1 INFO-VAX	Thu, 30 Jun 2005	Volume 2005 : Issue 363       Contents: Alpha Basic 1.6   Alpha comeback gaining steam ...$ Re: Alpha comeback gaining steam ...$ Re: Alpha comeback gaining steam ...$ Re: Alpha comeback gaining steam ...$ Re: AMD vs Intel: links to Compaq/HP# ANN: MBMON - a mailbox copy utility 1 Command line equivalent of "Compact This Folder"? 5 Re: Command line equivalent of "Compact This Folder"? ! Re: DCL enhancement more verb/FTP  Re: DS10L losing time  Re: DS10L losing time  Re: DS10L losing time  Re: DS10L losing time  Re: DS10L losing time  Re: FreeVMS 0.1.7 3 Re: Intel neuters Montvale, Itanic screams in alarm B Re: Note to the person at pool-70-110-191-53.phil.east.verizon.net Re: OpenVMS Hobby Program  Re: Revival of Alpha?  Re: Revival of Alpha?  shareable image, psect and data # Re: shareable image, psect and data  Re: Speaking of promoting VMS # RE: US military and confidentiality # Re: US military and confidentiality # Re: US military and confidentiality # Re: US military and confidentiality # RE: US military and confidentiality # Re: US military and confidentiality # Re: US military and confidentiality  Re: VMS Beginners FAQ on line  Re: VMS Beginners FAQ on line  Re: VMS Beginners FAQ on line B Re: WORKING_SET.COM gives all COM states when run at high priorityB Re: WORKING_SET.COM gives all COM states when run at high priorityB Re: WORKING_SET.COM gives all COM states when run at high priorityB Re: WORKING_SET.COM gives all COM states when run at high priorityB Re: WORKING_SET.COM gives all COM states when run at high priorityB Re: WORKING_SET.COM gives all COM states when run at high priorityB Re: WORKING_SET.COM gives all COM states when run at high priorityB Re: WORKING_SET.COM gives all COM states when run at high priorityB Re: WORKING_SET.COM gives all COM states when run at high priorityB Re: WORKING_SET.COM gives all COM states when run at high priority  F ----------------------------------------------------------------------    Date: 30 Jun 2005 08:12:52 -0700 From: carfer@gmail.com Subject: Alpha Basic 1.6C Message-ID: <1120144372.672046.249260@o13g2000cwo.googlegroups.com>    Hi guys,  @ Several months ago I posted an Alpha Basic bug when using Xfloat numbers in Def functions.   4 I've been told the problem is solved in 1.6 release.  G As a hobbiest user is quite difficult to me to get this media, but I've F reading, in this forum, that if a friend has it......I can borrow from him.  A So friends, has any one of you this kit available, near Barcelona ! (spain), or any other suggestion?    TIA    Carlos Fernandez   ------------------------------    Date: 30 Jun 2005 10:17:10 -0700 From: bob@instantwhip.com ) Subject: Alpha comeback gaining steam ... B Message-ID: <1120151830.616693.42250@g14g2000cwa.googlegroups.com>  0 maybe I wasn't so out of touch after all as some here thought ...  ) http://www.theinquirer.net/?article=24284    ------------------------------   Date: 30 Jun 2005 17:32:11 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)- Subject: Re: Alpha comeback gaining steam ... + Message-ID: <3iiokrFlqa0gU1@individual.net>   B In article <1120151830.616693.42250@g14g2000cwa.googlegroups.com>, 	bob@instantwhip.com writes:2 > maybe I wasn't so out of touch after all as some > here thought ... > + > http://www.theinquirer.net/?article=24284  >   4 Ahhhhh.....   If only it were true.    Sigh.........   bill   --  J 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: Thu, 30 Jun 2005 13:29:52 -0400 6 From: Brad Hamilton <brMadAhaPmiSlton@coMmcAasPt.Snet>- Subject: Re: Alpha comeback gaining steam ... 0 Message-ID: <o7udnVqWC_KNsVnfRVn-gw@comcast.com>   bob@instantwhip.com wrote:2 > maybe I wasn't so out of touch after all as some > here thought ... > + > http://www.theinquirer.net/?article=24284  >   F You'll notice that the "rumors" mentioned in the first paragraph link G back to the earlier "rumor" article published in the Inquirer.  Hardly  I suggestive of a big push back to Alpha.  Sounds more like a ploy to make    their rumor-mongering seem true.   --   Bradford J. Hamilton "All opinions are my own" * "Lose the MAPS, and replace '-at-' with @"   ------------------------------  % Date: Thu, 30 Jun 2005 13:53:18 -0400 % From: JF Mezei <jfmezei@teksavvy.com> - Subject: Re: Alpha comeback gaining steam ... , Message-ID: <42C43187.734DBF6F@teksavvy.com>   Brad Hamilton wrote:G > You'll notice that the "rumors" mentioned in the first paragraph link H > back to the earlier "rumor" article published in the Inquirer.  HardlyJ > suggestive of a big push back to Alpha.  Sounds more like a ploy to make" > their rumor-mongering seem true.    A And this may be a very smart ploy. It it grows enough to gain the F attention of Hurd, then Hurd will start to wonder why Alpha was killed= if it was so good, and realise that a very old Alpha is still D outperforming that IA64 thing and understand why customers are stillE prefering to buy Alpha and PaRisc systems over that IA64 thing. While H PARisc is safe for a few more years, Alpha systems are slated for end ofE sale very soon, and Hurd really must extend Alpha systems sales to at F least 2007 when supposedly great things will happen to HP's platforms.  B Realistically, I can seen existing EV7 selling through 2007, or HPB delivering EV79, or even just a speeded up EV7 done through better# manufacturing having higher yields.   G However that article is an attention grabber if that rumour does gather  steam.    ? It also makes one ask the question "could Alpha be restarted ?"   H If a lot of leading edge work done before the murder is still consideredE leading edge, perhaps EV8 isn't that far behind. (certaintly not that  far behind IA64)  D However, it would really take a miracle to see HP adopt Alpha as itsF primary platform for enterprise systems for VMS, HP-UX and Tandem. But> it would be a very wise move from a competitive point of view,5 especially if HP-UX on Alpha was really Digital Unix.    ------------------------------  % Date: Thu, 30 Jun 2005 11:12:53 +0200 3 From: Michael Unger <spam.to.unger@spamgourmet.com> - Subject: Re: AMD vs Intel: links to Compaq/HP + Message-ID: <3ihrjpFl8euoU1@individual.net>   & On 2005-06-29 23:58, "JF Mezei" wrote:  # > The AMD official complaint is at: ^ > http://www.amd.com/us-en/assets/content_type/DownloadableAssets/AMD-Intel_Full_Complaint.pdf > H > It is a scanned document, 48 pages (so I can't copy interesting text).           ^^^^^^^^^^^^^^^^  G It isn't. According to the document's metadata is has been written with G Microsoft Word, named "Complaint.doc", and converted to PDF using Adobe F Distiller 6.0 for Windows (on 29-Jun-2005, 15:09, local time offset ofE -5:00 to UTC) to be readable with Acrobat (Reader) V5.x and up, i.e.,  PDF version 1.4.  F > It is a very readable document, almost like a novel, especially whenJ > they start listing all sorts of events when intel intervened to prevenst/ > computer manufacturers from buying AMD chips.   $ To quote from the introductory part:   <quote> F 2. Just like Standard Oil and Alcoa before it, for over a decade IntelC has unlawfully maintained its monopoly by engaging in a relentless, C worldwide campaign to coerce customers to refrain from dealing with  AMD. Among other things,  C * Intel has forced major customers into exclusive or near-exclusive  deals;? * it has conditioned rebates, allowances and market development F funding on customers’ agreement to severely limit or forego entirely purchases from AMD; = * it has established a system of discriminatory, retroactive, E first-dollar rebates triggered by purchases at such high levels as to ? have the practical and intended effect of denying customers the B freedom to purchase any significant volume of processors from AMD;A * it has threatened retaliation against customers introducing AMD > computer platforms, particularly in strategic market segments;< * it has established and enforced quotas among key retailers: effectively requiring them to stock overwhelmingly, if notC exclusively, Intel-powered computers, thereby artificially limiting  consumer choice;@ * it has forced PC makers and technology partners to boycott AMD  product launches and promotions;? * and it has abused its market power by forcing on the industry D technical standards and products which have as their central purpose+ the handicapping of AMD in the marketplace.   J 3. Intel’s economic coercion of customers extends to all levels – from; large computer-makers like Hewlett-Packard and IBM to small F system-builders to wholesale distributors to retailers such as CircuitE City. All face the same choice: accept conditions that exclude AMD or D suffer discriminatory pricing and competitively crippling treatment.E In this way, Intel has avoided competition on the merits and deprived H AMD of the opportunity to stake its prices and quality against Intel’s( for every potential microprocessor sale. </quote>   > [...]    Michael    --  ; Real names enhance the probability of getting real answers. 5 My e-mail account at DECUS Munich is no longer valid.    ------------------------------  % Date: Thu, 30 Jun 2005 11:19:58 -0500 - From: Hunter Goatley <goathunter@goatley.com> , Subject: ANN: MBMON - a mailbox copy utility( Message-ID: <42C41BAE.80101@goatley.com>  ? Thanks to Ian Miller, MBMON can now be found in my VMS freeware  archive.   File:           mbmon.zip B Description:    A progam to copy messages written to a mailbox and>                  display and/or record them for later display. Version:        V1.05 Author:         Ian Miller <miller@Encompasserve.org>  Architecture:   Alpha  Size:           279 blocks Language:       C 8 URL:            http://eisner.encompasserve.org/~miller/ Released:       24-JUN-2005    Thanks, Ian!   http://www.process.com/openvms/   5 ftp://ftp.process.com/vms-freeware/fileserv/mbmon.zip : http://vms.process.com/ftp/vms-freeware/fileserv/mbmon.zip  1 ftp://ftp.tmk.com/vms-freeware/fileserv/mbmon.zip 6 http://www.tmk.com/ftp/vms-freeware/fileserv/mbmon.zip   And on the other mirrors.      Hunter ------9 Hunter Goatley, Process Software, http://www.process.com/ ; <goathunter@GOATLEY.COM>     http://www.goatley.com/hunter/    ------------------------------  # Date: Thu, 30 Jun 2005 14:04:53 GMT 0 From: spud_demon@e.thundermaker.net (Spud Demon): Subject: Command line equivalent of "Compact This Folder"?/ Message-ID: <9_Swe.15$ss3.3416@news.uswest.net>   K Onsite I use Mozilla over IMAP to read my VMS mail.  I have filters set up. 6 Offsite I use Yahmail, which runs as a CGI under CSWS.  J Yahmail doesn't have a filter capability, but I know that if I run MozillaL and then "Compress This Folder" on Inbox, Yahmail gets the filter changes.  I This is true even if I compress it from a different IMAP connection, e.g. 5 the Mozilla on my Powermac, which has no filters set.   F I can leave Mozilla running and set it to check the inbox often, but IH haven't figured out a way to tell it to compress the folder on a regularL basis.  If I knew how to do it from the command line I could make it a batch job.  Any hints?  . --Spud Demon		spud_demon -at- thundermaker.net   ------------------------------  # Date: Thu, 30 Jun 2005 17:40:15 GMT + From: Jeff Chimene <jchimene@earthlink.net> > Subject: Re: Command line equivalent of "Compact This Folder"?C Message-ID: <38Wwe.11823$hK3.9219@newsread3.news.pas.earthlink.net>    Spud Demon wrote: M > Onsite I use Mozilla over IMAP to read my VMS mail.  I have filters set up. 8 > Offsite I use Yahmail, which runs as a CGI under CSWS. > L > Yahmail doesn't have a filter capability, but I know that if I run MozillaN > and then "Compress This Folder" on Inbox, Yahmail gets the filter changes.  K > This is true even if I compress it from a different IMAP connection, e.g. 7 > the Mozilla on my Powermac, which has no filters set.  > H > I can leave Mozilla running and set it to check the inbox often, but IJ > haven't figured out a way to tell it to compress the folder on a regularN > basis.  If I knew how to do it from the command line I could make it a batch > job.  Any hints? > 0 > --Spud Demon		spud_demon -at- thundermaker.net  F Well, this is a forum about VMS, not Mozilla. So, here's one answer in> Perl: http://search.cpan.org/~kjohnson/NetxAP-0.02/Net/IMAP.pm  5 You might try asking your question on MozillaZine.org    Cheers,  jec    ------------------------------    Date: 30 Jun 2005 07:22:38 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) * Subject: Re: DCL enhancement more verb/FTP3 Message-ID: <DJz$tSN4xqut@eisner.encompasserve.org>   ` In article <42C35B89.60BFF059@comcast.net>, David J Dachtera <djesys.nospam@comcast.net> writes: > Bob Koehler wrote: >  > Does this help?   D    Yes.  Remote open is somthing that FTP cannot do.  As long as youH    don't mix reads and writes, or do positioning, FTP can look like it's    doing it.   ------------------------------    Date: 30 Jun 2005 07:17:10 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: DS10L losing time3 Message-ID: <WvOlS$NFT0rg@eisner.encompasserve.org>   S In article <42c30c96@duster.adelaide.on.net>, "Gremlin" <not-here@all.mate> writes:  > Hi All > N > My DS10L can't seem to keep time properly - it gets being by about 1 minute M > a day, so, after a couple of weeks, 15 minutes behind etc.  Is this a sign  O > of something broken/about to break?  I don't remember the "old" h/w I worked   > on being like this.   G    I've always found that VMS keeps better time than most OS, and VAXen B    kept better tie than Alphas, but I find a little drift on everyH    system.  At home and at work I use NTP on all my systems, independent    of architecture or OS.    ------------------------------  % Date: Thu, 30 Jun 2005 09:22:29 -0500 & From: Thomas Wirt <twnews@kittles.com> Subject: Re: DS10L losing timeE Message-ID: <4090f$42c40026$4367aba2$20334@msgid.meganewsservers.com>    Gremlin wrote: > Hi All > N > My DS10L can't seem to keep time properly - it gets being by about 1 minute M > a day, so, after a couple of weeks, 15 minutes behind etc.  Is this a sign  O > of something broken/about to break?  I don't remember the "old" h/w I worked   > on being like this.   H The time on your running system is kept by counting CPU ticks.  This is H not a very accurate way to keep time.  Sometimes a busy system does not I count ticks correctly and you loose time.  "SET TIME" with no qualifiers  F causes VMS to read and update the system time from the systems clock.  Below is an example from Help.             2.$ SET TIME             $ SHOW TIME #             14-DEC-2001 03:21:27.53   E             The SET TIME command in this example sets the system time F             according to the time-of-year clock. The SHOW TIME command3             requests a display of the current time.   G I recommend that you synchronize with an accurate Internet time source  I such as the USNO every day.  This way you do not rely on the accuracy of  F the system clock either.  Any newer TCP/IP stack that you have should  include NTP.   Hope that helps. --     Thomas Wirt  Systems Manager  Kittle's Home Furnishings  Indianapolis, IN   ------------------------------  % Date: Thu, 30 Jun 2005 09:42:42 -0500 & From: Thomas Wirt <twnews@kittles.com> Subject: Re: DS10L losing timeE Message-ID: <62fc8$42c404e3$4367aba2$20684@msgid.meganewsservers.com>    Gremlin wrote:   > Hi All > N > My DS10L can't seem to keep time properly - it gets being by about 1 minute M > a day, so, after a couple of weeks, 15 minutes behind etc.  Is this a sign  O > of something broken/about to break?  I don't remember the "old" h/w I worked   > on being like this.   I On a running system the time is kept by counting the CPU ticks.  This is  F not a very accurate way to keep time.  The "set time" command with no I qualifiers can be used to have VMS read and use the time from the system  I clock (same source of time as on boot).  See the example below from Help.              2.$ SET TIME             $ SHOW TIME #             14-DEC-2001 03:21:27.53   E             The SET TIME command in this example sets the system time F             according to the time-of-year clock. The SHOW TIME command3             requests a display of the current time.   E I recommend that you synchronize daily with a reliable Internet time  D source such as USNO (they keep official US time by law).  Any newer I version of TCP should include NTP which can be used to do a sync.  We do  I a single sync each night followed by a "set time/cluster".  This insures  9 that our cluster is mostly in sync and has accurate time.    Hope this helps. --     Thomas Wirt  Systems Manager  Kittle's Home Furnishings  Indianapolis, IN   ------------------------------  # Date: Thu, 30 Jun 2005 15:55:38 GMT # From: hoff@hp.nospam (Hoff Hoffman)  Subject: Re: DS10L losing time2 Message-ID: <_BUwe.7859$nM5.4079@news.cpqcorp.net>  S In article <42c30c96@duster.adelaide.on.net>, "Gremlin" <not-here@all.mate> writes:   - :My DS10L can't seem to keep time properly...   F   It's a computer, not a chronograph.  It's traditionally precise, but   not necessarily accurate.   F   If you need high(er) accuracy -- as opposed to precision -- you needF   to synchronize with an external time-base, either via a network timeE   protocol, or via a hardware time-base receiver, or a modem dial-up,    or...   M :                                          - it gets being by about 1 minute  L :a day, so, after a couple of weeks, 15 minutes behind etc.  Is this a sign N :of something broken/about to break?  I don't remember the "old" h/w I worked  :on being like this.  D   The OpenVMS FAQ has a section on this particular topic, within theG   FAQ chapter on time and time-keeping.  Also on maintaining the clock. G   (This topic arises with some regularity here in the 'group, hence its    inclusion in the FAQ.)  G   There's no specification on the Alpha software clock accuracy that I  H   am aware of -- this is a common misconception.  The software clock canF   and does drift, which is why packages such as NTP and DTSS are used.G   Third-party hardware options include cesium clocks and GPS receivers.   F   A minute-a-day isn't a markedly huge drift in the software time, butI   I would still check for hardware errors or for other high-IPL activity.         N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq N  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------  % Date: Thu, 30 Jun 2005 12:40:10 -0500 & From: Thomas Wirt <twnews@kittles.com> Subject: Re: DS10L losing timeE Message-ID: <d9386$42c42e7d$4367aba2$23556@msgid.meganewsservers.com>    Gremlin wrote: > Hi All > N > My DS10L can't seem to keep time properly - it gets being by about 1 minute M > a day, so, after a couple of weeks, 15 minutes behind etc.  Is this a sign  O > of something broken/about to break?  I don't remember the "old" h/w I worked   > on being like this.   I On a running system the time is kept by counting the CPU ticks.  This is  F not a very accurate way to keep time.  The "set time" command with no I qualifiers can be used to have VMS read and use the time from the system  I clock (same source of time as on boot).  See the example below from Help.              2.$ SET TIME             $ SHOW TIME #             14-DEC-2001 03:21:27.53   E             The SET TIME command in this example sets the system time F             according to the time-of-year clock. The SHOW TIME command3             requests a display of the current time.   E I recommend that you synchronize daily with a reliable Internet time  D source such as USNO (they keep official US time by law).  Any newer I version of TCP should include NTP which can be used to do a sync.  We do  I a single sync each night followed by a "set time/cluster".  This insures  9 that our cluster is mostly in sync and has accurate time.    Hope this helps.     Thomas Wirt    ------------------------------    Date: 30 Jun 2005 09:13:23 -0700' From: "marius  popa" <mapopa@gmail.com>  Subject: Re: FreeVMS 0.1.7C Message-ID: <1120148003.925438.150530@g43g2000cwa.googlegroups.com>   , niice , time to see how it works in bochs ;)   ------------------------------   Date: 30 Jun 2005 17:26:28 GMT$ From: "Doc." <doc@openvms-rocks.com>< Subject: Re: Intel neuters Montvale, Itanic screams in alarm7 Message-ID: <Xns9685C5DD5161Ddcovmsrox@212.100.160.126>   " %NEWS-I-NEWMSG, Dr. Dweeb wrote in5 news:42c41e15$0$67262$157c6196@dreader2.cybercity.dk     <snip>  * > I did not learn anything from this post.   <snip>  E > A credible link would be a start, rather than an apparent repeat of  > JF's old posts.   D I have no idea how credible you consider WikiNews, but the following0 does link to a few other articles on the topic.   i http://en.wikinews.org/wiki/U.S._wants_to_impose_its_%22no-fly%22_list_on_most_Canadian_domestic_flights    ? Admittedly, I don't want to encourage JF to contribute to WN...      Doc. --  G OpenVMS:     Eight out of ten hackers prefer *other* operating systems. G http://www.openvms-rocks.com    Deathrow Public-Access OpenVMS Cluster.    ------------------------------  % Date: Thu, 30 Jun 2005 09:15:38 -0400 4 From: "Peter Weaver" <newsgroup@weaverconsulting.ca>K Subject: Re: Note to the person at pool-70-110-191-53.phil.east.verizon.net 8 Message-ID: <jgSwe.3433$Ai.543911@news20.bellglobal.com>  8 "John E. Malmberg" <wb8tyw@qsl.network> wrote in message+ news:yemdneUw3-Kv4l7fRVn-qg@adelphia.com...  > Peter Weaver wrote: H > > Whoever is at  pool-70-110-191-53.phil.east.verizon.net please run a virus E > > checker, you have the Mytob virus and you have my address in your  address E > > book. So far you have sent me 23 copies of the virus, but it does 
 nothing to
 > > VMS :) > I > It may have also picked it up off of the internet, not just the address 4 > book.  So the system owner may not be seeing this. > ...   G Possible, but the address that is receiving this has not been published J anywhere on the web. The address was used in communications with people at HP and people in COV.    ------------------------------  # Date: Thu, 30 Jun 2005 07:09:23 GMT & From: "D Henigman" <dhenigman@shaw.ca>" Subject: Re: OpenVMS Hobby Program. Message-ID: <DUMwe.1835571$6l.309749@pd7tw2no>  . <susan_skonetski@hotmail.com> wrote in message< news:1119740832.130195.64800@z14g2000cwz.googlegroups.com... > Dear Newsgroup,  > H > Talk to me about the OpenVMS Hobbyist Program.  I really can't explain; > why I am asking, you just have to trust me in a good way.  > F > THE PROGRAM IS NOT AT RISK AT ALL I AM DOING PERSONAL RESEARCH - VMS* > Engineering is commited to this program. > F > I remember when we were at DECUS and Pat J. got up and asked the DECC > listens panel about the concept, there was a huge applause and it H > seemed like the right thing to do. I have to be honest and say all theE > friends of the DFW LUG knew what was coming and we were holding our F > breath. So the DFW LUG, Dave Cathy and John Wisniewski in particularH > took on the work of doing pretty much the whole program.  And now Dave( > alone.  Anyway I am making myself sad. > H > I have a few questions for this newsgroup.  You can reply here or send5 > me mail either to my hotmail account or my hp mail.  > 0 > Do you find the VMS Hobbyist Program valuable? > D Extremely valuable. I especially like the products commonly found onL Unix/Linux that HP/CPQ and others have been porting to VMS. Apache/PHP/MySQL2 etc..(now if the GUI's could be ported as well :-)  ' > How do you use your Hobbyist license?  > J Education mostly. As a web and database (mysql) server for my son's sports1 teams. And as a safe place to back up my home PC!   F > Since the license needs to be renewed each year, how do you do that?  . Montagar/email. Cut and paste the ones I need.   > Do you get the CD's?  : I have once or twice but don't need to upgrade very often.   > D > Give me one improvement you would make - be realistic (every hobbyH > costs something) even if you collected worms you would need a place to > put them.   9 I would like to see the Galaxy license added to the list. B That is a subject I would like to do more in than just read about.   > " > What type of machine do you use?  I A quad processor AS4100 but generally only run with 2 CPUs since I have a J hard time putting much of a load on it. Saves power and I have spares if a CPU board dies.      > I > The hobbyist program is currently available on VAX, Alpha and Integrity A > how do you think folks will use the hobby license on Integrity?  > I When I can find them and spare parts on Ebay. This is a hobby but I don't 6 want a box I can't fix or replace easily (ie cheaply).   >  > Thanks for your time.  >  > Sue  >    ------------------------------  % Date: Thu, 30 Jun 2005 10:33:37 +0200  From: S <soterroatyahoodotcom> Subject: Re: Revival of Alpha?& Message-ID: <42c3ae5e$1@news1.ethz.ch>   Dave Froble wrote:H > Excuse me.  I must have missed the announcement of Macro-32 and BLISS E > compilers for Power (And anything else 'John' wants to run VMS on).   I The guys from FreeVMS seem to be able to do something about it, so it is  	 doable...    S    ------------------------------  # Date: Thu, 30 Jun 2005 14:05:17 GMT * From: "FredK" <fred.nospam@nospam.dec.com> Subject: Re: Revival of Alpha?1 Message-ID: <x_Swe.7853$pR5.790@news.cpqcorp.net>   . "Tom Linden" <tom@kednos.com> wrote in message" news:opss25kykpzgicya@hyrrokkin...H > On Tue, 28 Jun 2005 09:49:37 -0400, John Smith <a@nonymous.com> wrote: >  > L > Well, it may not be too far-fetched, Fred.  Linux was booted onto ARC-onlyK > platforms using MILO, MIni LOader,  from a FAT partition and then linload  >   H The magic that has to do with booting is more complicated than where theI bits come from.  The VMS boot phase is also a bit more complex than Linux K or most UNIXes, as it is a highly modular executive that is effectively put D together on the fly during boot based on the HW and SYSGEN settings.I Linux and most UNIXes have a more monolithic kernel that contains all the 4 things needed for booting on the platform within it.  G If a system implements EFI, then the VMS bootstrap would remain largely K unchanged in basic operation, but would still require architecture-specific ? changes.  If it is not based on EFI (along with various PAL/SAL  infrastructure) E then the work would be much more involved (I shudder to contemplate a  BIOS based boot).   J In terms of the Alpha side of things, each platform family and potentiallyG family member would need a specific image built that knows the specific I topology of the system, and explicit knowledge of the core logic chipset.   G None of these things are beyond the scope of mere mortals, however they G would require a substantial subset of the VMS build environment (and in D the IPF case, the EFI toolkit and VMS specific code for it), someoneA knowedgeable in HW/FW and booting, and someone to hold their hand F while they learn to crawl.  More importantly, a business case would beJ needed to develop the build infrastructure, licensing, and support for 3rdK parties to do this - people seem to think that "opening" part or all of the  OS is free - it is not.   J But before any of this is remotely interesting for anything that is not anF Alpha or Itanium - there is the first hurdle which is MACRO-32, BLISS,C C (that is compatable with DECC), a LINKER, and an Image Activator. D Then comes the second hurdle, the executive code that implements theC core features needed - like changing modes, ASTs, register sets and + call frames, process context swapping, etc.   I While the "bootstrap" was a big hurdle in moving to IPF, in terms of time J and effort - it was gated by these other even larger challenges.  RememberF that we *booted* VMS on IPF a very looong time ago.  Long, long before we released V8.0 on it.    ------------------------------  % Date: Thu, 30 Jun 2005 08:09:56 -0400 + From: Chip Coldwell <coldwell@gmail.nospam> ( Subject: shareable image, psect and data@ Message-ID: <Pine.LNX.4.61.0506300745510.6874@frank.harvard.edu>  H Suppose I have a program that defines an external variable such as this:   int foo;  
 int main() {     foo = 1;     return 0; }   E After compiling this source file, I ANALYZE/OBJECT/GSD the resulting  E object file and find two GSD subrecords associated with the variable    "foo", one PSECT and one SYMBOL:  -  	9)  Program Section Definition (EGSD$C_PSC) ?  		alignment: 16-byte boundary                      <-- psect 8   		attribute flags:   			(0)  EGPS$V_PIC           0   			(1)  EGPS$V_LIB           0   			(2)  EGPS$V_OVR           1   			(3)  EGPS$V_REL           1   			(4)  EGPS$V_GBL           1   			(5)  EGPS$V_SHR           0   			(6)  EGPS$V_EXE           0   			(7)  EGPS$V_RD            1   			(8)  EGPS$V_WRT           1   			(9)  EGPS$V_VEC           0   			(10) EGPS$V_NOMOD         0   			(11) EGPS$V_COM           0   		allocation: 4 (%X'00000004')   		symbol: "FOO"  /  	10)  Global Symbol Specification (EGSD$C_SYM)   		data type: DSC$K_DTYPE_Z (0)   		symbol flags:  			(0)  EGSY$V_WEAK          0   			(1)  EGSY$V_DEF           1   			(2)  EGSY$V_UNI           0   			(3)  EGSY$V_REL           1   			(4)  EGSY$V_COMM          0   			(5)  EGSY$V_VECEP         0   			(6)  EGSY$V_NORM          0   		psect: 8   		value: 0 (%X'00000000')  		symbol: "FOO"  I I found the following wise words in the Linker Utility Manual to explain   the first PSECT subrecord:  I "Several Digital programming languages, including VAX C and VAX FORTRAN,  K implement certain external variables as program sections with the overlaid  H (OVR), global (GBL), and relocatable (REL) attributes.  When the linker I processes these object modules, it overlays the program sections so that  G the various object modules that reference the variable access the same   virtual memory."  E (I'm on Alpha, not VAX, but the behavior in this respect seems to be  	 similar.)   0 If another module references the variable "foo":   extern int foo;    void incfoo(void)  { 
     foo++: }   4 it creates only the SYMBOL GSD subrecord (no PSECT):  5          1)  Global Symbol Specification (EGSD$C_SYM) -                  data type: DSC$K_DTYPE_Z (0)                   symbol flags:4                          (0)  EGSY$V_WEAK          04                          (1)  EGSY$V_DEF           04                          (2)  EGSY$V_UNI           04                          (3)  EGSY$V_REL           04                          (4)  EGSY$V_COMM          04                          (5)  EGSY$V_VECEP         04                          (6)  EGSY$V_NORM          0                  symbol: "FOO"  K Now for my question.  If I want to create a shareable image in which "foo"  J is an universal symbol, I must put it in the SYMBOL_VECTOR that I pass to J the linker utility.  What should go after the equals?  From analyzing the E first module, you could make a case for either FOO=PSECT or FOO=DATA  H (since both PSECT and SYMBOL GSD subrecords exist).  From analyzing the H second module, it looks like it should be FOO=DATA (since only a SYMBOL J GSD subrecord exists).  Or do all of these subrecords end up resolving to , the same memory so it really doesn't matter?   Chip   --   Charles M. "Chip" Coldwell Turn on, log in, tune out    ------------------------------  # Date: Thu, 30 Jun 2005 17:35:50 GMT # From: hoff@hp.nospam (Hoff Hoffman) , Subject: Re: shareable image, psect and data2 Message-ID: <W3Wwe.7869$m16.3606@news.cpqcorp.net>  n In article <Pine.LNX.4.61.0506300745510.6874@frank.harvard.edu>, Chip Coldwell <coldwell@gmail.nospam> writes:  I   You have indicated your proposed solution and some associated research, J   but haven't indicated your particular goal -- some additional backgroundI   here on the particular problem you seek to solve could be quite useful, F   particularly if we are to provide you with alternatives and options.  	 :int foo;   ?   Per traditional C, this is a source-module-level (file scope)    declaration...   :extern int foo;  B   An explicit external, and traditionally referencible outside the&   source module...  A "true" external.  C   The C compiler has an /EXTERN_MODEL specification, which can come    into play.  G   Take a look at the declaration scope for whatever C instantiation and C   compiler version you are using, too -- external, file, or module.   C   With shareable images, the LINKER adds what is sometimes called a D   UNIVERSAL, a transfer vector, or a symbol vector.  This allows theD   declaration to be visible to other and subsequent LINK operations.  L :Now for my question.  If I want to create a shareable image in which "foo" K :is an universal symbol, I must put it in the SYMBOL_VECTOR that I pass to   :the linker utility.    E   As a start on shareable images, see the Shareable Image Cookbook in F   the left navigation of the Ask The Wizard area, and see topic (2486)1   in the main section of the Ask The Wizard area.      N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq N  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------  % Date: Thu, 30 Jun 2005 08:45:39 -0400 # From: "John Smith" <a@nonymous.com> & Subject: Re: Speaking of promoting VMS4 Message-ID: <42c3ea83$1_1@spool9-west.superfeed.net>   Main, Kerry wrote: >> -----Original Message----- + >> From: John Smith [mailto:a@nonymous.com]  >> Sent: June 29, 2005 7:47 PM >> To: Info-VAX@Mvb.Saic.Com) >> Subject: Re: Speaking of promoting VMS  >> >> Main, Kerry wrote:  >>>> -----Original Message----- - >>>> From: John Smith [mailto:a@nonymous.com]   >>>> Sent: June 28, 2005 9:35 PM >>>> To: Info-VAX@Mvb.Saic.Com' >>>> Subject: Speaking of promoting VMS  >>>> >>>>F >>>> Sybase sent a broadcast e-mail today, promoting the use of Linux. >>>> >>>>@ >>>> HP does nothing other than produce VMS 'success' stories on >>>> glossy paper B >>>> that nobody sees, or get posted on a web site that they don't >>>> drive traffic >>>> to. >>>>? >>>> HP does nothing other than preach to the existing customer  >>>> base to say that = >>>> VMS is right for mission-critical computing, except when  >>>> Cerner or OM Group % >>>> drops a customer into their lap.  >>>> >>>> >>>> >>>> >>>> Sybase's e-mail below >>>> ----------------------  >>>>? >>>> Think open-source software can't cut it in the enterprise?  >>>>B >>>> Still consider Linux short on availability, manageability and >>>> scalability?  >>>>7 >>>> Unsure about vendor support and user endorsements?  >>>>= >>>> Attend this Web seminar and hear how Sybase and IBM have  >> teamed up to > >>>> deliver a complete end-to-end Linux-based data management >>>> solution that? >>>> meets demanding enterprise computing requirements for high  >>>> performance and >>>> low ownership cost. >>>> >>>> The Right Move ( >>>> Mission-Critical Solutions on Linux >>>> >>>  >>> = >>> Mmm.. Yep, Linux for the enterprise .. It is certainly an  >> option. ForE >>> DB vendors who license by CPU (and core) lots of smaller boxes is C >>> great as it means extra revenue for them - especially if the DB G >>> clustering options are included since there is additional costs per  >>> cpu involved.  >>> : >>> Not sure how Sybase licenses its cluster option, but I >> suspect that it >>> will also be cpu based.  >>> ? >>> However, one does need to consider the effort and resources 
 >> to QA/Test : >>> the Vendor recommended monthly *security* patches in a >> similar matter E >>> as Windows. QA/test and patch roll-out groups will love the extra 	 >>> work.  >>> D >>> Course, how important is security in most enterprises these days >>> anyway eh? >>>  >>> Reference:: >>> https://www.redhat.com/archives/enterprise-watch-list/4 >>> [click on thread for each month and add them up] >>> G >>> Last I counted, there was approx 140+ since Jan of this year (2005) D >>> alone. June so far has 27. Then count all of the ones in 2004 .. >>> ( >>> Gotta love the industry hype though. >> >>	 >> Kerry,  >>? >> As Reagan said to Carter during the debates, "Well there you  >> go again.". >> >  > G > Mmmm.. Now why does the thought "pot calling the kettle black come to  > mind?" >  > :-)  > G > John - senior mgmt in HP pays next to zero attention to newsgroups .. F > While there is certainly some good idea's that do filter out and up,F > the noise level is just way to high for them to pay any attention toB > them. Imagine trying to keep up with OpenVMS, NSK, HP-UX, Linux,@ > Windows newsgroups and every one of them playing the arm chair > quarterback position.  > A > Now that is not to say they do not listen to Customers or field D > employees - they host numerous Cust events around the globe in theF > form of Executive Councils, TAF's, Account Consulting sessions, freeF > on site migration workshops, Customer / ISV visits etc. In addition,H > BCS supports and funds the Cust Bootcamp and twice per year AmbassadorF > meetings in Nashua where they get first hand feedback from up to 200H > Ambassadors from around the globe (and believe me, Ambassadors are notB > shy about voicing their opinions to who ever from mgmt is in the > room).  F The Executive sessions are more like let's get together after a day of6 fishing and tell lies about the big one that got away.  L How many times has a large account told HP "we're leaving VMS behind" and HPG moved heaven and earth to get that customer to remain on VMS - upto and H including twisting the arm of the ISV who says that the key app that theK account relies on is EOL on VMS? You guys have the muscle but you never use K it to help VMS - you jsut stand there and say "We have some lovely Linux or K Windows machines we can sell you that have *just* the right ISV app you are L looking for.' It may keep the account for HP but it does *zero* for the restL of us VMS users - it's just another lost account and just one more reason to' de-emphasize VMS in HP executive minds.     F > I am not saying that marketing could not be improved, but in generalA > (with some exceptions), the brand marketing strategy vs product E > marketing strategy is a much higher corp strategy than OpenVMS (and  > BCS for that matter).    Marketing improved??!!!??? You're joking, right?  It's bleedin' non-existent.     D > Perhaps it will change. I certainly do not know if it will or not.E > There are certainly proponents on both sides in the corp hierarchy.  > H > However, I know you are a great supporter of OpenVMS who is frustratedH > with the current marketing, but the constant negativity in the OpenVMSD > newsgroup is not going to change a general corp marketing strategy > that is way above OpenVMS.  2 The only way HP is going to get me to be quiet is:: a) a hit team (which some here I'm sure would support), or< b) actually do some effective advertising & marketing of VMSG As to which HP will do, I assign a higher probability to a) than to b).      --5 OpenVMS -  411,000 installed systems and not counting       O ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==---- S http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups K ----= East and West-Coast Server Farms - Total Privacy via Encryption =----    ------------------------------  % Date: Thu, 30 Jun 2005 07:35:45 -0400 ' From: "Main, Kerry" <kerry.main@hp.com> , Subject: RE: US military and confidentialityR Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB650F72@tayexc19.americas.cpqcorp.net>   > -----Original Message-----1 > From: JF Mezei [mailto:jfmezei@teksavvy.com]=20  > Sent: June 29, 2005 4:32 PM  > To: Info-VAX@Mvb.Saic.Com . > Subject: Re: US military and confidentiality >=20 > "Main, Kerry" wrote:A > > If I am a Customer concerned about high security, there is=20  > no way in any B > > way shape or manner that I am going to publicly acknowledge=20 > what I am D > > running as the base for whatever application environment I have. >=20H > This loses credibility when the same customer allows all other vendorsF > to announce a sale to them, but when it is for VMS, the vendor never= > bothers asking permsission to the customer, assuming the=20  > response would
 > be "NO". >=20  @ JF - you are simply making things up to suit your own agenda.=20  G Advertising wins for secure environments like DOD is simply not done by F any vendor. Sure, Microsoft can state general things like "DOD will beE using Windows for an application to store information and make things C more efficient..". They may even be allowed to state an environment F within DOD, but you can bet it is not one which has any elevated levelE of security. It likely also would take them 6 months to a year to get E all the approvals - even to get a mention of an insecure environment.   F Heck, I work a lot with the Federal Govt here in Canada and since 9-11H and a few high profile internal audits, they are extremely nervous aboutD any perceived vendor bias and/or public attention. Its gotten to theH point that some managers do not let their staff receive any vendor mugs,; pens, free pizza lunches (when talking about business) etc.   E I find it really hard to believe that this is not also the same thing  that is happening in the US.  @ > And consider that for those military departments who really=20 > rely on VMS,A > it is now to their advantage to get HP to market those sales=20  > and marketH > VMS since not doing so jeoperdizes the long term viability of VMS, theH > availability fo software and more importantly, experienced VMS people. >=20  G Do you think the govt does not have strict rules about vendor bias when F speaking or being quoted publicly in any type of an official capacity?  
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------   Date: 30 Jun 2005 11:40:01 GMT( From: bill@cs.uofs.edu (Bill Gunshannon), Subject: Re: US military and confidentiality+ Message-ID: <3ii40hFkcoc0U1@individual.net>   3 In article <lz9dSnkRMaod@eisner.encompasserve.org>, 0 	Kilgallen@SpamCop.net (Larry Kilgallen) writes:a > In article <fLCdnZcNRsCvt17fRVn-ow@mpowercom.net>, "Jack Peacock" <peacock@simconv.com> writes:  > L >> Except employees have to be recruited to maintain any facility, even the N >> secret ones.  How can you hire anyone if you can't list the skills needed? L >> The location can be left out, but if there really were a significant VMS I >> military presence then there would be contractors advertising for VMS   >> admins/programmers. >  > And there are, all the time. > = >> for a job).  If there are VMS ads they are well-concealed.  > F > Not if you look at Dice.com.  Sometimes it seems that 1/3 of the VMS, > postings are military and 1/3 are banking.  @ How are they military?  I wasn't aware that Dice.com was part ofD USAREC.  Seriously though, I have been resgistered with dice.com andA monster.com for several years.  I haven't found anything of value B about either of them.  LMCO (one of my favorite example companies)@ has hundreds of jobs in the IT business available, none of which appear on either place,    bill   --  J 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: Thu, 30 Jun 2005 08:53:49 -0400 # From: "John Smith" <a@nonymous.com> , Subject: Re: US military and confidentiality4 Message-ID: <42c3ec6d$1_2@spool9-west.superfeed.net>   Main, Kerry wrote: >> -----Original Message----- / >> From: JF Mezei [mailto:jfmezei@teksavvy.com]  >> Sent: June 29, 2005 4:32 PM >> To: Info-VAX@Mvb.Saic.Com/ >> Subject: Re: US military and confidentiality  >> >> "Main, Kerry" wrote: > >>> If I am a Customer concerned about high security, there is >> no way in any? >>> way shape or manner that I am going to publicly acknowledge  >> what I amD >>> running as the base for whatever application environment I have. >>A >> This loses credibility when the same customer allows all other B >> vendors to announce a sale to them, but when it is for VMS, theH >> vendor never bothers asking permsission to the customer, assuming the >> response would  >> be "NO".  >> > ? > JF - you are simply making things up to suit your own agenda.  > F > Advertising wins for secure environments like DOD is simply not doneC > by any vendor. Sure, Microsoft can state general things like "DOD C > will be using Windows for an application to store information and E > make things more efficient..". They may even be allowed to state an E > environment within DOD, but you can bet it is not one which has any E > elevated level of security. It likely also would take them 6 months B > to a year to get all the approvals - even to get a mention of an > insecure environment.  > H > Heck, I work a lot with the Federal Govt here in Canada and since 9-11D > and a few high profile internal audits, they are extremely nervousE > about any perceived vendor bias and/or public attention. Its gotten D > to the point that some managers do not let their staff receive anyE > vendor mugs, pens, free pizza lunches (when talking about business)  > etc. > G > I find it really hard to believe that this is not also the same thing  > that is happening in the US. > > >> And consider that for those military departments who really >> rely on VMS, ? >> it is now to their advantage to get HP to market those sales 
 >> and market E >> VMS since not doing so jeoperdizes the long term viability of VMS, E >> the availability fo software and more importantly, experienced VMS 
 >> people. >> > D > Do you think the govt does not have strict rules about vendor biasC > when speaking or being quoted publicly in any type of an official  > capacity?     J This is quite true. I've been on the receiving end of this with one app weL built - at the time and for a long while it was the one and only solution inI the marketplace that a specific government dept./agency thought was great G and liked it to be widely used among capital markets participants, they J could not and would not open their mouths to recommend us to anyone. TheirG words to us were, "We cannot be seen favoring or promoting one vendor".    --F OpenVMS - The never-advertised operating system with the dwindling ISV base.       O ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==---- S http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups K ----= East and West-Coast Server Farms - Total Privacy via Encryption =----    ------------------------------   Date: 30 Jun 2005 13:17:57 GMT( From: bill@cs.uofs.edu (Bill Gunshannon), Subject: Re: US military and confidentiality+ Message-ID: <3ii9o5Flolf7U1@individual.net>   4 In article <42c3ec6d$1_2@spool9-west.superfeed.net>,& 	"John Smith" <a@nonymous.com> writes: > Main, Kerry wrote: >>H >> I find it really hard to believe that this is not also the same thing >> that is happening in the US.   @ Once again, with the exception of certain black projects all DODC (and I am sure other gov't procurements, but I only have first-hand D experience with DOD) procurements are public knowledge and the RFP'sD are required to be published in the CBD.  Explain to me why HP can'tD announce it is bidding on and later that it won some of these RFP's?  E >> Do you think the govt does not have strict rules about vendor bias D >> when speaking or being quoted publicly in any type of an official >> capacity? >  > L > This is quite true. I've been on the receiving end of this with one app weN > built - at the time and for a long while it was the one and only solution inK > the marketplace that a specific government dept./agency thought was great I > and liked it to be widely used among capital markets participants, they L > could not and would not open their mouths to recommend us to anyone. TheirI > words to us were, "We cannot be seen favoring or promoting one vendor".  >   < NUtt hat only means the gov't can't say "Company A is better: than sliced bread".  There is no bias in saying "Company A: bid on RFP 5X389275B and was selected and awarded Contract< Q5693D786 for which they will provide 10 HP XP1000's running9 OpenVMS Version 8.3."  No bias, just a statement of fact. > And then HP can say "We won Contract Q5693D786 and here is the0 reason why we were chosen #1 for this proposal."  @ Unless, HP never bids on any of these, not even as the equipment provider to someone like LMCO.   bill   --  J 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: Thu, 30 Jun 2005 10:27:23 -0400 ' From: "Main, Kerry" <kerry.main@hp.com> , Subject: RE: US military and confidentialityR Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB650F7E@tayexc19.americas.cpqcorp.net>    -----Original Message----- $ > From: bill@triangle.cs.uofs.edu=20A > [mailto:bill@triangle.cs.uofs.edu] On Behalf Of Bill Gunshannon  > Sent: June 30, 2005 9:18 AM  > To: Info-VAX@Mvb.Saic.Com . > Subject: Re: US military and confidentiality >=206 > In article <42c3ec6d$1_2@spool9-west.superfeed.net>,( > 	"John Smith" <a@nonymous.com> writes: > > Main, Kerry wrote: > >>B > >> I find it really hard to believe that this is not also the=20 > same thing! > >> that is happening in the US.  >=20B > Once again, with the exception of certain black projects all DODE > (and I am sure other gov't procurements, but I only have first-hand F > experience with DOD) procurements are public knowledge and the RFP'sF > are required to be published in the CBD.  Explain to me why HP can'tF > announce it is bidding on and later that it won some of these RFP's? >=20  E Bill, you do not seem to understand how the RFP stuff actually works.   H Vendors do not announce "we are bidding on XYZ RFP". It is always a gameA where you try to figure out who the competition is and what their E strategy will be.  Some of the games include submitting clarification B questions (these are allowed and the crown responds to all biddersD without any indication who asked the question) which will put one of; your perceived competitors in a tougher position to respond 
 successfully.   F RFP responding is always a big cat-n-mouse game and if not successful,@ no vendor will admit it lost or even participated in the bid.=20  G > >> Do you think the govt does not have strict rules about vendor bias F > >> when speaking or being quoted publicly in any type of an official > >> capacity? > >=20 > >=20A > > This is quite true. I've been on the receiving end of this=20  > with one app we B > > built - at the time and for a long while it was the one and=20 > only solution in> > > the marketplace that a specific government dept./agency=20 > thought was great ; > > and liked it to be widely used among capital markets=20  > participants, they@ > > could not and would not open their mouths to recommend us=20 > to anyone. TheirA > > words to us were, "We cannot be seen favoring or promoting=20  > one vendor". > >=20 >=20> > NUtt hat only means the gov't can't say "Company A is better< > than sliced bread".  There is no bias in saying "Company A< > bid on RFP 5X389275B and was selected and awarded Contract> > Q5693D786 for which they will provide 10 HP XP1000's running; > OpenVMS Version 8.3."  No bias, just a statement of fact. @ > And then HP can say "We won Contract Q5693D786 and here is the2 > reason why we were chosen #1 for this proposal." >=20B > Unless, HP never bids on any of these, not even as the equipment  > provider to someone like LMCO. >=20  G Just because you do not read about it in the press, assuming HP and its D partners does not bid OpenVMS (and win occasionally) on specific DOD@ contracts is ridiculous. DOD is likely one of the biggest single1 Customers that any vendor could potentially have.   ? And contrary to some of the allegations here, many DOD folks do E understand the issues of security when dealing with commodity off the D shelf server platforms and their vendor recommended monthly security patches.   Regards   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------   Date: 30 Jun 2005 15:03:14 GMT( From: bill@cs.uofs.edu (Bill Gunshannon), Subject: Re: US military and confidentiality+ Message-ID: <3iifthFlqa5jU1@individual.net>   R In article <FD827B33AB0D9C4E92EACEEFEE2BA2FB650F7E@tayexc19.americas.cpqcorp.net>,* 	"Main, Kerry" <kerry.main@hp.com> writes: >  >  -----Original Message----- % >> From: bill@triangle.cs.uofs.edu=20 B >> [mailto:bill@triangle.cs.uofs.edu] On Behalf Of Bill Gunshannon >> Sent: June 30, 2005 9:18 AM >> To: Info-VAX@Mvb.Saic.Com/ >> Subject: Re: US military and confidentiality  >>=20 7 >> In article <42c3ec6d$1_2@spool9-west.superfeed.net>, ) >> 	"John Smith" <a@nonymous.com> writes:  >> > Main, Kerry wrote:  >> >> C >> >> I find it really hard to believe that this is not also the=20 
 >> same thing " >> >> that is happening in the US. >>=20 C >> Once again, with the exception of certain black projects all DOD F >> (and I am sure other gov't procurements, but I only have first-handG >> experience with DOD) procurements are public knowledge and the RFP's G >> are required to be published in the CBD.  Explain to me why HP can't G >> announce it is bidding on and later that it won some of these RFP's?  >>=20  > G > Bill, you do not seem to understand how the RFP stuff actually works.   G Kerry, actually, I do.  I have done it from both sides.  I have written D RFP's and I have helped put together bids on them.  I haven't always& been in this academic ivory tower! :-)   > J > Vendors do not announce "we are bidding on XYZ RFP". It is always a gameC > where you try to figure out who the competition is and what their G > strategy will be.  Some of the games include submitting clarification D > questions (these are allowed and the crown responds to all bidders  @ Sorry, we don't have any crowns down here and the original topic was regarding the US DOD.   F > without any indication who asked the question) which will put one of= > your perceived competitors in a tougher position to respond  > successfully.   , That part is pretty much the same down here.   > H > RFP responding is always a big cat-n-mouse game and if not successful,B > no vendor will admit it lost or even participated in the bid.=20  B We don't care about the ones HP looses.  We want to hear about the> ones they win.  Down here, the RFP and in most cases the final@ contract are public information.  If you want, I can probably go@ out ont he web and find ancient reports of some of the ones that@ the companies I worked for did during my tenure with them.  None@ of it was secret and after winning we tended to plaster them allC over every headline we could grab.  It was free advertising for the B company and it tended to rub the competitors' noses in it as well.   Try google on "ulana2".    > H >> >> Do you think the govt does not have strict rules about vendor biasG >> >> when speaking or being quoted publicly in any type of an official  >> >> capacity?  >> >=20  >> >=20 B >> > This is quite true. I've been on the receiving end of this=20 >> with one app weC >> > built - at the time and for a long while it was the one and=20  >> only solution in ? >> > the marketplace that a specific government dept./agency=20  >> thought was great< >> > and liked it to be widely used among capital markets=20 >> participants, they A >> > could not and would not open their mouths to recommend us=20  >> to anyone. Their B >> > words to us were, "We cannot be seen favoring or promoting=20 >> one vendor".  >> >=20  >>=20 ? >> NUtt hat only means the gov't can't say "Company A is better = >> than sliced bread".  There is no bias in saying "Company A = >> bid on RFP 5X389275B and was selected and awarded Contract ? >> Q5693D786 for which they will provide 10 HP XP1000's running < >> OpenVMS Version 8.3."  No bias, just a statement of fact.A >> And then HP can say "We won Contract Q5693D786 and here is the 3 >> reason why we were chosen #1 for this proposal."  >>=20 C >> Unless, HP never bids on any of these, not even as the equipment ! >> provider to someone like LMCO.  >>=20  > I > Just because you do not read about it in the press, assuming HP and its F > partners does not bid OpenVMS (and win occasionally) on specific DODB > contracts is ridiculous. DOD is likely one of the biggest single3 > Customers that any vendor could potentially have.   C Then why doesn't HP come right out and show it!!!  That's the point C people are trying to make here.  The information isn't secret, it's E public "BY LAW" (at least down here). If this is valuable advertising E opportunity and HP doesn't use it, what conclusion should be derived? 4 Either they don't win any or they really don't care.   > A > And contrary to some of the allegations here, many DOD folks do G > understand the issues of security when dealing with commodity off the F > shelf server platforms and their vendor recommended monthly security
 > patches.  > As well we should.  But that is still side-stepping the issue.  @ Why doesn't HP take advantage of the tremendous free advertisingB availability they would get from publishing powerful news releases/ about these (supposed) major DOD contract wins?    bill   --  J 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: Thu, 30 Jun 2005 13:02:41 -0400 % From: JF Mezei <jfmezei@teksavvy.com> , Subject: Re: US military and confidentiality, Message-ID: <42C425AD.79741658@teksavvy.com>   John Smith wrote: I > and liked it to be widely used among capital markets participants, they L > could not and would not open their mouths to recommend us to anyone. TheirI > words to us were, "We cannot be seen favoring or promoting one vendor".     G This si quite different from the vendor getting permission to mention a ; sale to company XXX without revealing any specific details.     C When HP brags about Tandem machines at NASDAQ, do they give details > about what they really do ? No. "They just process millions of transactions".    C When Microsoft brags about selling to the US military, do you think H Microsoft reveales any details the military don't want revealed ? No. ItF isn't the military bragging about microsoft, it is microsoft bragging.  B There are instances howvere where outfits will brag about choosingG microsoft. I still remember the vancouver airport authority bragging in C its annual report about having standardized all purchasing of IT on F Microsoft software only, thus saving the organisation much money. (how< can removing the RFP process and competitive bidding save an organisation money ?).  F This was undoubdedly some statement with a goal of telling wall streetF analysts that YVR Inc had gone "mainstream" and followed all the otherE sheep.  But I saw it as a terrible management decision and certaintly B gave the airport an image of having far less safety. I don't trustC windows to handle the systems that help a 747 land at that airport. E (while the policy may have been intended only for office systems, the 1 wording had not restricted it to office systems).    ------------------------------    Date: 30 Jun 2005 08:15:49 -0700! From: susan_skonetski@hotmail.com & Subject: Re: VMS Beginners FAQ on lineC Message-ID: <1120144549.539795.154090@z14g2000cwz.googlegroups.com>    I think its a great idea.    sue    ------------------------------  # Date: Thu, 30 Jun 2005 15:45:28 GMT # From: hoff@hp.nospam (Hoff Hoffman) & Subject: Re: VMS Beginners FAQ on line2 Message-ID: <ssUwe.7858$nM5.1172@news.cpqcorp.net>  L In article <42C2E127.9000606@spam.com>, Didier MORANDI <no@spam.com> writes: :Did you know that one?  : 9 :http://seqaxp.bio.caltech.edu/www/vms_beginners_faq.html   B   This and other introductory and tutorial-related URLs are listed@   in the OpenVMS FAQ, in the section on finding introductory and   tutorial information.   = :That could be another subject for the VMS Technical Journal?   D   I wrote a finding-support-resources article a couple of VTJs back.B   Looks like a couple of the URLs from that article are now stale,+   but the bulk of the text still applies...   B     http://h71000.www7.hp.com/openvms/journal/v2/articles/atw.html  C   An update to that is reasonable fodder for the blog, and/or for a *   new or an updated VTJ article, I expect.   :I volunteer to write it.   @   Do submit the article proposal via the VTJ website, of course.  (     http://www.hp.com/go/openvms/journal  N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq N  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------  % Date: Thu, 30 Jun 2005 09:00:39 -0700 ' From: David Mathog <mathog@caltech.edu> & Subject: Re: VMS Beginners FAQ on line+ Message-ID: <da14v7$fu6$1@naig.caltech.edu>    Didier MORANDI wrote:  > Did you know that one? > : > http://seqaxp.bio.caltech.edu/www/vms_beginners_faq.html  B 1.  seqaxp has been defunct for many years now. Its name has been ? aliased to the replacement machine so as not to break old links = like this one. If you cite this URL in the future please use  ( http://saf.bio.caltech.edu/...  instead.  E 2.  The replacement machine was a small Sun server running Solaris 8. B It would have been linux but one key piece of software at the time  was not yet available for linux.  # 3.  The companion document is here:   ,    http://saf.bio.caltech.edu/vms_sheet.html   Regards,   David Mathog mathog@caltech.edu   ------------------------------    Date: 30 Jun 2005 00:20:51 -0700 From: dooleys@snowy.net.auK Subject: Re: WORKING_SET.COM gives all COM states when run at high priority C Message-ID: <1120116051.235432.184540@f14g2000cwb.googlegroups.com>   
 AEF wrote: > Hello, > D > WORKING_SET.COM, which is in SYS$EXAMPLES, lists all processes andI > various working set statistics for each. I recently found out that when H > you run it at high priority, most or all of the states show up as COM.) > The state is determined by the function  >  >     f$getjpi("","STATE") > H > The higher the priority, the higher the percentage of states which are > reported as COM. > H > Why should the result of this function be dependent on the priority at > which it is run? > 5 > (Tested on VMS v6.1 and v6.2, both on VAX systems.)  > 1 > (On some systems it may be called WORKSET.COM.)  > A This is not my experience - I see processes in LEF and HIB states * I suspect the command f$getjpi("","STATE")# should read f$getjpi('pid',"state") 2 as you loop through pids sing pid = f$pid(context)= f$getjpi("","STATE") returns the state of the current process    >From help lex f$getjpi    pid   @ Specifies the process identification (PID) number of the processA for which information is being reported. Specify the pid argument A as a character string expression. You can omit the leading zeros.   = If you specify a null string (""),  the current PID number is  used.   < You cannot use an asterisk (*)  or percent sign (%) wildcard? character to specify the pid argument in the F$GETJPI function, < as you can with the $GETJPI system service. To get a list of7 process identification numbers, use the F$PID function.    Phil   ------------------------------    Date: 30 Jun 2005 07:20:05 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) K Subject: Re: WORKING_SET.COM gives all COM states when run at high priority 3 Message-ID: <XoYP0pNZ1tji@eisner.encompasserve.org>   j In article <1120087981.900047.325720@f14g2000cwb.googlegroups.com>, "AEF" <spamsink2001@yahoo.com> writes: >  > Hello, > D > WORKING_SET.COM, which is in SYS$EXAMPLES, lists all processes andI > various working set statistics for each. I recently found out that when H > you run it at high priority, most or all of the states show up as COM.) > The state is determined by the function  >  >     f$getjpi("","STATE") > H > The higher the priority, the higher the percentage of states which are > reported as COM. > H > Why should the result of this function be dependent on the priority at > which it is run? > C    Because at high priority you get the CPU even though those other 3    processes are COM and at low priority you don't.   @    And getting information from another process means the kernelF    dispatches an AST to that process, which must become COM to execute    the AST.    ------------------------------    Date: 30 Jun 2005 07:35:56 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) K Subject: Re: WORKING_SET.COM gives all COM states when run at high priority 3 Message-ID: <jAAnv8lFLa86@eisner.encompasserve.org>   i In article <1120105148.505681.16650@o13g2000cwo.googlegroups.com>, "AEF" <spamsink2001@yahoo.com> writes:  > E > I don't see how this can be. If $getjpi ran in the target process's F > context, wouldn't its result go into a symbol in that target processH > and screw it up? Also, if the target process needs to run $getjpi thanH > I'd expect to see processes all show up as CUR, since they are running
 > as you say.   H    No.  The kernel queues an AST to the target process which runs in theH    context of that process and returns to a kernel routine that deliversD    the info to the requesting process.  Because of the timing of theH    data collection vs. the scheduler updating the state information only@    one process per CPU will show up as CUR, and that will be theF    requesting process (some nice VMS Engineer made it so decades ago).   ------------------------------    Date: 30 Jun 2005 08:03:07 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) K Subject: Re: WORKING_SET.COM gives all COM states when run at high priority 3 Message-ID: <D7z$F2RQx3fZ@eisner.encompasserve.org>   i In article <1120105148.505681.16650@o13g2000cwo.googlegroups.com>, "AEF" <spamsink2001@yahoo.com> writes:  >  > " > David B Sneddon - bigpond wrote:  = >> Does $GETJPI not operate within the context of the process > >> you want the information about?  In which case that process3 >> would need to "run" to return the information...   ; Correct, for those items not available from non-paged pool.   E > I don't see how this can be. If $getjpi ran in the target process's F > context, wouldn't its result go into a symbol in that target process  H No, $GETJPI knows nothing about "symbols" since they are DCL constructs.E $GETJPI executes special kernel ASTs to get information from P1 space  of other processes.   H > and screw it up? Also, if the target process needs to run $getjpi thanH > I'd expect to see processes all show up as CUR, since they are running
 > as you say.   H No, CUR is a process that has work to do and currently has a CPU devotedH to it.  All processes will show up as CUR only on a multiprocessor where$ there are as many CPUs as processes.   ------------------------------    Date: 30 Jun 2005 06:29:53 -0700$ From: "AEF" <spamsink2001@yahoo.com>K Subject: Re: WORKING_SET.COM gives all COM states when run at high priority C Message-ID: <1120138193.202168.136920@o13g2000cwo.googlegroups.com>    Larry Kilgallen wrote:k > In article <1120105148.505681.16650@o13g2000cwo.googlegroups.com>, "AEF" <spamsink2001@yahoo.com> writes:  > >  > > $ > > David B Sneddon - bigpond wrote: > ? > >> Does $GETJPI not operate within the context of the process @ > >> you want the information about?  In which case that process5 > >> would need to "run" to return the information...  > = > Correct, for those items not available from non-paged pool.  > G > > I don't see how this can be. If $getjpi ran in the target process's H > > context, wouldn't its result go into a symbol in that target process > J > No, $GETJPI knows nothing about "symbols" since they are DCL constructs.G > $GETJPI executes special kernel ASTs to get information from P1 space  > of other processes.  > J > > and screw it up? Also, if the target process needs to run $getjpi thanJ > > I'd expect to see processes all show up as CUR, since they are running > > as you say.  > J > No, CUR is a process that has work to do and currently has a CPU devotedJ > to it.  All processes will show up as CUR only on a multiprocessor where& > there are as many CPUs as processes.    ? Thank you all for your helpful answers. But I have some further 
 questions:  G This alls sound like subtle issues of timing and internals. Later I may A ask more specific questions to individual replies. Thank you all.   B For now I ask: Why does SHOW SYSTEM work "properly" at priority 15C while WORKSET.COM doesn't? What's different about SHOW SYSTEM? SHOW B SYSTEM, when run at priority 15, doesn't show all processes in COMF states like WORKSET.COM does. My first guess would be that SHOW SYSTEMF uses $GETJPI and that WORKSET.COM uses F$GETJPI which uses $GETJPI (orC SYS$GETJPI?). So why the difference in behavior? I don't know, so I  ask.  
 Thanks again.    ------------------------------  # Date: Thu, 30 Jun 2005 14:01:48 GMT " From:   VAXman-  @SendSpamHere.ORGK Subject: Re: WORKING_SET.COM gives all COM states when run at high priority 0 Message-ID: <00A460E3.3157FC3E@SendSpamHere.ORG>  j In article <1120098848.856048.170530@g44g2000cwa.googlegroups.com>, "AEF" <spamsink2001@yahoo.com> writes: >  >  >Jeff Cameron wrote:! >> On 6/29/05 4:33 PM, in article ? >> 1120087981.900047.325720@f14g2000cwb.googlegroups.com, "AEF" " >> <spamsink2001@yahoo.com> wrote: >> >> > >> > Hello,  >> >G >> > WORKING_SET.COM, which is in SYS$EXAMPLES, lists all processes and L >> > various working set statistics for each. I recently found out that whenK >> > you run it at high priority, most or all of the states show up as COM. , >> > The state is determined by the function >> > >> >     f$getjpi("","STATE")  >> >K >> > The higher the priority, the higher the percentage of states which are  >> > reported as COM.  >> >K >> > Why should the result of this function be dependent on the priority at  >> > which it is run?  >> >8 >> > (Tested on VMS v6.1 and v6.2, both on VAX systems.) >> >4 >> > (On some systems it may be called WORKSET.COM.) >> > >> > Thanks. >> >N >> This is because you arte at a high priority and when ever you are executingN >> the function f$getjpi, you will always be at the CUR or current state, thatI >> is the one process on the system that has the CPU above all others, so E >> anyone else that wants the CPU at that time will show up as COM or H >> computable. You are preempting all lower processes that want the CPU. >> >> Jeff  > A >Well, talk about the act of measuring affecting the measurement!   $ It's like the uncertainty principle. --  K VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM              5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  # Date: Thu, 30 Jun 2005 14:05:35 GMT " From:   VAXman-  @SendSpamHere.ORGK Subject: Re: WORKING_SET.COM gives all COM states when run at high priority 0 Message-ID: <00A460E3.B8C7F82E@SendSpamHere.ORG>  d In article <42C36821.2070605@bigpond.com>, David B Sneddon - bigpond <dbsneddon@bigpond.com> writes: >AEF mentioned in passing: >>   >> Jeff Chimene wrote: >>  
 >>>AEF wrote:  >>> 
 >>>>Hello, >>>>F >>>>WORKING_SET.COM, which is in SYS$EXAMPLES, lists all processes andK >>>>various working set statistics for each. I recently found out that when J >>>>you run it at high priority, most or all of the states show up as COM.+ >>>>The state is determined by the function  >>>> >>>>    f$getjpi("","STATE") >>>>J >>>>The higher the priority, the higher the percentage of states which are >>>>reported as COM. >>>>J >>>>Why should the result of this function be dependent on the priority at >>>>which it is run? >>> I >>>Because your process running working_set.com is using all the CPU? All 9 >>>other processes are forced to non-CUR states, e.g. COM  >>> = >>>For extra credit, code up a real-time CPU-burner that also K >>>allocates/frees a few mebibytes of VM. Fire off a few of those and watch  >>>with working_set.com  >>> 
 >>>Cheers, >>>jec >>   >>  D >> Thanks for your rapid response! But some of my processes are idle7 >> terminal sessions. Why are they waiting for the CPU?  > ; >Does $GETJPI not operate within the context of the process < >you want the information about?  In which case that process1 >would need to "run" to return the information...   A It depends upon the information that you want to gather.  Some of A what $GETJPI gathers can be obtained from JIB, PCB or PHD.  If it B is something in the process context, that needs to be gethered viaB an AST and that would require that the process becomes computable.   --  K VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM              5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  % Date: Thu, 30 Jun 2005 10:20:00 -0400  From: "Jilly" <jilly@hp.com>K Subject: Re: WORKING_SET.COM gives all COM states when run at high priority , Message-ID: <42c3ff99$1@usenet01.boi.hp.com>  0 "AEF" <spamsink2001@yahoo.com> wrote in message = news:1120138193.202168.136920@o13g2000cwo.googlegroups.com...  >  > Larry Kilgallen wrote:L >> In article <1120105148.505681.16650@o13g2000cwo.googlegroups.com>, "AEF" # >> <spamsink2001@yahoo.com> writes:  >> > >> >% >> > David B Sneddon - bigpond wrote:  >>@ >> >> Does $GETJPI not operate within the context of the processA >> >> you want the information about?  In which case that process 6 >> >> would need to "run" to return the information... >>> >> Correct, for those items not available from non-paged pool. >>H >> > I don't see how this can be. If $getjpi ran in the target process'sI >> > context, wouldn't its result go into a symbol in that target process  >>K >> No, $GETJPI knows nothing about "symbols" since they are DCL constructs. H >> $GETJPI executes special kernel ASTs to get information from P1 space >> of other processes. >>K >> > and screw it up? Also, if the target process needs to run $getjpi than K >> > I'd expect to see processes all show up as CUR, since they are running  >> > as you say. >>K >> No, CUR is a process that has work to do and currently has a CPU devoted K >> to it.  All processes will show up as CUR only on a multiprocessor where ' >> there are as many CPUs as processes.  >  > A > Thank you all for your helpful answers. But I have some further  > questions: > I > This alls sound like subtle issues of timing and internals. Later I may C > ask more specific questions to individual replies. Thank you all.  > D > For now I ask: Why does SHOW SYSTEM work "properly" at priority 15E > while WORKSET.COM doesn't? What's different about SHOW SYSTEM? SHOW D > SYSTEM, when run at priority 15, doesn't show all processes in COMH > states like WORKSET.COM does. My first guess would be that SHOW SYSTEMH > uses $GETJPI and that WORKSET.COM uses F$GETJPI which uses $GETJPI (orE > SYS$GETJPI?). So why the difference in behavior? I don't know, so I  > ask. >  > Thanks again.  >   J SHOW SYSTEM does not use $GETJPI to gather it's info.  In a perfect world K F$GETJPI would have been written such that it would only issue AST(s) when  M absolutely needed but I suspect that is has common routines that it uses for  5 most of the items and those routines do issue an AST.    Jilly    ------------------------------    Date: 30 Jun 2005 11:45:47 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) K Subject: Re: WORKING_SET.COM gives all COM states when run at high priority 3 Message-ID: <Drb1ItliyY2I@eisner.encompasserve.org>   j In article <1120138193.202168.136920@o13g2000cwo.googlegroups.com>, "AEF" <spamsink2001@yahoo.com> writes: > D > For now I ask: Why does SHOW SYSTEM work "properly" at priority 15E > while WORKSET.COM doesn't? What's different about SHOW SYSTEM? SHOW D > SYSTEM, when run at priority 15, doesn't show all processes in COMH > states like WORKSET.COM does. My first guess would be that SHOW SYSTEMH > uses $GETJPI and that WORKSET.COM uses F$GETJPI which uses $GETJPI (orE > SYS$GETJPI?). So why the difference in behavior? I don't know, so I  > ask.  C    SHOW SYSTEM dosn't use $GETJPI.  SHOW SYSTEM will only show data H    which is available in the kernel and simply reads that data directly.F    $GETJPI can get data which is only in process space as well as dataH    in the kernel so it delivers an AST to the target process as a single0    general way to make sure it can get the data.   ------------------------------    Date: 30 Jun 2005 10:19:19 -0700$ From: "AEF" <spamsink2001@yahoo.com>K Subject: Re: WORKING_SET.COM gives all COM states when run at high priority B Message-ID: <1120151959.149199.55920@z14g2000cwz.googlegroups.com>   dooleys@snowy.net.au wrote:  > AEF wrote:
 > > Hello, > > F > > WORKING_SET.COM, which is in SYS$EXAMPLES, lists all processes andK > > various working set statistics for each. I recently found out that when J > > you run it at high priority, most or all of the states show up as COM.+ > > The state is determined by the function  > >  > >     f$getjpi("","STATE") > >  [...] C > This is not my experience - I see processes in LEF and HIB states     F Did you bump up your priority to 15? My question was specific to that.    , > I suspect the command f$getjpi("","STATE")% > should read f$getjpi('pid',"state") 4 > as you loop through pids sing pid = f$pid(context)? > f$getjpi("","STATE") returns the state of the current process  >  > >From help lex f$getjpi  >  > pid  > B > Specifies the process identification (PID) number of the processC > for which information is being reported. Specify the pid argument C > as a character string expression. You can omit the leading zeros.  > ? > If you specify a null string (""),  the current PID number is  > used.     F Yes, I know. I just quickly typed the relevant function. Yes, it needs a proper argument.  G I mentioned WORKSET.COM and the relevant function. I thought that would G be enough. To really be specific and accurate, I would have had to post E the entire procedure! Well, releavant portions, but I didn't think it  was worth the trouble.   [...]    ------------------------------   End of INFO-VAX 2005.363 ************************                                                                                                                                                                                                                                                                                                                                                                                                                                                wus%]pu\%&pѳ$,yZO1r~TdG
 Gd"-cH	HĥP#;IEu`]opto8XPO{ަ`F]V튥QXv	FlE^[~+Y15
)ߋ(p5kW#PX/HV׹ǵp״|\hCٳ7*f;O*6g- i4`5jV%1jq+CXqzx|ev>#]{'$Vkⴊ	N"V,.6@aGhzv"4:joʅG:e}l
@e-J<5D3_ӃHlaؔOUr@=B0sbh`{roҴ,,!!qd@uר2%BT{1M̯JޑG427&
I
eYkL; cB,26<G i0ⶔf$J}h0`nQë̻̓J..'s:9.mVlkҫg"3f*}&[7<A_RDA)gpx+ф܏uF!x eb%HPz5q5$Xks_Sn?7yÌ+6(N,.. "FK~+Gif4P%u)
Om,w1d' wd^OtrlD9D`DDδO̘6*>Le֋`k쾄&y"la
K?ʏsI#躭7$Nˀt(Ӎ{vvx4y\VdmB{}#I<`}dà5sjԖ|df5U֠~Y/Q\ *%N%kD ,R	HoxkJk@5<za*zKY98t1Bz@Ia