1 INFO-VAX	Mon, 10 Apr 2006	Volume 2006 : Issue 199       Contents:- Re: Anybody got the kit for BASIC and COBOL ? - Re: Anybody got the kit for BASIC and COBOL ?  DECnet IV routing errors Re: DECnet IV routing errors. Re: Getting the remote address for SET DISPLAY( How does a fix become a published patch?, Re: How does a fix become a published patch?, Re: How does a fix become a published patch?, Re: How does a fix become a published patch?' Re: OpenVMS is the worlds best desktop!  Re: OpenVMS on zx2000 or zx6000 % Re: possible to avoid a shadow merge? * Re: Quorum, locks and application question* Re: Quorum, locks and application question. Re: SMTP: stop sending "no such user" messages. Re: SYS$COMMON:[SYSEXE]TCPIP$SMTP_RECV_RUN.COM. Re: SYS$COMMON:[SYSEXE]TCPIP$SMTP_RECV_RUN.COM Re: TCPIP$SMTP_COMMON  Re: TCPIP$SMTP_POSTMASTER_ALIAS  Re: TCPIP$SMTP_POSTMASTER_ALIAS G Re: The Minimum You Need to Know to Be an OpenVMS Application Developer ! Re: Using AST completion routines  Re: Using the Serial Terminal   F ----------------------------------------------------------------------  # Date: Mon, 10 Apr 2006 15:07:39 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>6 Subject: Re: Anybody got the kit for BASIC and COBOL ?1 Message-ID: <%wu_f.6093$y04.669@news.cpqcorp.net>    Schnootling wrote:  I > Well, maybe it isn't available. But if it isn't available, then why the G > heck is it part of the layered products list we get from HP ?  Why do I > we get a big, long list of products if we really don't get some of them  > ?   F    The products aren't available on the hobbyist distro, due to space H constraints on the disks, and due to an attempt to sort out and include A the products that are the most interesting to the most hobbyists.   I    A full distro of all OpenVMS layered products has been hanging in the  G range of a dozen CD disks, or more.  OpenVMS itself is now occupying a  G half-dozen disks or so for a given architecture, between the docs, the  K OpenVMS distro, the core LPs disk, the Open Source Tools, and the Freeware.   G    Y'all might get a little cranky if the hobbyist distro were on DVD,  I too -- IDE/ATAPI is a comparatively recent addition in the Alpha product  G line (Personal Workstation series, and later), and as SCSI DVD readers  I are rather rare devices.  And we're over-filling a single DVD, of course.   E    I've not seen an HP official statement on acquiring these product  F kits; on how hobbyists are intended to gain access to the kits not in 7 the hobbyist distro.  (Not that I can recall, that is.)    ------------------------------    Date: 10 Apr 2006 09:28:04 -0700; From: "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> 6 Subject: Re: Anybody got the kit for BASIC and COBOL ?C Message-ID: <1144686484.766330.272950@v46g2000cwv.googlegroups.com>    Chuck wrote: > Hello Peter, Dave, and Ian,  > E > I just updated my Google profile with information about my location  > within Washington State. >  > My mailing/living address is > 1008 Central, S.E. > Olympia, Washington 98501  > H > I tried the http://vmsone.com link, but the connection timed out the 3& > separate times I tried to access it. > ? > I very much to appreciate y'all taking the time to assist me,  > Chuck   E vmsone.com is registered to the late, great John Wisniewski.  I don't 3 know who would be taking care of it now, if anyone.    ------------------------------  % Date: Mon, 10 Apr 2006 10:56:34 -0400 % From: "Chris Moore" <no.one@no.where> ! Subject: DECnet IV routing errors 9 Message-ID: <jmu_f.1610$%U2.137568@news20.bellglobal.com>   L Yesterday morning it looks like a local network problem was encountered that- had the DECnet-IV nodes in a serious problem.   K This has probably been asked before, but I'm just wondering what exactly it A means when a Phase IV routing node loses its adjacency to itself?     4 %%%%%%%%%%% OPCOM 9-APR-2006 00:59:48.24 %%%%%%%%%%%! Message from user DECNET on KOJAK ! DECnet event 4.18, adjacency down . From node 2.59 (KOJAK), 9-APR-2006 00:59:48.22E Circuit ISA-0, Dropped by adjacent node, Adjacent node = 2.59 (KOJAK)   4 %%%%%%%%%%% OPCOM 9-APR-2006 00:59:48.24 %%%%%%%%%%%! Message from user DECNET on KOJAK  DECnet event 4.15, adjacency up . From node 2.59 (KOJAK), 9-APR-2006 00:59:48.23+ Circuit ISA-0, Adjacent node = 2.59 (KOJAK)    Thanks Chris    ------------------------------  # Date: Mon, 10 Apr 2006 16:48:50 GMT A From: "Colin Butcher" <colin_DOT.butcher_AT@xdelta_DOT.co_DOT.uk> % Subject: Re: DECnet IV routing errors = Message-ID: <S%v_f.50963$wl.17892@text.news.blueyonder.co.uk>   L An adjacency down/up is usually because the hello packets stopped being seemJ for long enough to trigger an adjacency down event, then the hello packetsI started to be seen again. How many physical connections does this machine L have to the network? There should only be one connection to each LAN segmentG and there should only be routing between the LAN segments, no bridging, G otherwise you'll have a duplicate MAC address if you're using Phase IV. @ Which nodes are DECnet routers? Some kind of diagram might help.   --     Hope this helps, Colin. ) colin DOT butcher AT xdelta DOT co DOT uk E It's not mine, but I like this definition: Legacy = stuff that works. 0 "Chris Moore" <no.one@no.where> wrote in message3 news:jmu_f.1610$%U2.137568@news20.bellglobal.com... I > Yesterday morning it looks like a local network problem was encountered  that/ > had the DECnet-IV nodes in a serious problem.  > J > This has probably been asked before, but I'm just wondering what exactly itC > means when a Phase IV routing node loses its adjacency to itself?  >  > 6 > %%%%%%%%%%% OPCOM 9-APR-2006 00:59:48.24 %%%%%%%%%%%# > Message from user DECNET on KOJAK # > DECnet event 4.18, adjacency down 0 > From node 2.59 (KOJAK), 9-APR-2006 00:59:48.22G > Circuit ISA-0, Dropped by adjacent node, Adjacent node = 2.59 (KOJAK)  > 6 > %%%%%%%%%%% OPCOM 9-APR-2006 00:59:48.24 %%%%%%%%%%%# > Message from user DECNET on KOJAK ! > DECnet event 4.15, adjacency up 0 > From node 2.59 (KOJAK), 9-APR-2006 00:59:48.23- > Circuit ISA-0, Adjacent node = 2.59 (KOJAK)  >  > Thanks > Chris  >    ------------------------------  # Date: Mon, 10 Apr 2006 15:39:45 GMT F From: lederman@star.enet.dec.DISABLE-JUNK-EMAIL.com (Bart Z. Lederman)7 Subject: Re: Getting the remote address for SET DISPLAY . Message-ID: <5%u_f.39$zb2.121@news.oracle.com>  / The following works for me on V7.3-2 and V8.2-1   , $ remnode = F$TRNLNM ("SYS$REM_NODE") - "::"L $ SETDISP     :== "SET DISPLAY /CREATE /NODE = " 'remnode /TRANSPORT = tcpip  < In this environment, I happen to know that the connection is always going to be TCPIP.    Bart.    ------------------------------  % Date: Mon, 10 Apr 2006 07:40:49 -0400 + From: Steve Matzura <number6@speakeasy.net> 1 Subject: How does a fix become a published patch? 8 Message-ID: <sifk32dghgml5t6h03r8fe8rqd22pl91p8@4ax.com>  > Four-and-a-half years ago, my employer, through me and anotherB coworker, reported a problem to HP regarding the creation of imageD savesets of system disks on volumes starting at and larger than 9gb.= It took them nearly four years to figure it out, with about a A year-and-a-half of no action mostly due to us not pushing it, but A eventually the problem was found.  Seems there was a problem with B putting VMB.EXE out further than LBN 2^24-1 (= 16,777,215) becauseC BACKUP wasn't using all 32 bits (not to mention a quadword) to mark ; this location.  Anyway, the problem was determined and some   replacement images issued to us.  F Shortly after we reported success with the replacement images, I fullyE expected to see them incorporated into a patch, and then to have that A patch included in a mandatory service update. No such luck. Was I D supposed to file some special paperwork with HP to make this happen?E I remember in the old days, you could file an SPR and either see your @ problem show up in a patch kit or at least an article in the VAXC Professional soon thereafter.  I figure, if this problem plagued us F for as long as it did, others must be experiencing it, too, and should  be told there *is* a fix for it.   ------------------------------    Date: 10 Apr 2006 08:13:22 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) 5 Subject: Re: How does a fix become a published patch? 3 Message-ID: <DEIEgucjA5hV@eisner.encompasserve.org>   f In article <sifk32dghgml5t6h03r8fe8rqd22pl91p8@4ax.com>, Steve Matzura <number6@speakeasy.net> writes:@ > Four-and-a-half years ago, my employer, through me and anotherD > coworker, reported a problem to HP regarding the creation of imageF > savesets of system disks on volumes starting at and larger than 9gb.? > It took them nearly four years to figure it out, with about a C > year-and-a-half of no action mostly due to us not pushing it, but C > eventually the problem was found.  Seems there was a problem with D > putting VMB.EXE out further than LBN 2^24-1 (= 16,777,215) becauseE > BACKUP wasn't using all 32 bits (not to mention a quadword) to mark = > this location.  Anyway, the problem was determined and some " > replacement images issued to us. > H > Shortly after we reported success with the replacement images, I fullyG > expected to see them incorporated into a patch, and then to have that C > patch included in a mandatory service update. No such luck. Was I F > supposed to file some special paperwork with HP to make this happen?  @ You don't mention VMS versions, but my presumption would be thatA newer versions of Backup were fixed and older no-longer-supported  versions did not get the fix.   G > I remember in the old days, you could file an SPR and either see your B > problem show up in a patch kit or at least an article in the VAXE > Professional soon thereafter.  I figure, if this problem plagued us H > for as long as it did, others must be experiencing it, too, and should" > be told there *is* a fix for it.  ) The four and a half years may be the key.    ------------------------------  # Date: Mon, 10 Apr 2006 14:28:36 GMT # From: Beach Runner <bob@nospam.com> 5 Subject: Re: How does a fix become a published patch? < Message-ID: <oYt_f.141425$g47.63496@tornado.tampabay.rr.com>   Larry Kilgallen wrote:  h > In article <sifk32dghgml5t6h03r8fe8rqd22pl91p8@4ax.com>, Steve Matzura <number6@speakeasy.net> writes: > @ >>Four-and-a-half years ago, my employer, through me and anotherD >>coworker, reported a problem to HP regarding the creation of imageF >>savesets of system disks on volumes starting at and larger than 9gb.? >>It took them nearly four years to figure it out, with about a C >>year-and-a-half of no action mostly due to us not pushing it, but C >>eventually the problem was found.  Seems there was a problem with D >>putting VMB.EXE out further than LBN 2^24-1 (= 16,777,215) becauseE >>BACKUP wasn't using all 32 bits (not to mention a quadword) to mark = >>this location.  Anyway, the problem was determined and some " >>replacement images issued to us. >>H >>Shortly after we reported success with the replacement images, I fullyG >>expected to see them incorporated into a patch, and then to have that C >>patch included in a mandatory service update. No such luck. Was I F >>supposed to file some special paperwork with HP to make this happen? >  > B > You don't mention VMS versions, but my presumption would be thatC > newer versions of Backup were fixed and older no-longer-supported  > versions did not get the fix.  >  > G >>I remember in the old days, you could file an SPR and either see your B >>problem show up in a patch kit or at least an article in the VAXE >>Professional soon thereafter.  I figure, if this problem plagued us H >>for as long as it did, others must be experiencing it, too, and should" >>be told there *is* a fix for it. >  > + > The four and a half years may be the key. ) How do you know the problem still exists?    ------------------------------  # Date: Mon, 10 Apr 2006 16:25:53 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>5 Subject: Re: How does a fix become a published patch? 1 Message-ID: <lGv_f.6098$054.813@news.cpqcorp.net>    Larry Kilgallen wrote:h > In article <sifk32dghgml5t6h03r8fe8rqd22pl91p8@4ax.com>, Steve Matzura <number6@speakeasy.net> writes:A >> Four-and-a-half years ago, my employer, through me and another E >> coworker, reported a problem to HP regarding the creation of image G >> savesets of system disks on volumes starting at and larger than 9gb.       re: VMB.EXE above and BACKUP   B    As problem reports arrive into engineering, they are generally G prioritized and reviewed.   Key here are considerations like supported  C configurations, and simple things like the volume of reports for a  H particular problem, and the availability of a work-around.  Isolated or B comparatively rare problems (and those that are difficult to work F around) will typically receive a point fix, where critical or core or B widespread problems will variously see a side-built image for the F immediate customer(s) involved, and will then end up in the queue for + the next UPDATE or similar ECO kit release.   I    As each problem is worked on, the source code changes are logged into  H the next OpenVMS release, and the change is also at least built against E the original OpenVMS release stream, and is then possibly integrated  E into the remedial update stream associated with the original release.   D    If you are planning on attending the bootcamp, I am presenting a @ session on how the source code control system is configured and 9 operates, and it touches on the remedial release streams.   H    In this particular case, BACKUP was looking for a pattern that dates I back to the PDP-11/RSX-11 days as part of determining whether or not the  H disk was bootable, and it just doesn't hold on larger disks.  (I'm also D not particularly certain if there are disks officially supported by I OpenVMS VAX that are this large.)  And in this case, the work-around (if  I you don't have the fixed BACKUP image) is to restore the disk as normal,  H and then issue a WRITEBOOT operation.  You can also move the image down / below the threshold, and re-WRITEBOOT, as well.   H    The underlying OpenVMS volume structure (across VAX, Alpha and IA-64 H architectures) is limited to a longword block address, so you won't see C disks larger than 1 TB or (if and as the current sign bit "fun" is  I resolved) 2 TB on OpenVMS any time soon -- I'm not aware of plans to add  C quadword offsets within BACKUP and into all the other parts of the  D operating system and the layered products and the applications that  would be involved here, too.  E    BACKUP on OpenVMS I64 uses an existing (and undocumented) API for  F probing and for managing the boot blocks, and that API does deal with D larger offsets.  Now as to how BACKUP stores that detail within the D saveset, I don't know.  This API didn't make it back to OpenVMS VAX.  E    To find out if you have one of these high-located VMB.EXE images,  F issue a DUMP/HEADER/BLOCK=END=0 of SYS$SYSTEM:VMB.EXE, and see if the G Map Area LBN retrieval pointer has any bits set in the highest byte of  H the LBN longword.  If the bits are set, then you either need the BACKUP F image update, the WRITEBOOT workaround, or you need to reallocate the B image (and contiguously, /NOMOVE) in lower-addressed disk storage.   ------------------------------    Date: 10 Apr 2006 02:36:41 -0700- From: "Andrew" <andrew_harrison@symantec.com> 0 Subject: Re: OpenVMS is the worlds best desktop!C Message-ID: <1144661801.399179.157820@i39g2000cwa.googlegroups.com>    Bill Gunshannon wrote:E > In article <1144367292.054966.137130@e56g2000cwe.googlegroups.com>,  > 	paco.linux@gmail.com writes:  > > How are the Solaris, > C > Sun makes money selling hardware.  They are not just a low margin B > Wintel box company.  Solaris just went free so we haven't reallyE > had a chance to see the effect this will have on either the company  > or the OS yet. >   G Sun has never charged for Solaris on SPARC. The change is that they did C have a paid for licensing model for Solaris x86 for commerical use. E That has now changed so that you can choose to pay for support but do  not have to pay for the OS.   G This is a similar model to the RedHat subscription service type pricing E except that Solaris x86 is available regardless of whether you have a @ support contract  unlikle RedHat and Solaris support is cheaper.  G It remains to be seen if Sun can make money on Solaris x86, however you B have to remember that Solaris x86 comes form the same code base asE Solaris SPARC which Sun have to do for their SPARC servers. So we are A talking about covering the incremental cost of supporting the x86 , platform not the cost supporting a whole OS.  G Its pretty clear that Sun want to capture developers, students etc onto F Solaris make it easy for them to learn and develop new products on theE platform and then charge them to support the platform when it is used  to deploy commercial apps.  C >From Sun's perspective if the deployment platform is SPARC/Solaris C rather than x86/Solaris then even better assuming that x86 does not  mean Sun's x86 boxes.    regards  Andrew Harrison    ------------------------------  # Date: Mon, 10 Apr 2006 11:01:01 GMT 5 From: rdeininger@mindspringdot.com (Robert Deininger) ( Subject: Re: OpenVMS on zx2000 or zx6000[ Message-ID: <rdeininger-1004060701030001@dialup-4.233.173.169.dial1.manchester1.level3.net>   I In article <1144637521.533825.142550@i40g2000cwc.googlegroups.com>, "Rich # Jordan" <jordan@ccs4vms.com> wrote:   H >Probably because itanium workstations went the way of itanium desktops,F >itanium laptops, and dodo-birds... servers don't need fancy graphics.  E Pretty much correct.  This is the only real impact I have seen of the A decision to stop Itanium work in HP's workstation group.  The AGP  backplane went away.  I The rx2600 family started in the workstation group, and the AGP backplane F was developed specifically for graphics support.  When the workstationI group decided to focus exclusively on non-Itanium systems, the rx2600 was H transferred to the entry-level systems group, which developed the rx2620 product.  9 Leaving the AGP backplane behind was partly the result of H misunderstanding, I believe.  It seemed to take the HP-UX folks almost aG year to realize they'd left behind their only high-performance graphics I platform for HP-UX.  The eventually noticed the Radeon 7500 card that VMS H already supported on rx2620 (and the rest of the entry-level family) andE added HP-UX support.  But the AGP backplane was not added at the same > time; I think it was fairly unknown to the small server folks.  I That AGP backplane was due for a re-design in any case.  I don't think it F could take advantage of the higher bus speeds in current systems.  ButD when rx2620 was  in the pipeline, AGP was already pretty much over. A Graphics hardware was already moving to PCI-Express.  Further AGP I development work at that time -- and the work would have been significant # -- was viewed as a poor investment.    ------------------------------  # Date: Mon, 10 Apr 2006 14:25:29 GMT # From: Beach Runner <bob@nospam.com> . Subject: Re: possible to avoid a shadow merge?< Message-ID: <tVt_f.141389$g47.11291@tornado.tampabay.rr.com>   Michael Moroney wrote:  T > helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: >  > G >>Under what conditions will shutting down and rebooting the satellite  J >>cause a merge on this disk (or any other disks it can't dismount due to J >>open files (assuming the applications with the open files can't be shut + >>down prior to shutting down the system)).  >  > I > If I remember, it was a bug that when a certain file (either SYSUAF or  L > RIGHTSLIST) resided on a disk other than the system disk, that disk wasn'tK > dismounted properly on shutdown, and you'd get shadow merges if it was a   > shadowset. > B > I don't remember when it was fixed or even if it was ever fixed.   It was the pagefile    ------------------------------  # Date: Mon, 10 Apr 2006 14:58:46 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>3 Subject: Re: Quorum, locks and application question 2 Message-ID: <Gou_f.6091$n04.4368@news.cpqcorp.net>   JF Mezei wrote: G > Say I have an application which runs on every node of the cluster and H > each instance uses locks to assert itself and tell the other instances( > about its ability to service requests. > B > Does the application need to worry about quorum issues at all ?   D    Your application need only particularly concern itself with your D so-called "quorum issues" if active processing during a quorum wait I state is required.  That's semi-feasible for some contexts, but somewhat  F tricky as the code involved has to be running at moderately-high IPL. F If your application can simply stall for the the duration of the wait F state (as is the default behaviour for applications), then the simple   answer to your question is "no".  H    When a node or a lobe exits, clustering will clean up any locks left * dangling, which is the simple answer here.  E    Do realize that you can potentially lose the contents of the lock  G value block(s) involved if the copy of that data happened to be on the  H failed node or in the failed lobe, though you'll get a return status to I indicate that the values are now untrusted.  (As is usual with a failure  E condition, if the data is not written out to disk or to other static  I storage, then the best case result is that the memory-cached data ceases  G to exist.  The worst case, of course, is if some part of the data here  C actually does exist in static storage.  Then you've got "skewage".)   H    Cluster fail-over questions aren't easy ones to answer, particularly  the open-ended ones.   ------------------------------  % Date: Mon, 10 Apr 2006 13:28:50 -0400 / From: "William Webb" <william.w.webb@gmail.com> 3 Subject: Re: Quorum, locks and application question I Message-ID: <8660a3a10604101028t24bb6fdarfbfef6738812c208@mail.gmail.com>   = On 4/8/06, Richard Maher <maher_rj@hotspamnotmail.com> wrote:  > Hi JF, > H > > If node A is disconnected, is it correct to assume that applicationsH > > running on the other nodes will immediatly see the loss of the locks( > > hel[d] by their peer that ran on A ? > K > I'd say that if the applications on other nodes had lock requests pending L > for the resources held by their peers on Node A then the short answer is = yes L > depending on what you mean by "immediately". (And they should revalidate = the  > Value Block if necessary)  >  > Consider this conundrum: - > I > Q. How does a Rdb Monitor process faithfully implement it's FREEZE lock K > mechanism so as to protect the ACID properties of a database transaction?  > J > A. Stuffed if I know. (There was a discussion in ITRC about this a while > back)  > L > 1) Process A on Node X has entered a Cayman Island transaction for $1B an= d H > will have to rollback 'cos a commit-time constraint will stop it going
 > through.  & Would that be one billion US$ or CI$ ?   ; - )    WWWebb   ------------------------------  + Date: Mon, 10 Apr 2006 14:53:41 +0000 (UTC)  From: david20@alpha2.mdx.ac.uk7 Subject: Re: SMTP: stop sending "no such user" messages ( Message-ID: <e1drhl$l6$1@news.mdx.ac.uk>  \ In article <443980D7.662D8A65@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes:D >> > Since many names can point to one address, but one address onlyM >> > translates back to one name, could it be legitimate that the translation & >> > of an address doesn't match HELO? >  >Yes.  > 7 >All you, as a receiver, can do is just make sure that:  >  >(1) >HELO smtp.chocolate.com   >can translate to *an* IP  > 	 >and that  >(2)G >the IP address from which the call is coming can be reverse translated  >to an IP address. >  >  > H >For (1), you canot even "validate" the IP address. Consider a large ISPF >with 10 SMTP servers. They may all use "HELO SMTP.CHOCOLATE.COM"  andI >you can translatre SMTP.CHOCOLATE.COM to *an* IP but it doesn't garantee I >that it will translate to the same IP that is being used. (for instance, I >it may tranmslate to the first server, but currently, you call is coming  >from the 3rd server). >   K Note that checking the HELO argument is about the only check that the RFCs  7 explicitly say cannot be used to reject a mail message.   
 From RFC 1123    " <  The HELO receiver MAY verify that the HELO parameter reallyB  corresponds to the IP address of the sender. However the receiverM  MUST NOT refuse to accept a message, even if the sender's HELO command fails   verification. .  .  . J  Note also that the HELO argument is still required to have valid <domain>O  syntax, since it will appear in a Received: line; otherwise, a 501 error is to 	  be sent.    "   
 David Webb Security team leader CCSS Middlesex University     >For (2) > H >All this check does is ensure that the call is coming from an IP adressH >that is serious enough to have a reverse translation. You cannot reallyG >validate it. You are simply eliminating calls coming from IP addresses ; >that aren't serious enough to have a reverse translation.   > E >Consider a canadian bank that has outsourced its corporate email toa C >telco. You'd be getting an email from a user@bank.ca coming from a L >server smtp.telus.ca. But this is perfectly valid so you have to accept it.   ------------------------------    Date: 10 Apr 2006 00:19:41 -0700 From: Bart.Zorn@gmail.com 7 Subject: Re: SYS$COMMON:[SYSEXE]TCPIP$SMTP_RECV_RUN.COM C Message-ID: <1144653581.378628.299620@i40g2000cwc.googlegroups.com>   C Indeed, and they do it in every and all tcpip*.com files. They even D found the "case_sensitive" argument in F$TRNLNM. I doubt if they can explain why they use it.  F During the last OpenVMS TUD in the Netherlands, Guy Peleg told me thatF he is now also working with the TCPIP team. Hopefully that will result+ in some quality improvements in this field!   	 Bart Zorn    ------------------------------  % Date: Mon, 10 Apr 2006 03:39:32 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 7 Subject: Re: SYS$COMMON:[SYSEXE]TCPIP$SMTP_RECV_RUN.COM , Message-ID: <443A0B9A.C5321ECC@teksavvy.com>   Bart.Zorn@gmail.com wrote:H > During the last OpenVMS TUD in the Netherlands, Guy Peleg told me that- > he is now also working with the TCPIP team.   B Is there a TCPIP Team left ? Or just a few consultants left strewn around the world ?  C Am wondering if it woud be wiser for HP to just pay Process to take E TCPIP Services over and merge it into Multinet. In the end, isn't VMS $ going to end up at Process anyways ?   ------------------------------    Date: 10 Apr 2006 06:43:18 -0700 From: Bart.Zorn@gmail.com  Subject: Re: TCPIP$SMTP_COMMONB Message-ID: <1144676598.791784.56140@i40g2000cwc.googlegroups.com>  G I have experimented with this logical (I am not sure which version, but E most likely V7.3-2 with V5.4, ECO2 or something). The result was then 3 that if I defined TCPIP$SMTP_COMMON to anything but B SYS$SPECIFIC:[TCPIP$SMTP], SMTP just refused to start, without anyD sensible error message (I seem to remember $STATUS was 1). I gave up" soon. It was not important enough.  	 Bart Zorn    ------------------------------  + Date: Mon, 10 Apr 2006 13:06:53 +0000 (UTC)  From: david20@alpha2.mdx.ac.uk( Subject: Re: TCPIP$SMTP_POSTMASTER_ALIAS) Message-ID: <e1dl9d$rrp$1@news.mdx.ac.uk>   \ In article <4437FB7A.C87134B4@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes:0 >Phillip Helbig---remove CLOTHES to reply wrote: >>  G >> Does defining TCPIP$SMTP_POSTMASTER_ALIAS have any effect other than < >> setting the From: address of things like bounce messages? > F >That is the primary purpose. And from the receiving system's point ofD >view, if it gets a message from "postmaster" destined to an unknown. >user, it SHOULD ignore it and not bounce it.  > K NO. It should only ignore it and not bounce it if the envelope from address J of the incoming message is null see RFC 1123. (all bounce messages an SMTP; server generates should have null envelope from addresses). M Messages from postmaster which don't have null envelope from addresses should ( be treated like all other mail messages.        
 David Webb Security team leader CCSS Middlesex University      D >What you need to have though on your SMTP server node is a MAIL SET$ >FORWARD/user="postmaster" VALIDUSER > G >When there are real problems and someone sends a valid complaint about I >your system, then tend to send it to "postmaster", so you need to ensure @ >that you get it. In many cases, lack of response to a letter to3 >postmaster gets that system lincluded in the RBLs.    ------------------------------  + Date: Mon, 10 Apr 2006 13:14:25 +0000 (UTC)  From: david20@alpha2.mdx.ac.uk( Subject: Re: TCPIP$SMTP_POSTMASTER_ALIAS) Message-ID: <e1dlnh$rrp$2@news.mdx.ac.uk>   w In article <e191i0$opf$1@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: 6 >In article <4437FB7A.C87134B4@teksavvy.com>, JF Mezei( ><jfmezei.spamnot@teksavvy.com> writes:  > 2 >> Phillip Helbig---remove CLOTHES to reply wrote: >> >  I >> > Does defining TCPIP$SMTP_POSTMASTER_ALIAS have any effect other than > >> > setting the From: address of things like bounce messages? >>  H >> That is the primary purpose. And from the receiving system's point ofF >> view, if it gets a message from "postmaster" destined to an unknown0 >> user, it SHOULD ignore it and not bounce it.  > H >So this might solve the problem of double bounces.  I was getting emailD >to non-existent users, then bouncing it from TCPIP$SMTP, which thenI >bounced back to me since the recipient (apparent sender of the spam) had C >been forged.  So if I bounce it from postmaster to a non-existent  E >recipient, the remote system should NOT bounce it back to me, right?  >   L If DEC TCPIP Service's SMTP server is not sending bounces with null envelope* from addresses then it is severely broken.  
 David Webb Security team leader CCSS Middlesex University    H >Of course, this doesn't solve the real problem that the bounce is sent E >in the first place, but I can solve this by moving to ALPHA for the  H >cluster alias and have the Symbiont-Checks-Deliverability set to false. > H >This is not ideal, since this would also reject stuff where the sender H >had just mistyped a real email address, i.e. not spam to a non-existentG >address.  Would a real person sending email see the "not deliverable"  	 >message?  > F >> What you need to have though on your SMTP server node is a MAIL SET& >> FORWARD/user="postmaster" VALIDUSER > > >Yes, I have that (and still have one for TCPIP$SMTP as well). > I >> When there are real problems and someone sends a valid complaint about K >> your system, then tend to send it to "postmaster", so you need to ensure B >> that you get it. In many cases, lack of response to a letter to5 >> postmaster gets that system lincluded in the RBLs.  > G >Right.  In my case, though, I send emails through someone else's well  J >maintained SMTP server with fixed IP address.  I started using this when I >I moved to a dynamic IP address, since many sites disallow whole blocks  . >of what they know to be dynamic IP addresses. >    ------------------------------  % Date: Mon, 10 Apr 2006 09:11:07 -0400 ) From: "Neil Rieck" <n.rieck@sympatico.ca> P Subject: Re: The Minimum You Need to Know to Be an OpenVMS Application Developer9 Message-ID: <MPs_f.1547$%U2.125064@news20.bellglobal.com>   , "Ian Miller" <ijm@uk2.net> wrote in message = news:1144060325.985123.164610@z34g2000cwc.googlegroups.com... ' > There is now a review of this book at ; > http://www.openvms.org/stories.php?story=06/04/02/4453076  > H One of the best VMS programming books I've ever come across is "Writing L VAX/VMS Applications Using Pascal" by Theo de Klerk published in 1991 (BTW, M my copy is not for sale :-). I learned quite a bit of VMS application theory  K in this book and implemented many of the ideas into BASIC, C  and C++. The  K reason I mention this book is that I just received my copy of "The Minimum  K You Need to Know to be an OpenVMS Developer" and it is clearly in the same  G league as the Pascal book. This is a major effort and should be on the  K bookshelf of every beginner through intermediate level OpenVMS application  
 developer.   Just my 2-cents worth.  
 Neil Rieck Kitchener, Waterloo, Cambridge,  Ontario, Canada.   ------------------------------    Date: 10 Apr 2006 07:59:11 -0700$ From: "roger" <rogerntucker@msn.com>* Subject: Re: Using AST completion routinesC Message-ID: <1144681150.967797.106830@g10g2000cwb.googlegroups.com>   G As mentioned, but maybe not very clear, is that AST routines are called C (fired) asynchronously with respect to the main code, so you cannot A change the return value because in effect you have two threads of A control.  The main thread and the AST thread.  If you want to NOT G report an error from a system service, the easiest way to do this is to D just check for that specific status.  This happens all the time, forF example.  If (!$VMS_STATUS_SUCCESS(status) && status != RMS$_EOF) when6 checking for errors other than end of file.  The macro9 $VMS_STATUS_SUCCESS is in stsdef.h if you want to use it.    ------------------------------   Date: 9 Apr 2006 23:46:27 -0700 $ From: "Wilm" <w4.boerhout@planet.nl>& Subject: Re: Using the Serial TerminalB Message-ID: <1144651587.056064.48950@i39g2000cwa.googlegroups.com>  G >>>There is a hack using /SYSPASSWORD that accomplishes the same thing. # >>>I think this is what he's doing.    Indeed!   C The command has been in the startup file since the middle eighties, " maybe the hack was necessary then.G I sure wasn't aware that it was a hack, just that it evoked the desired  behaviour. You learn every day.    /Wilm    ------------------------------   End of INFO-VAX 2006.199 ************************