used a real live vote as a playground for you to test" > your HTTP and browser skills :-) > (and you were not alone).   0 What is not specifically prohibited, is allowed.  H If you put anything on my system, such as a cookie, then your system is E also fair game for anything.  I had a suspicion, and I verified same.   E As for voting, to what purpose, other than to show you have too much gI free time, would it be to stuff the poll with a million or so votes?  To '1 do so would only get all votes for VMS discarded.y  9 As the editor has stated, his purpose has been fulfilled.   B > Elections Canada considered allowing internet voting for federalE > elections, but decided against it for similar reasons. The issue isrF > about conserving your vote's anonymity, while ensuring you vote onlyH > once and of course that you cannot vote on the internet and then go to1 > the local physical voting office to vote again.t   ------------------------------   Date: 15 Apr 05 17:14:13 EDT) From: cook@wvnvms.wvnet.edu (George Cook)a- Subject: Re: Operating System Security Survey ! Message-ID: <qN6LnfoO$FZE@wvnvms>c  ^ In article <Xns963961F3BD076dcovmsrox@212.100.160.126>, "Doc." <doc@openvms-rocks.com> writes:# > %NEWS-I-NEWMSG, JF Mezei wrote inl< > news:1113510820.4371e5ca84a0d0760f08fb16af50e0b1@teranews  > G >> And to whoemever stuffed the ballots: shame on you; you deprived VMStF >> of a legitimate place, albeit less extraordinary. Had VMS gotten 1%I >> more than the others, it would have been a greater achievement becauseh& >> that vote would have been credible. > K > I'll point out now that I went back to the poll with a different browser eM > after Didier's comment about being able to vote multiple times.  I did not  A > write any script or anything like that to stuff the ballot box.h  B I originally had intended to vote only once, but when someone here@ indicated they had a problem voting with a particular browser, IC decided to test VMS Mosaic (as its developer) to see if it could be<
 used to vote.a     George Cook: WVNETo   ------------------------------  + Date: Sat, 16 Apr 2005 02:47:54 +0000 (UTC)e7 From: moroney@world.std.spaamtrap.com (Michael Moroney)c- Subject: Re: Operating System Security Surveyo( Message-ID: <d3pucq$bmc$1@pcls4.std.com>  6 "Peter Weaver" <newsgroup@weaverconsulting.ca> writes:  = >But then the best joke was allowing people to vote "Windows"t7 >for the "Which operating system is more secure?" poll.n  J It would be kind of interesting if someone ran a similar poll asking whichJ is the most secure operating system and listed Linux, BSD, VMS, commercialJ Unixes, mainframe OS's and some obscure ones - and not mention Windows at I all.  No explanation, just no mention of Windows as if it was obvious it  J shouldn't be mentioned.  It would be interesting to see the feedback that  would generate...  -- k -Mike    ------------------------------  # Date: Sat, 16 Apr 2005 03:27:02 GMT 6 From: "Kenneth Farmer" <kfarmer@NOSPAM.spyderbyte.com>- Subject: Re: Operating System Security Surveyr> Message-ID: <aC%7e.64308$9v2.1651618@twister.southeast.rr.com>  @ "Peter Weaver" <newsgroup@weaverconsulting.ca> wrote in message & news:3can5nF6md6s0U1@individual.net... > Dave Froble wrote: >> JF Mezei wrote: >>> Mitch Wagner wrote:n >>>eG >>>> As of now, we have 2,303 responses. I have no idea how many actualrB >>>> people this corresponds to; looks like several people in this: >>>> newsgroup voted more than once. A lot more than once. >>>r >>>i4 >>> Ok, 921 responses for VMS is perhaps a bit high. >>G >> I think you underestimate the participants of this newsgroup.  Those3E >> still here are the hard core VMS people.  Others left long ago.  A"E >> high percentage of c.o.v responding to such a poll would be what Il" >> would expect, not an aberation. >cM > Don't forget Sue sent the address out to her distribution list which would b > include a number ofnI > people who do not frequent COV. Less than an hour ago I was talking to a > someone who mentioned seeingK > the poll in Sue's note but had not been to COV to see any of this thread.     M I also posted it on OpenVMS.org.  It had well over one hundred views.  I can -2 only assume many of them participated in the poll.     Ken   % _____________________________________m" Kenneth R. Farmer <>< 336-736-73760 http://www.OpenVMS.org | http://dcl.OpenVMS.org    ------------------------------  # Date: Sat, 16 Apr 2005 05:26:48 GMTn  From: John Santos <john@egh.com>- Subject: Re: Operating System Security Survey * Message-ID: <sm18e.2573$c93.1264@trnddc08>   Larry Kilgallen wrote:V > In article <13cxl2mzy5dpk.dlg@wagmail.com>, Mitch Wagner <mitch@wagmail.com> writes: >  > J >>Likewise, our Voting Booth polls are designed to get people thinking andB >>generate discussion. Which is what this one has certainly done.  >  > J > Please do not be under the illusion that participants here are otherwise  > reticent to discuss things :-)  G True.  Unfortunately, people here are discussing polling methods rathersB than OS security.  Though, Larry, you could explain why you always+ disable cookies to get us back on track :-)o   --   John Santosr Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Fri, 15 Apr 2005 21:24:35 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net>! Subject: OT: Requiem for a Feline + Message-ID: <42607763.16E8ABA3@comcast.net>   7 http://home.earthlink.net/~djesys/personal/cuddles.htmlm   -- n David J Dachtera dba DJE Systemsw http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page:t" http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/h   Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------    Date: 15 Apr 2005 11:05:12 -0700 From: qkumbazu21@gmail.com  Subject: Secure wireless networkC Message-ID: <1113588312.815390.206500@o13g2000cwo.googlegroups.com>b  D I secure my wireless and now i cant use it, it says that is a secureG network. What im suppost to do to be able to get online, or disable it?v Please help    ------------------------------    Date: 15 Apr 2005 15:35:52 -0700 From: jordan@ccs4vms.com4 Subject: Re: Unix - VMS compatible encryption methodB Message-ID: <1113604552.132982.63630@g14g2000cwa.googlegroups.com>  G Got a test going today.  GPG current on a wintel peecee, and my ancientnF restored from backup PGP V2.6.2 ported by David G. North running underE VMS V6.2 on my VAXstation.  Created the keys, exported the public keypD to the GPG machine, which imported it (using --openpgp option).  GPGD encrypted a file, we transferred it back to the VAX, both binary andE ascii armored.  When we try to decrypt it we get the following error:n  E Unsupported packet format: - you need a newer version of PGP for this  file.r  C We're reviewing encryption options on the GPG side to see if we caneF come up with a magic combination, but if anyone out there who has done8 this knows the magic incantations, please post.  Thanks!   Rich   ------------------------------    Date: 15 Apr 2005 16:03:20 -0700 From: bob@instantwhip.come4 Subject: Re: Unix - VMS compatible encryption methodB Message-ID: <1113606200.155450.89150@o13g2000cwo.googlegroups.com>  . we are running pgp 2.6.3i on alpha vms 7.1 and. the only only problem we had is that binary on3 unix gpg will not deencrypt but it will on windoze,s( so ascii armor is needed on unix gpg ...   ------------------------------    Date: 15 Apr 2005 16:34:25 -0700 From: jordan@ccs4vms.com4 Subject: Re: Unix - VMS compatible encryption methodC Message-ID: <1113608065.069094.137470@l41g2000cwc.googlegroups.com>s  F Well, we tried both binary and ascii armor output from wintel gpg, andD the VAX PGP 2.6.2 decline both of them with the packet error.  As itD turns out we (at least currently) don't need to encrypt files on VMSD and send them to the Unix box; its a one way street if we can get itB working.  The 2.6.2 install has the IDEA encryption in it, which ID understand was the subject of later controversy and removal; some ofD the GPG docs imply that may be a compatibility issue but I've got to
 read more.   ------------------------------    Date: 15 Apr 2005 16:15:30 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)o? Subject: Re: Updated VMS information please keep for future useh3 Message-ID: <jl$kXhqQGxk9@eisner.encompasserve.org>n  f In article <1113576765.279785.80320@z14g2000cwz.googlegroups.com>, susan_skonetski@hotmail.com writes:G > Dear Newsgroup, I did have to remove the jobs since I do not have thefH > ok to post them on the web (Ireland, 2 in NY of of Monster and one web > site)  >  > Warm Regards,i > SueI  E    OBTW, we probably have a job for anyone who want to maintain a VMSfE    6.1 based real-time application, with some AIX on the side.  Send      me your resume.   ------------------------------  % Date: Fri, 15 Apr 2005 16:12:57 -0700d, From: Ken Fairfield <my.full.name@intel.com>: Subject: Re: What do you expect from DIR/SIZ=(ALLO,USED) ?+ Message-ID: <d3phpq$kin$1@news01.intel.com>h  = Peter 'EPLAN' LANGSTOEGER writes in a message that hasn't yet' appeared on my server:  = > What should one expect to see when entering the DCL command  > # > $ DIRECTORY/SIZE=(ALLOCATED,USED)C >  > 1) The same as with /SIZE=ALLoC > Note: On V7.3 there were only ALL, ALLOCATED and USED, but now onh% > V7.3-2 and V8.2 there is also UNITSo > ) > 2) The allocated and used size reversede >  > 3) Only the Used size  >  > 4) Only the Allocated size >  > 5) A command syntax errore > 1 > I didn't expect 3) but this is how DCL behaves.  >  > What do you think ?   B      I actually think this _should_ be (5), a command syntax error; (even with the "ALLOCATED" changed to "ALLOCATION"), but...r  <     a) HELP DIRECTORY /SIZE entry does NOT support a LIST of@        options, i.e., there is no "/SIZE=(keyword[,...])", there=        is "/SIZE[=option]", i.e., there is only the choice ofo9        _single_ keyword options ALL, ALLOCATION and USED.   A     b) OTOH, DCL allows listing a (non-positional) qualifier moren@        than once on the command line, and the qualifier (and itsD        options) for the *last* occurrence override those encounteredA        earlier on the command line.  For example, it is perfectly6        correct to type,i  4              $ Delete/Confirm Junk.Lis;*  /NoConfirm  @        and have the command execute with no confirmation prompt.  A     c) Finally, as near as I can tell, the CLD makes no syntacticpD        distinction between when lists of keywords are allowed versusB        when a only single keywords is allowed.  (The best that can1        be done is to add a disallow clause, e.g.,o  /           disallow any2 (ALL, ALLOCATION, USED)c  3        but the CLD for DIRECTORY does not do that.)a  D      Given that no syntax error "should" be generated given the CLD,C then we fall back to sequential processing, along the lines of (b),s and get the answer (3).r  D      Anyone want to file an SPR against DIRECTORY for not generating a syntax error?  :-)          -Kenm -- s6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfield ! D1C Automation VMS System SupportW" who:   kenneth dot h dot fairfield where: intel dot com   ------------------------------  % Date: Fri, 15 Apr 2005 20:08:27 -0400m> From: "Chris Moore" <paynoattentionto@themanbehindthe.curtain>D Subject: Re: [OpenVMS] What do you expect from DIR/SIZ=(ALLO,USED) ?9 Message-ID: <2IY7e.6657$MZ2.959383@news20.bellglobal.com>   I Tried this on systems from 7.1 thru 7.3-1 (some VAX, some Alpha), got theh same result in each case.i  K Directory listing used the last parameter of the /SIZE list ==> i.e. answers #3      C "Peter 'EPLAN' LANGSTOEGER" <peter@langstoeger.at> wrote in message $ news:425ec0ec@NEWS.LANGSTOEGER.AT...= > What should one expect to see when entering the DCL command' >l# > $ DIRECTORY/SIZE=(ALLOCATED,USED)b >h > 1) The same as with /SIZE=ALLfC > Note: On V7.3 there were only ALL, ALLOCATED and USED, but now ond% > V7.3-2 and V8.2 there is also UNITSs >a) > 2) The allocated and used size reversedf >a > 3) Only the Used sizes >W > 4) Only the Allocated size >a > 5) A command syntax errorS >C >s1 > I didn't expect 3) but this is how DCL behaves.' >r > What do you think ?  >r > TIA. >  > -- n > Peter "EPLAN" LANGSTOEGERf' > Network and OpenVMS system specialistr > E-mail  peter@langstoeger.atH > A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  # Date: Fri, 15 Apr 2005 18:48:43 GMT 3 From: hammond@not@peek.ssr.hp.com (Charlie Hammond)n# Subject: Re: [PCSI] Feature requeste1 Message-ID: <f0U7e.3985$J_4.595@news.cpqcorp.net>i  - In article <42600153$1@NEWS.LANGSTOEGER.AT>,  8 peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) writes:  H >May I ask for a new PCSI feature, where the version number of a productF >is allowed to be part of the product name (in addition to /VERSION=). ..H >I often do Copy&Paste of kits and would like to be able to easily enter >t$ >	$ PRODUCT INSTALL VMS732_SYS-V0600 >- >Currently I can onlyo >e' >	$ PRODUCT INSTALL VMS732_SYS/VERS=6.0  >or A >	$ PRODUCT INSTALL VMS732_SYS and answer the question, which kit  > < >which means I have to paste and then edit the command line.   Peter --  C I think you are asking to be allowed to include the fully qualifiedcF product name, similar to the kit file name, and to have PCSI parse out< the PRODUCER, BASE_SYSTEM, NAME and VERSION.  e.g., entering  ,     $ PRODUCT INSTALL HP-AXPVMS-VMS730-V0600   would be equivalent to  K     $ PRODUCT INSTALL /PRODUCER=HP /BASE_SYSTEM AXPVMS-VMS730 /VERSION=V6.0y  B In theory, this would be a problem if product names are allowed toH contain the "-" character.  In practice, how woudl this parse if NOT ALL& of the elements are included?  i.e. in  )     $ PRODUCT INSTALL AXPVMS-VMS730-V0600i  G how does PCSI know that AXPVMS is the BASE_SYSTEM and not the PRODUCER?   I Alternatively, how do you justify (in general) including VERSIN this way,s! but not PRODUCER and BASE_SYSTEM?e  ? It might be easier to allow /SOURCE to include the file name.    Would that address your issue?  F I'll forward this to the PCSI developers, but I suspect that they will9 need a formal request and justification to consider this.e   --  J       Charlie Hammond -- Hewlett-Packard Company -- Ft Lauderdale  FL  USAF           (hammond@not@peek.ssr.hp.com -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------    Date: 15 Apr 2005 22:27:20 +01006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)# Subject: Re: [PCSI] Feature request * Message-ID: <42603fc8@NEWS.LANGSTOEGER.AT>  g In article <f0U7e.3985$J_4.595@news.cpqcorp.net>, hammond@not@peek.ssr.hp.com (Charlie Hammond) writes: f >In article <42600153$1@NEWS.LANGSTOEGER.AT>, peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) writes:I >>May I ask for a new PCSI feature, where the version number of a productuG >>is allowed to be part of the product name (in addition to /VERSION=).s >nD >I think you are asking to be allowed to include the fully qualifiedG >product name, similar to the kit file name, and to have PCSI parse outv= >the PRODUCER, BASE_SYSTEM, NAME and VERSION.  e.g., entering  >n- >    $ PRODUCT INSTALL HP-AXPVMS-VMS730-V0600w >o >would be equivalent to  > L >    $ PRODUCT INSTALL /PRODUCER=HP /BASE_SYSTEM AXPVMS-VMS730 /VERSION=V6.0  G $ PRODUCT INSTALL /PRODUCER=HP /BASE_SYSTEM=AXPVMS VMS730 /VERSION=V6.0   = Nice idea. And for consistency it would be the only solution. C But I think, this would be difficult to implement (if not requiringiB to specify all parts of the kitfilename). And I think, it would beH easier to do what I asked for, as I personally only need the version ;-)  The same as with VMSINSTAL kits.  G btw There are usually no equally named products from different vendors.eG Base system might be an issue, but not currently (because I've no mixed.K architecture clusters and therefore the kits can't be in the same location)eN and I also hope, that "VMS" kits are getting more (and not the "xxxVMS" kits).  G But the full spec would (also) allow to get rid of the question for the-? right kit when entering the command line w/o any qualifiers ;-)o  ? btw: Do you perhaps know of a logical for the /OPTION=NOCONFIRMSN (like PCSI$LOG for /LOG and PCSI$TRACE for /TRACE and PCSI$$SAVE_RECOVERY_DATA for /SAVE_RECOVERY_DATA) ?  C >In theory, this would be a problem if product names are allowed to  >contain the "-" character.   H In theory, yes. But I think I read more than a decade ago (when PCSI wasM introduced), that dashes in productnames are forbidden for exact this reason.mM I don't think, that this has changed. But you would know whom to ask, right ?o  @ >It might be easier to allow /SOURCE to include the file name.   >Would that address your issue?   G No. PCSI$SOURCE is defined here for the location (directory) of all therI kits. I'd like to type as few chars as possible and only copy/paste where E needed, Redefining the logical or specifying another qualifier is notlN what I want. But I'm not the official/only user of VMS. What do others think ?  G >I'll forward this to the PCSI developers, but I suspect that they willi: >need a formal request and justification to consider this.  ) From a hobbyist ? You must be kidding ;-)n   Anyway, thanks.0   -Peter  9 PS: May I ask for another enhancement ? PRODUCT CHECK kit>I I don't like to find out during an install, that a kit file is incompleteEI or garbaged during transfer. Simple testing offline before would be nice.>I Currently, I do it with a PRODUCT COPY (seq to ref), but I hate to forgeto) to delete the directorytree afterwards...>  B PPS: Another enhancement ? PRODUCT REMOVE /FORCE to allow removingE a kit installed on a no longer available disk. Currently one needs toeK define the corresponding disk logical to another/system disk before REMOVE.j  I PPPS: Yes another enhancement ? PRODUCT CHECK/DATABASE[=file] to find outtF if the database (you know, the SYS$SYSTEM:*.PCSI$* files) is still ok. -- I Peter "EPLAN" LANGSTOEGERm% Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------    Date: 15 Apr 2005 14:09:46 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) K Subject: Re: [PCSI] PACKAGE vs COPY, was How to compress .PCSI to .PCSI$COMy3 Message-ID: <$tRkAPxdSnEP@eisner.encompasserve.org>d  Z In article <115vv9tsd1etd55@corp.supernews.com>, Dave Froble <davef@tsoft-inc.com> writes: > Larry Kilgallen wrote: > 7 >> There is no warranty for _anything_ on the freeware.a@ >> That is a major issue for strict security and accountability. > K > Warranty/Support and Security are two seperate and distinct issues.  One t > does not imply the other.g  C Well I did list them separately, but if you have a security breach,s who do you sue ?   ------------------------------  % Date: Fri, 15 Apr 2005 16:54:09 -0400a' From: Dave Froble <davef@tsoft-inc.com>:K Subject: Re: [PCSI] PACKAGE vs COPY, was How to compress .PCSI to .PCSI$COMr0 Message-ID: <1160afnnrbthua4@corp.supernews.com>   Larry Kilgallen wrote:\ > In article <115vv9tsd1etd55@corp.supernews.com>, Dave Froble <davef@tsoft-inc.com> writes: >  >>Larry Kilgallen wrote: >> >>7 >>>There is no warranty for _anything_ on the freeware.w@ >>>That is a major issue for strict security and accountability. >>K >>Warranty/Support and Security are two seperate and distinct issues.  One o >>does not imply the other.s >  > E > Well I did list them separately, but if you have a security breach,i > who do you sue ?  F Hell of a question.  Never thought of that.  Probably the entity that  broched your security.  G I haven't read the fine print on the licenses lately, but I have never  G read any claims in any VMS license, or SPD, or anywhere, that says the w> vendor of the OS would be responsible for any security breech.   Do you know of any such?   ------------------------------    Date: 15 Apr 2005 16:50:03 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)hK Subject: Re: [PCSI] PACKAGE vs COPY, was How to compress .PCSI to .PCSI$COMi3 Message-ID: <dQjOXH3$scEL@eisner.encompasserve.org>   Z In article <1160afnnrbthua4@corp.supernews.com>, Dave Froble <davef@tsoft-inc.com> writes:  I > I haven't read the fine print on the licenses lately, but I have never .I > read any claims in any VMS license, or SPD, or anywhere, that says the o@ > vendor of the OS would be responsible for any security breech. >  > Do you know of any such?  D License terms are mere the starting point for possible legal action.   ------------------------------   End of INFO-VAX 2005.211 ************************