1 INFO-VAX	Tue, 08 Aug 2006	Volume 2006 : Issue 439       Contents:* Re: ACCOUNTING Utility report presentation5 Re: Authentication for outgoing TCP/IP Services mail? 5 Re: Authentication for outgoing TCP/IP Services mail? 5 Re: Authentication for outgoing TCP/IP Services mail? 5 Re: Authentication for outgoing TCP/IP Services mail? 5 Re: Authentication for outgoing TCP/IP Services mail? 5 Re: Authentication for outgoing TCP/IP Services mail? 5 Re: Authentication for outgoing TCP/IP Services mail? 5 Re: Authentication for outgoing TCP/IP Services mail?  Re: Back at work Re: Back at work Re: Back at workO Compile and gone! (Was: Re: VMS>7.3 Linker Manual - I64 Short Data restriction)  jss job  Re: jss job . Official: Do not try to write UWSSs in C crap!2 Re: Official: Do not try to write UWSSs in C crap!2 Re: Official: Do not try to write UWSSs in C crap!. Process/RSA help vms based bank use smartcardsA Re: Under VMS, on an HSG80, can Raid Partition Size be Increased?  Updating firmware on DEGPA-TA  Re: USB numeric keypad anyone?P Re: V8.3 Apache and Javascript in HTML files? (was Re: VMS 7.3 vs 8.x - java in & Re: VMS backup: competition from Apple& Re: VMS backup: competition from Apple& Re: VMS backup: competition from Apple& Re: VMS backup: competition from Apple& Re: VMS backup: competition from Apple" Re: XP1000 USB port working in VMS  F ----------------------------------------------------------------------  % Date: Tue, 08 Aug 2006 11:09:55 -0500 + From: brandon@dalsemi.com (BRANDON, JOHN M) 3 Subject: Re: ACCOUNTING Utility report presentation 1 Message-ID: <06080811095515@dscis6-0.dalsemi.com>    andrewr@cornasys.com writes:D > > I have a requirement to set up the ACCOUNTING utility to extract0 > > reports on the OpenVMS resource utilisation.  / koehler@eisner.nospam.encompasserve.org writes: F >    I always used monitor/record and then monitor/input to meet these >    kinds of needs. > . >    Are you sure you're using the right tool?  " "Ian Miller" <ijm@uk2.net> writes:> > Back around 1987 I wrote a program that reads the accounting@ > records and writes it out in a fixed record definitions formatD > so it can be read with Datatrieve and other programs.  It's called? > ACC_CONVERT.MAR, and it's been in a number of DECUS and other D > software collections.  In 2000 I did a few touchups to make it run@ > better on Alpha.  It's 15 KBytes, so it's a little large for aB > Usenet News posting, but I suppose it could be done if you can't@ > find it on the Internet somewhere.  Or I can make it availableB > on a web page somewhere.  But try a web search first, the format@ > of the accounting record hasn't changed in a long time, so the* > program shouldn't have to change either.    O Would use of the ECP or T4 or TDC (replacement for ECP) packages be sufficient? N I use ECP and that allows for extract into a CSV file - port into Excell and I+ have a chart with CPU usage, memory, etc...   K I have found that (for the most part) I can use ECP instead of ACCOUNTING -  with a few excpetions.   Links to those products...9 http://h71000.www7.hp.com/openvms/products/ecp/index.html 9 http://h71000.www7.hp.com/openvms/products/tdc/index.html O http://h71000.www7.hp.com/OpenVMS/products/t4/t4_faq.html?jumpid=reg_R1002_USEN            John "REBOOT" Brandon  VMS Systems Administrator * firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  % Date: Tue, 08 Aug 2006 10:38:54 +0200 / From: Paul Sture <paul.sture.nospam@hispeed.ch> > Subject: Re: Authentication for outgoing TCP/IP Services mail?: Message-ID: <8ae7$44d84d9f$50db5015$24917@news.hispeed.ch>   Bob Koehler wrote:o > In article <415a8$44d4c9ff$50db5015$27406@news.hispeed.ch>, Paul Sture <paul.sture.nospam@hispeed.ch> writes: L >> Some time ago, my ISP changed their mail system so that I need to supply I >> authentication for outgoing mail (Good Move against spamming zombies).  >>B >> In Mac Mail, Thunderbird etc, there are settings for supplying $ >> authentication for outgoing mail. >>1 >> Does similar exist in TCP/IP Services for VMS?  >> > H >    That kind of thing can be fonud in most browsers.  Try reading your% >    email via Mozilla, works for me.  >   - I'd better say what triggered my question :-)   F I am using Graham Burley's FTP_NW.COM utility to fetch the latest ECO  notifications.   See   % http://www.encompasserve.org/~burley/    for details.  H I just thought it would be nice to be able to send notifications to the  outside world too.   --  
 Paul Sture   ------------------------------  * Date: Tue, 8 Aug 2006 10:16:54 +0000 (UTC) From: david20@alpha2.mdx.ac.uk> Subject: Re: Authentication for outgoing TCP/IP Services mail?, Message-ID: <eb9oam$3oj$1@south.jnrs.ja.net>  ` In article <1WJBg.1645$tG4.883@news.cpqcorp.net>, Hoff Hoffman <hoff-remove-this@hp.com> writes:  >david20@alpha2.mdx.ac.uk wrote: > H >> True enough but doesn't help if you want to run your own mail server. > ) >   SMTP mail servers (MTAs) use port 25.  > I >   SMTP mail clients (MUAs) use port 25 or (with SMTP-MTA ESMTP RFC2476  + >support) to the MSA listening on port 587.  > G >   I've suggested a modification to the SMTP transport to allow it to  I >connect into port 587, as the SMTP transport image for the OpenVMS MAIL  ! >utility (SMTP%) is the MUA here.  > I >   Basically either the SMTP transport client has to be modified to use  H >port 587, or there needs to be some sort of a translating gateway here G >that accepts connections on 25 and passes them along on 587.  Or some  G >other MAIL client need be used, or a custom SMTP MAIL transport image  J >need be created, or the ISP needs to designate this OpenVMS system as an  >MTA.  (Which it is.)  >    Hoff,   I If the ISP is using separate incoming and outgoing mail servers they can  J set things up to require authentication (using SASL and SMTP AUTH) on the  standard smtp port. N If the ISP is using the same mailhub for both incoming and outgoing mail then O they can set it up to accept authentication using SASL and SMTP AUTH on port 25 M but cannot obviously require it without compromising their ability to receive F mail from other organisations. In that case they would have to use the% submission port 587 for this purpose.   M I have asked PMDF to add an option into their product to support this type of K authentication (PMDF already supports SASL and SMTP AUTH and the submission D port for connections to it as a server but doesn't currently support9 authentication when passing mail out to a ISP smarthost).   2 Note other MTAs already provide this functionality   see for instance  6 http://www.sendmail.org/~ca/email/auth.html#smtpclient  $ Using sendmail as a client with AUTH    S http://postfix.state-of-mind.de/patrick.koetter/smtpauth/smtp_auth_mailservers.html   ( 16. SMTP Authentication for Mail servers     and   ) http://www.hserus.net/wiki/index.php/Exim             
 David Webb Security team leader CCSS Middlesex University   ------------------------------  * Date: Tue, 8 Aug 2006 10:29:56 +0000 (UTC) From: david20@alpha2.mdx.ac.uk> Subject: Re: Authentication for outgoing TCP/IP Services mail?, Message-ID: <eb9p34$42c$1@south.jnrs.ja.net>  O In article <op.tdxdrbpgzgicya@hyrrokkin>, "Tom Linden" <tom@kednos.com> writes: 0 >On Mon, 07 Aug 2006 10:32:21 -0700, JF Mezei  = > & ><jfmezei.spamnot@teksavvy.com> wrote: >  >> Hoff Hoffman wrote:J >>>    The other approaches are to seek and acquire ISP authorization for= > F >>> your SMTP mail server, or to use an alternate mail client (MSA) to >>> submit your email. >>J >> You do realise that you are effgectively suggesting one move off VMS ?= > $ >> ("use an alternate mail client"). > J >Nothing wrong with that, and it is a better solution, but then you could= >  = >  >use" >soyMail too, which is quite good. >>  K I'm not aware of any ISP SMTP authentication methods in SoyMail to send out N mail - you could set smtp-default-host to your ISPs mailhub but as far as I amM aware SoyMail can't be configured to respond to the authentication challenge. L As I recall the previous version Yahmail just sent mail out through the mail! server running on the VMS system.       G >> The point is that we want VMS to be modern and usable in the current J >> environment. SMTP is being tightened up in many ways to prevent spam. = >So J >> either HP re-hires engineers to do work on the TCPIP Services product,= > J >> or you open-source it NOW so that we (the users) can gradually add the= >  >> missing features. > ) >Or you use a third party product like MX   J PMDF doesn't currently support this option so I'd be surprised if MX does.  
 David Webb Security team leader CCSS Middlesex University   ------------------------------  * Date: Tue, 8 Aug 2006 10:55:43 +0000 (UTC) From: david20@alpha2.mdx.ac.uk> Subject: Re: Authentication for outgoing TCP/IP Services mail?, Message-ID: <eb9qje$4fk$1@south.jnrs.ja.net>  a In article <P8NBg.1669$HQ4.1151@news.cpqcorp.net>, Hoff Hoffman <hoff-remove-this@hp.com> writes:  >JF Mezei wrote: >> Hoff Hoffman wrote:I >>>    The other approaches are to seek and acquire ISP authorization for F >>> your SMTP mail server, or to use an alternate mail client (MSA) to >>> submit your email. >>  J >> You do realise that you are effgectively suggesting one move off VMS ? $ >> ("use an alternate mail client"). > F >   Eh?  There are various supported and freeware MUAs (mail clients) F >available for OpenVMS.  I certainly usually end up packaging several J >MUAs onto each Freeware, and the SWB (Mozilla) client for OpenVMS offers  >a MUA.  > G >> The point is that we want VMS to be modern and usable in the current I >> environment. SMTP is being tightened up in many ways to prevent spam.   > F >   So you're really and seriously asking for us to violate the RFCs? : >(That sort of approach would require justification, too.) >  Hoff,   J Modern mail servers can accept mail on both the standard smtp port and on K the submission port. However they can also be run on non-standard ports for K both sending and receiving mail. In this instance all outgoing mail will be K sent via the ISP's mailhub hence if the ISP requires use of the submission  M port that should not be a problem it is just sending on a non-standard port - K port 587 rather than port 25. The problem is the authentication challenge.    K The writers of PMDF (who have authored a number of the RFCs) have supported . use of non-standard ports pretty much forever.    G >   I can understand wanting to have this, but MTAs operate on port 25   >and not on port 587.     M Ports 25 and 587 are the standard default ports. Use of non-standard ports is  not banned by the RFCs.   3 >MTAs are not supposed to send to port 587.  Which  I >means you either use an MUA and operate as a client as the ISP expects,  F >or you negotiate with your ISP to provide authorized port 25 access. J >(And if it's not obvious, OpenVMS provides an MUA that operates with its B >MTA, and expects the MTAs to be connected per established norms.) >   L You are misinterpretting the established norms since sending to non-standardH ports and SMTP auth for connection to another mailserver is supported by2 numerous MTAs including Sendmail, Postfix and Exim    
 David Webb Security team leader CCSS Middlesex University      G >   I'd certainly not mind seeing an integrated MUA client for OpenVMS  C >(either via the existing SMTP% transport, or via some replacement  H >transport), for those that are using OpenVMS as a client behind an ISP J >and need to move to port 587 and into an MSA.  And it would not surprise J >me if there was such a construct around -- I certainly have the shell of G >the transport image around, as do others.  (But having an MTA connect  B >into an MSA is a non-starter without a whole lot more supporting  >justification.) >  >  >  >    ------------------------------  % Date: Tue, 08 Aug 2006 14:42:31 +0200 + From: Martin Vorlaender <mv@pdv-systeme.de> > Subject: Re: Authentication for outgoing TCP/IP Services mail?+ Message-ID: <4jrf5fF97ipbU1@individual.net>    Paul Sture wrote: H > I am using Graham Burley's FTP_NW.COM utility to fetch the latest ECO  > notifications. >  > See  > ' > http://www.encompasserve.org/~burley/  >  > for details. > J > I just thought it would be nice to be able to send notifications to the  > outside world too.  4 You do know Ken Farmer's OpenVMS Alert mailing list?  , http://www.openvms.org/pages.php?page=Alerts   cu, 	    Martin  --  D One OS to rule them all       | Martin Vorlaender  |  OpenVMS rules!7 One OS to find them           | work: mv@pdv-systeme.de J One OS to bring them all      |   http://www.pdv-systeme.de/users/martinv/> And in the Darkness bind them.| home: martin@radiogaga.harz.de   ------------------------------  # Date: Tue, 08 Aug 2006 13:19:16 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>> Subject: Re: Authentication for outgoing TCP/IP Services mail?1 Message-ID: <ob0Cg.1688$rB5.216@news.cpqcorp.net>    david20@alpha2.mdx.ac.uk wrote: J > are misinterpretting the established norms since sending to non-standardJ > ports and SMTP auth for connection to another mailserver is supported by4 > numerous MTAs including Sendmail, Postfix and Exim  ?    In this case, MUAs send to 25 or 587, while MTAs send to 25.   H    What other protocols and ports might be used here (and I've seen all H manner of non-default and non-standard ports used, and non-default port D schemes including doorbell schemes that unmask other ports) doesn't G really matter here, because the ISP is using 25 for its own transport,  @ and requesting 587 (the MSA port) for its clients; for the MUAs.  E    The RFCs are rather clear about not having MTAs connected to MSAs  G (for all that the RFCs can actually mean in reality -- more than a few  ; RFCs around have ambiguitities or run contrary to accepted  G implementations), but (as I recall) I did also mention the possibility    of a gateway between 25 and 587.  E    The intended configuration (from the perspective of the ISP) is a  I MUA, meaning you need to use a client package -- and these are available  G for OpenVMS, including the mail client inside the Mozilla suite.  (Now  I as for a MUA transport image, as I mentioned, that would be useful here,  I too.  Combine it with a POP or IMAP server daemon, and you'd have a full  H client-mail configuration -- rather than the existing SMTP server.  But % OpenVMS has neither of these pieces.)   H    Which means you are left to use an MUA on OpenVMS, or negotiate with C the ISP to open 25.  (Or to find a gateway to feed between the ISP  G environment and the server environment -- POP/IMAP into SMTP, and SMTP  I MTA into the MSA -- OpenVMS doesn't have SMTP AUTH on port 25, nor AFAIK   an MSA on 587.)   E    Now as for adding SMTP AUTH (and likely also MSA support), that's  H certainly an entirely reasonable request -- but like the discussions of I the ports this particular ISP is using -- it's not immediately relevant.  G   If/when the SMTP AUTH TCP/IP Services work (or the SMTP client work)  F is accepted as an enhancement (and regardless of what pieces might be F involved), the timeframe involved is inevitably past the horizon that < the original poster will be willing to tolerate.  Hence the G recommendation for a MUA that has this -- and which also happens to be  G the configuration clearly recommended by and expected by the ISP -- or  & to get port 25 opened for this server.  I    As is my norm for these sorts of newsgroup discussions, I'll pass the  : enhancement request along to the OpenVMS business manager.   ------------------------------  * Date: Tue, 8 Aug 2006 15:35:27 +0000 (UTC) From: david20@alpha2.mdx.ac.uk> Subject: Re: Authentication for outgoing TCP/IP Services mail?, Message-ID: <ebaavv$99e$1@south.jnrs.ja.net>  ` In article <ob0Cg.1688$rB5.216@news.cpqcorp.net>, Hoff Hoffman <hoff-remove-this@hp.com> writes:  >david20@alpha2.mdx.ac.uk wrote:K >> are misinterpretting the established norms since sending to non-standard K >> ports and SMTP auth for connection to another mailserver is supported by 5 >> numerous MTAs including Sendmail, Postfix and Exim  > @ >   In this case, MUAs send to 25 or 587, while MTAs send to 25. > N MTAs, MUAs and MSAs are all free to send to whatever ports they are configuredN with. The default standard port for smtp mail transmission whether from a MUA P or MTA is port 25. The default standard port for message submission from an MSA M is port 587. But those are just defaults. Any particular group of servers and < clients can privately agree to use other non-standard ports.       I >   What other protocols and ports might be used here (and I've seen all  I >manner of non-default and non-standard ports used, and non-default port  E >schemes including doorbell schemes that unmask other ports) doesn't  H >really matter here, because the ISP is using 25 for its own transport, A >and requesting 587 (the MSA port) for its clients; for the MUAs.  >   D Please refer to the links I provided for Sendmail, Postfix and Exim.H All these MTAs support the server providing a username and password whenH sending to mail servers which require SMTP AUTH from connecting systems.P (The ISP's mail server has no way of determining whether an incoming connection O is from an MUA or an MTA. The SMTP protocol provides no way for it to make that M distinction. What it can determine is whether the mail is from a user who it  H knows by requiring authentication using SMTP AUTH (According to how the F ISP's mail server is setup this may be on the standard SMTP port, The F submission port or some other non-standard port specified by the ISP).  F >   The RFCs are rather clear about not having MTAs connected to MSAs H >(for all that the RFCs can actually mean in reality -- more than a few < >RFCs around have ambiguitities or run contrary to accepted H >implementations), but (as I recall) I did also mention the possibility ! >of a gateway between 25 and 587.  >   G Please point to a reference to something in the RFCs which states that  J a system performing the functions of an MTA cannot also perform a message 2 submission. I am unaware of any such proscription.  9 Indeed rfc2821 section 2.3.3 talking about MUAs and MTAs   specifically warns   " J  However, while these terms are used with at least the appearance of greatF  precision in other environments, the implied boundaries between MUAs L  and MTAs often do not match common, and conforming, practises with internetF  mail. Hence, the reader should be cautious about inferring the strongM  relationships and responsibilities that might be implied if these terms were   used elsewhere. "   F And as I have pointed out many mail servers can perform this function.      
 David Webb Security team leader CCSS Middlesex University        F >   The intended configuration (from the perspective of the ISP) is a J >MUA, meaning you need to use a client package -- and these are available H >for OpenVMS, including the mail client inside the Mozilla suite.  (Now J >as for a MUA transport image, as I mentioned, that would be useful here, J >too.  Combine it with a POP or IMAP server daemon, and you'd have a full I >client-mail configuration -- rather than the existing SMTP server.  But  & >OpenVMS has neither of these pieces.) > I >   Which means you are left to use an MUA on OpenVMS, or negotiate with  D >the ISP to open 25.  (Or to find a gateway to feed between the ISP H >environment and the server environment -- POP/IMAP into SMTP, and SMTP J >MTA into the MSA -- OpenVMS doesn't have SMTP AUTH on port 25, nor AFAIK  >an MSA on 587.) > F >   Now as for adding SMTP AUTH (and likely also MSA support), that's I >certainly an entirely reasonable request -- but like the discussions of  J >the ports this particular ISP is using -- it's not immediately relevant. H >  If/when the SMTP AUTH TCP/IP Services work (or the SMTP client work) G >is accepted as an enhancement (and regardless of what pieces might be  G >involved), the timeframe involved is inevitably past the horizon that  = >the original poster will be willing to tolerate.  Hence the  H >recommendation for a MUA that has this -- and which also happens to be H >the configuration clearly recommended by and expected by the ISP -- or ' >to get port 25 opened for this server.  > J >   As is my norm for these sorts of newsgroup discussions, I'll pass the ; >enhancement request along to the OpenVMS business manager.  >    ------------------------------  # Date: Tue, 08 Aug 2006 16:36:36 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>> Subject: Re: Authentication for outgoing TCP/IP Services mail?2 Message-ID: <o43Cg.1704$YF5.1461@news.cpqcorp.net>   david20@alpha2.mdx.ac.uk wrote:   H > And as I have pointed out many mail servers can perform this function.  H    Then it would be reasonable to load and run one, if this is required.  H    Or to use a variant MUA, and particularly one that does support port ' 587 and SMTP-MTA SMTP AUTH submissions.   /    Or to clear the port 25 access with the ISP.   "    Or to use a different platform.  I    If a particular MTA switches and gateways something (and modifies the  C SMTP headers, as is contraindicated in my reading of the RFCs) and  E connects into an MSA somewhere as part of passing along the message,  F well, bully for it.  The TCP/IP Services MTA does not presently offer G this capability, and the present OpenVMS MUA SMTP MAIL transport image  D assumes and performs non-AUTH port 25 operations, and (assuming the I enhancement request(s) that could/would/will arise here are accepted and  F engineered and released) the timeframe involved is likely well beyond ; that of the requirements of the initiator of this sequence.   I    There are options here.  Port 587 submissions via TCP/IP Services MTA  0 and MUA components are not among these, however.   ------------------------------   Date: 8 Aug 2006 06:54:15 +0100 2 From: "Dave Weatherall" <djw-nothere@nospam.nohow> Subject: Re: Back at work ? Message-ID: <DTiotGxQ0bj6-pn2-lusTEWrJUeod@dave2_os2.home.ours>   E On Mon, 7 Aug 2006 22:55:59 UTC, "Sue" <susan_skonetski@hotmail.com>   wrote:   > Dear Newsgroup,  > F > I just wanted to let everyone know that I am now back at work.  I amI > not 100% but really needed to get back to work.  Thank you for all your 
 > support. >  > Warm Regards, and big hugs,  >  > Sue  >   - Good to hear it Sue - keep on getting better.    --   Cheers - Dave W.   ------------------------------   Date: 8 Aug 2006 10:01:59 -0700  From: davidc@montagar.com  Subject: Re: Back at work B Message-ID: <1155056519.404494.96970@p79g2000cwp.googlegroups.com>  
 Sue wrote: > Dear Newsgroup,  > F > I just wanted to let everyone know that I am now back at work.  I amI > not 100% but really needed to get back to work.  Thank you for all your 
 > support. >  > Warm Regards, and big hugs,  >  > Sue   E Welcome back, Sue!  I'm sure you'll be back in the swing of things in < no time.  We'll take it easy on you for a while, though! :-)   ------------------------------   Date: 8 Aug 2006 10:17:54 -0700  From: davidc@montagar.com  Subject: Re: Back at work C Message-ID: <1155057473.999932.191600@p79g2000cwp.googlegroups.com>   
 Sue wrote: > Dear Newsgroup,  > F > I just wanted to let everyone know that I am now back at work.  I amI > not 100% but really needed to get back to work.  Thank you for all your 
 > support. >  > Warm Regards, and big hugs,  >  > Sue   E Welcome back, Sue!  I'm sure you'll be back in the swing of things in < no time.  We'll take it easy on you for a while, though! :-)   ------------------------------  $ Date: Tue, 8 Aug 2006 20:33:00 +08003 From: "Richard Maher" <maher_rj@hotspamnotmail.com> X Subject: Compile and gone! (Was: Re: VMS>7.3 Linker Manual - I64 Short Data restriction)1 Message-ID: <eba04u$n9t$1@news-02.connect.com.au>    Hi John,  A Thanks for the very detailed response. Doesn't look good does it.    > Is your application in COBOL?   K No, it's BASIC. The plans were put in place at this site over a year ago to J break the uber-image up into "n" shareable images but I was hoping that ifJ there was any justice in the world then something would've changed betweenA 8.1 and 8.3 to make bypassing the Short_Data segment possible :-(   I I find it difficult in these circumstances to argue against the shameless I Charon-Alpha reseller! Why don't HP just keep smacking VMS users over the D head until they finally wake up and smell the coffee? If you have toB completely re-engineer your application to upgrade to an unproven,I longevity-suspect boat-anchor then why don't you put that migration money  into something with a future?   I I personally dislike JF banging-on about that "IA64 Thing", but I tell ya  it's starting to grow on me.  J Dear God in heaven please let John have a refreshing holiday and bring himE back safe to start porting MACRO to 8086,I32,Pentium7 whatever it is!    Cheers Richard Maher  3 "John Reagan" <john.reagan@hp.com> wrote in message + news:DBIBg.1639$sx1.741@news.cpqcorp.net...  > Richard Maher wrote: > > Hi,  > > D > > When could one expect to see a new version of the Linker Utility	 Reference K > > Manual available with VMS. The one on the Web and Doc CDs appears to be  VMS H > > 7.3 April 2001. All the I64 stuff I've seen is in one of the 8.2 NewJ > > Features Manual (Which would make sense if 8.2 was the first supported@ > > itanium version :-) but are we gonna get a new set with 8.3? > / > Yes, an updated linker manual comes with 8.3.  >  > > I > > Also, Can anyone shed any light of possible solutions/work-arounds to  the K > > 4MB limit on this Short Data segment thing? (Something other than "Your 3 > > app's crap - re-engineer it!" would be nice :-)  > E > In general, filling up the 4MB short section results from compilers I > asking for too many address constants.  Although the short section also E > contains linker-created function descriptors and static user data 8  > bytes or smaller.  > J > As background, on Alpha, each routine is passed a routine-unique pointerJ > to the "linkage" section which contains its procedure descriptor and allE > the compiler requested address constants.  While there isn't a hard D > limit, the downside is that if multiple compilations need the sameA > address, there are many copies in the final image.  On I64, the H > compilers don't allocate space for address constants, but request thatH > the linker materialize the address for them.  It makes the linker moreB > complicated, but it also allows the linker to remove duplicates. > E > By convention (we didn't 'invent' the convention, it comes from the E > Intel Run-Time Conventions), this linker-created section of address J > constants (and function descriptors, etc.) is shared by the whole image.F >   Each routine has the address of this "short" section passed in R1.@ > The short section is accessed with the following instructions: > D >              ; Add offset from R1 to the location where the linkerK >              ; allocated the address.  This form of the 'add' instruction " >              ; is defined to be:) >              ;       add Ra = imm22, Rb ; >              ; where Rb must be either R0, R1, R2, or R3.  >              ;H >              ; When the compiler generates the instruction, it usuallyH >              ; puts a 0 in the 22-bit literal and issues a relocation.H >              ; The relocation tells the linker to create an address toF >              ; SYM, allocate it in the short section, and modify theH >              ; immediate operand of the instruction to have the offset2 >              ; from R1 to the allocated address. >              ;' >              add Rn = @ltoff(SYM), R1  > F >              ; Now that we have the address of the address, fetch it >              ; >              ld8 Rn = [Rn] > I >              ; We can now fetch the static data.  It might be something H >              ; readonly like a linker-materialized address or it might< >              ; be some user variable (8 bytes or smaller). >              ;7 >              ld8 Rn = [Rn]  ; fetch an 8-byte 'thing'  >  > > @ > > Specifically what, if anything, does the Option File command "SHORT_DATA=NO" % > > do for us? (Latent support only?)  >  > Doesn't do anything for you. >  > > L > > I have seen the /SEGMENT=(Short_data=WRITE) but another 65K is not gonnaH > > help us and it would be could to avoid any behavioural changes where
 > > possible.  > D > You can also try /SEGMENT=SYMBOL_VECTOR=NOSHORT (if you creating aG > shareable image), avoid /NONATIVE_ONLY (it makes function descriptors D > larger), and avoid pulling code sections from the default cluster.9 > However, I wouldn't expect much help from any of these.  >  > > L > > There seems to be a fair sprinkling of discussions on the web (Some withJ > > Linux and Windows) but it seems to come down to the compiler having to do it  > > or you're stuffed? > > 2 > > Will PSECT_ATTR=data,NOSHORT never be allowed? > > L > > Can we not CLUSTER/COLLECT out of the SHORT segment hand have the Linker fix ; > > up the "little" address references to real 32-bit ones?  > G > As you can see from my description above, the compiler generated code H > with explicit knowledge that the data in question (user static data orH > linker-created addresses/FDs) can be found from an offset from R1.  IfF > you tell the linker not to put that PSECT "near" R1, the code cannot > possibly reach it. > G > Given the model used, on I64 you can compile all your modules without @ > any guarantee that you can combine them into one applications. > D > Right now, I'm working on a COBOL issue where it generates addressJ > constants for code labels to use with the PERFORM statement.  I have twoI > customers who cannot link due to this.  I have prototype COBOL compiler E > read for testing by one of the customers now.  I'm off for a 3 week H > absense [2-wks vacation, 1-wk business trip].  I'll finish it when I'm	 > return.  > D > Now, for those of you who have read the Itanium ELF ABI to see theE > complete list of relocations defined, you might notice a handful of G > relocations that would let the linker fill in the 64-bit literal of a G > 'movl' (move with long literal) instruction.  Such a relocation would E > eliminate the 22-bit (4MB) short section by just not having a short H > section.  However, the compilers would have to generate different codeH > and the linker would have to implement them.  I don't believe that anyF > Itanium linker (Windows, Linux, HP-UX, OpenVMS) implement the largerH > relocations or do any compilers produce them.  It makes the code quiteF > position dependent and would require a $h!t load of image activationB > time relocations especially for shareable images built that way! > E > Is your application in COBOL?  Other than COBOL, I've yet to see an E > application that won't link on OpenVMS I64.  Of course, knowing the A > above information, it would be trivial to hand-construct (well, 7 > machine-construct) an application that wouldn't link.  >  > --  
 > John Reagan 7 > HP Pascal/{A|I}MACRO/COBOL for OpenVMS Project Leader  > Hewlett-Packard Company    ------------------------------   Date: 8 Aug 2006 02:47:34 -0700 $ From: "blackdog" <cljlk@hotmail.com> Subject: jss jobB Message-ID: <1155030454.107029.281720@75g2000cwc.googlegroups.com>  ? I have a jss job, I want to hold it for couple days not to run.  How to hold the job?   Thanks   ------------------------------  % Date: Tue, 08 Aug 2006 14:31:44 +0200 + From: Martin Vorlaender <mv@pdv-systeme.de>  Subject: Re: jss job+ Message-ID: <4jreh8F8nel1U1@individual.net>    blackdog schrieb: A > I have a jss job, I want to hold it for couple days not to run.  > How to hold the job?  ; I have no idea what a jss job is, so I may be off by miles:   , - if it's a process (shown by $SHOW SYSTEM):#    $ SET PROCESS /SUSPEND /ID=<pid>   $    Resume it by $SET PROCESS /RESUME  9 - if it's a batch job (shown by $SHOW QUEUE /BATCH/ALL) : !    SET ENTRY <entry-number> /HOLD   "    Resume it by SET ENTRY /RELEASE   HTH,	    Martin  --  D One OS to rule them all       | Martin Vorlaender  |  OpenVMS rules!7 One OS to find them           | work: mv@pdv-systeme.de H One OS to bring them all      | http://www.pdv-systeme.de/users/martinv/> And in the Darkness bind them.| home: martin@radiogaga.harz.de   ------------------------------  $ Date: Tue, 8 Aug 2006 20:47:43 +08003 From: "Richard Maher" <maher_rj@hotspamnotmail.com> 7 Subject: Official: Do not try to write UWSSs in C crap! 1 Message-ID: <eba10i$ocl$1@news-02.connect.com.au>    Hi,   J I couldn't find Hoffs poison-pen letter about calling RTLs from inner-modeK in any of the new manuals. (Do you have a refernce Steve or did it miss the ) cut?) but I did find this jiucy titbit: -    From: L http://h71000.www7.hp.com/doc/83final/4548/4548pro_012.html#itn_priv_shr_sec   Note  L ---------------------------------------------------------------------------- ----J On I64, HP recommends that you protect the entire image, rather than partsE of the image (that is, individual image segments). Partial protection J requires that you verify that all data to be protected is in the protectedL segment. Compilers for I64 put data in different types of sections. By doingD so, it becomes difficult to control protection setting. For example,F compilers put some data into short data sections. The linker then mustJ collect these sections into short data segments, which cannot be collectedK into user-defined clusters (the only clusters that you can protect with the I linker option). That is, for partially protected images, you need control H over that location that the compiler puts all your data. The compiler ofE your choice might not offer a reliable method to do so; therefore; HP ' recommends protecting the entire image. L ---------------------------------------------------------------------------- ------  L John Reagan may jump in and tell me MACRO can't be trusted to behave either, but in his absence: -   L "Use MACRO you stinking peasants! If you want to wear sandles and use C then# piss-off to Unix where you belong."    Regards Richard Maher   ; PS. Don't forget that $analyse/image on all of those images I Installed/Protected and if you see DECC$SHR whaddaya think it means? Yep. - compile-and-go just like rolling of a log :-(    ------------------------------   Date: 8 Aug 2006 07:31:25 -0700   From: "Ian Miller" <ijm@uk2.net>; Subject: Re: Official: Do not try to write UWSSs in C crap! C Message-ID: <1155047485.301735.161980@b28g2000cwb.googlegroups.com>   + Do you mean this section curiously entitled   9 "3.5.2.1 Saving System Dumps" from the V8.3 release notes   H http://h71000.www7.hp.com/doc/83final/6677/6677pro_003.html#saving_dumps     V8.3  F The following changes should be made to the paragraph in Section 31.2,= "Writing a Privileged Routine (User-Written System Service)":   F "As a protected image, your program does not have the entire operatingG system programming environment at its disposal. Unless a module has the E prefix SYS$ or EXE$, you must avoid calling it from an inner mode. In D particular, do not call LIB$GET_VM or LIB$RET_VM from an inner mode.B You can call OpenVMS RMS routines from executive mode but not from
 kernel mode."   G LIB$GET_VM should be LIB$FREE_VM. You cannot call these LIBRTL routines C directly, and you cannot call any routines that might now or in the C future call these routines indirectly. This includes other routines A within LIBRTL and the user-mode C library, among other libraries.    ------------------------------  # Date: Tue, 08 Aug 2006 16:23:29 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>; Subject: Re: Official: Do not try to write UWSSs in C crap! 1 Message-ID: <5U2Cg.1703$TG5.524@news.cpqcorp.net>   F    Watching you pound your head against the UWSS construction and its H limitations isn't particularly entertaining.  (At least I don't find it * entertaining.  Others might or might not.)  D    Writing UWSS modules isn't easy, and particularly if the calling E environment for the UWSS isn't what would be considered trustworthy.  G Sharing an address space with untrusted code isn't usually secure when  B your security-relevant data is (also) (potentially) writable from H user-mode code.   If all critical data and all critical code is stashed G behind the memory management protections and if the data transitioning  B into inner-mode can be probed and verified, life is rather better.  I    The relative difficulties here are the reason why I mentioned various  I other approaches when this discussion started, and why I usually mention  I alternatives whenever the UWSS discussion arises.  It's the reason why I  E also usually look to use pseudo device drivers or pseudo drivers and  - ACPs, or server processes (daemons) and such.   D    Is it impossible to secure one of these UWSS images?  No.  Is it  easy?  Not necessarily.       --   F    And I don't know that the full update for the UWSS references into I RTLs made it in; I'd reported that back in April or so, and didn't catch  D the implicit assumptions around the implementation of EXE$ and SYS$ E calls back then.  (The kernel versus the UWSS implementation for the  H routine; I too have created SYS$ calls that are outside the kernel, and E I'm not the only engineer around to have created these.  Usage which  : unfortunately and obviously leads to some confusion here.)   ------------------------------   Date: 8 Aug 2006 10:07:33 -0700  From: bob@instantwhip.com 7 Subject: Process/RSA help vms based bank use smartcards C Message-ID: <1155056852.986386.156410@m79g2000cwm.googlegroups.com>   9 http://www.openvms.org/stories.php?story=06/08/07/5053848    ------------------------------   Date: 8 Aug 2006 07:50:43 -0700   From: "KGB" <kgbarkes@gmail.com>J Subject: Re: Under VMS, on an HSG80, can Raid Partition Size be Increased?C Message-ID: <1155048643.520416.271970@b28g2000cwb.googlegroups.com>   D > Not meaning to give you a red-ass, but If you're not Kevin Barkes,7 > then you're not *the* KGB as far as VMS is concerned.  >  > ref:  kgb.com   F Thank you, WIlliam. Although Erik Estrada and I are still more or less4 under the radar now, it's nice to be remembered. ;-)   Regards,   KGB    -----  Kevin G. Barkes % Email: kgb@kgb.com | Web: www.kgb.com  KGB Report:  http://www.kgb.com Commentwear by KGB:  http://www.commentwear.com National Temperature Index: # http://nationaltemperatureindex.com  DCL Dialogue on line:  http://www.kgb.com/dcl.html  Random Quotations Generator: http://www.goodquotations.com " Over 12,000 searchable quotations.   ------------------------------  $ Date: Tue, 8 Aug 2006 09:35:49 -0400C From: "David Turner, Island Computers US Corp" <dbturner@icusc.com> & Subject: Updating firmware on DEGPA-TA8 Message-ID: <hq0Cg.7146$dQ4.1578@bignews1.bellsouth.net>  K Anyone know how to do the firmware update (or at least anyone know the file + name and where to find it) for the DEGPA-TA  Old style Gigabit card?    David    --     David B Turner Island Computers US Corp 2700 Gregory St, Suite 180 Savannah GA 31404  Tel: 912 447 6622 X201 Cell: 912 447 6622 X251  Fax: 912 201 0402  Email: dbturner@islandco.com Web: http://www.islandco.com% ===================================== < All orders are subject to the following terms and conditions. of sale. These should be read before ordering.% http://www.islandco.com/warranty.html    ------------------------------  % Date: Tue, 08 Aug 2006 11:12:45 +0200 / From: Paul Sture <paul.sture.nospam@hispeed.ch> ' Subject: Re: USB numeric keypad anyone? ; Message-ID: <10aa3$44d8558e$50db5015$26520@news.hispeed.ch>     VAXman- @SendSpamHere.ORG wrote:b > In article <KwHBg.1632$tA4.256@news.cpqcorp.net>, Hoff Hoffman <hoff-remove-this@hp.com> writes: >> >> Ken Fairfield wrote:  >>F >>>    So Hoff, do you have a terminal emulator that allows you to useA >>> all of the LK463 keys within EVE (under Windows XP or newer)? D >>   I've been using the (available on the OpenVMS Freeware) VTstar L >> terminal emulator package for some eons now.  I haven't tried it with an  > ? > Can you relay this to the person that provided this package:  ! > How about a Mac OS X version?    >   	 Seconded.    --  
 Paul Sture   ------------------------------  # Date: Tue, 08 Aug 2006 12:52:11 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>Y Subject: Re: V8.3 Apache and Javascript in HTML files? (was Re: VMS 7.3 vs 8.x - java in  2 Message-ID: <%N%Bg.1687$BA5.1242@news.cpqcorp.net>   patrick jankowiak wrote:  K > After FTP-ing some hundreds of HTML files over, I noticed that once they  K > got on the system live, I could use any web browser and view source, and  H > one short line of javascript had been inserted in each HTML file near I > the beginning. I verified this by FTP-ing the files back to myself and  , > looking at them, sure enough it was there. ....J > Was VMS 8.3 set up wrong with some default I don't know about or should H > it be playing with my HTML files? Did Apache do this? Has anyone else  > seen this?  F    OpenVMS itself doesn't have this HTML Javascript insertion option, D barring manually running a perl script, DCL procedure or other such  against the HTML files.   G    Some more details, please?   What was the Javascript?  Which Apache?   I    Does the Javascript appear if you TYPE the files on OpenVMS?  And was  : the line in the source for the file; in the original copy?  B    Was any HTML editor used?  (In my experience, Adobe Macromedia H Dreamweaver doesn't insert HTML into the file without explicit request, B but I've used other HTML packages which gratuitously reformat and - otherwise insert changes into the HTML file.)   A    I've certainly seen various Microsoft Windows packages insert  B Javascript into the data on the fly, usually as part of a Windows = security package.  One of the Norton packages was using this.   F    It's not typical that Apache inserts anything into its HTML, but I F expect its possible.  (Apache could probably make you your coffee, if B you connected your hardware and loaded the right options into the C .htbeverage file; remember to enable the RSS feed into the brewer.)    ------------------------------  $ Date: Tue, 8 Aug 2006 06:40:14 -0400) From: "Neil Rieck" <n.rieck@sympatico.ca> / Subject: Re: VMS backup: competition from Apple < Message-ID: <44d868fc$0$24210$9a6e19ea@news.newshosting.com>  : "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message& news:44D823BB.8EE02AD7@teksavvy.com...J > Apple is having its litte party in San Francisco. Announced for its nextC > version of the OS: a new backup technology called "Time Machine".  >  [...snip...] > B > And while HP apologists may still be supporting that IA64 thing,B > consider how quickly Apple has transitioned to the 8086. Its newF > machine, the MacPro completes its transition in some 220 days. (lessA > than a year for full product revamp of their servers, desktops,  > laptops).  > I If memory serves, last year Steve Jobs admitted Apple had been running an K Intel "skunk works" for more than 5 years. So the time-frames you reference  may not mean what you think.  K > It uses the new Intel Duo Core, and supports 64 bit computing. MacOS will ; > allow 64 bit and 32 bit applications to run concurrently.  > K We all know about Moore's law, and we all know the chip makers have changed I course, so does anyone think that desktops won't have 32 cores by 2012 or  2014?    2006    | 2 cores  2007.5 | 4 cores 2009    | 8 cores  2010.5 | 16 cores  2012    | 32 cores  D HP claims they need IA-64 because they wanted to run big 32 + 64 CPUH machines. Desktop core numbers will go up and new features will be addedK which makes me wonder how much new-feature development will take place on a  lower volume product line?  
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.8 http://www3.sympatico.ca/n.rieck/links/cool_openvms.html9 http://www3.sympatico.ca/n.rieck/links/openvms_demos.html    ------------------------------  $ Date: Tue, 8 Aug 2006 06:43:45 -0400) From: "Neil Rieck" <n.rieck@sympatico.ca> / Subject: Re: VMS backup: competition from Apple < Message-ID: <44d869cf$0$24175$9a6e19ea@news.newshosting.com>  ; "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message  & news:44D823BB.8EE02AD7@teksavvy.com...J > Apple is having its litte party in San Francisco. Announced for its nextC > version of the OS: a new backup technology called "Time Machine".  >  [...snip...] >  > B > And while HP apologists may still be supporting that IA64 thing,B > consider how quickly Apple has transitioned to the 8086. Its newF > machine, the MacPro completes its transition in some 220 days. (lessA > than a year for full product revamp of their servers, desktops, J > laptops). It uses the new intel Duo Core, and supports 64 bit computing.F > MacOS will allow 64 bit and 32 bit applications to run concurrently. > I What I find more interesting is Apple's adoption of Open Source software. 5 http://www.theinquirer.net/default.aspx?article=33533   
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.8 http://www3.sympatico.ca/n.rieck/links/cool_openvms.html9 http://www3.sympatico.ca/n.rieck/links/openvms_demos.html    ------------------------------  * Date: Tue, 8 Aug 2006 12:42:21 +0000 (UTC) From: david20@alpha2.mdx.ac.uk/ Subject: Re: VMS backup: competition from Apple , Message-ID: <eba0rd$66h$1@south.jnrs.ja.net>  h In article <44d868fc$0$24210$9a6e19ea@news.newshosting.com>, "Neil Rieck" <n.rieck@sympatico.ca> writes:; >"JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message ' >news:44D823BB.8EE02AD7@teksavvy.com... K >> Apple is having its litte party in San Francisco. Announced for its next D >> version of the OS: a new backup technology called "Time Machine". >>
 >[...snip...]  >>C >> And while HP apologists may still be supporting that IA64 thing, C >> consider how quickly Apple has transitioned to the 8086. Its new G >> machine, the MacPro completes its transition in some 220 days. (less B >> than a year for full product revamp of their servers, desktops, >> laptops). >>J >If memory serves, last year Steve Jobs admitted Apple had been running anL >Intel "skunk works" for more than 5 years. So the time-frames you reference >may not mean what you think.  > L >> It uses the new Intel Duo Core, and supports 64 bit computing. MacOS will< >> allow 64 bit and 32 bit applications to run concurrently. >>L >We all know about Moore's law, and we all know the chip makers have changedJ >course, so does anyone think that desktops won't have 32 cores by 2012 or >2014? >  >2006    | 2 cores >2007.5 | 4 cores  >2009    | 8 cores >2010.5 | 16 cores >2012    | 32 cores  >   ) What will a desktop PC do with 32 cores ? 0 What 32 things will it be running concurrently ?  
 David Webb Security team leader CCSS Middlesex University      E >HP claims they need IA-64 because they wanted to run big 32 + 64 CPU I >machines. Desktop core numbers will go up and new features will be added L >which makes me wonder how much new-feature development will take place on a >lower volume product line?  >  >Neil Rieck  >Kitchener/Waterloo/Cambridge, >Ontario, Canada. 9 >http://www3.sympatico.ca/n.rieck/links/cool_openvms.html : >http://www3.sympatico.ca/n.rieck/links/openvms_demos.html >  >  >  >    ------------------------------  # Date: Tue, 08 Aug 2006 14:17:32 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>/ Subject: Re: VMS backup: competition from Apple 1 Message-ID: <021Cg.1692$Wn5.855@news.cpqcorp.net>    david20@alpha2.mdx.ac.uk wrote:   + > What will a desktop PC do with 32 cores ? 2 > What 32 things will it be running concurrently ?  I    The windowing system, the anti-malware software, IPv6, encryption, HD  I codecs, the great grandson of the CSS content protection scheme, the the  G WinFTTP link (a really dumb FTTP card), IEEE 802.11zzzz galactic WLAN,  D the 10Gb SAN into the petabyte storage farm, the fully-adaptive and A self-patching operating system (which obviously uses those other  H processors to stage the updates and to dynamically trans-bootstrap into = the new environment), a fully-parallelized office suite with  G cranial-scan I/O; I'm sure we can find a few processes and a few tasks   to fill the available cycles.    ------------------------------   Date: 8 Aug 2006 08:33:54 -0700 ' From: "toby" <toby@telegraphics.com.au> / Subject: Re: VMS backup: competition from Apple C Message-ID: <1155051234.916037.226030@m79g2000cwm.googlegroups.com>    Neil Rieck wrote: < > "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message( > news:44D823BB.8EE02AD7@teksavvy.com...L > > Apple is having its litte party in San Francisco. Announced for its nextE > > version of the OS: a new backup technology called "Time Machine".  > >  > [...snip...] > > D > > And while HP apologists may still be supporting that IA64 thing,D > > consider how quickly Apple has transitioned to the 8086. Its newH > > machine, the MacPro completes its transition in some 220 days. (less  G Well, they've done more difficult and more dramatic transitions before,  and equally smoothly.   C > > than a year for full product revamp of their servers, desktops, 
 > > laptops).  > > K > If memory serves, last year Steve Jobs admitted Apple had been running an M > Intel "skunk works" for more than 5 years. So the time-frames you reference  > may not mean what you think.  G This was an obvious deduction long before the Intel line was announced.    > M > > It uses the new Intel Duo Core, and supports 64 bit computing. MacOS will = > > allow 64 bit and 32 bit applications to run concurrently.  > > M > We all know about Moore's law, and we all know the chip makers have changed K > course, so does anyone think that desktops won't have 32 cores by 2012 or  > 2014? ...   A Sun leads this area. IMHO Apple should have partnered with Sun or  Transmeta and not Intel. :)    --T    > F > HP claims they need IA-64 because they wanted to run big 32 + 64 CPUJ > machines. Desktop core numbers will go up and new features will be addedM > which makes me wonder how much new-feature development will take place on a  > lower volume product line? >  > Neil Rieck > Kitchener/Waterloo/Cambridge,  > Ontario, Canada.: > http://www3.sympatico.ca/n.rieck/links/cool_openvms.html; > http://www3.sympatico.ca/n.rieck/links/openvms_demos.html    ------------------------------  % Date: Tue, 08 Aug 2006 08:56:22 -0700 4 From: Alan Frisbie <Usenet02_REMOVE@Flying-Disk.com>+ Subject: Re: XP1000 USB port working in VMS % Message-ID: <1155052484.275099@smirk>    Alan Frisbie wrote: < > Forrest Kenney has asked me to announce that some fixes he= > made for some other USB devices now allows the built-in USB ? > port on the XP1000 to work.   You will need to craft an entry  > for it in sys$use_config.dat.  > = > He didn't say explicitly, but I assume this code will be in ' > VMS v8.3 or in a soon-to-come update.    Two corrections:  0    That should be sys$system:sys$user_config.dat  "    The code is in F8.3 and beyond.  = Also, as I discovered with my XP1000, there is an environment = variable, usb_enable, that must be set to ON (mine was set to < OFF.   If it is set to OFF, SHOW CONFIGURATION will show the! Embedded USB port as not enabled.   6 I now have a working USB rocket launcher protecting my VMS system.   :-)    Alan   ------------------------------   End of INFO-VAX 2006.439 ************************