1 INFO-VAX	Tue, 23 Oct 2001	Volume 2001 : Issue 590       Contents:
 Re: <None> ADV:CREDIT CARD PROCESSING Advanced Server  Re: ALGOL and 1022 database ( Alphacide poll results:  Where are they? Another PR debacle Re: BACKUP ACCVIO  Re: BACKUP ACCVIO  Re: BCC support in VMSMAIL Re: BCC support in VMSMAIL Re: BCC support in VMSMAIL Re: BCC support in VMSMAIL Re: BCC support in VMSMAIL RE: BCC support in VMSMAIL Re: BCC support in VMSMAIL Re: BCC support in VMSMAIL Re: BCC support in VMSMAIL Re: BCC support in VMSMAIL Re: BCC support in VMSMAILG Browsing a VMS FTP site, with Browser, fails to display names correctly K Re: Browsing a VMS FTP site, with Browser, fails to display names correctly K Re: Browsing a VMS FTP site, with Browser, fails to display names correctly K Re: Browsing a VMS FTP site, with Browser, fails to display names correctly P Re: Browsing a VMS FTP site, with Browser, fails to display names correctly corrP Re: Browsing a VMS FTP site, with Browser, fails to display names correctly corrP Re: Browsing a VMS FTP site, with Browser, fails to display names correctly corr$ Re: Compaq: VMS is alive and kicking$ Re: Compaq: VMS is alive and kicking$ RE: Compaq: VMS is alive and kicking$ RE: Compaq: VMS is alive and kicking$ Re: Compaq: VMS is alive and kicking$ Re: Compaq: VMS is alive and kicking$ Re: Compaq: VMS is alive and kicking$ RE: Compaq: VMS is alive and kicking$ Re: Compaq: VMS is alive and kicking$ Re: Compaq: VMS is alive and kicking$ Re: Compaq: VMS is alive and kicking$ Re: Compaq: VMS is alive and kicking$ Re: Compaq: VMS is alive and kicking4 Re: Converting an ODS-2 to an ODS-5 file system disk copy files in DCLedit  Re: DECNET ping equivalent? I Re: dynamic_cast failure in Compaq C++ version 6.3 on VMS on DEC Alpha... P Re: dynamic_cast failure in Compaq C++ version 6.3 on VMS on DEC Alpha... Alpha. Re: ECC  Re: ECC  Re: ECC @ Re: Fat Binaries? (was: Re: WRITEBOOT and dual-boot bootblocks?) fixing Error Count...  Re: fixing Error Count...  FTP timing out with OPCOM msg ! Re: FTP timing out with OPCOM msg ) Re: Fwd: Compaq: VMS is alive and kicking B Fwd: OpenVMS VMS721_DCL-V0100 Alpha V7.2-1 DCL ECO Summary[ONHOLD]F Re: Fwd: OpenVMS VMS721_DCL-V0100 Alpha V7.2-1 DCL ECO Summary[ONHOLD] Get a Macintosh ! Re: How know last system restart? ! Re: How know last system restart? ! Re: How know last system restart? ! Re: How know last system restart? ! Re: How know last system restart? ! Re: How know last system restart?   How to enable security audit...?$ Re: How to enable security audit...?$ Re: How to enable security audit...?$ RE: How to enable security audit...?( Re: HSZ40, KZPSA, AlphaStation 200 4/233( Re: HSZ40, KZPSA, AlphaStation 200 4/233 Re: I am changing jobs Re: I am changing jobs Re: I am changing jobs RE: I am changing jobs& Re: Major VMS reseller says VMS legacy& Re: Major VMS reseller says VMS legacy Re: Mitnick Uncaged  re: Mitnick Uncaged  Re: Mitnick Uncaged 
 mode nibble?? ) Re: New to OpenVMS - Looking for software ) Re: New to OpenVMS - Looking for software ) Re: New to OpenVMS - Looking for software ) Re: New to OpenVMS - Looking for software ) Re: New to OpenVMS - Looking for software > SAMBA VMS and Windows XP - Re: Windows XP reality check pleaseB Re: SAMBA VMS and Windows XP - Re: Windows XP reality check please Re: SET FILE/ATTRIB in C/ Shared file access under NT and VMS question...  Single or Multiple Sys Disks? ! Re: Single or Multiple Sys Disks? ! Re: Single or Multiple Sys Disks? ! Re: Single or Multiple Sys Disks?  Socket problem-URGENT " Re: SuperDLT on older (EV5) Alphas" Re: SuperDLT on older (EV5) Alphas; Re: TELNET sessions not closed, limit exceeded, how to fix? 3 The Navy competition may not be Microsoft, but Palm # RE: Windows XP reality check please # Re: Windows XP reality check please # Re: Windows XP reality check please # Re: Windows XP reality check please # Re: Windows XP reality check please # Re: Windows XP reality check please # Re: Windows XP reality check please # Re: Windows XP reality check please  Working sets Re: Working sets Re: Working sets Re: Working sets' Re: WRITEBOOT and dual-boot bootblocks? ' Re: WRITEBOOT and dual-boot bootblocks? ' Re: WRITEBOOT and dual-boot bootblocks? ' Re: WRITEBOOT and dual-boot bootblocks? ' Re: WRITEBOOT and dual-boot bootblocks? ' RE: WRITEBOOT and dual-boot bootblocks? / Re: X server and SET DISPLAY on different port? # Re: [MOZILLA] Burner of CPU time!!! # Re: [MOZILLA] Burner of CPU time!!! # Re: [MOZILLA] Burner of CPU time!!! # Re: [MOZILLA] Burner of CPU time!!! # Re: [MOZILLA] Burner of CPU time!!! # Re: [MOZILLA] Burner of CPU time!!!   F ----------------------------------------------------------------------    Date: 23 Oct 2001 08:39:14 -0500- From: koehler@encompasserve.org (Bob Koehler)  Subject: Re: <None> 3 Message-ID: <t$CIv$l9vw1M@eisner.encompasserve.org>   q In article <PRVA7.1078737$ai2.81685906@bin2.nnrp.aus1.giganews.com>, "Bill Todd" <billtodd@metrocast.net> writes:   I > What I've read suggests that the activation may be tied to the BIOS (in M > which case it will survive replacement of other hardware) or to some set of M > hardware components, not all of which need be present.  So single-component C > upgrades may not be a problem, but more ambitions revamps may be.  > L > The general reaction has not been positive, but whether it will suffice to" > stifle sales remains to be seen.  E    Right now I think there's a problem of lack of information.  Which B    "set of hardware components"?  And why should I have to contact    Microsoft if I change them?   ------------------------------   Date: Tue, 23 Oct 2001 00:23:29 & From: signup1@ecommercechargecards.com# Subject: ADV:CREDIT CARD PROCESSING ' Message-ID: <231.193398.783631@unknown>   : **********************************************************; To be removed from further mailings please respond to this  ( email with "remove" in the subject line.: **********************************************************   Dear Friend,  < Discover how you can accept credit cards directly from your ; website without ever having to purchase or lease expensive  8 credit card equipment. If you are interested in learning= more please click on the link below. INSTANT ACCOUNT SET-UPS, ; LOWEST RATES AND COMPLETE SHOPPINGCART SOLUTIONS AVAILABLE!    http://www.webmktplace.com    % Feel free to call us @ 1.800.288.7363   , If you prefer you can respond to this email@  & mailto:signup@ecommercechargecards.com  2 Please include NAME, PHONE# and best time to call.       IBS/PBS 
 7657 Winnetka  Canoga Park Ca   ------------------------------    Date: 23 Oct 2001 06:07:37 -0700) From: P.Young@unsw.EDU.AU (Patrick Young)  Subject: Advanced Server= Message-ID: <55f85d77.0110230507.20ca85fe@posting.google.com>    Hi,   J A quick question regarding Advanced Server. I'm not new to OpenVMS however= am somewhat new to Advanced Server "license tips and tricks".   H Some time ago we purchased 40 x PWLMXXXCA07.02 which has worked well forK our web maintainers to "map drive letters" from PCs to update CSWS (Apache) L web content. We are using these under OpenVMS 7.2-1 and Advanced Server 7.2a  H We seem to have run out of these now and want to do a number of upgradesN anyway during November/December concerning OpenVMS, Advanced Server, and CSWS.  D In relation to this move we have just purchased 50 x PWLMXXXCA07.03.  N The problem we have is that our web service is expanding - ie: within the next7 year we will want to use around 90 connection licenses.   . Can you trade in the PWLMXXXCA07.02 license???  K Sorry if I'm not making much sense here - anything realated to M$ causes me  much grief.    ------------------------------  ! Date: Tue, 23 Oct 01 07:07:16 GMT  From: jmfbahciv@aol.com $ Subject: Re: ALGOL and 1022 database+ Message-ID: <9r3f7q$iti$1@bob.news.rcn.net>   / In article <3BD40BEF.E3BBE416@duff-family.com>, (    P Duff <pduff@duff-family.com> wrote: >Larry Kilgallen wrote:  >>  F >> In article <1vjastcqtciaedqoi6mu08fp95tq9o6fls@4ax.com>, Alan Greig >> <a.greig@virgin.net> writes:  >>  E >> > I think I recall reading many years ago that 1022 was the serial B >> > number of the KL-10 the database was originally developed on. >>  I >> Whereas I believe it means the software ran on (PDP-)10s and (PDP-)20s  >> too (2).  1022. > B >I don't monitor this group at all, in fact, I didn't know that itF >existed until a former 1022 co-worker pointed it out to me yesterday.5 >Pursuant to this message and to the thread at large, 7 >a common guess that the name is borne of 10s and 20s,  ? >is undone by the fact that when 1022 was developed there were   >no DEC20s, only 10s. :(> >I worked w/ SH co-founders Andy Garland and Charlie Houseman @ >for about 10 years. Although I asked them and other co-workers,2 >AFAIK they never revealed the source of the name. > 8 >Re: expiration date, it's not entirely straightforward 7 >to defeat the date checking mechanism in that version, B >since the date is checked in several stages, at different places 5 >in the code, some of which happened at random times. 8 >If memory serves, there was one customer whose license = >required a copy that never expired, for whom I was directed  5 >to write a patch to JFCL out the expiration checks.  5 >No, I didn't keep a copy of the patch, so don't ask.  >   = So use FILDDT on the EXE to find the DATE UUO.  Usually there : will be just one with the date stored somewhere.  JFCL out9 the MOVEM.  If there's more than one...eh...I don't know. < I think I'd try to JFCL out the UUO and the next instruction (the error return).    /BAH  ' Subtract a hundred and four for e-mail.    ------------------------------  # Date: Tue, 23 Oct 2001 06:44:03 GMT * From: "Bill Todd" <billtodd@metrocast.net>1 Subject: Alphacide poll results:  Where are they? ? Message-ID: <Tk8B7.569359$Lw3.35223120@news2.aus1.giganews.com>   K While subsequent events may have pushed themselves into the foreground, I'm F still curious what the results of Terry's Alphacide poll (conducted at< tru64.org) were, since I don't recall seeing them published.   - bill   ------------------------------    Date: 23 Oct 2001 11:15:05 -0500 From: rivie@cougar.no.domain Subject: Another PR debacle 3 Message-ID: <slrn9tb5u5.7qi.rivie@cougar.no.domain>   H According to http://www.theinquirer.net/22100117.htm Red Hat is droppingB Linux on the Alpha because "Compaq is doing away with the platform soon".   --  
 Roger Ivie ivie@cc.usu.edu     > -----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----A http://www.newsfeeds.com - The #1 Newsgroup Service in the World! > -----==  Over 80,000 Newsgroups - 16 Different Servers! =-----   ------------------------------  # Date: Tue, 23 Oct 2001 12:35:10 GMT 8 From: "Steven Santinelli" <Santinelli@smscompNOSPAM.com> Subject: Re: BACKUP ACCVIO; Message-ID: <2udB7.1219$EO1.159072@news1.news.adelphia.net>   
 Hey Paul -  H I'd report it to CSC... They probably all ready have a fix they can send ya...      Steve     - "Paul Sture" <paul@sture.ch> wrote in message % news:VA.00000493.2b57f98f@sture.ch...  > Alpha VMS V7.2-1H1 > G > Before I report this to CSC, anyone experiencing BACKUP ACCVIOs? I've I > had it more than once now. Nothing relevant in the hardware error logs.  > C > One of my backups failed with an access violation on mounting the C > second tape volume. Volume 2 was not initialized by the operation * > (still had a saveset from August on it). >  > From the logfile:  >  > (23:12) $!  > (23:12) $    backup/noassist -( >           DSA127:[*...]/since=backup -, >           /ignore=(interlock,label_proc) -4 >           $1$MUA25:DSA127_SUN.bck /media=compact -E >           /label=(03WSG0,03WSG1,03WSG2,03WSG3,03WSG4,03WSG5,03WSG6) 2 > %BACKUP-I-RESUME, resuming operation on volume 2= > %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual < > address=0000000056415366, PC=0000000000000000, PS=00000000* > (00:03) $    bck_stat   = "4/%X1000000C" > (00:03) $    set noverify 9 > Status of this backup-operation :        (Severity : 4)  >  > Other details: > . > Tape type TZ88 on HSJ50-AX controller HSJ27.I > Disk being backed up is a 2 member shadowset. Each member consists of 3 < > striped (9 GB) disks on HSJ50-AX controllers HSJ29, HSJ31./ > Firmware version V57J-1 on these controllers.  >  > ___  > Paul Sture
 > Switzerland  >    ------------------------------   Date: 23 Oct 2001 08:19:21 CDT= From: wayne@tachysoft.xxx.062469.killspam.00be (Wayne Sewell)  Subject: Re: BACKUP ACCVIO. Message-ID: <oRiMdDNO4y6S@tachxxsoftxxconsult>  N In article <VA.00000493.2b57f98f@sture.ch>, Paul Sture <paul@sture.ch> writes: > Alpha VMS V7.2-1H1 > H > Before I report this to CSC, anyone experiencing BACKUP ACCVIOs? I've I > had it more than once now. Nothing relevant in the hardware error logs.  > D > One of my backups failed with an access violation on mounting the D > second tape volume. Volume 2 was not initialized by the operation * > (still had a saveset from August on it). >  > From the logfile:  >  > (23:12) $!  > (23:12) $    backup/noassist -( >           DSA127:[*...]/since=backup -, >           /ignore=(interlock,label_proc) -4 >           $1$MUA25:DSA127_SUN.bck /media=compact -E >           /label=(03WSG0,03WSG1,03WSG2,03WSG3,03WSG4,03WSG5,03WSG6) 2 > %BACKUP-I-RESUME, resuming operation on volume 2= > %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtual < > address=0000000056415366, PC=0000000000000000, PS=00000000* > (00:03) $    bck_stat   = "4/%X1000000C" > (00:03) $    set noverify 9 > Status of this backup-operation :        (Severity : 4)  >  > Other details: > . > Tape type TZ88 on HSJ50-AX controller HSJ27.I > Disk being backed up is a 2 member shadowset. Each member consists of 3 < > striped (9 GB) disks on HSJ50-AX controllers HSJ29, HSJ31./ > Firmware version V57J-1 on these controllers.  >   I I assume you are current with your backup patches.  There have been a few F accvios in backupshr.  I remember one case where the accvio was in theM condition handler.  The backup got some *other* error, such as input file not M found, and the bug in the condition handler essentially *converted* the error 
 to an accvio!    --  O =============================================================================== M Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxx : http://www.tachysoft.xxx/www/tachyon.html and wayne.html  K change .xxx to .com in addresses above, assuming you are not a spambot  :-) O =============================================================================== H Randolph Duke (in Trading Places): "Mother always said you were greedy.". Mortimer Duke: "She meant it as a compliment!"   ------------------------------  # Date: Tue, 23 Oct 2001 08:17:33 GMT L From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")# Subject: Re: BCC support in VMSMAIL 8 Message-ID: <00A03EE8.F4AEDB36@SSRL04.SLAC.STANFORD.EDU>  K In article <3BD3EC5E.E683DA66@compaq.com>, Guy Peleg <guy.peleg@compaq.com>  writes:    >  >Hello All,  > B >Lately I have been playing with the idea of adding BCC support to	 >VMSMAIL. / >I would like to get your feedback about this :  >  >Do we really need this ?   J First, despite the crankiness of the rest of this note, I want to say thatI I really appreciate your participating in this forum and giving customers * a say in the development process.  Thanks.  G If I were still on VMSMAIL I would really like this, but I (and all my  L command-line users) are using PMDF MAIL.  BCC gets used a fair amount here; I probably not daily, but regularly, and there are cases where nothing else  will answer the need.   I >Do we need to have  BCC support sending messages from VMSMAIL to VMSMAIL  >or VMSMAIL to Exchange only ?  N If the support isn't for everything - which probably means for every supportedL mail transport - it isn't worth doing.  Picture a user sending a message viaJ BCC; if it fails, or (worse) if the messages go through but the recipient I list isn't suppressed, that user will lose all faith in VMS MAIL and use   something else.     K I would rather see you fix the peculiarities in To: and CC: processing with , addresses that involve different transports.    J (First, the distinction between To: and CC: is lost and everything ends upK as a To: address.  Second, the only addresses a recipient on another system N sees are the ones on that system, so if they do the equivalent of a reply/all,M it only gets their local recipients, not the whole list it was sent to.  This K wreaked havoc around here, and was the reason why we made all our users useS PMDF MAIL instead.)    -- Alan     O ===============================================================================e0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056eM  Physical mail to: SSRL -- SLAC BIN 69, PO BOX 4349, STANFORD, CA  94309-0210 O ===============================================================================w   ------------------------------  % Date: Tue, 23 Oct 2001 15:15:42 +0200r& From: Guy Peleg <guy.peleg@compaq.com># Subject: Re: BCC support in VMSMAILe* Message-ID: <3BD56D7E.281B8BDA@compaq.com>    Thank you all for your feedback.8 Aside from relplies here I have received about 50 mails.  K General response is : we want BCC and  make sure we support all transports.s  S Well majority have spoken we will have BCC support soon. As soon as the developmentk  3 work is over I will post a note in this conference.qR I know there is a lot of work to be done in VMSMAIL, however we need to start with
 something.   Guy   * Alan Winston - SSRL Admin Cmptg Mgr wrote:  M > In article <3BD3EC5E.E683DA66@compaq.com>, Guy Peleg <guy.peleg@compaq.com>M	 > writes:a >k > >g
 > >Hello All,M > >aD > >Lately I have been playing with the idea of adding BCC support to > >VMSMAIL.i1 > >I would like to get your feedback about this :i > >t > >Do we really need this ?n >3L > First, despite the crankiness of the rest of this note, I want to say thatK > I really appreciate your participating in this forum and giving customers , > a say in the development process.  Thanks. >iH > If I were still on VMSMAIL I would really like this, but I (and all myM > command-line users) are using PMDF MAIL.  BCC gets used a fair amount here;iK > probably not daily, but regularly, and there are cases where nothing elsee > will answer the need.- >0K > >Do we need to have  BCC support sending messages from VMSMAIL to VMSMAILy  > >or VMSMAIL to Exchange only ? >tP > If the support isn't for everything - which probably means for every supportedN > mail transport - it isn't worth doing.  Picture a user sending a message viaK > BCC; if it fails, or (worse) if the messages go through but the recipientoJ > list isn't suppressed, that user will lose all faith in VMS MAIL and use > something else.n >0M > I would rather see you fix the peculiarities in To: and CC: processing with . > addresses that involve different transports. >cL > (First, the distinction between To: and CC: is lost and everything ends upM > as a To: address.  Second, the only addresses a recipient on another system P > sees are the ones on that system, so if they do the equivalent of a reply/all,O > it only gets their local recipients, not the whole list it was sent to.  ThiseM > wreaked havoc around here, and was the reason why we made all our users use  > PMDF MAIL instead.): >n	 > -- Alanl > Q > ===============================================================================e2 >  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUO >  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056DO >  Physical mail to: SSRL -- SLAC BIN 69, PO BOX 4349, STANFORD, CA  94309-0210oQ > ===============================================================================5   ------------------------------    Date: 23 Oct 2001 07:02:24 -0700( From: bob@instantwhip.com (Bob Ceculski)# Subject: Re: BCC support in VMSMAILc= Message-ID: <d7791aa1.0110230602.3d8b992c@posting.google.com>   f "Rich Jordan" <rjordan@mindspring.com> wrote in message news:<9r2nu6$9gs$2@slb6.atl.mindspring.net>... > Guy,L >      BCC would be a nice addition, but what our customers really need is aM > supported, functional smtp/mime compliant mailer without having to nuy PMDFoL > to get it.  Exchange is absolutely not an issue anywhere we work (even theK > PC side uses a better system than that) but working compatibility (if not I > full functionality) with older versions of VMSmail is necessary (if theBJ > changes you make are not backported).  In other words if someone sends aM > BCC'd mail to someone else on a V6.2 system, nothing screwy happens.   I'll I > see the earlier mention of the PMDF mail program, and raise you a fully G > functional and supported Pine (or equivalent) for VMS (with BCC :)...! > H > BTW, I've asked Process if there's any way they could sell/license theN > client programs only (assuming they could be made to work without PMDF beingJ > there) but it doesn't look like that will happen (at least I never heard > back about it).R > 
 > Rich Jordan8 > > > Guy Peleg wrote in message <3BD3EC5E.E683DA66@compaq.com>... > >r
 > >Hello All,t > > D > >Lately I have been playing with the idea of adding BCC support to > >VMSMAIL.l1 > >I would like to get your feedback about this :s > >  > .a > .e > .o > >Guy Peleg > >OpenVMS Engineering  O you can get mime parsing and attachment separation in vms pine!  imap4 servicespO from tcpware ip stack works great w/vmsmail ... but like this poster is saying, K we need to be able to use vmsmail as a screening tool for all our smtp mailsM which means mime parsing and attachment separation ... you don't need to looknL at pmdf as an example but even vms pine does this nicely ... if you could atL least give us vms pine mime and attachment functionality that would be a bigF improvement!  where has this been?  if pine can do it why not vmsmail?   ------------------------------    Date: 23 Oct 2001 07:03:47 -0700( From: bob@instantwhip.com (Bob Ceculski)# Subject: Re: BCC support in VMSMAIL = Message-ID: <d7791aa1.0110230603.2a6d2644@posting.google.com>*  X Guy Peleg <guy.peleg@compaq.com> wrote in message news:<3BD3EC5E.E683DA66@compaq.com>... > Hello All, > C > Lately I have been playing with the idea of adding BCC support to 
 > VMSMAIL.0 > I would like to get your feedback about this : >  > Do we really need this ?J > Do we need to have  BCC support sending messages from VMSMAIL to VMSMAIL > or VMSMAIL to Exchange only ?p > / > Thoughts comments and complaints are welcome.  >  > Guy Pelego > OpenVMS Engineeringm  O you can get mime parsing and attachment separation in vms pine!  imap4 serviceseI from tcpware ip stack works great w/vmsmail ... bcc would be nice, but we-H need to be able to use vmsmail as a screening tool for all our smtp mailM which means mime parsing and attachment separation ... you don't need to look1L at pmdf as an example but even vms pine does this nicely ... if you could atL least give us vms pine mime and attachment functionality that would be a bigF improvement!  where has this been?  if pine can do it why not vmsmail?   ------------------------------  + Date: Tue, 23 Oct 2001 14:27:02 +0000 (UTC)f From: david20@alpha1.mdx.ac.uk# Subject: Re: BCC support in VMSMAILe+ Message-ID: <9r3unm$5ou$1@aquila.mdx.ac.uk>e  h In article <d7791aa1.0110230603.2a6d2644@posting.google.com>, bob@instantwhip.com (Bob Ceculski) writes:Y >Guy Peleg <guy.peleg@compaq.com> wrote in message news:<3BD3EC5E.E683DA66@compaq.com>...eP >you can get mime parsing and attachment separation in vms pine!  imap4 servicesJ >from tcpware ip stack works great w/vmsmail ... bcc would be nice, but weI >need to be able to use vmsmail as a screening tool for all our smtp mailmN >which means mime parsing and attachment separation ... you don't need to lookM >at pmdf as an example but even vms pine does this nicely ... if you could atiM >least give us vms pine mime and attachment functionality that would be a big.G >improvement!  where has this been?  if pine can do it why not vmsmail?D    G I thought the public domain version of Pine for VMS (as opposed to the i0 version shipped with PMDF) was very out of date.  B PMDF mail is a good example of what can be done with extending theD functionality of VMS Mail since it is such a close clone of VMSMAIL.  
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------  % Date: Tue, 23 Oct 2001 09:57:35 -0400)* From: WILLIAM WEBB <WWEBB1@email.usps.gov># Subject: RE: BCC support in VMSMAIL - Message-ID: <0033000039230021000002L012*@MHS>i  * =0ABut were they "To:" or were they "Cc:"?   :^)    WWWebb   > -----Original Message-----1 > From: Info-VAX-Request@Mvb.Saic.Com at INTERNETe) > Sent: Tuesday, October 23, 2001 9:36 AMhD > To: Webb, William W Raleigh, NC; Info-VAX@Mvb.Saic.Com at INTERNET% > Subject: RE: BCC support in VMSMAILo >1 >e" > Thank you all for your feedback.: > Aside from relplies here I have received about 50 mails. >.= > General response is : we want BCC and  make sure we support  > all transports.  >c= > Well majority have spoken we will have BCC support soon. As 
 > soon as the 
 > developmenta > 5 > work is over I will post a note in this conference. > > I know there is a lot of work to be done in VMSMAIL, however > we need to start > with something.o >  > Guye >d, > Alan Winston - SSRL Admin Cmptg Mgr wrote: >c8 > > In article <3BD3EC5E.E683DA66@compaq.com>, Guy Peleg > <guy.peleg@compaq.com> > > writes:t > >d > > >T > > >Hello All,h > > >iF > > >Lately I have been playing with the idea of adding BCC support to
 > > >VMSMAIL.h3 > > >I would like to get your feedback about this :  > > >  > > >Do we really need this ?t > >r= > > First, despite the crankiness of the rest of this note, I  > want to say that< > > I really appreciate your participating in this forum and > giving customers. > > a say in the development process.  Thanks. > >h> > > If I were still on VMSMAIL I would really like this, but I
 > (and all my = > > command-line users) are using PMDF MAIL.  BCC gets used a  > fair amount here;l: > > probably not daily, but regularly, and there are cases > where nothing else > > will answer the need.i > >e: > > >Do we need to have  BCC support sending messages from > VMSMAIL to VMSMAIL" > > >or VMSMAIL to Exchange only ? > > > > > If the support isn't for everything - which probably means > for every supportedc: > > mail transport - it isn't worth doing.  Picture a user > sending a message viaA? > > BCC; if it fails, or (worse) if the messages go through but- > the recipientv? > > list isn't suppressed, that user will lose all faith in VMSs > MAIL and use > > something else.s > >-? > > I would rather see you fix the peculiarities in To: and CC:w > processing withe0 > > addresses that involve different transports. > > ; > > (First, the distinction between To: and CC: is lost andi > everything ends up= > > as a To: address.  Second, the only addresses a recipient$ > on another system.7 > > sees are the ones on that system, so if they do theh > equivalent of a reply/all,> > > it only gets their local recipients, not the whole list it > was sent to.  This= > > wreaked havoc around here, and was the reason why we made1 > all our users user > > PMDF MAIL instead.)e > >i > > -- AlanV > >I > > H > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=I =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= - =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3Dz5 > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3Dt4 > >  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDU: > >  Disclaimer: I speak only for myself, not SLAC or SSRL > Phone:  650/926-30568 > >  Physical mail to: SSRL -- SLAC BIN 69, PO BOX 4349, > STANFORD, CA  94309-0210 > > H > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=I =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=s- =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=5 > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D0 >=   ------------------------------    Date: 23 Oct 2001 08:00:10 -0700) From: P.Young@unsw.EDU.AU (Patrick Young)e# Subject: Re: BCC support in VMSMAIL = Message-ID: <55f85d77.0110230700.5355d624@posting.google.com>c  X Guy Peleg <guy.peleg@compaq.com> wrote in message news:<3BD3EC5E.E683DA66@compaq.com>... > Hello All, > C > Lately I have been playing with the idea of adding BCC support to_
 > VMSMAIL.0 > I would like to get your feedback about this : >   E OpenVMS mail stalled some time way back - I'm on PMDF. I'll guess youkF will need to talk to the PMDF folks and setup some sort of agreement -I any other effort would be a waste of time and a duplication of effort forc no good reason.a   Many thanks anyway.    ------------------------------    Date: 23 Oct 2001 10:12:16 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)o# Subject: Re: BCC support in VMSMAILw3 Message-ID: <jBSFDRa9e1Gi@eisner.encompasserve.org>   h In article <d7791aa1.0110230603.2a6d2644@posting.google.com>, bob@instantwhip.com (Bob Ceculski) writes:  A > you can get mime parsing and attachment separation in vms pine!   H What Guy was discussing was the features that come built-in to VMS mail.D No added software, meaning it will always be there at any client oneE may visit (if they are running a sufficiently recent version of VMS).S  % Besides, Guy knows how to capitalize.S   ------------------------------  % Date: Tue, 23 Oct 2001 11:11:25 -0400O; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com>0# Subject: Re: BCC support in VMSMAIL0$ Message-ID: <3bd588d6$1@news.si.com>  B >"Need" is a relative term.  To have any hope of appearing to be a& >useful mail program, BCC is required.  I Interesting thought.  Could you explain how BCC is useful for anything ati all other than SPAM? --A Brian Tillman                   Internet: tillman_brian at si.comSA Smiths Aerospace                          tillman at swdev.si.come= 3290 Patterson Ave. SE, MS      Addresses modified to prevent < Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  % Date: Tue, 23 Oct 2001 11:13:42 -0400d; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com> # Subject: Re: BCC support in VMSMAIL=$ Message-ID: <3bd5895e$1@news.si.com>  L >BCC is also useful for sending a group mail where you don't want to run the) >risk of one recipient spamming the rest./  $ Ah!  This I can understand.  Thanks. --A Brian Tillman                   Internet: tillman_brian at si.comnA Smiths Aerospace                          tillman at swdev.si.com== 3290 Patterson Ave. SE, MS      Addresses modified to preventc< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  + Date: Tue, 23 Oct 2001 15:30:45 +0000 (UTC)U From: david20@alpha2.mdx.ac.uk# Subject: Re: BCC support in VMSMAIL + Message-ID: <9r42f5$6uh$1@aquila.mdx.ac.uk>0  i In article <55f85d77.0110230700.5355d624@posting.google.com>, P.Young@unsw.EDU.AU (Patrick Young) writes: Y >Guy Peleg <guy.peleg@compaq.com> wrote in message news:<3BD3EC5E.E683DA66@compaq.com>...o
 >> Hello All,I >> dD >> Lately I have been playing with the idea of adding BCC support to >> VMSMAIL.d1 >> I would like to get your feedback about this :a >>   > F >OpenVMS mail stalled some time way back - I'm on PMDF. I'll guess youG >will need to talk to the PMDF folks and setup some sort of agreement -nJ >any other effort would be a waste of time and a duplication of effort for >no good reason. >  >Many thanks anyway.  O Unfortunately that is probably not an option any longer. Looking at the public eO documentation on PMDF to get ideas of what is possible is one thing looking at   code is another. MK Up until March last year Compaq could have probably done all sorts of deals > with Innosoft with respect to PMDF mail and PMDF Pine for VMS.E Innosoft didn't really want to be in the mail client business anyway.s  : But on March 7th 2000 Innosoft were acquired by SUN   see   & http://www.innosoft.com/press/smi.html  I Although subsequently Process took over future development and support off= PMDF as I understand it SUN still retains the rights to PMDF. N (For instance I understand the agreement between Process and Sun is what stopsC Process from producing a "cut down" PMDF for the hobbyist program).l  
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------    Date: 23 Oct 2001 03:29:20 -0700# From: nye_osborne@hotmail.com (Nye)lP Subject: Browsing a VMS FTP site, with Browser, fails to display names correctly= Message-ID: <3bcd2eea.0110230229.5288d858@posting.google.com>=  
 Hi VMS Folks,=  < I'm having trouble FTPing (via Browser) from a WinNT PC to aE VMS-hosted FTP site, maybe ALL VMS sites, but's hard to tell what I'ml connecting to.  ; Both MSIE and Netscape fail to work, through a Proxy Server   7 Directories and files are displayed with FULL file-name=$ characteristics, eg name.DIR;VERSION  E When I try to select the link the browser errors, BECAUSE it's trying C to select the full VMS filename, file-type, version number and all.cC I can manually delete the unnecessary bits from the browser address7B window and then successfully connect (both directories and files).  A Any suggestions on what to reconfigure for MS Proxy Server or thei	 Browser ?        Nyek   ------------------------------    Date: 23 Oct 2001 07:59:11 -0700< From: alphaman-nixspam@hsv.sungardtrust.com (Aaron Sakovich)T Subject: Re: Browsing a VMS FTP site, with Browser, fails to display names correctly= Message-ID: <8af17fe1.0110230659.5babf04b@posting.google.com>   h nye_osborne@hotmail.com (Nye) wrote in message news:<3bcd2eea.0110230229.5288d858@posting.google.com>... > Hi VMS Folks,m > > > I'm having trouble FTPing (via Browser) from a WinNT PC to aG > VMS-hosted FTP site, maybe ALL VMS sites, but's hard to tell what I'md > connecting to. > = > Both MSIE and Netscape fail to work, through a Proxy Serveri > 9 > Directories and files are displayed with FULL file-name & > characteristics, eg name.DIR;VERSION > G > When I try to select the link the browser errors, BECAUSE it's tryingoE > to select the full VMS filename, file-type, version number and all.iE > I can manually delete the unnecessary bits from the browser addresseD > window and then successfully connect (both directories and files). > C > Any suggestions on what to reconfigure for MS Proxy Server or the  > Browser ?v > 	 >     Nyes  D That's standard behavior from MS -- their response to my bug-post onE the IE5 beta was "eh, okay, we just won't support VMS ftp servers..."r  D The only alternative I've seen is to replace the UCX ftp server (I'mE presuming you're using UCX) with the MadGoat server (or other vendorss> ftp server which can provide U*X-like ftp dirs to the client).   Sorry, Aaronf   ------------------------------  % Date: Tue, 23 Oct 2001 17:10:39 +0200f, From: Theo Jakobus <Theo.Jakobus@iaf.fhg.de>T Subject: Re: Browsing a VMS FTP site, with Browser, fails to display names correctly) Message-ID: <3BD5886F.1080700@iaf.fhg.de>e  
 Nye wrote:   > Hi VMS Folks,n > > > I'm having trouble FTPing (via Browser) from a WinNT PC to aG > VMS-hosted FTP site, maybe ALL VMS sites, but's hard to tell what I'mo > connecting to. > = > Both MSIE and Netscape fail to work, through a Proxy Server  > 9 > Directories and files are displayed with FULL file-name & > characteristics, eg name.DIR;VERSION    Q OpenVMS allows to setup the ftp account which doesn't show the file version, the =! logical is:  TCPIP$FTP_NO_VERSIONS3 For more information (there are a lot of options):  I http://www.openvms.compaq.com:8000/73FINAL/6526/6526pro_028.html#ftp_chap    > G > When I try to select the link the browser errors, BECAUSE it's trying=E > to select the full VMS filename, file-type, version number and all.-E > I can manually delete the unnecessary bits from the browser addresstD > window and then successfully connect (both directories and files).    K There are ftp programs as shareware available which are able to handle the d2 OpenVMS files like: WS_FTP http://www.ipswitch.com       Regards, -- g  ; *********************************************************** ; *                                                         *m; *  Theo Jakobus                                           *.; *  Fraunhofer-Institut fuer Angewandte Festkoerperphysik  *t; *  Tullastr. 72                                           *p; *  D-79108 Freiburg                                       * ; *  Germany                                                *b; *  Phone:   +49-(0)761-5159-325                           *t; *  FAX :    +49-(0)761-5159-200                           *l; *  e-mail:  Theo.Jakobus@iaf.fhg.de                       *a; *  http://www.iaf.fhg.de                                  *e; *                                                         *f; ***********************************************************m   ------------------------------  + Date: Tue, 23 Oct 2001 16:17:02 +0100 (GDT)o) From: Andy Harper <Andy.Harper@kcl.ac.uk>IT Subject: Re: Browsing a VMS FTP site, with Browser, fails to display names correctly: Message-ID: <SIMEON.10110231602.X@kings-ut-srv1.kcl.ac.uk>  - On 23 Oct 2001 07:59:11 -0700 Aaron Sakovich f. <alphaman-nixspam@hsv.sungardtrust.com> wrote:  j > nye_osborne@hotmail.com (Nye) wrote in message news:<3bcd2eea.0110230229.5288d858@posting.google.com>... > > Hi VMS Folks,g > > @ > > I'm having trouble FTPing (via Browser) from a WinNT PC to aI > > VMS-hosted FTP site, maybe ALL VMS sites, but's hard to tell what I'm  > > connecting to. > > ? > > Both MSIE and Netscape fail to work, through a Proxy Serveri > > ; > > Directories and files are displayed with FULL file-names( > > characteristics, eg name.DIR;VERSION > > I > > When I try to select the link the browser errors, BECAUSE it's tryingyG > > to select the full VMS filename, file-type, version number and all.iG > > I can manually delete the unnecessary bits from the browser address F > > window and then successfully connect (both directories and files).  '   Your problem stems from three things.aG      1. No standard way for an ftp server to send back a file list in a.!         machine-parseable format.EE      2. UCX FTP's insistence on sending a format that's completely atgH         odds with the most other ftp servers, that tend to try and mimic          the basic unix ls formatF      3. Microsoft's inability to recognize that other software exists &         and adjust theirs accordingly.  LH I have seen some ftp clients that try various heuristic methods to make ? sense of the formats (WS_FTP for the PC is a good one; you can rG configure a range for formats, or let it try to auto-detect). UCX have eC chosen one that is not impossible to parse, just so far out of the t! ordinary that clients can't cope.n  E I would suggest that digital/compaq/hp look into this and provide at eC least a configurable display format that resembles the rest of the vF world, rather than all of us trying to convince the rest of the world G to handle the UCX format. Either that, or don't use the UCX FTP server.a    F > That's standard behavior from MS -- their response to my bug-post onG > the IE5 beta was "eh, okay, we just won't support VMS ftp servers..."h  "   Well, that seems fairly typical.   > F > The only alternative I've seen is to replace the UCX ftp server (I'mG > presuming you're using UCX) with the MadGoat server (or other vendorsc@ > ftp server which can provide U*X-like ftp dirs to the client).  F     I concur. The madgoat FTP server has generally been very good and =     certainly has a switch to make the output more unix like.   E     Of course, it's just as hard to get the rest of the UCX users to  !     switch their FTP servers too.t  H     Perhaps the best long-time approach is for the IETF to come up with F     a standard for the FTP file list format, though they would likely G     argue that ftp is dying out and been replaced by http. Perhaps one a     of them can comment ?t   Regards,   ----------------------! Andy Harper B.Sc., M.B.C.S, C.Engw Systems and Mail Manager Kings College London   ------------------------------  % Date: Tue, 23 Oct 2001 12:43:40 -0400w- From: JF Mezei <jfmezei.spamnot@videotron.ca>mY Subject: Re: Browsing a VMS FTP site, with Browser, fails to display names correctly corro, Message-ID: <3BD59E3B.8674E855@videotron.ca>  
 Nye wrote:9 > Directories and files are displayed with FULL file-name-& > characteristics, eg name.DIR;VERSION  N From a philosophical point of view it is a problem with Microsoft's software. M The FTP client must not make any assumption on the file name and structure ofoN the remote host and should be able to transform a file name to something which) is acceptable to the local client system.   M From a practical point of view, you need to ask the system manager of the VMS-I system at the other end to define a logical name which will cause the FTPe, server to hide the version numbers from you.    , for instance: $DEFINE TCPIP$FTP_NO_VERSION 1  L I am not sure if it works, but you migt try to put the above command in your$ LOGIN.COM on the remote VMS system.   8 So create a text file on your system with only one line  $DEFINE TCPIP$FTP_NO_VERSION 1  J and then transfer it to the VMS system, assuming you have your own accountL there. When you next start the FTP, (it may take some time since the processG may continue to live for some time after you have stopped your client).3  L (Note: if your account already has a LOGIN.COM file, you should fetch it andQ edit it. Add the $DEFINE command to an area that is executed as non-interactive.)   N I am not 100% sure that the DEFINE command will work for a single user though.   ------------------------------  # Date: Tue, 23 Oct 2001 17:22:33 GMTs1 From: "Mark D. Jilson" <jilly@clarityconnect.com> Y Subject: Re: Browsing a VMS FTP site, with Browser, fails to display names correctly corrs2 Message-ID: <3BD5A769.B76F9993@clarityconnect.com>  F Microsoft is ignoring the RFC (959 I believe) and will not change it's7 behavior.  TCPIP Services for VMS is following the RFC.I   JF Mezei wrote:  >  > Nye wrote:; > > Directories and files are displayed with FULL file-names( > > characteristics, eg name.DIR;VERSION > O > From a philosophical point of view it is a problem with Microsoft's software.IO > The FTP client must not make any assumption on the file name and structure offP > the remote host and should be able to transform a file name to something which+ > is acceptable to the local client system.d > O > From a practical point of view, you need to ask the system manager of the VMS K > system at the other end to define a logical name which will cause the FTP . > server to hide the version numbers from you. > . > for instance: $DEFINE TCPIP$FTP_NO_VERSION 1 > N > I am not sure if it works, but you migt try to put the above command in your% > LOGIN.COM on the remote VMS system.w > 9 > So create a text file on your system with only one line:  > $DEFINE TCPIP$FTP_NO_VERSION 1 > L > and then transfer it to the VMS system, assuming you have your own accountN > there. When you next start the FTP, (it may take some time since the processI > may continue to live for some time after you have stopped your client).d > N > (Note: if your account already has a LOGIN.COM file, you should fetch it andS > edit it. Add the $DEFINE command to an area that is executed as non-interactive.)D > P > I am not 100% sure that the DEFINE command will work for a single user though.   -- =D Jilly	- Working from Home in the Chemung River Valley - Lockwood, NY0 	- jilly@clarityconnect.com			- Brett Bodine fan. 	- Mark.Jilson@Compaq.com			- since 1975 or so, 	- http://www.jilly.baka.com               -   ------------------------------  # Date: Tue, 23 Oct 2001 17:35:36 GMT ' From: Don Sykes <anonymous@pacbell.net>2Y Subject: Re: Browsing a VMS FTP site, with Browser, fails to display names correctly corr3+ Message-ID: <3BD5AA9A.F48E6890@pacbell.net>D  
 Interesting. D> When I try ftp'ing from my Browser (Netscape 4.7) on NT to VMSF anonymous, I see multiple versions, but without the version part (;n).H When i download either one, I always get the newest ver. Other than that it works fine. u* Note: TCPIP$FTP_NO_VERSION is not defined.   Donf  
 Nye wrote: >  > Hi VMS Folks,o > > > I'm having trouble FTPing (via Browser) from a WinNT PC to aG > VMS-hosted FTP site, maybe ALL VMS sites, but's hard to tell what I'mn > connecting to. > = > Both MSIE and Netscape fail to work, through a Proxy Servero > 9 > Directories and files are displayed with FULL file-namew& > characteristics, eg name.DIR;VERSION > G > When I try to select the link the browser errors, BECAUSE it's tryingtE > to select the full VMS filename, file-type, version number and all.oE > I can manually delete the unnecessary bits from the browser address-D > window and then successfully connect (both directories and files). > C > Any suggestions on what to reconfigure for MS Proxy Server or thea > Browser ?e > 	 >     Nyeg   ------------------------------    Date: 23 Oct 2001 08:17:20 -0500+ From: young_r@encompasserve.org (Rob Young) - Subject: Re: Compaq: VMS is alive and kickingi3 Message-ID: <E8bI9d5oc451@eisner.encompasserve.org>e  q In article <yt4B7.1082659$ai2.82435708@bin2.nnrp.aus1.giganews.com>, "Bill Todd" <billtodd@metrocast.net> writes:i   > K > The only reference I can find to EV7's memory bandwidth is a 16-month-oldsL > article by Paul DeMone at realworldtech.com, which pegs it at 12.8 GB/sec.J > Since the processors have high-bandwidth interconnects as well, a singleL > processor might well be able to exploit memory connected directly to otherK > processors over this interconnect and achieve a total memory bandwidth ofeH > 44.8 GB/sec (the figure in his article last June) - but clearly if oneN > processor is doing so, the others aren't accessing their own local memory atK > all, so using the 12.8 GB/sec figure per processor seems more reasonable.  >   A 	Variation on that might be to make that number 32 GB/sec.  Since @ 	each CPU has two memory controllers on die, one could be active> 	serving remote memory requests and flip-flop as pages for theH 	local cpu need that controller (starvation avoidance).  That 44.8 GB/s H 	is best case when remote CPU is idling on memory requests.  Unlikely.   	So realistic best case may be:p    2 			6.4 GB/s  -  local memory controller aggregate.2 		4 * 6.4 GB/s      -  through network controller.  D 	Of course it gets trickier than this.  You can't have 4 active CPUsE 	with 4 active network ports , there aren't enough memory controllers B 	to support that.  4 CPUs would have 8 combined memory controllers( 	so average per-CPU memory bandwidth is:   			(8 * 6.4) / 4  4 	or 12.8 GB/sec bursting to 44.8 GB/s as seen above.  A 	Of course it gets trickier than that as some (how much?) networko> 	port activity is for remote L2 hits adding to your bandwidth.   				Robe   ------------------------------    Date: 23 Oct 2001 08:19:54 -0500+ From: young_r@encompasserve.org (Rob Young)s- Subject: Re: Compaq: VMS is alive and kickinga3 Message-ID: <zuBd9kq8ieIF@eisner.encompasserve.org>o  u In article <g05B7.145922$vq.33917340@typhoon.ne.mediaone.net>, "Terry C. Shannon" <terryshannon@mediaone.net> writes:A >  > N > Good point. While early Marvel collateral showed 128-way systems, the recentN > roadmaps cite 64P boxes. I am told that the latest PID reflects 128P systemsJ > but don't know for sure. Given the Marvel design, if you can build a 64PN > system, you ought to be able to scale it to 128P simply by adding additional > 8P Marvel building blocks. >   < 	Maybe super-special customers are seeking a config with 128 	processors.   				Roby   ------------------------------    Date: 23 Oct 2001 09:12:30 -0500- From: koehler@encompasserve.org (Bob Koehler)t- Subject: RE: Compaq: VMS is alive and kicking 3 Message-ID: <QHcoOONEsZ19@eisner.encompasserve.org>m   In article <BE56C50EA024184DAF48F0B9A47F5CF4010D711A@kaoexc01.americas.cpqcorp.net>, "Main, Kerry" <Kerry.Main@compaq.com> writes: > JF,  > H > In the past, that was a marketing term by most Wintel vendors, but theB > reality today is that the "industry standard" or, in some cases,D > "perceived" industry standard technologies that many Customers areG > looking for are for example : XML, LDAP, X.509, TCPIP, HTTP(S), J2EE, F > Java, SOAP, X.500 (yes - still very much in use) etc. (and many more > "eTerms" evolving)...   C    You have to remember the Compaq and HP both think they're in thenG    bussiness to manufacture computers.  They're making the same mistake-C    DEC did of not realizing it's the software they do or don't make.H    which sells the computers.  Unless there's evidence otherwize, always;    take "industry standard" from these folks as "hardware".@      .   ------------------------------  % Date: Tue, 23 Oct 2001 09:41:32 -0400r+ From: "Main, Kerry" <Kerry.Main@compaq.com>e- Subject: RE: Compaq: VMS is alive and kickingoT Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4010D7120@kaoexc01.americas.cpqcorp.net>   Rob,  ? >>> Maybe super-special customers are seeking a config with 128e
 processors.<<   G While there will always be a market for high end "see how many CPUs yousG can cram in a single box", the challenge will always be "how do you geta9 applications to take advantage of this many processors?".7  = The alternative to scaling up (big SMP) is always scaling outs4 (clustering). Each has advantages and disadvantages.  H However, the "sweet spot" for high end servers is likely much lower thanH 64 processors. This might be the reason why IBM have apparently recently changed their philosophy -  C http://news.cnet.com/news/0-1003-200-7407646.html "IBM unveiled itssC top-end Unix server Thursday, marking Big Blue's departure from thecD industry's assumption that more processors makes a better computer."   Regards,  
 Kerry Main Senior Consultant  Compaq Canada Corp.  Professional Servicesn Voice: 613-592-4660a Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----2 From: Rob Young [mailto:young_r@encompasserve.org] Sent: October 23, 2001 9:20 AM To: Info-VAX@Mvb.Saic.Comn- Subject: Re: Compaq: VMS is alive and kickingh    H In article <g05B7.145922$vq.33917340@typhoon.ne.mediaone.net>, "Terry C., Shannon" <terryshannon@mediaone.net> writes: >=20 >=20G > Good point. While early Marvel collateral showed 128-way systems, the  recentF > roadmaps cite 64P boxes. I am told that the latest PID reflects 128P systemsoF > but don't know for sure. Given the Marvel design, if you can build a 64PMC > system, you ought to be able to scale it to 128P simply by addingr
 additional > 8P Marvel building blocks. >=20  < 	Maybe super-special customers are seeking a config with 128 	processors.   				Roby   ------------------------------    Date: 23 Oct 2001 09:55:35 -0500+ From: young_r@encompasserve.org (Rob Young)o- Subject: Re: Compaq: VMS is alive and kickingU3 Message-ID: <MR2$JyJWXmSV@eisner.encompasserve.org>   a In article <E8bI9d5oc451@eisner.encompasserve.org>, young_r@encompasserve.org (Rob Young) writes:ks > In article <yt4B7.1082659$ai2.82435708@bin2.nnrp.aus1.giganews.com>, "Bill Todd" <billtodd@metrocast.net> writes:3 >  >> 0L >> The only reference I can find to EV7's memory bandwidth is a 16-month-oldM >> article by Paul DeMone at realworldtech.com, which pegs it at 12.8 GB/sec.SK >> Since the processors have high-bandwidth interconnects as well, a singleeM >> processor might well be able to exploit memory connected directly to otheriL >> processors over this interconnect and achieve a total memory bandwidth ofI >> 44.8 GB/sec (the figure in his article last June) - but clearly if oneeO >> processor is doing so, the others aren't accessing their own local memory atoL >> all, so using the 12.8 GB/sec figure per processor seems more reasonable. >> a > C > 	Variation on that might be to make that number 32 GB/sec.  SincesB > 	each CPU has two memory controllers on die, one could be active@ > 	serving remote memory requests and flip-flop as pages for theJ > 	local cpu need that controller (starvation avoidance).  That 44.8 GB/s J > 	is best case when remote CPU is idling on memory requests.  Unlikely.  ! > 	So realistic best case may be:i >  > 4 > 			6.4 GB/s  -  local memory controller aggregate.4 > 		4 * 6.4 GB/s      -  through network controller. > F > 	Of course it gets trickier than this.  You can't have 4 active CPUsG > 	with 4 active network ports , there aren't enough memory controllerstD > 	to support that.  4 CPUs would have 8 combined memory controllers* > 	so average per-CPU memory bandwidth is: >  > 			(8 * 6.4) / 4 > 6 > 	or 12.8 GB/sec bursting to 44.8 GB/s as seen above. > C > 	Of course it gets trickier than that as some (how much?) network @ > 	port activity is for remote L2 hits adding to your bandwidth. >   A 	Hey.. it may get even trickier than this.  Suppose memory accessa= 	is sequential.  A smart controller might just pre-fetch into F 	a small region of L2 such that it would free up the memory controllerA 	for random access, plus there are 4 network ports and two memory-? 	controllers, the network ports being less of a bottleneck than D 	waiting on memory latency and a free memory controller.  IncreasingB 	broadcast traffic a bit I suppose but seems more of a win... i.e.= 	that remote memory fetch actually becomes a remote L2 fetch.    					Rob   ------------------------------  % Date: Tue, 23 Oct 2001 08:05:45 -0700 ' From: David Mathog <mathog@caltech.edu>m- Subject: Re: Compaq: VMS is alive and kickingn+ Message-ID: <3BD58749.E25710EE@caltech.edu>    JF Mezei wrote:e   >n >vJ > So far, IA64 stands to be just another Alpha, restricted to high end andM > prevented from competing directly against the 8086. And while one can argueeN > that if you look long enough into the future, you can see IA64 replacing theN > 8086 on the desktop, you also have to consider that you cannot look that far > into the future for VMS.  H I agree with most of what you said, but think that there's a substantial probability thatP the AMD Hammer line will be the successor to the x86 on the desktop.   This will beM horrificly bad news for Intel, because that same CPU line will work in largerd servers, whichL will force the Itanium (small market, high unit cost) to compete against the Hammer (large market,iN low unit cost), and AMD could conceivably price very low and still make a good profit on everyaG chip, yet force Intel to sell Itanium at or even below cost.  (Ie, do an Compaq.)    The VMS spinN on this being that by betting the farm on Itanium and not covering their butts withN a (clandestine?) Hammer port, the VMS crew will most likely find their product once1 again tied to the most expensive platform around.   K That said, politics at Chompaq and not technical or even market issues willo
 determine theeQ fate of the OS.  Just as it has for the last 10 years.  The OS is cursed - doomedu
 forever to= be crippled by the acts of incompetent (or worse) management.s   Regards,   David Mathog mathog@caltech.edu   ------------------------------  % Date: Tue, 23 Oct 2001 11:40:20 -0400e( From: Hamlyn Mootoo <univms@bigfoot.com>- Subject: Re: Compaq: VMS is alive and kicking + Message-ID: <3BD58F64.4583D3C2@bigfoot.com>    "Main, Kerry" wrote: >   [snip]e > H > Course, if one defines proprietary as "developed and maintained by oneI > vendor only", then all OS's are proprietary, but that is a never ending  > rat hole discussion.  C I beg to differ. Linux is most certainly NOT PROPRIETARY, as it is,mA using your defintion, NOT "developed and maintained by one vendor F only".  Further, since it is open source, YOU can develop and maintain0 it YOURSELF if you wish, as long as you care to.   HM >  > :-)  > 
 > Regards, >  > Kerry Main > Senior Consultanti > Compaq Canada Corp.i > Professional Servicesh > Voice: 613-592-4660  > Fax  :  819-772-7036 > Email: Kerry.Main@Compaq.com >  > -----Original Message-----6 > From: JF Mezei [mailto:jfmezei.spamnot@videotron.ca]  > Sent: October 22, 2001 4:37 PM > To: Info-VAX@Mvb.Saic.Comi/ > Subject: Re: Compaq: VMS is alive and kickinge >  > john nixon wrote:nG > > This kind of article is exactly the kind we need.  Comments such as. > yourst( > > and J.F.'s do not help worth a twit. > F > Digital spent 8 years trying to kill VMS. Compaq spend the following > yearsFF > ignoring (at best) VMS. And Compaq has seriously considered outright	 > killingBB > VMS not too long ago. Getting folks to trust that VMS' future is > anything morerI > than a florified maintenance mode without any marketing will talk a lotn > more- > than just a few good articles in the press.p > D > Winkler, Carly and Curly continue to state that they will focus on
 > industryH > standard stuff. And that has more weight on one's option on the future > of VMSC > within the Carly&Curly show than any single article in the press.p > F > The primary problem isn't with the public's perception of VMS, it is > withF > Compaq and HP's perception of VMS. You can't sustantially change the
 > public'sC > perception before you fix the owning company's perception of VMS.. > F > Didn't that article state that VMS might have to give up some of its
 > niches ?E > VMS has given up international funds transfers, one of its former 6 	 > niches. F > And Compaq didn't mind because it expects most of these customers to > choose anNI > NT solution which is becoming available. Another niche might go becausec > HP-UXr. > would be the choice solution for that niche. > C > Oracle is implementing its own clustering capabilities so even onv	 > systemso? > without decent clustering, customers might still benefit froml
 > clustering.e > = > VMS cannot live on clustering alone for very long. It needsc > applications.a > * > > For years, the complaint was that  VMSJ > > was technically the best, but they could not buy a positive comment in > thee
 > > press. > G > That is because they did not want positive comments in the press. Thej	 > messageoJ > from Compaq/HP is pretty clear: if you're an existing large key customer > ofI > VMS, you need not worry, we'll support you for years to come, but don'tF > expect@ > VMS to grow out of its few remaining niches or get any form of
 > substantialt6 > marketing because VMS is not an industry standard OS   ------------------------------  % Date: Tue, 23 Oct 2001 11:48:44 -0400e+ From: "Main, Kerry" <Kerry.Main@compaq.com>l- Subject: RE: Compaq: VMS is alive and kicking T Message-ID: <BE56C50EA024184DAF48F0B9A47F5CF4010D7122@kaoexc01.americas.cpqcorp.net>   Hamlyn,h  @ >>> I beg to differ. Linux is most certainly NOT PROPRIETARY,<<<  4 I knew that statement would draw a Linux fan out :-)  F So, I can take any ISV package that states support for Red Hat and runH it with absolutely no issues on SuSE Linux? Will that ISV vendor supportC me when I call him and report a problem with their application when % running on another Linux platform?=20I  H Do Linux ISV application vendors provide source code for their Customers to maintain on their own?n  7 [please - not another "what is proprietary?" thread ..]u   Regards,  
 Kerry Main Senior Consultant$ Compaq Canada Corp.  Professional Servicest Voice: 613-592-4660u Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----/ From: Hamlyn Mootoo [mailto:univms@bigfoot.com]e Sent: October 23, 2001 11:40 AMp To: Info-VAX@Mvb.Saic.ComN- Subject: Re: Compaq: VMS is alive and kickingl         "Main, Kerry" wrote: >=20  [snip]e >=20H > Course, if one defines proprietary as "developed and maintained by oneB > vendor only", then all OS's are proprietary, but that is a never ending > rat hole discussion.  C I beg to differ. Linux is most certainly NOT PROPRIETARY, as it is,nA using your defintion, NOT "developed and maintained by one vendor9F only".  Further, since it is open source, YOU can develop and maintain0 it YOURSELF if you wish, as long as you care to.   HM >=20 > :-)y >=20
 > Regards, >=20 > Kerry Main > Senior ConsultantT > Compaq Canada Corp.a > Professional Servicese > Voice: 613-592-4660  > Fax  :  819-772-7036 > Email: Kerry.Main@Compaq.com >=20 > -----Original Message-----6 > From: JF Mezei [mailto:jfmezei.spamnot@videotron.ca]  > Sent: October 22, 2001 4:37 PM > To: Info-VAX@Mvb.Saic.Como/ > Subject: Re: Compaq: VMS is alive and kicking  >=20 > john nixon wrote:IG > > This kind of article is exactly the kind we need.  Comments such ass > yours ( > > and J.F.'s do not help worth a twit. >=20F > Digital spent 8 years trying to kill VMS. Compaq spend the following > years F > ignoring (at best) VMS. And Compaq has seriously considered outright	 > killingSB > VMS not too long ago. Getting folks to trust that VMS' future is > anything moretE > than a florified maintenance mode without any marketing will talk ae loto > more- > than just a few good articles in the press.n >=20D > Winkler, Carly and Curly continue to state that they will focus on
 > industryH > standard stuff. And that has more weight on one's option on the future > of VMSC > within the Carly&Curly show than any single article in the press.  >=20F > The primary problem isn't with the public's perception of VMS, it is > withF > Compaq and HP's perception of VMS. You can't sustantially change the
 > public'sC > perception before you fix the owning company's perception of VMS.n >=20F > Didn't that article state that VMS might have to give up some of its
 > niches ?E > VMS has given up international funds transfers, one of its former 6f	 > niches.nF > And Compaq didn't mind because it expects most of these customers to > choose anoA > NT solution which is becoming available. Another niche might goh becauseb > HP-UXi. > would be the choice solution for that niche. >=20C > Oracle is implementing its own clustering capabilities so even ono	 > systemst? > without decent clustering, customers might still benefit from 
 > clustering.h >=20= > VMS cannot live on clustering alone for very long. It needst > applications.a >=20* > > For years, the complaint was that  VMSG > > was technically the best, but they could not buy a positive commentr in > theh
 > > press. >=20G > That is because they did not want positive comments in the press. The 	 > messageuA > from Compaq/HP is pretty clear: if you're an existing large keys customer > ofC > VMS, you need not worry, we'll support you for years to come, buti don'tc > expect@ > VMS to grow out of its few remaining niches or get any form of
 > substantial 6 > marketing because VMS is not an industry standard OS   ------------------------------  % Date: Tue, 23 Oct 2001 12:40:58 -0400e5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>w- Subject: Re: Compaq: VMS is alive and kickingh1 Message-ID: <x5hB7.822$RL6.8791@news.cpqcorp.net>l  = JF Mezei wrote in message <3BD490C3.DA6BF5AF@videotron.ca>...b >wJ >Agreed. They can't scuttle VMS *yet*. But the writing is on the wall that theyJ >intend to scuttle it as soon as possible and get customer to move over toE >"industry standard" platforms. If Compaq/HP want to streamline theire product-J >line, it is only fair to expect OS consolidation after they are done with thep >server consolidation. >u    J The only "writing on the wall" right now is being scribbled in crayon by a few people in this group.t  7 You guys take reading chicken entrails to a new height.e   ------------------------------  % Date: Tue, 23 Oct 2001 12:45:01 -0400 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>h- Subject: Re: Compaq: VMS is alive and kickingt1 Message-ID: <j9hB7.823$RL6.9139@news.cpqcorp.net>'  I Well, you "could" but I think that at some point you might not be able tooJ call it "Linux" anymore.  From what I understand, the *code* itself can beH used without emcuberance, and you can do whatever you want to it.  But IH believe the copyright to the Linux "brand" identity police would preventH something with substantial changes to it from being marketed and sold as "Linux".    A Hamlyn Mootoo wrote in message <3BD58F64.4583D3C2@bigfoot.com>...s >@ >n >"Main, Kerry" wrote:  >> > [snip] >>I >> Course, if one defines proprietary as "developed and maintained by one J >> vendor only", then all OS's are proprietary, but that is a never ending >> rat hole discussion.W >eD >I beg to differ. Linux is most certainly NOT PROPRIETARY, as it is,B >using your defintion, NOT "developed and maintained by one vendorG >only".  Further, since it is open source, YOU can develop and maintainI1 >it YOURSELF if you wish, as long as you care to.l >r >HMh >> >> :-) >> >> Regards,  >>
 >> Kerry Main) >> Senior Consultant >> Compaq Canada Corp. >> Professional Services >> Voice: 613-592-4660 >> Fax  :  819-772-7036y >> Email: Kerry.Main@Compaq.comD >> >> -----Original Message-----f7 >> From: JF Mezei [mailto:jfmezei.spamnot@videotron.ca]s! >> Sent: October 22, 2001 4:37 PMe >> To: Info-VAX@Mvb.Saic.Com0 >> Subject: Re: Compaq: VMS is alive and kicking >> >> john nixon wrote:H >> > This kind of article is exactly the kind we need.  Comments such as >> yours) >> > and J.F.'s do not help worth a twit.v >>G >> Digital spent 8 years trying to kill VMS. Compaq spend the following1 >> yearsG >> ignoring (at best) VMS. And Compaq has seriously considered outrighto
 >> killingC >> VMS not too long ago. Getting folks to trust that VMS' future isd >> anything moreJ >> than a florified maintenance mode without any marketing will talk a lot >> morel. >> than just a few good articles in the press. >>E >> Winkler, Carly and Curly continue to state that they will focus onc >> industryaI >> standard stuff. And that has more weight on one's option on the futuree	 >> of VMSiD >> within the Carly&Curly show than any single article in the press. >>G >> The primary problem isn't with the public's perception of VMS, it isc >> with-G >> Compaq and HP's perception of VMS. You can't sustantially change thea >> public'seD >> perception before you fix the owning company's perception of VMS. >>G >> Didn't that article state that VMS might have to give up some of itsC >> niches ? F >> VMS has given up international funds transfers, one of its former 6
 >> niches.G >> And Compaq didn't mind because it expects most of these customers ton >> choose anJ >> NT solution which is becoming available. Another niche might go because >> HP-UX/ >> would be the choice solution for that niche.0 >>D >> Oracle is implementing its own clustering capabilities so even on
 >> systems@ >> without decent clustering, customers might still benefit from >> clustering. >>> >> VMS cannot live on clustering alone for very long. It needs >> applications. >>+ >> > For years, the complaint was that  VMSsK >> > was technically the best, but they could not buy a positive comment ino >> the >> > press.t >>H >> That is because they did not want positive comments in the press. The
 >> messageK >> from Compaq/HP is pretty clear: if you're an existing large key customer- >> of-J >> VMS, you need not worry, we'll support you for years to come, but don't	 >> expectrA >> VMS to grow out of its few remaining niches or get any form ofs >> substantial7 >> marketing because VMS is not an industry standard OS4   ------------------------------  % Date: Tue, 23 Oct 2001 19:06:44 +0200k& From: John McLean <mcleanj@dplanet.ch>- Subject: Re: Compaq: VMS is alive and kickingi* Message-ID: <3BD5A3A4.2B131521@dplanet.ch>   Bob Koehler wrote: >  > In article <BE56C50EA024184DAF48F0B9A47F5CF4010D711A@kaoexc01.americas.cpqcorp.net>, "Main, Kerry" <Kerry.Main@compaq.com> writes: > > JF,p > >sJ > > In the past, that was a marketing term by most Wintel vendors, but theD > > reality today is that the "industry standard" or, in some cases,F > > "perceived" industry standard technologies that many Customers areI > > looking for are for example : XML, LDAP, X.509, TCPIP, HTTP(S), J2EE,>H > > Java, SOAP, X.500 (yes - still very much in use) etc. (and many more > > "eTerms" evolving)...1 > E >    You have to remember the Compaq and HP both think they're in theiI >    bussiness to manufacture computers.  They're making the same mistake E >    DEC did of not realizing it's the software they do or don't makeeJ >    which sells the computers.  Unless there's evidence otherwize, always= >    take "industry standard" from these folks as "hardware".h >    I'll agree with on that.  9 Take a Bloomberg report dated 23 Oct, dateline Houston at Q http://quote.bloomberg.com/fgcgi.cgi?T=marketsquote99_news.ht&s=AO9T5WBWlQ29tcGFx   H entitled "Compaq's Capellas Hopes Sale to Hewlett-Packard Is No Misstep" (by Peter J. Brennan)v  H The article claims that stockholders are starting to agree with CapellasG because the stock price is rising.  I doubt that; I suspect they simplysH think there is a buck to be made and that Compaq was underpriced (as did1 Parnassus Investments according to the article). v     Some interesting paragraphs -e  F "At the beginning of last quarter, Capellas talked about restructuringH Compaq for the third time since he became CEO in July 1999. He had pared? distributors and merged the company's corporate and consumer PCl divisions."t  D Three times.. in two years ?   At this kind of frequency it's called+ rearranging the deck chairs on the Titanic.e     and ...o  H "Capellas aimed to sell more services and to reduce PCs to about a third@ of sales from 45 percent in the second quarter. He found that he. couldn't do that without selling the company."  B Doh !  How about selling the high-end stuff... you know, where theE profit margins are reasonable... you know, the operating systems that-: your company creates, so you don't split profit with M$...     and ...h  D "Yet, Hewlett-Packard will be increasingly dependent on selling PCs,C which now account for 21 percent of total sales and will rise to 33 F percent when the acquisition closes. That disturbs many investors, who@ say PCs will hold back earnings growth at the combined company."  ' Sounds like they are smarter than Mike.n     and ...   F "Capellas, 47, was unavailable to comment, spokesman Arch Currid said.F The company paid its CEO $29.7 million in cash and stock last year. He4 will be Hewlett-Packard's president after the sale."  F and $29.7 million is a lot more than they spent on advertising VMS ... and Unix ... and Tandem.    k  D VMS alive and kicking ?  More like "Barely alive and been thoroughly kicked by its owner" r     John McLeani   ------------------------------  % Date: Tue, 23 Oct 2001 19:28:01 +0200t& From: John McLean <mcleanj@dplanet.ch>- Subject: Re: Compaq: VMS is alive and kickinga* Message-ID: <3BD5A8A1.2E10CFFD@dplanet.ch>   Fred Kleinsorge wrote: > ? > JF Mezei wrote in message <3BD490C3.DA6BF5AF@videotron.ca>...  > >iL > >Agreed. They can't scuttle VMS *yet*. But the writing is on the wall that > theyL > >intend to scuttle it as soon as possible and get customer to move over toG > >"industry standard" platforms. If Compaq/HP want to streamline theirc	 > productqL > >line, it is only fair to expect OS consolidation after they are done with > the  > >server consolidation. > >. > L > The only "writing on the wall" right now is being scribbled in crayon by a > few people in this group.a > 9 > You guys take reading chicken entrails to a new height.-    C Not so.  Take Gartner Report "HP/Compaq Merger: Hard Choices on the / Server Side" by George Weiss, dated 6 Sep 2001.   @ It concludes "Enterprises should carefully assess new-generationH systems, such as blades, for upgrade and support continuity, and migrateE OpenVMS and other legacy software without awaiting the outcome of ther
 acquisition."b     and K http://techupdate.zdnet.com/techupdate/stories/main/0,14179,2817135,00.htmlrE on October 10 from the same writer as "VMS is alive and kicking" (ie.d David Berlind) ..o  H 'For example, Compaq has customers running Tru64 Unix and OpenVMS on theA Alpha architecture and, now that Alpha has been discontinued, the A company's plan was to migrate them to Tru64 or Windows running ono	 Itanium.'   
 and later ...i  H 'According to Gartner's McGuckin, "All roads lead to HP-UX or Windows XPE on Itanium." In other words, existing Tru64 and OpenVMS customers are[C now looking at HP-UX (instead of Tru64) or Windows XP as their onlyM	 options.'o    E I don't agree with J-F about the scuttling of VMS but I do agree with E him on the point that a customers perception is heavily influenced bydE the people who are doing the sales.  If Compaq's marketing department.F has anything to do with it VMS will be like hardporn magazines in someE countries, only available in brown-paper wrapper to people who reallyi know where to ask for it.      John McLeang   ------------------------------  % Date: Tue, 23 Oct 2001 13:48:52 -0400a5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>t- Subject: Re: Compaq: VMS is alive and kickingc1 Message-ID: <a5iB7.826$RL6.9644@news.cpqcorp.net>c  % Terry C. Shannon wrote in message ...  >n6 >"Bill Todd" <billtodd@metrocast.net> wrote in message? >news:yt4B7.1082659$ai2.82435708@bin2.nnrp.aus1.giganews.com...  >>@ >> Robert Deininger <rdeininger@mindspring.com> wrote in messageI >> news:rdeininger-2210011353550001@user-2ivea8a.dialup.mindspring.com...cL >> > In article <HMVA7.113479$%B.8809579@bin1.nnrp.aus1.giganews.com>, "Bill* >> > Todd" <billtodd@metrocast.net> wrote: >> > >> >F >> > > Aside from his obvious propensity for marketeer-speak, to stateF >> > > that EV7 has "memory access speeds 300 to 500 times faster than >today'sH >> > > offerings" ranks pretty high on the technical-incompetence meter. >> > >> >7 >> > But that isn't quite what the article said, is it?. >>L >> Since I cut the quote directly from the article, I'm reasonably sure that >ith >> is. >> >> >H >> > "Somewhere in the 2004 timeframe, Compaq will ship 64-processor andK >> > perhaps 128-processor Alpha systems based on the EV7 processor family.eK >> > Code-named "Marvel," these are, Gorham says, "big honkin' systems withcI >> > lots of memory" that are very scalable, very modular, and very fast,l >witheJ >> > memory access speeds 300 to 500 times faster than today's offerings." >> >F >> > He said EV7 processor _family_, not EV7.  That probably means the future) >> > EV79, or whatever they will call it.n >>K >> The EV79 is projected to be mostly a process-shrink of the EV7, in whichfJ >> case while its clock rate may improve there's no reason to believe that >itsK >> memory-access latency will improve (though unrelated memory developmentsuJ >> over time could help a bit).  I'm not a memory sub-system expert, but IK >> could imagine that memory bandwidth could improve by as much as the EV79CJ >> clock rate improves - but this is more like a factor of 2 than a factor of >> several hundred.i >.H >That seems correct. Whether Mark misspoke or was misinterpreted remains >uncertain.Y >< >> >> >L >> > He also talked about 64- or128-CPU systems 3 years or so in the future. >IK >> > think there is plenty of room to speed up the entire memory subsystem.L >>< >> There is - but hardly by 2.5 or more orders of magnitude. >r
 >No doubt! >r >> >>   IK >> > don't have the specs handy, but perhaps someone has numbers available.nJ >> > What is the maximum per-CPU memory bandwidth on today's EV7 chips, in >theK >> > ideal situation where the memory subsystem has tons of cache and never K >> > makes the CPU wait?  How much better will EV79 be?  Now multiply these5 >> > numbers by 64 or 128. >>I >> I could have missed the 128-processor Alphas when I looked at the roadM >mapsmE >> in the recent CETS presentations:  were they still visible (I onlyV	 >remember. >> 64P configurations)?: >SF >Good point. While early Marvel collateral showed 128-way systems, the recentE >roadmaps cite 64P boxes. I am told that the latest PID reflects 128P  systemslI >but don't know for sure. Given the Marvel design, if you can build a 64PdB >system, you ought to be able to scale it to 128P simply by adding
 additional >8P Marvel building blocks.  >v    L The design architecturally is for up to 256 EV7 chips.  Cable routing on theH platform limits it on the Marvel box.  Our roadmap is only for up to 64p= right now, but 128p might be possible I've heard unofficially   J But I'll bet if you wanna buy a couple terrabytes of memory, someone might throw in the extra CPUs ;-) ;-)-   ------------------------------  % Date: Tue, 23 Oct 2001 13:56:30 -0400 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> = Subject: Re: Converting an ODS-2 to an ODS-5 file system disk 1 Message-ID: <kciB7.827$RL6.9577@news.cpqcorp.net>0  K The simple truth is that most VAX customers are running the oldest possibleCH version that they can safely get away with.  Lots of V5.1, with the restJ mostly dragged forward by cluster version compatability issues with Alpha.  L Backporting the 64-bit Alpha code to VAX would be a rather large undertakingD for no real return on the investment (there are no NEW customers for	 VAX/VMS).   J The concept of "porting" to IA32 is not novel, but it would be a large and problematic job.      E Tim Llewellyn wrote in message <3BD4DF2E.3D94A2CB@cableinet.co.uk>...a >C >y >David Froble wrote: >hG >> To ask VAX users if they want stability, and equate a 'Yes' for thatoJ >> question to be an automatic 'No' for new features would show tremendousI >> bias.  I don't know what questions were asked.  I pretty sure that VMSeK >> software maintenance costs do not differ between VAX and Alpha.  I thinkMH >> that the VAX customers on software maintenance are getting the shaft. >  > C >you have a point, but then, if, say, the IPF port of OpenVMS ALphatH >was augmented by a port to VAX and explicit support of 32 as well as 64E >bit architectures (which might actually be the best 1st step towardsnB >vms on 32 bit Intel platforms, if anyone cared), then the upgradeE >to, say, VMS 8.0 ne SuperOpenVmsThatWillFinallyRuleTheITWorld, wouldmB >probably be to much risk for those hanging onto their VAXen. Some@ >of them havn't even upgraded to VMS 6, as I'm sure you realize. >i> >Some might argue that those still running on VAX have skimped@ >on their TCO budget's by failing to keep up to date. However as7 >most of them probably were actively discouraged at thec >buisiness/marketting = >level by their vendor from continuing investment in VMS thent< >I agree with your basic point that things should be done to; >bring them back on a positive heading with respect to VMS.m >tD >However, all this may well be moot due to architercural differencesE >between the platforms, and/or the development and testing budget(s).r >w >regards >--u >Tim.Llewellyn@cableinet.co.uk > C >Standard disclaimer applies. My views in no way represent those of " >my employers or service provider.   ------------------------------    Date: 23 Oct 2001 01:46:05 -0700  From: itchimily@yahoo.com (Eran) Subject: copy files in DCLedit= Message-ID: <7f503f55.0110230046.6eccff50@posting.google.com>    Hi,nC I try to copy files fron one directory to another using the DCLeditM tool.o@ I need to do it in DOS environment and therefore I can't use the "copyFiles" command. Any suggestions? Thanks   ------------------------------    Date: 23 Oct 2001 06:02:30 -0700( From: bob@instantwhip.com (Bob Ceculski)$ Subject: Re: DECNET ping equivalent?= Message-ID: <d7791aa1.0110230502.4b50fa61@posting.google.com>t  j Tim Llewellyn <tim.llewellyn@cableinet.co.uk> wrote in message news:<3BD43520.99F0284A@cableinet.co.uk>... > Paul Repacholi wrote:i > > 9 > > Tim Llewellyn <tim.llewellyn@cableinet.co.uk> writes:  > > J > > > but does it reach out and "touch" the requested node like ping does?H > > > Maybe I am wrong and it does, can't test at the moment as I am VMS > > > deprived.h >  jF > > > Anyway, you don't get packet loss figures like you do with ping. > >  > > OK then,$ > > $ mc ncp sho node <foo> counters > > if you will. > > B > > When the node come up, it hellos the adjacent nodes. It or oneG > > of its adjacent nodes MUST be a router except in the case of a fulleD > > adjacent net. So the status is propergated via the routing info. > > 6 > > This is a real network, 'reachable' is for real ;) > > > OK, I was being unduly pedantic and I am sorry. Steve SpiresB > I believe did post about DTS and counters earlier in the thread. > A > I do fondly remember the days when once you go the MAIL> promptiG > back after a send you knew the mail had been delivered to the remote s > mailbox over DECNET.  B why remember it?  use it!  we use tcpware for phase iv over ip andA smtp over decnet mail ... if you look a little while, you usuallyiC can find the products you need for vms, if not, write it!  there is B no excuse not to be running vms now or in the immediate future ...6 unless you want mr nimdas IIS nightmare ... good luck!   ------------------------------  % Date: Tue, 23 Oct 2001 09:08:28 -0400 , From: "Kenneth Block" <krblock@computer.org>R Subject: Re: dynamic_cast failure in Compaq C++ version 6.3 on VMS on DEC Alpha...1 Message-ID: <gZdB7.809$RL6.8888@news.cpqcorp.net>i  G V6.3 had one bug with dynamic_cast. It was specific to casting from oneTI standard library type to annother: strstream to stream, I beleieve. If in A your example, you are casting user types, then it is not related.k  J If you can reduce the problem down to 100 lines of code, you can send that to compaq_cxx.bugs@compaq.com.   ------------------------------  # Date: Tue, 23 Oct 2001 17:37:11 GMTn0 From: Monty Brandenberg <mcbinc@ne.mediaone.net>Y Subject: Re: dynamic_cast failure in Compaq C++ version 6.3 on VMS on DEC Alpha... Alpha.i/ Message-ID: <3BD5AB0A.80A01287@ne.mediaone.net>a   "James E. Melter" wrote: > F > cout<<"====================================================="<<endl;: > cout<<"pParent type is "<<typeid(*pParent).name()<<endl;F > cout<<"====================================================="<<endl; > H > and the type is correct so the downcast should work, but the assertionE > still fails in this code immediately following(Note that Child is at! > subclass of Parent, obviously):S > 4 >     Child *pChild= dynamic_cast<Child *>(pParent); >     assert (pParent!= NULL);   assert (pChild != NULL);  ?????o  ; If your assert was firing, then pChild would have been 0 as'9 well.  And I believe it's perfectly fine to invoke typeid < on what looks like a dereferenced null pointer (someone have? chapter/verse?).  And on the pedantic side, zero is the definedn non-object:e   assert (0 != pChild);o   or   assert (pChild);   -- uM Monty Brandenberg, Software Consultant                              MCB, Inc.wM mcbinc@world.std.com                                          P.O. Box 426188 M mcbinc@ne.mediaone.net                              Cambridge, MA  02142-0021g 617.864.6907   ------------------------------    Date: 23 Oct 2001 10:17:10 +0200G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>s Subject: Re: ECCH Message-ID: <y47ktmixdl.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  . Paul Repacholi <prep@prep.synonet.com> writes:  A > VaxVMS would examine the page tables and: reload the page if it>= > was RO, Kill the process if it was process private, or elsei
 > the system.s    Right, that's how I remember it.   	Jan   ------------------------------  % Date: Tue, 23 Oct 2001 14:39:17 -0000o- From: wspencer@ap.nospam.org (Warren Spencer)o Subject: Re: ECC7 Message-ID: <914365C24warrenspencer1977@207.126.101.97>4  J moskal@zk3.dec.com (Andy Moskal) wrote in <3BD0A992.B351D471@zk3.dec.com>:  
 -- snip --  B >Alpha systems with ECC memory survive single bit errors.  So will	 >Itanium.-D >You can build a Alpha system or a Itanium system with Parity memory >insteadH >of ECC - in which case the system will crash on a simgle bit error.  It >isrB >pretty unlikely you will ever see this on any server from Compaq.  
 -- snip --  F Question:  In an Alpha 1000A I bought (for my employer) in 1996, each ; memory bank required five simms - 4 for data and 1 for ECC.u  B In an Alpha 1000A purchased 2 years later, the ECC sockets on the K motherboard were gone - 4 simms required per bank rather than five.  I had 0M boldly assumed Compaq had done away with ECC on this model of Alpha Server.  1 Was I incorrect?   ws   --     Warren Spencer' Senior Software Engineer (not a writer)s The Associated Press  L ** My employer does not necessarily agree with my statements - neither do I  **   ------------------------------  % Date: Tue, 23 Oct 2001 16:03:32 +0100 * From: "Richard Brodie" <R.Brodie@rl.ac.uk> Subject: Re: ECC+ Message-ID: <9r40s4$hu4@newton.cc.rl.ac.uk>e  : "Warren Spencer" <wspencer@ap.nospam.org> wrote in message1 news:914365C24warrenspencer1977@207.126.101.97...i   >rG > Question:  In an Alpha 1000A I bought (for my employer) in 1996, eacho= > memory bank required five simms - 4 for data and 1 for ECC.n > C > In an Alpha 1000A purchased 2 years later, the ECC sockets on theeL > motherboard were gone - 4 simms required per bank rather than five.  I hadM > boldly assumed Compaq had done away with ECC on this model of Alpha Server.n > Was I incorrect?  O I believe so. IIRC the difference was that the later memory modules had the ECC O on the module itself. The 4 + 1 business always seemed a kludge, although therei0 was no doubt a good reason to do it at the time.   ------------------------------  # Date: Tue, 23 Oct 2001 16:34:31 GMTt2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)I Subject: Re: Fat Binaries? (was: Re: WRITEBOOT and dual-boot bootblocks?) 1 Message-ID: <r_gB7.821$RL6.9088@news.cpqcorp.net>o   In article <rdeininger-2210012042210001@user-2ive6ab.dialup.mindspring.com>, rdeininger@mindspring.com (Robert Deininger) writes:c@ :Come to think of it, VMS "fat binaries" would be very useful...C :Is this a practical thing to implement in a future version of VMS?g  I   OpenVMS Engineering has looked at the requirements both in the context  J   of VAX and Alpha systems and in the context of Alpha binaries generated C   to use specific Alpha microprocessor features, and it would mean  I   additional parallel image activation paths and image headers, extensivewF   rework or a rewrite of the linker, of the compilers, and of various H   related tools (DECset PCA, SDA, ANALYZE/OBJECT, etc), the addition of H   file system support for threaded files, a tool to strip the binaries, H   etc.  We have also looked at variations, including shadow directories H   and search lists, etc.  At support issues.  We have further looked at G   the increase in the size of the distribution kits, as the "fat" kits lI   will be (larger than) the aggregate size of the existing VAX and Alpha t   kits.r  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------    Date: 23 Oct 2001 04:56:04 -0700! From: dodi_abx@hotmail.com (dodi)  Subject: fixing Error Count...< Message-ID: <fd561df.0110230356.2f1c9a03@posting.google.com>  T Is there someone can help me, about how to check/prevent our disks from error count.   ------------------------------    Date: 23 Oct 2001 09:39:57 -0500- From: koehler@encompasserve.org (Bob Koehler)t" Subject: Re: fixing Error Count...3 Message-ID: <275PH6NUV$Rl@eisner.encompasserve.org>   ` In article <fd561df.0110230356.2f1c9a03@posting.google.com>, dodi_abx@hotmail.com (dodi) writes:J > Is there someone can help me, about how to check/prevent our disks from  > error count.      Turn them off.u  E    You've got to give us at least a clue to what kind of disks you'reaH    using if you want real help.  Soft errors on older disks were routineF    and routinely ignored.  Hard errors mean it's time to buy somethingE    new.  Very old versions of VMS would log an error if you mounted at4    write locked disk without the /nowrite qualifier.  ;    So how are we supposed to figure out what you are doing?o   ------------------------------    Date: 23 Oct 2001 08:10:56 -07001 From: jensenc@augsburgfortress.org (Chris Jensen) & Subject: FTP timing out with OPCOM msg= Message-ID: <8ab621e2.0110230710.499eed1b@posting.google.com>   E My company has two OpenVMS 7.2 servers. We are getting occasional ftpI; service timeouts on both with the following OPCOM messages:   8 %%%%%%%%%%%  OPCOM  16-OCT-2001 23:01:15.60  %%%%%%%%%%%$ Message from user INTERnet on ALPHA25 INTERnet ACP AUXS failure  Status = %SYSTEM-F-DUPLNAM  	 8 %%%%%%%%%%%  OPCOM  16-OCT-2001 23:01:15.62  %%%%%%%%%%%$ Message from user INTERnet on ALPHA2$ INTERnet ACP Timeout Idle FTP Server  ? The above message would indicate that there is a duplicate namesE somewhere. These systems have the imaginative hostnames of ALPHA1 andoA ALPHA2, and at one time they switched identities when the systems-E swapped roles as the production and dev servers. I thought we changedFF the names everywhere they needed to be, but it looks like the old name may be lingering somewhere.g  3 Any help would be most appreciated. Thanks! --Chris.   ------------------------------  % Date: Tue, 23 Oct 2001 18:57:37 +0200r= From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <noone@dummy.com>.* Subject: Re: FTP timing out with OPCOM msg) Message-ID: <3BD5A181.C185045D@dummy.com>   = The DUPLNAM error can occure if one process tries to set it'sg? process name to a name that already exists. Check that there is.; no "SET PROC/NAME=xxx" in the LOGIN.COM of the account usedt9 to logon through FTP. In that case this could happen whenr( the second user tries to logon with FTP.   Jan-Erik Sderholm.e   Chris Jensen wrote:s > : > %%%%%%%%%%%  OPCOM  16-OCT-2001 23:01:15.60  %%%%%%%%%%%& > Message from user INTERnet on ALPHA27 > INTERnet ACP AUXS failure  Status = %SYSTEM-F-DUPLNAM. >>   ------------------------------  % Date: Tue, 23 Oct 2001 10:50:26 +0100h% From: Alan Greig <a.greig@virgin.net>h2 Subject: Re: Fwd: Compaq: VMS is alive and kicking8 Message-ID: <p4fattk23oiq7rp8d14buvuvtk98qajjdh@4ax.com>  E On Mon, 22 Oct 2001 05:11:26 -0400, Ken Robinson <kenrbnsn@rbnsn.com>  wrote:   >nH >This is a very interesting article that came via the ZDNet Tech update:  F Yes, it is positive and that's got to be good but the original articleF this is a follow-up to quotes Gartner as saying that, after talking toD Carly, they believe that VMS will be phased out. The Compaq VMS team obviously say the opposite.e   -- Alan   ------------------------------  % Date: Tue, 23 Oct 2001 10:55:05 -0400t- From: "John Eisenschmidt" <jeisensc@aaas.org>eK Subject: Fwd: OpenVMS VMS721_DCL-V0100 Alpha V7.2-1 DCL ECO Summary[ONHOLD]o+ Message-ID: <sbd54ca8.050@AAASMTA.aaas.org>i  F This is a MIME message. If you are reading this text, you may want to F consider changing to a mail reader or gateway that understands how to ( properly handle MIME multipart messages.   --=_FAA019F8.E081CD09r* Content-Type: text/plain; charset=US-ASCII+ Content-Transfer-Encoding: quoted-printables Content-Disposition: inline     Man, it is not my week for ECOs.   --=_FAA019F8.E081CD09  Content-Type: message/rfc822   Received: from science.aaas.orgp5 	by AAASMTA.aaas.org; Tue, 23 Oct 2001 10:36:08 -0400aP Received: from list.support.compaq.com (list.support.compaq.com [192.208.35.17])= 	by science.aaas.org (8.11.3/8.9.1) with SMTP id f9NEr5h02177b9 	for <jeisensc@aaas.org>; Tue, 23 Oct 2001 10:53:05 -0400t+ Date: Tue, 23 Oct 2001 08:30:02 -0600 (MDT)i= From: system PRIVILEGED account <root@stage1.cxo.cpqcorp.net>>^ Message-Id: <LYRIS-14242-31982-2001.10.23-07.29.27--jeisensc#aaas.org@list.support.compaq.com>B To: "OpenVMS Patch Mailing List" <openvms@list.support.compaq.com>G Subject: OpenVMS VMS721_DCL-V0100 Alpha V7.2-1 DCL ECO Summary[ON HOLD] G List-Unsubscribe: <mailto:leave-openvms-14242L@list.support.compaq.com>o  Reply-To: ECO-Queries@compaq.com  O *******************************************************************************lO *                                                                             *sO *                     This is a newly released patch...                       *iO *                                                                             *iO *  Online links can be found at                                               *of *    http://ftp.support.compaq.com/patches/public/vms/axp/v7.2-1/dec-axpvms-vms721_dcl-v0100--4.READMEO *******************************************************************************o    E TITLE: OpenVMS VMS721_DCL-V0100 Alpha V7.2-1 DCL ECO Summary[ON HOLD]-  = NOTE:  An OpenVMS saveset or PCSI installation file is storede;        on the Internet in a self-expanding compressed file.O  2?        For OpenVMS savesets, the name of the compressed savesete<        file will be kit_name.a-dcx_vaxexe for OpenVMS VAX or@        kit_name.a-dcx_axpexe for OpenVMS Alpha. Once the OpenVMS>        saveset is copied to your system, expand the compressedF        saveset by typing RUN kitname.dcx_vaxexe or kitname.dcx_alpexe.  4C        For PCSI files, once the PCSI file is copied to your system,aC        rename the PCSI file to kitname-dcx_axpexe.pcsi, then it can H        be expanded by typing RUN kitname-dcx_axpexe.pcsi.  The resultantK        file will be the PCSI installation file which can be used to installC        the ECO.>  aE Copyright (c) Compaq Computer Corporation 2001.  All rights reserved.a   New Kit Date     :  19-OCT-2001o Modification Date:  23-OCT-2001'3 Modification Type:  Kit placed on Engineering Hold.   : This kit has been placed on hold by OpenVMS Engineering.     PROBLEM STATEMENT:  J After installing the VMS721_DCL-V0100 kit the f$cvtime() function does not work.    PROBLEM SYMPTOM:  M The behavior of fcvtime() before installation of the VMS721_DCL-V0100 kit is:    $ WRITE SYS$OUTPUT f$cvtime()e 2001-10-22 13:10:08.99  L The behavior of fcvtime() after installation of the VMS721_DCL-V0100 kit is:   $ WRITE SYS$OUTPUT f$cvtime()v  B   Improperly handled condition, bad stack or no handler specified.1     Signal arguments:   Number = 0000000000000005I1                         Name   = 000000000000000C 1                                  0000000000000000n1                                  0000000000000004S1                                  000000007AFE39ACe1                                  0000000000000012    WORKAROUND:e  I In order to revert back to the pre-VMS721_DCL-V0100 behavior, perform them following steps:  O - from a privileged account rename the file SYS$COMMON:[SYSEXE]DCL.EXE_OLD;1 top0   be the latest SYS$COMMON:[SYSEXE]DCL.EXE file.   - Execute the command:  7    INSTALL REPLACE SYS$SYSTEM:DCL.EXE/SHARE/OPEN/HEADERo   ALTERNATIVE KIT:  M There is no alternative kit.  OpenVMS Engineering is developing a solution to(M this problem.  When this solution is available it will ship in a new DCL kit.m   OP/SYS:     OpenVMS Alpha    COMPONENT:  DCL   ' SOURCE:     Compaq Computer Corporation-   ECO INFORMATION:  $      ECO Kit Name:  VMS721_DCL-V01007                     DEC-AXPVMS-VMS721_DCL-V0100--4.PCSI /      ECO Kits Superseded by This ECO Kit:  Nonef*      ECO Kit Approximate Size:  800 Blocks*      Kit Applies To:  OpenVMS Alpha V7.2-1*      System/Cluster Reboot Necessary:  No 0      Rolling Re-boot Supported:  Not Applicable $      Installation Rating:  INSTALL_3E                            3 - To be installed on all systems running7C                                the listed versions of OpenVMS whichgG                                are experiencing the problems described.m        Kit Dependencies:  H        The following remedial kit(s), or later, must be installed BEFORE         installation of this kit:            VMS721_UPDATE-V0300          VMS721_PCSI-V0100  =        In order to receive all the corrections listed in thissA        kit, the following remedial kits should also be installed:             None    ECO KIT SUMMARY:  3 An ECO kit exists for DCL on OpenVMS Alpha V7.2-1. s+ This kit addresses the following problems: K  + PROBLEMS ADDRESSED IN VMS721_DCL-V0100 KIT:   C   o  Nested commands  reset  the  SYS$COMMAND  input  stream  after       completing.  +           Images Affected:  [SYSEXE]DCL.EXEp      C   o  The F$CONTEXT command returns the wrong status.   See  examplea      below:i    F              $ dummy=f$context("process",ctx,"uic","nobrackets","eql")8              %DCL-S-NORMAL, normal successful completion  +           Images Affected:  [SYSEXE]DCL.EXEa    C   o  The F$DEVICE lexical function  will  not  find  some  devices.oB      Even though the TLZ10 is a supported device, the DCL command:  :            $ WRITE SYS$OUTPUT F$DEVICE("*","TAPE","TLZ10")  ,      results in the following error message:  ?            %DCL-W-IVKEYW, unrecognized keyword - check validity             and spellingo  +           Images Affected:  [SYSEXE]DCL.EXEe    C   o  The F$CVTIME command returns IVATIME for legal arguments.  The C      command  should  return  the zero hour of the date included in-)      the input_time argument of F$CVTIME:1  &              AJAX> b=f$cvtime("24-",,)               AJAX> SHOW SYMBOL b+                B = "2001-04-24 00:00:00.00".  A      If the F$CVTIME command is preceded with a certain F$EXTRACTp)      command, the F$CVTIME command fails:t  *              AJAX> a=f$extract(0,1,"1234")&              AJAX> b=f$cvtime("24-",,)8              %DCL-W-IVATIME, invalid absolute time - use4                 DD-MMM-YYYY:HH:MM:SS.CC format \24-\  ,            Images Affected:  [SYSEXE]DCL.EXE    C   o  The DEFINE command corrupts the image context registers, which-:      prevents $CONTINUE.  See partial error display below:             $ CREATE a.a           adsf;kjf;fkj            Interrupt             $ DEFINE aaaaa bbbbo           $ CREATE  A             Improperly handled condition, bad stack or no handlert              specified.=;               Signal arguments:   Number = 0000000000000005 ;                                   Name   = 000000000000000C 9                                          0000000000010000=;                                            7FFAD25200000004d9                                          7FFAD25200000004a;                                            0000000000000012b               Register dump:<               R0  = 0000000000000000  R1  = 000000007FFAD788<               R2  = 0000000000000000  R3  = 0000000000000004<               R4  = 0000000000000000  R5  = 0000000000000000<               R6  = 000000007FFAD778  R7  = 000000007FFAD778<               R8  = 000000007FFABD48  R9  = 000000007FFABF50<               R10 = 000000007FFADFC0  R11 = 000000007FFCE3E0<               R12 = 0000000000000000  R13 = 000000007AFD66A0<               R14 = FFFFFFFF823EE1C0  R15 = 000000007AFD5990<               R16 = 000000000000001B  R17 = 000000007FFAD778<               R18 = 000000007FFAD778  R19 = 0000000000000000<               R20 = 0000000000000000  R21 = FFFFFFFFFFFFFFA8<               R22 = 0000000000000000  R23 = 0000000000000001<               R24 = 0000000000000000  R25 = 0000000000000000<               R26 = 000000007FFADFC0  R27 = 0000000000000008<               R28 = 7FFAD25200000004  R29 = 7B0296A60000000B<               SP  = 000000007FFABD05  PC  = 7FFAD25200000004$               PS  = 0500000000000012E           %SYSTEM-F-ACCVIO, access violation, reason mask=00, virtuale           address=0000000000000           000C, PC=0000000000000012, PS=00000002  +           Images Affected:  [SYSEXE]DCL.EXE9       INSTALLATION NOTES:a  @ No reboot is necessary after successful installation of the kit.   INSTALLATION INSTRUCTIONS:  C Install this kit with the POLYCENTER Software installation  utility1C by logging into the SYSTEM account, and typing the following at thee DCL prompt:r  4 PRODUCT INSTALL VMS721_DCL /SOURCE=[location of Kit]  B The kit location may be a tape drive, CD, or a disk directory that contains the kit.l  > Additional help on installing PCSI kits can be found by typing) HELP PRODUCT INSTALL at the system prompt   . Scripting of Answers to Installation Questions  H   o  During installation, this kit will ask and require user response toH      several  questions.   If  you  wish to automate the installation ofH      this kit and avoid having to provide responses to these  questions,H      you must create a DCL command procedure that includes the following      definitions and commands:  -      1.  Define logical NO_ASK$BACKUP as TRUEe  F      2.  If a reboot is required, define logical NO_ASK$REBOOT as TRUE  H      3.  Add the following qualifiers to the PRODUCT INSTALL command and/          add that command to the DCL procedure.   )            /PROD=DEC/BASE=AXPVMS/VER=V1.0<    (      4.  De-assign the logicals assigned  H      For example, a sample command file to install the  VMS721_DCL-V0100      kit would be:  $      $ DEFINE/SYS NO_ASK$BACKUP TRUE      $!7<      $ PROD INSTALL VMS721_DCL/PROD=DEC/BASE=AXPVMS/VER=V1.0      $!n!      $ DEASSIGN/SYS NO_ASK$BACKUPd      $ exit      ---a= You are currently subscribed to openvms as: jeisensc@aaas.orgeQ To unsubscribe send a blank email to leave-openvms-14242L@list.support.compaq.comh   --=_FAA019F8.E081CD09--d   ------------------------------  % Date: Tue, 23 Oct 2001 12:30:40 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)O Subject: Re: Fwd: OpenVMS VMS721_DCL-V0100 Alpha V7.2-1 DCL ECO Summary[ONHOLD] L Message-ID: <rdeininger-2310011230400001@user-2ive7bq.dialup.mindspring.com>  ? In article <sbd54ca8.050@AAASMTA.aaas.org>, "John Eisenschmidt", <jeisensc@aaas.org> wrote:  H > This is a MIME message. If you are reading this text, you may want to H > consider changing to a mail reader or gateway that understands how to * > properly handle MIME multipart messages.    F Sigh.  Your newsreader/mailer has gotten away from you, John.  Time to slap it silly.  G It looks like your forwarded messages are being converted to nasty MIMEbI attachments.  And your plain text is turning into quoted-printable, whicho0 is actually a misspelling for QUOTED-UNREADABLE.  & Please, plain text only in comp.os.vms   -- a Robert Deininger rdeininger@mindspring.comb   ------------------------------    Date: 23 Oct 2001 08:48:27 -0500- From: koehler@encompasserve.org (Bob Koehler)  Subject: Get a Macintosh3 Message-ID: <EyATvNqKJidn@eisner.encompasserve.org>n  a In article <tt8etjetciki7f@corp.supernews.com>, Michael Zarlenga <zarlenga@conan.ids.net> writes:d  E > Soon I will booting Linux as well as ME.  With luck, I will be able 6 > to use that O/S for most of my time spent on the PC.  G    My kid boots Linux on his iMac all the time.  Sometimes he goes back C    and boots MacOS, and sometimes he can get it to run under Linux.u  E    Having recently got my first Mac OS X system, I think I've got theeF    much better solution.  Anytime I really feel like using UNIX I justF    bring up the shell.  The rest of the time I'm using a Mac that acts    like a Mac.  No rebooting.-  C    I'm no fan of UNIX, but Apple is the first vendor to deliver thel9    promise of hiding the pains of UNIX behing a good GUI.f   ------------------------------    Date: 23 Oct 2001 02:46:28 -0700# From: torpeman@hotmail.com (Benito)0* Subject: Re: How know last system restart?= Message-ID: <cb4a5b25.0110230146.7bac4a91@posting.google.com>0   Thank you, everybody.g   Regards.  h torpeman@hotmail.com (Benito) wrote in message news:<cb4a5b25.0110220837.4839931f@posting.google.com>... > Hello, > G > Im new in VMS and i need to know the last system boot date/time, theb& > VMS is OpenVMS AXP Version V7.1-1H2. >  > Can anybody help me. >  > Thanks & Regards.e   ------------------------------  % Date: Tue, 23 Oct 2001 08:50:14 -0400_  From: jamese@beast.dtsw.army.mil* Subject: Re: How know last system restart?0 Message-ID: <01102308501434@beast.dtsw.army.mil>  O spamsink2001@yahoo.com (Alan E. Feldman) wrote on 22 Oct 2001 19:20:45 -0700 iny1 <b096a4ee.0110221820.1fe9317@posting.google.com>:s  = > Michael Zarlenga <zarlenga@conan.ids.net> wrote in message o- > news:<tt8tvr67rh594e@corp.supernews.com>...r& > > Tom Linden <tom@kednos.com> wrote:0 > > : Now, if I wanted to define in my login.comA > > : uptime== the below text, how do I escape the double quotes?h. > > :> $ write sys$output f$getsyi("boottime") > > : > > uptime :== "write sys$output f$getsyi("""boottime""")" > > 5 > > Once inside "s, you have to use 2 ""s to get 1 ".M >  > Then why did you use 3?0  I That's what I wondered, so I tried both ways and got only one difference.s Can anyone explain this?  <     $ uptime :== "write sys$output f$getsyi("""boottime""")"       $ show symbol uptimeK       UPTIME == "write sys$output f$getsyi("BOOTTIME")"	  ! NOTE Upper CaseN       $ uptime     10-OCT-2001 18:23:17.00U  :     $ uptime :== "write sys$output f$getsyi(""boottime"")"       $ show symbol uptime7       UPTIME == "write sys$output f$getsyi("boottime")"        $ uptime     10-OCT-2001 18:23:17.00s  3 The "==" approach also gives lower case "boottime":m  9     $ uptime == "write sys$output f$getsyi(""boottime"")"h       $ show symbol uptime7       UPTIME == "write sys$output f$getsyi("boottime")"v       $ uptime     10-OCT-2001 18:23:17.00e  : Ed James                           ed.james@telecomsys.com5 TeleCommunications Systems, Inc.   voice 410-295-1919t; 2024 West Street, Suite 300              800-810-0827 x1919e5 Annapolis, MD 21401-3556           fax   410-280-1094o   ------------------------------    Date: 23 Oct 2001 09:20:07 -0500- From: koehler@encompasserve.org (Bob Koehler)D* Subject: Re: How know last system restart?3 Message-ID: <ieboxKx0aVB2@eisner.encompasserve.org>   | In article <3bd44e35$0$18120$6e365a64@newsreader02.highway.telekom.at>, "Ren Schelbaum" <rene.schelbaum@datakom.at> writes: > N > 1.) if you have accounting enabled a "acc/fu/type=sysinit" will give you all	 > rebootsS  B Not if you have the good sense to migrate security data to offline storage.  M > 2.) if not, do a "show system". you will have the uptime in the first line.a3 > A little calculation will give you the boot-time.-      The lexicals are easier.,   ------------------------------  % Date: Tue, 23 Oct 2001 11:14:27 -0400c  From: jamese@beast.dtsw.army.mil* Subject: Re: How know last system restart?0 Message-ID: <01102311142742@beast.dtsw.army.mil>  < Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> wrote on( Tue, 23 Oct 2001 15:36:26 +0200 (MET) in0 <01K9UFCJJTEM90VPIH@sysdev.deutsche-boerse.com>:  G > Please CC to the group, since my posts aren't getting through at the  	 > moment.a > J > -------------------8<--------------------------------------------------- >  > > Can anyone explain this? >  > Yes. > @ > >     $ uptime :== "write sys$output f$getsyi("""boottime""")" > >  > >     $ show symbol uptimeN > >      UPTIME == "write sys$output f$getsyi("BOOTTIME")"   ! NOTE Upper Case > >  a > >     $ uptime > >     10-OCT-2001 18:23:17.004 > > > > >     $ uptime :== "write sys$output f$getsyi(""boottime"")" > >  > >     $ show symbol uptime; > >       UPTIME == "write sys$output f$getsyi("boottime")"w > >  > >     $ uptime > >     10-OCT-2001 18:23:17.000 > I > Completely clear.  The :== (as opposed to ==) says "what follows is to cC > be treated as a character string.  (An effect of this is that thenG > right-hand side of such an assignement does not have to be in quotes,8* > unless you want to preserve case etc.)   > K > After encountering a single quote, DCL will say "what comes is a string; oI > I'll preserve the case" until it comes to another single quote.  If it .H > comes to a double quote, it will replace it to a single quote.  Thus, ( > the first example gets tokenized into: > 2 >    `write sys$output f$getsyi("'  boottime  `")' > K > where I have used single quotes to denote the strings resulting from the 7J > parsing and the spaces are merely for clarity.  As you can see, boottimeJ > is not part of such a string, and so gets upcased.  The result is then, 4 > as you say, write sys$output f$getsyi("BOOTTIME"). > 0 > In the second example, the parsing results in: > * >    write sys$output f$getsyi("boottime") > C > In this case, the argument is enclosed in quotes, so doesn't get *J > upcased.  I think the key is to think of the first example as resulting I > in three tokens, which due to the fact that you used :== instead of ==  7 > were then re-combined into a single character string.  > 7 > > The "==" approach also gives lower case "boottime":e > > = > >     $ uptime == "write sys$output f$getsyi(""boottime"")": > J > Right.  In this case, the quotes say "this is a string", so you get the G > same result as in the second example above, in which the quotes were * > superfluous.   Thanks, Phillip. V  : Ed James                           ed.james@telecomsys.com5 TeleCommunications Systems, Inc.   voice 410-295-1919s; 2024 West Street, Suite 300              800-810-0827 x1919 5 Annapolis, MD 21401-3556           fax   410-280-1094e   ------------------------------  % Date: Tue, 23 Oct 2001 16:10:15 -0000 / From: Michael Zarlenga <zarlenga@conan.ids.net>A* Subject: Re: How know last system restart?/ Message-ID: <ttb5j78j84a9cc@corp.supernews.com>   / Alan E. Feldman <spamsink2001@yahoo.com> wrote:e9 :> uptime :== "write sys$output f$getsyi("""boottime""")"  :> p4 :> Once inside "s, you have to use 2 ""s to get 1 ".   : Then why did you use 3?i  6 My Windows keyboard driver must've malfunctioned.  ;')   --   -- Mike Zarlenga   ------------------------------  % Date: Tue, 23 Oct 2001 16:10:57 -0000 / From: Michael Zarlenga <zarlenga@conan.ids.net>e* Subject: Re: How know last system restart?/ Message-ID: <ttb5khthb68if5@corp.supernews.com>D  ( Carl Perkins <carl@gerg.tamu.edu> wrote:( : The easiest way is not to do so, i.e.:  4 : $ uptime :== Write Sys$Output F$GetSYI("BOOTTIME")  C : If you really want to use the plain old "==" definition operator,e4 : you can escape the double quotes by doubling them:  7 : $ uptime == "Write Sys$Output F$GetSYI(""BOOTTIME"")"C   : I prefer the former.   I would, too, if it worked.1   -- : -- Mike Zarlenga   ------------------------------    Date: 23 Oct 2001 04:41:14 -0700! From: dodi_abx@hotmail.com (dodi) ) Subject: How to enable security audit...? < Message-ID: <fd561df.0110230341.5a39d3ad@posting.google.com>  : Is there someone can assist me about the subject writen...D The Audit_server is already started, but when I entered "show audit" there is a message:0- System security audits currently enabled for:    FILE access:     Success:     delete 2     BYPASS:      read,write,execute,delete,control      Other:       create,deaccessB "%SHOW-W-NOAUDITING, security auditing disabled; no events will be logged".* Althought I've restart the Audit_server...  8 Pls..I do appreciate..If you can tell me the solution... Thx.   ------------------------------  % Date: Tue, 23 Oct 2001 16:01:55 +0400/4 From: "Ruslan R. Laishev" <Laishev@SMTP.DeltaTel.RU>- Subject: Re: How to enable security audit...? 0 Message-ID: <3BD55C33.21AC462A@SMTP.DeltaTel.RU>   Hi ! 	Have a look to OPCOM...
 $repl /enableS< $subm /user=system/nolog sys$system:startup /param=("AUDIT")     dodi wrote:r > < > Is there someone can assist me about the subject writen...F > The Audit_server is already started, but when I entered "show audit" > there is a message:O/ > System security audits currently enabled for:  >   FILE access: >     Success:     delete74 >     BYPASS:      read,write,execute,delete,control" >     Other:       create,deaccessD > "%SHOW-W-NOAUDITING, security auditing disabled; no events will be
 > logged"., > Althought I've restart the Audit_server... > : > Pls..I do appreciate..If you can tell me the solution... > Thx.   --   Cheers, F +OpenVMS [Sys|Net] HardWorker .......................................+E  Russia,Delta Telecom Inc,                    Cel:  +7 (901) 971-3222OE  191119,St.Petersburg,Transportny per. 3                     116-3222 F +http://starlet.deltatel.ru ................. SysMan rides HailStorm +   ------------------------------   Date: 23 Oct 2001 11:57:01 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)R- Subject: Re: How to enable security audit...?a0 Message-ID: <9r3lud$ra7$1@n.ruf.uni-freiburg.de>  ` In article <fd561df.0110230341.5a39d3ad@posting.google.com>, dodi_abx@hotmail.com (dodi) writes:; >Is there someone can assist me about the subject writen...gE >The Audit_server is already started, but when I entered "show audit"s >there is a message:. >System security audits currently enabled for: >  FILE access:r >    Success:     delete3 >    BYPASS:      read,write,execute,delete,controls! >    Other:       create,deaccessAC >"%SHOW-W-NOAUDITING, security auditing disabled; no events will bei	 >logged".n+ >Althought I've restart the Audit_server...0  K Once the server is running (which it is in your case) you need to configureI= it. See "HELP SET AUDIT", especially "HELP SET AUDIT/ENABLE".e   Regards,    Christoph Gartmann   H -- --------------------------------------------------------------------+H | Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452 |H | Immunbiologie                                                        |H | Postfach 1169                 Internet: gartmann@immunbio.mpg.de     |H | D-79011  Freiburg, FRG                                               |H +--------- http://www.immunbio.mpg.de/home/english/menue.html ---------+   ------------------------------  % Date: Tue, 23 Oct 2001 08:45:44 -0400c* From: WILLIAM WEBB <WWEBB1@email.usps.gov>- Subject: RE: How to enable security audit...?(- Message-ID: <0033000039219454000002L042*@MHS>a   =0A$ HELP SET AUDIT/ALARM  $ HELP SET AUDIT/ENABLES   You gotta use both.    WWWebb   > -----Original Message-----1 > From: Info-VAX-Request@Mvb.Saic.Com at INTERNETe) > Sent: Tuesday, October 23, 2001 7:48 AMUD > To: Webb, William W Raleigh, NC; Info-VAX@Mvb.Saic.Com at INTERNET+ > Subject: How to enable security audit...?  >  > < > Is there someone can assist me about the subject writen...F > The Audit_server is already started, but when I entered "show audit" > there is a message:a/ > System security audits currently enabled for:  >   FILE access: >     Success:     delete04 >     BYPASS:      read,write,execute,delete,control" >     Other:       create,deaccessD > "%SHOW-W-NOAUDITING, security auditing disabled; no events will be
 > logged"., > Althought I've restart the Audit_server... > : > Pls..I do appreciate..If you can tell me the solution... > Thx. >=   ------------------------------  % Date: Tue, 23 Oct 2001 09:04:58 -040004 From: John Malmberg <Malmberg@dskwld.zko.dec.compaq>1 Subject: Re: HSZ40, KZPSA, AlphaStation 200 4/233 4 Message-ID: <3BD56AFA.8050605@dskwld.zko.dec.compaq>   Marty Kuhrt wrote:  H > I recently got ahold of an HSZ40 with some disk shelves and thought it     Known good?C  F > would be nice to set up a SCSI cluster with the HSZ in the middle atE > home.  I've done this with HSZ50s and KZPSA controllers in 4100s at1 > work without problem.       E > At home I have a pair of AlphaStation 200 4/233s, each with a KZPSA0G > and the appropriate Y cables, etc.  The SCSI ID for machine A's KZPSA D > is 5, machines B's is 6, and the HSZ is 7.  Both boxes are runningJ > V7.3, have the latest console firmware V7.0-9 and KZPSA firmware, A11.  2 > The HSZ is running V31-F of the firmware (IIRC). > G > When I try to get this to work the machines will crash at the console5H > level, or they won't show any of the disks hanging off the HSZ (I have: > a couple of units defined at the HSZ SHOW UNIT level).   > C > The crashes usually have a footprint starting with something like= > 5 > Avanti System Machine Check Through Vector 00000660 . > logout frame address 0x6048 code 0x100000205    : In a simpler hardware configuration, will the system boot?D Bascially start with just one controller and no disks and add things one at a time until it breaks.  D Have you got a good light and possibly a magnifying glass and tripleD checked all of your 68 pin SCSI connectors?  It is extremely easy to bend a pin or two.   Is all the termination correct?i  E I do not know if the AlphaStations you mentioned above are ones whereyF it can matter what order the options are what PCI slots.  Do you have  the hardware manuals for them?   -John  malmberg@dskwld.zko.dec.compaq Personal Opinion Onlys   ------------------------------    Date: 23 Oct 2001 12:03:40 -0500+ From: kuhrt@encompasserve.org (Marty Kuhrt)r1 Subject: Re: HSZ40, KZPSA, AlphaStation 200 4/233 3 Message-ID: <DzY5NXzaDyqV@eisner.encompasserve.org>o  k In article <3BD56AFA.8050605@dskwld.zko.dec.compaq>, John Malmberg <Malmberg@dskwld.zko.dec.compaq> writes:i > Marty Kuhrt wrote: > I >> I recently got ahold of an HSZ40 with some disk shelves and thought itU > 
 > Known good?a >   A Was when I removed it from production at work a month or two ago.D  G >> would be nice to set up a SCSI cluster with the HSZ in the middle at F >> home.  I've done this with HSZ50s and KZPSA controllers in 4100s at >> work without problem.    > I just checked the recently retired config at work and it is aA KZPSA (A11) in a 2100, hooked up to a HSZ40 (V31Z).  I should try C a single, straight through connection from the KZPSA to the Z40 an t see if it makes a difference.e   >  > G >> At home I have a pair of AlphaStation 200 4/233s, each with a KZPSA hH >> and the appropriate Y cables, etc.  The SCSI ID for machine A's KZPSAE >> is 5, machines B's is 6, and the HSZ is 7.  Both boxes are running K >> V7.3, have the latest console firmware V7.0-9 and KZPSA firmware, A11.  u3 >> The HSZ is running V31-F of the firmware (IIRC).0  3 I checked again, the home HSZ is running V31Z, too.b   >> 0H >> When I try to get this to work the machines will crash at the consoleI >> level, or they won't show any of the disks hanging off the HSZ (I have ; >> a couple of units defined at the HSZ SHOW UNIT level).  s >> tD >> The crashes usually have a footprint starting with something like >> m6 >> Avanti System Machine Check Through Vector 00000660/ >> logout frame address 0x6048 code 0x100000205r >  > < > In a simpler hardware configuration, will the system boot?F > Bascially start with just one controller and no disks and add things  > one at a time until it breaks.  I Both systems boot fine and form a two node cluster, with KZPSAs, but not b hooked up to the HSZ.  -   > F > Have you got a good light and possibly a magnifying glass and tripleF > checked all of your 68 pin SCSI connectors?  It is extremely easy to > bend a pin or two. > ! > Is all the termination correct?e >   B It is worth double checking.  I've replaced the terminators at theA ends of the Ys, and made sure the resistors are removed from KZs.o  G > I do not know if the AlphaStations you mentioned above are ones where H > it can matter what order the options are what PCI slots.  Do you have   > the hardware manuals for them? >   C (wheat grain sized light bulb goes on over his head) You know, lastvA time I got this to work, I think the KZPSA was in the second sloti@ down on the AS200.  In this case they're both in the "top" slot.   ------------------------------    Date: 23 Oct 2001 06:51:44 -0700( From: bob@instantwhip.com (Bob Ceculski) Subject: Re: I am changing jobsg= Message-ID: <d7791aa1.0110230551.17098715@posting.google.com>-  k "Sue Skonetski" <susan.skonetski@compaq.com> wrote in message news:<U%YA7.776$RL6.8375@news.cpqcorp.net>...s > Dear Newsgroup,s > N > I just wanted to let you know that I am changing jobs.  As of last week I amN > now part of the OpenVMS Technical Field Programs group and no longer part of > the OpenVMS Marketing group. > L > I will still be doing pretty much the same the same thing as before, but I > just wanted to let you knowS >  > Warm Regards,. >  > Sue   I the defcon9 "unhackable" article was great!  this needs to get out to the M IIS, nimda num pc crowd and to gartner that there is a secure web alternativejI to IIS, and its not unix, its vms!  why not circulate this article aroundtJ to the press, vendors, the web, everywhere?  why is compaq mngt (capellas)N keeping this a secret?  this is the vms promotion of a lifetime at the perfectO time gartner has just hammered IIS, and this article, besides the places myselfeO and a few others have posted or emailed, has been silenced to the openvms times F site ... am i wrong about this? whats wrong your marketing department?M are they being told to keep silent or do they really understand anything more  than windows and pcs?t   ------------------------------  % Date: Tue, 23 Oct 2001 11:14:47 -0400t( From: Hamlyn Mootoo <univms@bigfoot.com> Subject: Re: I am changing jobso+ Message-ID: <3BD58967.821300D4@bigfoot.com>   > I am curious as to why the change in title if your duties willC substantially be the same.  Is Compaq phasing out its VMS marketing E group? To be honest I did not know that you worked for that group--orNG that it existed for that matter (not being sarcastic, I honestly didn't H know there was a group specifically for VMS marketing).  Was this change6 voluntary or just part of an organizational reshuffle?   HM   Sue Skonetski wrote: >  > Dear Newsgroup,o > N > I just wanted to let you know that I am changing jobs.  As of last week I amN > now part of the OpenVMS Technical Field Programs group and no longer part of > the OpenVMS Marketing group. > L > I will still be doing pretty much the same the same thing as before, but I > just wanted to let you knowt >  > Warm Regards,v >  > Sue5   ------------------------------    Date: 23 Oct 2001 10:17:55 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)0 Subject: Re: I am changing jobs-3 Message-ID: <IQQOH4+Fgr1b@eisner.encompasserve.org>h  h In article <d7791aa1.0110230551.17098715@posting.google.com>, bob@instantwhip.com (Bob Ceculski) writes:  K > the defcon9 "unhackable" article was great!  this needs to get out to the2O > IIS, nimda num pc crowd and to gartner that there is a secure web alternative K > to IIS, and its not unix, its vms!  why not circulate this article around - > to the press, vendors, the web, everywhere?   B Just make sure it is sent by somebody who knows how to capitalize.   ------------------------------  % Date: Tue, 23 Oct 2001 09:40:43 -0500i+ From: Christopher Smith <csmith@amdocs.com>x Subject: RE: I am changing jobs-L Message-ID: <3B55D7F383B0D31197D9009027541CBF1170DDB3@cmiexch1.cmi.itds.com>  K As I understand it, it's not the whole department, only some of it.  Sue is2I certainly not the problem, but could be a solution if somebody would turno
 her loose. :)    Regards,   Chris1  ! Christopher Smith, Perl Developer9 Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");  'o  t   > -----Original Message-----8 > From: bob@instantwhip.com [mailto:bob@instantwhip.com]  = > the defcon9 "unhackable" article was great!  this needs to " > get out to the@ > IIS, nimda num pc crowd and to gartner that there is a secure  > web alternativet= > to IIS, and its not unix, its vms!  why not circulate this   > article around= > to the press, vendors, the web, everywhere?  why is compaq r > mngt (capellas)t9 > keeping this a secret?  this is the vms promotion of a   > lifetime at the perfecto@ > time gartner has just hammered IIS, and this article, besides  > the places myself @ > and a few others have posted or emailed, has been silenced to  > the openvms timeseH > site ... am i wrong about this? whats wrong your marketing department?7 > are they being told to keep silent or do they really s > understand anything more > than windows and pcs?  >    ------------------------------    Date: 23 Oct 2001 05:57:55 -0700( From: bob@instantwhip.com (Bob Ceculski)/ Subject: Re: Major VMS reseller says VMS legacyd= Message-ID: <d7791aa1.0110230457.6c24005c@posting.google.com>h  a JF Mezei <jfmezei.spamnot@videotron.ca> wrote in message news:<3BB55A82.783947C9@videotron.ca>...h > "David J. Dachtera" wrote:D > > So, it is up to you, it is up to me, it is up to all of us to beH > > positive about VMS wherever we can. If THEY won't "sell" it, then we" > > should sell it for them, IMHO. > J > That is how VMS has survived since Palmer. Pushed and saved by customersP > decrying Digital/Compaq's handling of VMS, forcing the company to postpone the) > death announcement over and over again.i > O > But with the way Compaq announced the death of Alpha, I have given up. What'seO > the point ? Let them decide on their product suite. If the want to ignore VMSsJ > or kill it, why waste energy trying to oppose it ? At best you will only > postpone their decision. > I > What is the point of busting your own butt, and not only get nothing intO > return, but get a bad reputation because you go against the vendor's wishes ?  > L > If HP wants VMS revenus, it should work to get them. Why should it rely onA > volunteers doing all the work so that HP can get some profits ?s  C vms will survive thanks to the military and government (not all ourSB govt employees are stupid thank God!) ... how long it survives mayB depend on the itanic port ... it will be ported, and if intel usesD alpha logic and compilers to create a stable cpu, i believe vms willA thrive as new apps will be ported ... if intel screws up, it willnC take the whole customer base to demand compaq (hp?) start producing8A alpha again (ev8) ... every vms user needs to be vocal about this.B and remain true ... right now is the worst time to port to anotherA platform (do you want your IIS, mr. nimda?) and with ev7 vms will C stay feasible for at least 9 more years ... everyone needs to emailu compaq (i do)!   ------------------------------    Date: 23 Oct 2001 14:20:30 -0000= From: Doc.Cypher <Use-Author-Supplied-Address-Header@[127.1]>a/ Subject: Re: Major VMS reseller says VMS legacye5 Message-ID: <20011023142030.8886.qmail@nym.alias.net>c  " -----BEGIN PGP SIGNED MESSAGE-----  9 On 23 Oct 2001, bob@instantwhip.com (Bob Ceculski) wrote:p   <snip>  D >vms will survive thanks to the military and government (not all ourC >govt employees are stupid thank God!) ... how long it survives maytC >depend on the itanic port ... it will be ported, and if intel usesOE >alpha logic and compilers to create a stable cpu, i believe vms will B >thrive as new apps will be ported ... if intel screws up, it willD >take the whole customer base to demand compaq (hp?) start producingB >alpha again (ev8) ... every vms user needs to be vocal about thisC >and remain true ... right now is the worst time to port to anothertB >platform (do you want your IIS, mr. nimda?) and with ev7 vms willD >stay feasible for at least 9 more years ... everyone needs to email >compaq (i do)!f  = bob ... might ... i ... suggest ... reading ... rfc ... 1855.r  % http://www.faqs.org/rfcs/rfc1855.htmle     Doc. - -- e6 The bigger the humbug, the better people will like it.K ~ Phineas Taylor Barnum.                              http://vmsbox.cjb.netl   -----BEGIN PGP SIGNATURE-----l Version: 2.6.2  @ iQEVAwUBO9Sk8sriC3SGiziTAQG1Vgf/bMvs+hIAAZv5cTgVSbhX2Ebhci49LdXO@ UbE1nwk8IErtKM4wXGXUnYmBb6hy3VuQrOQ779ePJiMgiKxxY49DchdOZaoN3UcL@ Gw2gIcSSUMLi5r1UiK3wM8hQGzehy+uqbAu151iL8d0V9ETZ32wawBtrklJX9ZmY@ Bj0L5DiMIdLL1PlKqHh8A37l5O1BH2GVkZCGX9DnwUwdzUeEkt1121EHfjj5JW45@ flm8diMgy3eUc8yDjJn9CTICCeWQOXFwx8X4/NmgiuZcTbN2EA7CwBgkbO5DwJDo8 T5lzOpED1ydRwgViBY+VqGiKfj7c/gUGohJZZubcoKkLiU7RK3rC7w== =hZF/a -----END PGP SIGNATURE-----i   ------------------------------    Date: 23 Oct 2001 04:59:46 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)r Subject: Re: Mitnick Uncaged3 Message-ID: <yB3O1dr5PhyM@eisner.encompasserve.org>   d In article <3BD4EA3F.EDF650C@cableinet.co.uk>, Tim Llewellyn <tim.llewellyn@cableinet.co.uk> writes:  G > yes, but back on topic, Peter, if you were running VMS you could haveh > given hertH > OPER and a decent helpdesk application ages ago and if she isn't a spy > really (:-))J > then that would probably keep your wife happy and out of your UAF etc at > the same time. > H > All in the best of humour, (I hope) obviously. Execept that the decent > helpdesk r( > application seems to be sadly lacking.  ? I think the PTR program DEQ uses for Field Test is pretty good,s$ with clients for DCL, HTTP and Mail.  6 The problem is that it is not available for customers.# Digital has it now (and you don't).R   ------------------------------  % Date: Tue, 23 Oct 2001 11:51:42 +0100@/ From: Nigel Arnot <sysmgr@maxwell.ph.kcl.ac.uk>0 Subject: re: Mitnick Uncaged6 Message-ID: <00A03F41.8BF7E6AB.8@maxwell.ph.kcl.ac.uk>   > 2 > Chuck McCrobie <mccrobie@cablespeed.com> writes: > J > > But only a fool would put on computer.  My saying "If its on computer,K > > its not secure."  Nothing, short of keeping it to yourself, is secure -SC > > one can merely raise the level of effort required to obtain it.o > O > Using a filesystem that encrypts everything with a sufficiently large RSA/...lM > key, and the key or passphrase to it in your head, will raise that level to0O > requiring torture of the owner. At that point, any other method is as or less 	 > secure.L >  > 	Jan   Not.  H You can park a van full of radio gear outside and read everything that'sJ displayed on a commercial monitor screen courtesy of the stray radio wavesG that CRTs emit. I believe that it's also possible, though trickier, forbA LCD panels. Keyboards (even the wired sort) are also vulnerable.    I If you can break in, you can plant a "bug" inside the PC, and use that to9H record keystrokes. Download by radio or, if you are cheeky, via the sameK internet software that the PC's owner is online with! And of course, unless-O the PC's owner is using something like Tripwire, you can also bug the software.E  L So how sure are you that the RSA key hasn't found its way into the swapfile?I Sure, folks that know about security software know how to avoid that. But I if you're using a "canned" product, it may be written by incompetents, or K it may be deliberately equipped with a backdoor courtesy of some governmentsF agency. And of course, what happens to the data after your filesystem - decrypts it and passes it to an application?    J I can think of plenty more possibilities. And I'm not in the spy business,@ I'm quite sure that "they" have things that I'd never guess at.   K One spy fiction I read recently speculated that torture is obsolete. ModerneG way is portrayed as sedate and blindfold the person to be interrogated YK and wire him up to a lot of sensors. Brainwave and physiological responses eG to questions heard betray his true answers even if the suspect refuses   to open his mouth. h   	Yours,a
 		Nigel Arnotr- 		NRA@MAXWELL.PH.KCL.AC.UK                   t  7 		"In the beginning there was nothing, which exploded."-   ------------------------------    Date: 23 Oct 2001 12:56:51 +0200G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>o Subject: Re: Mitnick UncagedH Message-ID: <y4wv1mpqto.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  1 Nigel Arnot <sysmgr@maxwell.ph.kcl.ac.uk> writes:u  N > So how sure are you that the RSA key hasn't found its way into the swapfile?  C Doesn't matter, the contents of the swapfile are encrypted as well.l   	Jan   ------------------------------    Date: 23 Oct 2001 07:08:25 -0700& From: snyderslake@yahoo.com (nuallain) Subject: mode nibble??= Message-ID: <352b9262.0110230608.73ad9568@posting.google.com>a  B Could someone help to descibe what a mode nibble's function is?  ID understand that the mode nibble is the first 4 bits on the op code. D What does this do?  Does it describe the number of operands that theA op code will be handling so as to allocate enough memory for this  number of operands?f   Thanks.h   ------------------------------  % Date: Tue, 23 Oct 2001 10:03:30 +0100T% From: Alan Greig <a.greig@virgin.net>a2 Subject: Re: New to OpenVMS - Looking for software8 Message-ID: <qgcattom9sia3p3ecomobi91phr2fu0mki@4ax.com>  B On Mon, 22 Oct 2001 18:24:16 -0700, "Brad Hughes" <brad@tgsmc.com> wrote:   >,A >And then there's the bandwidth issue:  since capital letters aredC >larger they use up more bandwidth.  It's a waste of electrons.  :)g  C What is really sad is that 99% of CEOs and 90% of CIOs/IT Directors 4 would probably accept it as true if Gartner said it.   >Brada   -- Alan   ------------------------------  % Date: Tue, 23 Oct 2001 11:43:42 +0200y= From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <noone@dummy.com>e2 Subject: Re: New to OpenVMS - Looking for software) Message-ID: <3BD53BCE.B497A87E@dummy.com>d  $ In a PC store, looking at a laptop :  . "That 2.2 kg, is that with a full hard disk ?"   Jan-Erik Sderholm.0   ------------------------------  % Date: Tue, 23 Oct 2001 19:47:30 +0010A% From: paddy.o'brien@zzz.tg.nsw.gov.auY2 Subject: Re: New to OpenVMS - Looking for software5 Message-ID: <01K9UOZKBIGI006R6I@tgmail.tg.nsw.gov.au><   Alan,d  C >On Mon, 22 Oct 2001 18:24:16 -0700, "Brad Hughes" <brad@tgsmc.com>  >wrote:  >a >>B >>And then there's the bandwidth issue:  since capital letters areD >>larger they use up more bandwidth.  It's a waste of electrons.  :) >>D >What is really sad is that 99% of CEOs and 90% of CIOs/IT Directors5 >would probably accept it as true if Gartner said it.l  J It is also sad that 102% of IT staff and management are appointed with no  (zero, zilch) knowledge of IT.  O Disclaimer: I do not work for our IT.  I use as a developer and manage various aL production/development machines -- all ****VMS****.  Possible disclaimer, I  might not have a job next week.h   Regards, Paddy   ------------------------------   Date: 23 Oct 2001 13:08:00 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)i2 Subject: Re: New to OpenVMS - Looking for software+ Message-ID: <9r3q3g$pt5$1@info.cs.uofs.edu>o  8 In article <qgcattom9sia3p3ecomobi91phr2fu0mki@4ax.com>,(  Alan Greig <a.greig@virgin.net> writes:E |> On Mon, 22 Oct 2001 18:24:16 -0700, "Brad Hughes" <brad@tgsmc.com> 	 |> wrote:t |>   |> >D |> >And then there's the bandwidth issue:  since capital letters areF |> >larger they use up more bandwidth.  It's a waste of electrons.  :) |> aF |> What is really sad is that 99% of CEOs and 90% of CIOs/IT Directors7 |> would probably accept it as true if Gartner said it.m  G It really is, considering the uppercase letters have less one bits than F their lowercase equivalents and thus use less electricty and therefore less bandwidth.c   bill   -- hJ 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>   i   ------------------------------  % Date: Tue, 23 Oct 2001 15:02:16 +0100s% From: Alan Greig <a.greig@virgin.net>.2 Subject: Re: New to OpenVMS - Looking for software8 Message-ID: <2ttatt0ajrmmjl1qpqo5nt0m5lcru6s379@4ax.com>  < On 23 Oct 2001 13:08:00 GMT, bill@triangle.cs.uofs.edu (Bill Gunshannon) wrote:  9 >In article <qgcattom9sia3p3ecomobi91phr2fu0mki@4ax.com>,r) > Alan Greig <a.greig@virgin.net> writes:aF >|> On Mon, 22 Oct 2001 18:24:16 -0700, "Brad Hughes" <brad@tgsmc.com>
 >|> wrote: >|>  >|> >tE >|> >And then there's the bandwidth issue:  since capital letters areIG >|> >larger they use up more bandwidth.  It's a waste of electrons.  :)e >|> G >|> What is really sad is that 99% of CEOs and 90% of CIOs/IT DirectorsO8 >|> would probably accept it as true if Gartner said it. >kH >It really is, considering the uppercase letters have less one bits thanG >their lowercase equivalents and thus use less electricty and therefore> >less bandwidth.  F Which is the opposite of the original 'claim' of course! And even then< we'd need to switch back to sixbit ascii to get the benefit.   >bill    -- Alan   ------------------------------  % Date: Tue, 23 Oct 2001 08:50:07 -0400 4 From: John Malmberg <Malmberg@dskwld.zko.dec.compaq>G Subject: SAMBA VMS and Windows XP - Re: Windows XP reality check pleaseo4 Message-ID: <3BD5677F.6070403@dskwld.zko.dec.compaq>   Carl Perkins wrote:a  H > I havn't done an activation. There are currently two XP systems on ourJ > net so one of the occasions was to mess with it to figure out how to getG > them to talk to the Samba server I'm running on VMS - which isn't toorJ > hard to do (but the connection is slow (the W2K buffer size negotiation E > patch might improve that, but I havn't gotten around to it yet) andrJ > occasionally disconnects requiring reentering the password, and possiblyF > username, for no known reason, usually when left running overnight).    C The W2K buffer size patch may not help.  It broke again with "/ME".?  F As far as I know, the fix for "/ME" is only in the SAMBA 2.2.x stream,% but someone could have backported it.-  C According to preliminary tests, it should not take much work to getmH SAMBA 2.0.8 to link and run on OpenVMS 7.3.  Earlier versions of OpenVMS will require a SMOP.   Samba 2.0.7 has the W2K patch.  H Samba 2.0.8 has a few small fixes + a manditory security update for UNIX systems.  B Samba 2.0.9 has a small change that is not compiled in on OpenVMS.  * Samba 2.0.10 ?  I have not looked at this.  F Samba 2.2.x needs a SMOP and a lot of testing and analysis.  I can getC a clean link, but with a lot of stubs for routines that need to be oB written or patched.  There were quite a few architectural changes.  I > (The changes needed to get Samba to work were all done via the securityh > policy thingy.)b    G What changes?  Enabling plain-text passwords is not required for SAMBA g 2.0.6 on OpenVMS.p   Is there something else?   -John. Malmberg@dskwld.zko.dec.compaq Personal Opinion Onlys   ------------------------------  % Date: Tue, 23 Oct 2001 19:42:48 +0200r& From: Michael Joosten <joost@c-lab.de>K Subject: Re: SAMBA VMS and Windows XP - Re: Windows XP reality check pleaseL$ Message-ID: <3BD5AC18.167E@c-lab.de>   John Malmberg wrote: >  > Carl Perkins wrote:m > J > > I havn't done an activation. There are currently two XP systems on ourL > > net so one of the occasions was to mess with it to figure out how to getI > > them to talk to the Samba server I'm running on VMS - which isn't too K > > hard to do (but the connection is slow (the W2K buffer size negotiationxG > > patch might improve that, but I havn't gotten around to it yet) andrL > > occasionally disconnects requiring reentering the password, and possiblyH > > username, for no known reason, usually when left running overnight). > F This 'overnight' problem sounds like what I'm suffering now. Only thatF it is 'evening' for me... About between 24:00 and 01:00, often the SMBF conncetions get stuck, Word (as usual) becomes rabid, Explorer freezesF and so on. Though I do not have to reenter the password, as the SAMBAs0 are integrated in the domain controller network.  C As you are listing the patches, which of them would apply to my/ouro, problem? Is it just that W2k patch of 2.0.7?    E > The W2K buffer size patch may not help.  It broke again with "/ME".- > H > As far as I know, the fix for "/ME" is only in the SAMBA 2.2.x stream,' > but someone could have backported it.e > E > According to preliminary tests, it should not take much work to getnJ > SAMBA 2.0.8 to link and run on OpenVMS 7.3.  Earlier versions of OpenVMS > will require a SMOP. >   > Samba 2.0.7 has the W2K patch. > J > Samba 2.0.8 has a few small fixes + a manditory security update for UNIX
 > systems. > D > Samba 2.0.9 has a small change that is not compiled in on OpenVMS. > , > Samba 2.0.10 ?  I have not looked at this. > H > Samba 2.2.x needs a SMOP and a lot of testing and analysis.  I can getD > a clean link, but with a lot of stubs for routines that need to beD > written or patched.  There were quite a few architectural changes. >    --  * Michael Joosten, SBS C-LAB, joost@c-lab.de* Fuerstenallee 11, 33094 Paderborn, Germany, Phone: +49 5251 606127, Fax: +49 5251 6060658 C-LAB is a cooperation of University Paderborn & SIEMENS   ------------------------------  # Date: Tue, 23 Oct 2001 08:48:11 GMTD, From: "Paul Dennis" <comedyox@earthlink.not>! Subject: Re: SET FILE/ATTRIB in C@C Message-ID: <f9aB7.4363$Sd.323515@newsread1.prod.itd.earthlink.net>.  D > :Does anyone have an example of how to do the equivalent of an SETH > :FILE/ATTRIB=(RFM:FIX etc...) in a C program (not by using system() or > :similar) ???"   sys$modify?    .pd.   ------------------------------   Date: 23 Oct 2001 17:54:59 GMT# From: casinoop2@aol.com (CasinoOp2) 8 Subject: Shared file access under NT and VMS question...9 Message-ID: <20011023135459.17436.00000819@mb-ch.aol.com>a  O I have a database on a VMS Alpha system. The main file is open shared using the. following Fortran statement:  J OPEN(UNIT=1,FILE='dkb100:[ana.bank]data.bnk',STATUS='OLD',ACCESS='DIRECT', RECL=25,@               RECORDTYPE='FIXED',ORGANIZATION='RELATIVE',SHARED)    L This works just fine under VMS. Many people can read/write simultaneously toK various records. But under NT, I have a Visual Basic program that open thisr# file using the following statement:e  O OPEN "I:\ana.bank\data.bnk" For Randon Access Read Write Shared As #1 Len = 100e  N Where "I" is mapped to the drive DKB100:. This works fine as long as no one onL the VMS side is accessing the same file, if they are then the file cannot beN opened by Visual Basic. Conversly, if VB has this open, then VMS gives a "file locked" error.  N It would really be nice if an NT program could access this file as well as VMS programs. Any idea out there?r     Clark Calkins, programmere   Schafer Corp   ccalkins@schaferlabs.com   ------------------------------  % Date: Tue, 23 Oct 2001 08:40:34 -0700i. From: Jack Trachtman <Jack.Trachtman@vmmc.org>& Subject: Single or Multiple Sys Disks?( Message-ID: <3BD58F71.1E80C13C@vmmc.org>  G After all these years managing stand alone VMS systems, I'll finally bei; getting to manage two clustered systems: 2xES45 and 3xES45.   B My question: How to choose between single system disk and multipleD disks.  The first is easier to manage, but the second allows rolling( upgrades w/less down time for the users.  + Could I get some thoughts on this?  Thanks.-   ------------------------------  + Date: Tue, 23 Oct 2001 16:35:56 +0000 (UTC)v From: david20@alpha1.mdx.ac.uk* Subject: Re: Single or Multiple Sys Disks?+ Message-ID: <9r469c$82e$1@aquila.mdx.ac.uk>t  Y In article <3BD58F71.1E80C13C@vmmc.org>, Jack Trachtman <Jack.Trachtman@vmmc.org> writes:lH >After all these years managing stand alone VMS systems, I'll finally be< >getting to manage two clustered systems: 2xES45 and 3xES45. >hC >My question: How to choose between single system disk and multipleaE >disks.  The first is easier to manage, but the second allows rolling ) >upgrades w/less down time for the users.t > , >Could I get some thoughts on this?  Thanks. >m  / Use one shadowed/hardware mirrored system disk.eK When it comes time for the upgrade unmirror the system disk so you have theh  original system disk and a copy.> Take one of the systems down and use that to upgrade the copy.G Boot that system from the upgraded copy and complete post upgrade tasksh% including applying necessary patches. < Shutdown other systems and boot them from the upgraded copy.  J Dismount original system disk and make it a mirror of the new system disk.     Cluster downtime zero.; System doing upgrade downtime - however long upgrade takes.i> Other cluster system's downtime - time to shutdown and reboot.  PJ Fallback if something goes wrong. Until you recreate the shadowset/mirror N after the upgrade you have both your original system disk and the new upgradedM system disk. Hence if you delay the final step you can run the systems on the-O new version for a while safe in the knowledge that you can switch back by just y( shutting down and rebooting the systems.  J Also the usual state will be that the system disk is mirrored and hence a L system disk failure at any time other than in the middle of the upgrade will' have no affect on cluster availability. L (Of course if your company has the money and you want to cover yourself evenK more you would have 3 disks in the system disks mirrorset. To guard againstoK the possibilty of the original system disk failing whilst you are doing the.	 upgrade).     
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------  % Date: Tue, 23 Oct 2001 12:45:24 -0400g2 From: rdeininger@mindspring.com (Robert Deininger)* Subject: Re: Single or Multiple Sys Disks?L Message-ID: <rdeininger-2310011245250001@user-2ive7bq.dialup.mindspring.com>  7 In article <3BD58F71.1E80C13C@vmmc.org>, Jack Trachtman   <Jack.Trachtman@vmmc.org> wrote:  I > After all these years managing stand alone VMS systems, I'll finally beD= > getting to manage two clustered systems: 2xES45 and 3xES45.r > D > My question: How to choose between single system disk and multipleF > disks.  The first is easier to manage, but the second allows rolling* > upgrades w/less down time for the users. > - > Could I get some thoughts on this?  Thanks.t  I You'll be talking about 5 system disks if you allow 1 per system, or 2 ifiJ you allow 1 per cluster.  Maintaining 3 extra system disks, with upgrades,E patches, layered products, and freeware seems like WAY too much work,  IMHO.   J If you're worried about downtime, you should be shadowing important disks, including system disks.d  H Will you be running the clusters as boot node + satellites, or will each' system have direct access to the disks?s  F You can handle rolling upgrades with common system disks.  Split off aC shadow member (or copy the system disk to a spare drive).  Take one J cluster member away from your users and perform the upgrade on the copy ofJ the system disk.  During the upgrade, all but one node continue to use theE original system disk; the node  performing the upgrade boots from thelG copy.  At the end, the newly-modified disk becomes the master copy, andh> you reboot the rest of the nodes so they use the new software.  E This kind of upgrade is easier if you have put all the cluster-commonrG "environemnt" files like SYSUAF (see SYLOGICALS.COM for a more completeyF list) on a separate disk, not the system disk.  Then you don't have toI worry about two copies of these getting de-syncronized while you upgrade.   I Depending on your environment, applications, etc., you can likely arrangeo+ a rolling upgrade with minimal user impact.i  G Bottom line: I would do it with one (shadowed) system disk per cluster.s   -- ' Robert Deininger rdeininger@mindspring.comr   ------------------------------  % Date: Tue, 23 Oct 2001 12:43:05 -0400e  From: norm.raphael@jamesbury.com* Subject: Re: Single or Multiple Sys Disks?4 Message-ID: <C2256AEE.005B8DD3.00@jklh21.valmet.com>  & How are the disks attached/configured?% Can you/will you want to shadow them?i. How often are you going to upgrade vs. manage?' How much down-time can you/users stand?         1 Jack.Trachtman@vmmc.org on 10/23/2001 11:40:34 AM   ) Please respond to Jack.Trachtman@vmmc.orgt   To:   Info-VAX@mvb.saic.com  cc:u' Subject:  Single or Multiple Sys Disks?u        G After all these years managing stand alone VMS systems, I'll finally bet; getting to manage two clustered systems: 2xES45 and 3xES45.a  B My question: How to choose between single system disk and multipleD disks.  The first is easier to manage, but the second allows rolling( upgrades w/less down time for the users.  + Could I get some thoughts on this?  Thanks.a   ------------------------------  % Date: Tue, 23 Oct 2001 14:51:41 +0300j From: mergoktas@erdemir.com.tr Subject: Socket problem-URGENT1 Message-ID: <C2256AEE.004128C1.00@erdemir.com.tr>   : I have tcp/ip server task which can accept 16 simultaneous; connections. Task is written in FORTRAN by using QIO systemdJ services. All I/O(s) are done by AST routines, accepting a new connection,. reading from a socket and writing to a socket.  0 DEC TCP/IP Services for OpenVMS VAX Version V3.3( on a VAX 4000-400 running OpenVMS V5.5-2  J My clients request some data from the server at every 5 seconds by sendingL 10 bytes long message. Server responds about 400 bytes data for each reques= t.  L The system works 2 or 3 hours without any problem but after that I have the=   following report from UCX.   UCX> sh dev bg585/fullL =A0Device_socket: bg585 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0LOCA=$ L =A0 =A0 =A0 =A0 =A0 =A0 =A0 REMOTEN =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Port: =A0 =A0 =A0=2  =A0 =A0 5500 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 1051L =A0 =A0 =A0 =A0 =A0 Type: STREAM =A0 =A0 =A0 =A0Host: =A0 =A0 CAL21 =A0 =A0="  =A0 =A0 =A0 =A0 =A0 165.156.2.162D =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 Service:  L =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0=D  =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0RECEIVE =A0 =A0 =A0 SENDL =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Queu=I ed I/O =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A00 =A0 =A0 =A0 =A0 =A0 =A0 0oL =A0 =A0 =A0 =A0Q0LEN =A0 =A0 =A0 =A0 0 =A0 =A0 =A0 =A0 =A0 =A0 Socket buffe=3 r bytes =A0 =A0 =A0 =A0 =A0 0 =A0 =A0 =A0 =A0 60019HL =A0 =A0 =A0 =A0QLEN =A0 =A0 =A0 =A0 =A00 =A0 =A0 =A0 =A0 =A0 =A0 Socket buf=1 fer quota =A0 =A0 =A0 16000 =A0 =A0 =A0 =A0 60000 L =A0 =A0 =A0 =A0QLIMIT =A0 =A0 =A0 =A00 =A0 =A0 =A0 =A0 =A0 =A0 Total buffer=5  alloc =A0 =A0 =A0 =A0 =A0 =A00 =A0 =A0 =A0 =A0 68712pL =A0 =A0 =A0 =A0TIMEO =A0 =A0 =A0 =A0 0 =A0 =A0 =A0 =A0 =A0 =A0 Total buffer=1  limit =A0 =A0 =A0 =A064000 =A0 =A0 =A0 =A0240000rL =A0 =A0 =A0 =A0ERROR =A0 =A0 =A0 =A0 0 =A0 =A0 =A0 =A0 =A0 =A0 Buffer or I/=5 O waits =A0 =A0 =A0 =A05610 =A0 =A0 =A0 =A0 =A0 =A0 0tL =A0 =A0 =A0 =A0OOBMARK =A0 =A0 =A0 0 =A0 =A0 =A0 =A0 =A0 =A0 Buffer or I/O =5 drops =A0 =A0 =A0 =A0 =A0 0 =A0 =A0 =A0 =A0 =A0 =A0 0-L =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0I/O =A completed =A0 =A0 =A0 =A0 =A0 =A0 =A05609 =A0 =A0 =A0 =A0 =A05611 L =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0Byte=7 s transferred =A0 =A0 =A0 =A0 56090 =A0 =A0 =A0 1572393e   =A0 Options: =A0REUSEADR" =A0 State: =A0 =A0ISCONNECTED PRIV =A0 RCV Buff: WAIT =A0 SND Buff: None  L When I have above picture from UCX, I still able to receive data from clien= tr side,s- but ofcourse can not send data to the client..  L Since the "Socket buffer bytes" exceeds the "Socket buffer quota", socket g= ets> inL SUSPENDED mode. If I close the socket and initiate a new connection from cl= ient sideE it starts working normally. But some time later same problem happens.h  J I analized the network traffic at packet level when the send buffer of the9 socket started to stack data. There was no packet at all.e  ? I think I have enough room for non-paged dynamic memory on VMS.f  L What can be the reason with socket settings or some VMS parameters or anyth= ings else ?L And what is the latest version of UCX that I can use on VAX 4000-400 runnin= g  OpenVMS V5.5-2?n Please help. URGENT. Thanks a lot in advance.   ------------------------------  $ Date: Tuesday, 23 OCT 2001 10:25 EDT. From: Bryan Jensen <bjj100@arlvax.arl.psu.edu>+ Subject: Re: SuperDLT on older (EV5) Alphas + Message-ID: <9r43m6$mba@r02n01.cac.psu.edu>9  = In article <8af17fe1.0110221143.4ac14675@posting.google.com>,aA    alphaman-nixspam@hsv.sungardtrust.com (Aaron Sakovich) writes:.> >JONESD@er6.eng.ohio-state.edu (David Jones) wrote in message < >     news:<9pifun$s9a$1@charm.magnus.acs.ohio-state.edu>...K >> We want to replace the trusty TZ88 DLT drive on our AS 800 with a CompaqsG >> SuperDLT drive (with an Ultra 2 SCSI interface card).  Compaq hasn'taM >> qualfied the drive with the AlphaServer 800, but have no reason to believea' >> it won't work.  The OS is VMS 7.2-1.r >> rN >> Has anybody tried with combination that experienced unforseen difficulties?L >> Would the 11 MB/sec transfer rate on the SDLT present a problem if backupN >> can't feed it that fast (I don't see anything in the specs about dual speed >> support). >-D >I'm interested too, but want to put it on an AlphaStation 500.  TheE >specs say that it will work in either LVD or SE mode; anyone know ifsD >the standard SE interface on the AS500 will fit the bill?  I'm also& >curious about data starvation issues.  > I've only had a SuperDLT drive for a short time, but here's my experience.   F I am using a Quantum LVD SuperDLT drive on a 400MHz AlphaServer 1000A G running VMS 7.2-1.  It has an KZPCA SCSI LVD/SE interface dedicated to s> the tape drive.  It's not fast enough to keep the tape moving.  F I have also tried it on a VAXstation 4000/90A (single ended of course) and found it to work ok.  F It seems to handle data starvation just fine (unlike 8mm and DAT tapesF which seem to eat the oxide off the tape when you don't feed them data quick enough).   Two problems I had (or have):   : 1)  Half of our order of 10 Compaq SDLT tapes arrived bad.F     As would be expected with any DLT drive, one resulted in a dropped!     leader necessitating service.t  @ 2)  It seems it may take days to restore a single file.  The VMSD     algorithm for skipping savesets seems to require multiple passesD     over the entirety of each saveset.  Of course, with only a writeD     lock and activity light, I don't really know what the tape drive
     is doing.n   ------------------------------  % Date: Tue, 23 Oct 2001 12:21:32 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)+ Subject: Re: SuperDLT on older (EV5) AlphaslL Message-ID: <rdeininger-2310011221330001@user-2ive7bq.dialup.mindspring.com>  8 In article <9r43m6$mba@r02n01.cac.psu.edu>, Bryan Jensen" <bjj100@arlvax.arl.psu.edu> wrote:     > Two problems I had (or have):  > < > 1)  Half of our order of 10 Compaq SDLT tapes arrived bad.H >     As would be expected with any DLT drive, one resulted in a dropped# >     leader necessitating service.i  I I haven't used the new SDLTs.  It sounds like they use the same mechanism4) as the older generations I discuss below.b  H It is a good idea to open the little door on the tape and visually checkB the leader.  Make sure it is in place on its little hook.  This isJ especially important for new tapes, since the rigors of shipping sometimesG bounce the hub out of its locked position.  If the hub turns a bit, themJ tape is probably loose enough to let the leader come off the hook.  If youH use such a tape without fixing it, you'll end up with an unhooked leader
 in the drive.   H The service folks recommend wearing gloves when re-threading an unhookedH leader in a DLT drive, to keep finger oils off the tape path.  You don'tE need more than a screwdriver and a bit of dexterity to do this.  It'sw% annoying, but shouldn't be expensive.'  D On OEM DLT drives, we have found that an unhooked leader while underF warranty results in the vendor's service tech explaining the procedureD over the phone, and telling us to do it ourselves.  (As if we didn'tH already know how.)  After the warranty expires, the same problem resultsF in the recommendation to send the drive in for an expensive, flat-rate service call.  Fat chance.  J If you're not comfortable fiddling with the innards of an expensive drive,G acquire an old TK50 drive and use it to practice re-threading leaders. aG The procedure is the same, but the newer drives have more rollers and aeE longer leader.  I wouldn't recommend that untrained folks fiddle witht* anything else besides the leader, however.  B > 2)  It seems it may take days to restore a single file.  The VMSF >     algorithm for skipping savesets seems to require multiple passesF >     over the entirety of each saveset.  Of course, with only a writeF >     lock and activity light, I don't really know what the tape drive >     is doing.a  F On a recent version of VMS, fast skip should work with SDLTs.  I don'tH know for sure that backup takes advantage of it.  If you are in a hurry,J you can use SET MAGTAPE/SKIP=FILE:nn to get to the right saveset quickly. F Under the right conditions, that DOES use fast skip, in my experience.  : Exactly what BACKUP command are you using that is so slow?   -- : Robert Deininger rdeininger@mindspring.com    ------------------------------  + Date: Tue, 23 Oct 2001 04:58:13 -0700 (PDT)-. From: Fabio Cardoso <fabiopenvms@yahoo.com.br>D Subject: Re: TELNET sessions not closed, limit exceeded, how to fix?@ Message-ID: <20011023115813.74233.qmail@web20203.mail.yahoo.com>   Lawrence  3 The CSC/Brazil gave me just the two files by email.a" It wasnt a polycenter based patch.     Regardsn   FC=20r  / --- Lawrence Bleau <bleau@umtof.umd.edu> wrote:y7 > I posted comp.os.vms about this problem last week.=20a > As a quick refresh, theT5 > problem is that the count of active incoming telnet  > sessions sometimes=20V5 > does not get decremented when a user logs out.  The8 > user's process and its TNa/ > device goes away, but the BG device stays.=20   > Eventually the telnet limit is/ > reached and no one can log in via that means.n >=20 > One respondant (Fabio CardosoI, > <fabiopenvms@yahoo.com.br>) suggested that+ > I replace the files TCPIP$INETACP.EXE ands > TCPIP$BGDRIVER.EXE.  He said:p >=20, > >I had this same problem a few months ago.4 > >The CSC gave me a TCPIP patch. You should contact
 > >Compaq. > >u, > >My operating system is OpenVMS 7.2 and my! > >TCPIP Services is 5.0A - ECO 3e > >i& > >I applied the TCPIP$INETACP.EXE and% > >TCPIP$BGDRIVER.EXE and SOLVED !!!!r >=206 > This person could not supply me with the name of the > patch, however.  I5 > checked Compaq's web site and noticed a patch namedl > TCPIPALP_E03A51 (fileh5 > DEC-AXPVMS-TCPIP_ECO-V0501-153-4.PCSI) that has the  > following descriptionc1 > in it under the heading "Corrections for COMPAQ  > TCP/IP Services V5.1 NET, > Images", ECO 2 updates, ECO L (web address >s@ http://ftp1.service.digital.com/public/vms/axp/v7.2-2/tcpip/5.1/ > tcpipalp_e03a51.README): >=20 > ECO Lm >=20
 > Problem:7 > =09When a Server process created for a Listen serviceg > exitsD5 > =09we fail to detect the process termination.  Thisi
 > can lead6 > =09to undeleteable BG devices hanging around as well > as hanging > =09client processes. >=20 > Solution:w6 > =09In the routine PROCESS_CREATE_LISTEN, we were not > using the=208 > =09proper termination mailbox in most cases.  A simple > fix to' > =09this routine resolves the problem.e >=20 > Reference:2 > =09PTR 70-5-1632 / CFS.82526 / Req Id: MGO67116A >=20 > ---0 >=205 > Does anyone know if this is the patch that does thej > trick? >=205 > The problem is, this patch is for OpenVMS 7.2-2 anda > TCPIP V5.1, while I'me- > running OpenVMS 7.1-2 and TCPIP 5.0A ECO 3.J >=204 > Does anyone know if the images for this particular > fix have been backported6 > to the level of VMS I'm running?  If so, where can I > find them?  If they.2 > are available only on a trial basis, I'll accept > them like that.  I'd/ > rather get such images from a trusted source,e > though, rather than someone's/ > anonymous ftp site.u >=206 > Otoh, the above ECO L may not be the correct fix.=20 > After all, it only8 > replaces TCPIP$INETACP.EXE, not TCPIP$BGDRIVER.EXE.=20 > Maybe someone else2 > knows which patch is correct, and can inform me? >=20 > Lawrence Bleau > University of Maryland$ > Physics Dept., Space Physics Group > 301-405-6223 > bleau@umtof.umd.edup     =3D=3D=3D=3D=3DnL =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D  F=E1bio dos Santos Cardoso OpenVMS System Manager Rio de Janeiro - Brazile fabiopenvms@yahoo.com.brL =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3De  2 __________________________________________________ Do You Yahoo!?, Make a great connection at Yahoo! Personals. http://personals.yahoo.com   ------------------------------    Date: 23 Oct 2001 04:52:18 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) < Subject: The Navy competition may not be Microsoft, but Palm3 Message-ID: <JXXHGr2+WXF2@eisner.encompasserve.org>t  * http://www.msnbc.com/news/646394.asp?cp1=1   ------------------------------  % Date: Tue, 23 Oct 2001 12:29:58 +0100 * From: Andrew Robinson <arobinson@hspg.com>, Subject: RE: Windows XP reality check pleaseM Message-ID: <CDA4BAD1E10ED41181AC00508B6051D3C3E420@grumpy.internal.hspg.com>u  K The corporate version still requires registering via M$oft + the use of theiL 'licensing fulfilment CD's' This does then allow multiple installations fromE the same CD. You need a separate license for each of the XP family ofa suites. K One point that is missing in this tread, is that apart from Apple, who took J years to work out what marketing is all about, (and I hope it wasn't/isn'tL too late) what is the alternative for the totally machine illiterate? I haveL 3 boys all of which could use my home PC before they could read. The pictureL of Noddy etc., was all they needed and they learnt what to do with the mouseK to make Noddy work. There was also a button they could use to speak and seenI their 89 year old Grandma through Video Conferencing.   If I could find ahG PC/machine with the same sort of family software and ease of use then Ie would change to it tomorrow.I I have a VMS box at home but in a child's eyes its unfriendly, doesn't do I anything good and requires a reasonable amount of Keyboard skills just toi
 get it going.eH Back to the question of XP however, its going to be a pain for corporateJ admins, so delay of installs for as long as possible. From a home PC view,K the majority of users I've seen, they couldn't give a flying fig what theirhJ PC was running, just as long as they could type snotty letters to the bankH manager, blow up the latest type of aliens and have a cute cuddly animalH running all over the screen. My Corporate users are almost the same they$ just don't get the aliens & animals.   Regardss  B Andrew Robinson (not in favour of m$oft for critical applications)   -----Original Message-----: From: Tim Llewellyn [mailto:tim.llewellyn@cableinet.co.uk] Sent: 23 October 2001 04:43e To: Info-VAX@Mvb.Saic.Com , Subject: Re: Windows XP reality check please      % Carl, you have some very good points.c  = However, don't you worry about what extra baggage they've hadhB to add to the kernel to implememt this scheme and their competence+ and implementing and testing it adequately?   e   Carl Perkins wrote:G >   D a whole load of reasons why Bob Koehler was being a bit too paranoid  about what this thread is about.      -- o Tim.Llewellyn@cableinet.co.uk  z  C Standard disclaimer applies. My views in no way represent those of e! my employers or service provider.l   ------------------------------  % Date: Tue, 23 Oct 2001 07:59:24 -0400n4 From: John Malmberg <Malmberg@dskwld.zko.dec.compaq>, Subject: Re: Windows XP reality check please2 Message-ID: <3BD55B9C.90403@dskwld.zko.dec.compaq>   John Vottero wrote:   I > I upgraded from 2000 to XP and it feels more like a downgrade.  I can'tgJ > think of one thing that's actually improved.  The one feature that I wasM > looking forward to was the ability to be logged on to more than one accountoK > and quickly switch between them.  Now I've found out that this only worksTF > for home and workgroup environments, if you're in a domain it is not > available.    F The ability to do so in a domain is latent at least as far back as NT = 3.51.  Resource kit utilities do a usable job of exposing it.S  I I did a session on that at CETS 2000, Covering what it did well and what n it did not so well.t  @ It is pretty trivial to set up, and work around the interesting F limitations where a program did not follow the right API for accessing the desktop.  F The list of programs that I have found that did not comply might have D something to do with why the feature never made it into a mainstream product.    L > One key change that I can think of is the product activation.  You have toL > activate within 60 days or it stops working.  I haven't activated yet so IK > don't know how much information they want.  I'm going to delay activating L > until after I add memory, otherwise I may have to activate twice and incur* > the wrath of the Redmond License Police. > A > I'm glad I didn't pay for this "upgrade".  (MSDN subscription).p    H I have sort of lost interest in keeping up with MSDN for the past year, G but usually the MSDN subscriptions contained built in activation codes  * that did not require product registration.  D After all they are restricted to development use only.  No business  support use allowed.  H Of course there seems to be all sorts of different MSDN deals available.   -JohnA   Personal Opinion Only  malmberg@dskwld.zko.dec.compaq   ------------------------------    Date: 23 Oct 2001 08:44:31 -0500- From: koehler@encompasserve.org (Bob Koehler) , Subject: Re: Windows XP reality check please3 Message-ID: <B$oqGDvvaP$Z@eisner.encompasserve.org>0  W In article <tt8dbab4ksr02a@news.supernews.com>, "John Vottero" <John@mvpsi.com> writes:   M > There is a home version of OpenVMS.  It just costs about 5 times as much as0
 > Windows.  C    Free is too expensive?  When I think of home version, I think ofA    hobbyist.   ------------------------------  % Date: Tue, 23 Oct 2001 11:59:13 -04000% From: "John Vottero" <John@mvpsi.com>A, Subject: Re: Windows XP reality check please/ Message-ID: <ttb4uj2r230kc4@news.supernews.com>   : "Bob Koehler" <koehler@encompasserve.org> wrote in message- news:B$oqGDvvaP$Z@eisner.encompasserve.org...c@ > In article <tt8dbab4ksr02a@news.supernews.com>, "John Vottero" <John@mvpsi.com> writes: > L > > There is a home version of OpenVMS.  It just costs about 5 times as much as > > Windows. > E >    Free is too expensive?  When I think of home version, I think off >    hobbyist. >0   home != hobbyist.    ------------------------------  % Date: Tue, 23 Oct 2001 12:10:58 -0400=% From: "John Vottero" <John@mvpsi.com>A, Subject: Re: Windows XP reality check please/ Message-ID: <ttb5kn4986vrfb@news.supernews.com>B  A "John Malmberg" <Malmberg@dskwld.zko.dec.compaq> wrote in message-, news:3BD55B9C.90403@dskwld.zko.dec.compaq... > John Vottero wrote:  >IK > > I upgraded from 2000 to XP and it feels more like a downgrade.  I can't L > > think of one thing that's actually improved.  The one feature that I wasG > > looking forward to was the ability to be logged on to more than oneE accountAG > > and quickly switch between them.  Now I've found out that this only  worksIH > > for home and workgroup environments, if you're in a domain it is not > > available. >r >cG > The ability to do so in a domain is latent at least as far back as NTd? > 3.51.  Resource kit utilities do a usable job of exposing it.  >I  K TopDesk and VDesk were both removed from the W2K Resource Kit.  "Run As..." < is not good enough (not that TopDesk and VDesk were either).  J > I did a session on that at CETS 2000, Covering what it did well and what > it did not so well.  >aA > It is pretty trivial to set up, and work around the interestingnH > limitations where a program did not follow the right API for accessing > the desktop. >aG > The list of programs that I have found that did not comply might havetF > something to do with why the feature never made it into a mainstream
 > product. >a >tK > > One key change that I can think of is the product activation.  You have1 toL > > activate within 60 days or it stops working.  I haven't activated yet so IdB > > don't know how much information they want.  I'm going to delay
 activatingH > > until after I add memory, otherwise I may have to activate twice and incurt, > > the wrath of the Redmond License Police. > >.C > > I'm glad I didn't pay for this "upgrade".  (MSDN subscription).a >c >mI > I have sort of lost interest in keeping up with MSDN for the past year,rH > but usually the MSDN subscriptions contained built in activation codes, > that did not require product registration. > E > After all they are restricted to development use only.  No businesso > support use allowed. >2  F The Server software is restricted to development use only.  You can do- anything you want with the workstation stuff.e   ------------------------------   Date: 23 Oct 2001 16:29:54 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)l, Subject: Re: Windows XP reality check please, Message-ID: <9r45u3$10i1$1@info.cs.uofs.edu>  / In article <ttb4uj2r230kc4@news.supernews.com>,E(  "John Vottero" <John@mvpsi.com> writes:= |> "Bob Koehler" <koehler@encompasserve.org> wrote in messageo0 |> news:B$oqGDvvaP$Z@eisner.encompasserve.org...C |> > In article <tt8dbab4ksr02a@news.supernews.com>, "John Vottero"g |> <John@mvpsi.com> writes:  |> >O |> > > There is a home version of OpenVMS.  It just costs about 5 times as muchc |> asa |> > > Windows.e |> >H |> >    Free is too expensive?  When I think of home version, I think of |> >    hobbyist./ |> > |>   |> home != hobbyist.  H I have been waiting for someone to point this out.  As a matter of fact,G there have been at least one person who has posted an articles recently'G regarding using VMS at home that was obviously not Hobbyist use. (Hint: : go back and read the thread about "DECNET over IP" again.)   bill   -- eJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   t   ------------------------------  % Date: Tue, 23 Oct 2001 12:53:54 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> , Subject: Re: Windows XP reality check please, Message-ID: <3BD5A0A0.C8E21CEB@videotron.ca>   Bob Koehler wrote:E >    Free is too expensive?  When I think of home version, I think ofa >    hobbyist.  N The hobbyist programme is great for many of us. However, it does not answer toN the needs of a small business (or a home business) who may want a REAL license of VMS that is affordable.   ------------------------------  % Date: Tue, 23 Oct 2001 19:37:20 +0200 & From: Michael Joosten <joost@c-lab.de>, Subject: Re: Windows XP reality check please$ Message-ID: <3BD5AAD0.41C6@c-lab.de>   Carl Perkins wrote:  >    > F > I don't know what swapping motherboards counts as under this scheme.E > It might require reactivation regardless of what else did or didn'tyE > change (and if it doesn't, it could still count for a good 3 things H > or more changeing since you usually also change at least the processorG > and memory when you change the motherboard). Again, this is not clear F > to me. I'm no XP expert - I've payed some attention to news about itG > and fiddled with the Pro version for a few minutes a couple of times.eH > I havn't done an activation. There are currently two XP systems on ourJ > net so one of the occasions was to mess with it to figure out how to getG > them to talk to the Samba server I'm running on VMS - which isn't toorI > hard to do (but the connection is slow (the W2K buffer size negotiationTE > patch might improve that, but I havn't gotten around to it yet) andsJ > occasionally disconnects requiring reentering the password, and possiblyF > username, for no known reason, usually when left running overnight). >   A I read in a German computer magazine that they tried to 'push thetB changes to the max'. Even changed the motherboard with 2 or cards.! Still, no reactivation required. rG It might be that the earlier betas were more picky about this, but with5D all the ramblings in the press I'd guess they made it more generous.     > E > As far as using it goes, my limited exposure indicates that it doesoD > boot faster and shut down faster, even compared to W2K. (Actually,G > shutting down faster is a good thing - turning the computer off whilesD > it is in the "saving data to disk" stage is one of the most commonH > methods of corrupting data on an NT system other than crashes. The oneG > time my OS was rendered unbootable on my home system was probably duewI > to this - it's a pity that the two dialog boxes are not more different,i9 > like say bright red for the "saving data" dialog box.) u  F There were some lockups with my office NT or W2K machines where I justC pressed the power button, and I never had any problems with trashedoH filesystem or the like. (knock wood...). On the other hand, with W2K nowG and our SAMBA SOlaris-based fileservers, Explorer needs to be restartedi6 every other day. Definitely worse than NT, I must say!       -- ,* Michael Joosten, SBS C-LAB, joost@c-lab.de* Fuerstenallee 11, 33094 Paderborn, Germany, Phone: +49 5251 606127, Fax: +49 5251 6060658 C-LAB is a cooperation of University Paderborn & SIEMENS   ------------------------------  % Date: Tue, 23 Oct 2001 16:01:01 +1000w8 From: "Jaime H. Barroco" <jbarroco@broadway.bigpond.com> Subject: Working sets 5 Message-ID: <rF7B7.284$c5.2244@newsfeeds.bigpond.com>o   All,,.   I have great   ------------------------------  % Date: Tue, 23 Oct 2001 17:20:25 +1000 / From: "Phil Howell" <phowell@snowyhydro.com.au>t Subject: Re: Working setsc2 Message-ID: <sU8B7.2427$G%4.166970@ozemail.com.au>  C "Jaime H. Barroco" <jbarroco@broadway.bigpond.com> wrote in messagen/ news:rF7B7.284$c5.2244@newsfeeds.bigpond.com...c > All,,  >  > I have great >  >n send button?   ------------------------------  % Date: Tue, 23 Oct 2001 11:59:05 -0400h2 From: rdeininger@mindspring.com (Robert Deininger) Subject: Re: Working setsdL Message-ID: <rdeininger-2310011159050001@user-2ive7bq.dialup.mindspring.com>  H In article <rF7B7.284$c5.2244@newsfeeds.bigpond.com>, "Jaime H. Barroco"& <jbarroco@broadway.bigpond.com> wrote:   > All,,p >  > I have great     It's a Boo!P  . In the midst of the word he was trying to say,& In the midst of his laughter and glee,% He softly and silently vanished away,o$ For the Snark was a Boojum, you see.  ( (from memory, probably not quite right.)   -- E Robert Deininger rdeininger@mindspring.comn   ------------------------------  % Date: Tue, 23 Oct 2001 12:45:11 -0400P  From: norm.raphael@jamesbury.com Subject: Re: Working setse4 Message-ID: <C2256AEE.005BBED6.00@jklh21.valmet.com>   Close enough for me. It made my day (so far).        3 rdeininger@mindspring.com on 10/23/2001 11:59:05 AMe  + Please respond to rdeininger@mindspring.coma   To:   Info-VAX@mvb.saic.coma4 cc:    (bcc: Norm Raphael/Worcester/Neles-Jamesbury) Subject:  Re: Working sets        H In article <rF7B7.284$c5.2244@newsfeeds.bigpond.com>, "Jaime H. Barroco"& <jbarroco@broadway.bigpond.com> wrote:   > All,,0 >o > I have great     It's a Boo!a  . In the midst of the word he was trying to say,& In the midst of his laughter and glee,% He softly and silently vanished away,M$ For the Snark was a Boojum, you see.  ( (from memory, probably not quite right.)   -- Robert Deininger rdeininger@mindspring.com    ------------------------------  % Date: Tue, 23 Oct 2001 11:06:38 +0100n* From: "Richard Brodie" <R.Brodie@rl.ac.uk>0 Subject: Re: WRITEBOOT and dual-boot bootblocks?, Message-ID: <9r3ffe$19fo@newton.cc.rl.ac.uk>  ? "Robert Deininger" <rdeininger@mindspring.com> wrote in messagenF news:rdeininger-2210012042210001@user-2ive6ab.dialup.mindspring.com...  I > As you said, there is no support for dual-architecture boot disks, so ImI > have trouble seeing how anyone could find dual-architecture boot blocks0	 > useful.l  G One could build a dual system disk with a 'nothing shared' model easilyb> enough, I imagine. I can't see many practical benefits though.   ------------------------------    Date: 23 Oct 2001 05:27:48 -0700) From: P.Young@unsw.EDU.AU (Patrick Young)a0 Subject: Re: WRITEBOOT and dual-boot bootblocks?= Message-ID: <55f85d77.0110230427.1dc9f1bf@posting.google.com>3  k hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) wrote in message news:<7e%A7.786$RL6.8716@news.cpqcorp.net>...r >  =E >   Does anybody actually use this dual-format boot block capability?= >   A I don't use this capability - I can only imagine a configuration oB consisting of a VAX and an Alpha both with removable (mobile) SCSIC racks and a disk moved between the two (with the addition explained: below). F I have always been a fan of the coolness of the possibilities (sorry).  C Over the weekend I had cause to setup a cluster here at home using eD a single system disk and both OpenVMS 7.2 and 7.3 running. The AlphaC PC164 SRM console simply *does not deal* with many disk controllers  (even DEC/Compaq supplied).i   I handled it with:   $ dir dk:[0,0]vms*comm*,sys*   Directory DK:[000,000]  E VMS$COMMON.DIR;1    VMS$COMMON_AXP_72.DIR;1       SYS0.DIR;1         r  SYS1.DIR;1          SYSEXE.DIR;1  B and did the appropriate set file/enter, constructed the cluster by hand and" booted the 2nd system using LANCP.  F Unless there are any shared SCSI Alpha/VAX clusters I don't know about I > have not personally come up against your question as an issue.   ------------------------------  % Date: Tue, 23 Oct 2001 14:10:28 +0100 4 From: John Laird <john@laird-towers.freeserve.co.uk>0 Subject: Re: WRITEBOOT and dual-boot bootblocks?8 Message-ID: <lsqattoag6oigocnvamaqr3m5ef665ptvt@4ax.com>  E On Mon, 22 Oct 2001 20:22:27 GMT, hoffman@xdelta.zko.dec.nospam (Hoffh Hoffman) wrote:k  F > This question is specific to the real usefulness of having both VAX < >  and Alpha boot block contents within the same boot block.  G Hoff, remind us which processors actually *need* the boot block definedeH in order to boot.  It's by no means all of them, is it ?  (Certainly for VAX.)   F [Having found the boot block on my 4000/60 system disk looked as if it) hadn't seen a writeboot for some time...]g   	Johna --  
 John Laird Yezerski Roper Ltd   ------------------------------  % Date: Tue, 23 Oct 2001 17:11:58 +0100a% From: "a.carlini" <arcarlini@iee.org>a0 Subject: Re: WRITEBOOT and dual-boot bootblocks?' Message-ID: <3BD596CE.9C8A6E05@iee.org>t   Richard Brodie wrote:pI > One could build a dual system disk with a 'nothing shared' model easilyd@ > enough, I imagine. I can't see many practical benefits though.  / A single CD that could boot on VAX and/or Alphat1 may be useful in some circumstances. For example, , the NetBSD folks might want to have a single/ bootable disk that could do VAX & Alpha & i386.w  , As a "production" feature, it seems to have  considerably less value.   AntonioV   -- S   ---------------R- Antonio Carlini             arcarlini@iee.org    ------------------------------  # Date: Tue, 23 Oct 2001 16:53:11 GMTx2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)0 Subject: Re: WRITEBOOT and dual-boot bootblocks?1 Message-ID: <XfhB7.824$RL6.9137@news.cpqcorp.net>r  o In article <lsqattoag6oigocnvamaqr3m5ef665ptvt@4ax.com>, John Laird <john@laird-towers.freeserve.co.uk> writes: F :On Mon, 22 Oct 2001 20:22:27 GMT, hoffman@xdelta.zko.dec.nospam (Hoff :Hoffman) wrote: : G :> This question is specific to the real usefulness of having both VAX -= :>  and Alpha boot block contents within the same boot block.0 :nH :Hoff, remind us which processors actually *need* the boot block definedI :in order to boot.  It's by no means all of them, is it ?  (Certainly fora :VAX.)  H   IIRC, all Alpha systems use the boot block.  All OpenVMS IA-64 systemsG   will use the boot block.  Various VAX systems -- specifically the VAX K   systems that do not have VMB built into console ROM and that also do not tG   require a console storage medium for bootstrap: the VAX-11/750 (usinguI   the ROM-based boot path and not the ROM-to-TU58-to-boot path), the VAX 'I   8200 and VAX 8300 series, the VAX 6000 series, the VAX 7000 series, thenK   VAX 10000 series.  (I'll ignore "fun" like VMBUVAX1P.EXE and VMB9AQ.EXE.)a  I   All MicroVAX systems and all (but one) VAXstation systems systems have a3   (old) versions of VMB built into the console ROM.i  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Tue, 23 Oct 2001 11:59:51 -0500w+ From: Christopher Smith <csmith@amdocs.com> 0 Subject: RE: WRITEBOOT and dual-boot bootblocks?L Message-ID: <3B55D7F383B0D31197D9009027541CBF1170DDB8@cmiexch1.cmi.itds.com>  E Out of curiosity, which VAXStation system is The One? :)  (Or are youdJ referring to the VAXStation which was an add-on for the 11/780 (I think)?)   Regards,   Chrisa  ! Christopher Smith, Perl Developerp Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");d 'n  r   > -----Original Message-----% > From: hoffman@xdelta.zko.dec.nospamp  > >   All MicroVAX systems and all (but one) VAXstation systems  > systems have l5 >   (old) versions of VMB built into the console ROM.    ------------------------------  % Date: Tue, 23 Oct 2001 12:33:42 -0400-( From: David Froble <davef@tsoft-inc.com>8 Subject: Re: X server and SET DISPLAY on different port?, Message-ID: <3BD59BE6.1060003@tsoft-inc.com>  $ Brian Schenkenberger, VAXman- wrote:  n > In article <3bd4350c$0$271$edfadb0f@dspool01.news.tele.dk>, "Jesper Naur" <jesper.naur@post.tele.dk> writes: > J >>Brian Schenkenberger, VAXman- <system@SendSpamHere.ORG> wrote in message, >>news:00A03E55.F5BC5B17@SendSpamHere.ORG... >>K >>>In article <00A03C1F.B4F4B513@SendSpamHere.ORG>, system@SendSpamHere.ORGa >>>u) >>(Brian Schenkenberger, VAXman-) writes:  >>G >>>>Is it possible to run the DECwindows server such that it will use as >>>> >>different0 >>I >>>>port (a port other than port 6000)?  If so, can a SET DISPLAY commandm >>>>	 >>then beG >>$ >>>>directed to this new port?  How? >>>> >>If you say >>< >>   $ set display/create/trans=tcpip/node=<node>/server=<n> >>; >>the applications will attempt to connect to port 6000+<n>t >> >>   Best regards  >>   Jesper Naur >> >  > How about port 100?  >  > --Q > VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM-
 >            -L >   "And of course, I'm a genius, so people are naturally drawn to my fiery K >   intellect.  Their admiration overwhelms their envy!" -- Calvin & Hobbesv >  >   0 What happens if you use a value of -5900 for n ?     -- c4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com6 T-Soft, Inc.  170 Grimplin Road  Vanderbilt, PA  15486   ------------------------------  % Date: Tue, 23 Oct 2001 08:24:51 +0200g7 From: Alain Chappuis <alain.chappuis@medecine.unige.ch>h, Subject: Re: [MOZILLA] Burner of CPU time!!!1 Message-ID: <3BD50D33.6B7719DE@medecine.unige.ch>t   Colin Blake a crit :i > I > The CPU consumed when "idle" depends upon what you're making Mozilla doaD > when its "idle". For example, leave Mozilla on a page with lots ofF > animated graphics, and you'll see that kind of CPU usage. But if youH > leave Mozilla on a "static" page, then CPU usage should come way down.  H No when Mozilla stay in my home page, I dont have any animated graphics,* and I dont have my mail box page opened...   Alain. -- iD  +----------------------+------------------------------------------+D  | Alain Chappuis       | Responsable: E-mail; cmu.unige.ch        |D  | Analyste             | WEB    : www.medecine, ebn, jid, Sifm    |D  | Universite de Geneve | E-mail : Alain.Chappuis@unige.ch         |D  | Centre Medical Univ. | Phone  : +41 (22) [70]25.073             |D  | 1, Rue Michel-Servet | FAX    : +41 (22) 347.33.34 ou 702.58.58 |K  | CH-1211 Geneve 4     | http://ebn.unige.ch/www/alain.html       |       nD  +----------------------+------------------------------------------+   ------------------------------    Date: 23 Oct 2001 08:57:18 -0500- From: koehler@encompasserve.org (Bob Koehler)P, Subject: Re: [MOZILLA] Burner of CPU time!!!3 Message-ID: <a836hcWoGyMt@eisner.encompasserve.org>e  W In article <3BD4310B.D5290886@theblakes.com>, Colin Blake <colin@theblakes.com> writes:sI > The CPU consumed when "idle" depends upon what you're making Mozilla dooD > when its "idle". For example, leave Mozilla on a page with lots ofF > animated graphics, and you'll see that kind of CPU usage. But if youH > leave Mozilla on a "static" page, then CPU usage should come way down. >   K    Yes, it is much lower, but it's still not 0, which is what it should be. B    Looks like Mozilla is polling for something, it should be event
    driven.   ------------------------------  # Date: Tue, 23 Oct 2001 15:16:50 GMTd' From: Colin Blake <colin@theblakes.com>e, Subject: Re: [MOZILLA] Burner of CPU time!!!- Message-ID: <3BD589E0.808D5649@theblakes.com>v  [ The UNIX code (from which the VMS version is built) expects to be able to select() on a set8[ of fd's that include the X connection number. On OpenVMS, the X connection number is NOT anTZ fd, and so this purely event driven model won't work for us without some major changes. SoZ we have to watch the fd's via a select, and also have to watch for X activity as well. TheW way this is currently done (so as to not have to rewrite major chunks of Mozilla) is toi[ select() for a certain amount of time and then check for X activity. This polling mechanismiQ does, of course, cost in terms of CPU. And the more frequently you poll, the moreiT responsive the browser feels, but the more it costs you in terms of CPU. The currentU settings were chosen to give good response without too much CPU usage. If you wish tonW change the poll timers, you can. Look in MOZILLA.COM and you'll see VMS_POLL_TIMER_MIN,nZ VMS_POLL_TIMER_MAX, and VMS_POLL_TIMER_DELTA (they are currently commented out because theV default values are the same as the values in MOZILLA.COM). The poll timer is a slidingW timer; it starts at the MIN value and increases (each time there is no activity) by thenZ DELTA until it reaches MAX. It never goes past MAX and goes back to MIN each time there is[ some activity. All values are in milli-seconds. A DELTA value of 0 indicates that the timertY should doubled each time there is no activity seen. Try playing. You may find some values  which work better for you.   ------------------------------  + Date: Tue, 23 Oct 2001 16:05:13 +0000 (UTC)  From: david20@alpha1.mdx.ac.uk, Subject: Re: [MOZILLA] Burner of CPU time!!!+ Message-ID: <9r44fp$7ei$1@aquila.mdx.ac.uk>l  W In article <3BD589E0.808D5649@theblakes.com>, Colin Blake <colin@theblakes.com> writes:s\ >The UNIX code (from which the VMS version is built) expects to be able to select() on a set\ >of fd's that include the X connection number. On OpenVMS, the X connection number is NOT an[ >fd, and so this purely event driven model won't work for us without some major changes. Soh[ >we have to watch the fd's via a select, and also have to watch for X activity as well. ThenX >way this is currently done (so as to not have to rewrite major chunks of Mozilla) is to\ >select() for a certain amount of time and then check for X activity. This polling mechanism( >does, of course, cost in terms of CPU.     N Do the rewrite and feed it back to the Mozilla developers so you won't have to keep rewriting it.O The load of Netscape 3.03 on VMS is bad enough anything worse will be unusable.gI From the sound of this performance (I haven't tried using Mozilla on VMS nL yet) if this isn't sorted out then you might as well not have bothered doing this work in the first place.   
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------  % Date: Tue, 23 Oct 2001 12:27:13 -0400 5 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>b, Subject: Re: [MOZILLA] Burner of CPU time!!!1 Message-ID: <DUgB7.820$RL6.8977@news.cpqcorp.net>   F VMS will address select() and fork() in a future version.  It's in ourA plans.  Since doing a select() as described is a widely used UNIXaK convention, it's not likely to get a warm reception as something to change.c  L david20@alpha1.mdx.ac.uk wrote in message <9r44fp$7ei$1@aquila.mdx.ac.uk>...: >In article <3BD589E0.808D5649@theblakes.com>, Colin Blake <colin@theblakes.com> writes:mK >>The UNIX code (from which the VMS version is built) expects to be able to  select() on a seteL >>of fd's that include the X connection number. On OpenVMS, the X connection number is NOT anJ >>fd, and so this purely event driven model won't work for us without some major changes. SooF >>we have to watch the fd's via a select, and also have to watch for X activity as well. The.J >>way this is currently done (so as to not have to rewrite major chunks of Mozilla) is toK >>select() for a certain amount of time and then check for X activity. This  polling mechanisme( >>does, of course, cost in terms of CPU. >1 >3L >Do the rewrite and feed it back to the Mozilla developers so you won't have to >keep rewriting it.iF >The load of Netscape 3.03 on VMS is bad enough anything worse will be	 unusable.1I >From the sound of this performance (I haven't tried using Mozilla on VMSuG >yet) if this isn't sorted out then you might as well not have botheredm doingl >this work in the first place. >- >David Webb- >VMS and Unix team leader0 >CCSS0 >Middlesex University    ------------------------------  + Date: Tue, 23 Oct 2001 17:01:41 +0000 (UTC)8 From: david20@alpha1.mdx.ac.uk, Subject: Re: [MOZILLA] Burner of CPU time!!!+ Message-ID: <9r47pl$8eg$1@aquila.mdx.ac.uk>c  i In article <DUgB7.820$RL6.8977@news.cpqcorp.net>, "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> writes:-G >VMS will address select() and fork() in a future version.  It's in oursB >plans.  Since doing a select() as described is a widely used UNIXL >convention, it's not likely to get a warm reception as something to change. ><  H I wasn't suggesting that the Mozilla writers needed to change it for anyI other platform just that you put in some ifdef VMS pieces of code to use   something better for VMS.h  I Having Mozilla unusable due to perfomance on VMS until VMS 7.4 / VMS 8 ori" whenever is not really acceptable.  
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------   End of INFO-VAX 2001.590 ************************