1 INFO-VAX	Sun, 17 Apr 2005	Volume 2005 : Issue 213       Contents: Re: Building an Alpha. Building an Alpha. Re: Building an Alpha.9 Communications between DECterm and DECW session manager ? = Re: Communications between DECterm and DECW session manager ? P Re: Do You Want To Know For Sure That You Are Going To Heaven? The reason some pP Re: Do You Want To Know For Sure That You Are Going To Heaven? The reason some pP Re: Do You Want To Know For Sure That You Are Going To Heaven? The reason some pP Re: Do You Want To Know For Sure That You Are Going To Heaven? The reason some pP Re: Do You Want To Know For Sure That You Are Going To Heaven? The reason some p Re: FA: VMS SW kits. Re: FA: VMS SW kits.( Re: How to disable outgoing FTP service?( Re: How to disable outgoing FTP service?( Re: How to disable outgoing FTP service? RE: Itanium ovms 8.2 migration* RE: Microsoft Joins the Battle for Itanium$ RE: Operating System Security Survey$ Re: Operating System Security Survey$ Re: Operating System Security Survey5 Re: TCPIP: why can an MX record not point to a CNAME?  two questions about MINICOPY  Re: two questions about MINICOPY" why no new AUDSRV patch for 7.3-2?  F ----------------------------------------------------------------------    Date: 16 Apr 2005 19:15:45 -0700* From: "Mister Q" <quodling@iprimus.com.au> Subject: Re: Building an Alpha. C Message-ID: <1113704145.900293.239030@g14g2000cwa.googlegroups.com>   E Oh, and another point. Attaching a console terminal? No real need for E graphics. So, coming out of the 9 pin male on the back of the system, F and going to a laptop with a serial port which also presents itself as
 a 9 Pin Male.   E So, ultimately I need a 9 pin female to 9 pin female. Not sure of the F wiring, (Modem, or null modem/crossover (and I have never seen a 9 pinA null modem)).  (I do have a couple of 9 pin female to 25 pin male C cables, and am digging through the junk boxes for 25-23 F-F, in the F optimistic hope that I can plug all of this together and make it work.   Q    ------------------------------    Date: 16 Apr 2005 18:55:36 -0700* From: "Mister Q" <quodling@iprimus.com.au> Subject: Building an Alpha. C Message-ID: <1113702936.655409.226450@g14g2000cwa.googlegroups.com>   E I have an AlphaStation 400 which I am planning to get up and running. G It has the standard Motherboard SCSI, coming out on a 50 Mpin connector  on the back of the system.  B I also have a BA364, which is one of the 4 scsi bays+CD  expansion@ boxes. It appears to have a 68 pin scsi connection (I have put 3! RZ29B-VA's and an TLZ07-VA in it.   F Two Questions - what cable do I need to take SCSI from one to 'tother?D And there are some switch packs in the back of the BA364 - any ideas what they do?    Q    ------------------------------  % Date: Sun, 17 Apr 2005 01:31:59 -0400 ' From: Dave Froble <davef@tsoft-inc.com>  Subject: Re: Building an Alpha. 0 Message-ID: <1163t6kd0vthk1b@corp.supernews.com>   Mister Q wrote: G > Oh, and another point. Attaching a console terminal? No real need for G > graphics. So, coming out of the 9 pin male on the back of the system, H > and going to a laptop with a serial port which also presents itself as > a 9 Pin Male.   F See the FAQ.  You'll need xmit, recv, and ground.  It must be a 'null ' modem', ie; xmit to recv, recv to xmit.   G > So, ultimately I need a 9 pin female to 9 pin female. Not sure of the H > wiring, (Modem, or null modem/crossover (and I have never seen a 9 pinC > null modem)).  (I do have a couple of 9 pin female to 25 pin male E > cables, and am digging through the junk boxes for 25-23 F-F, in the H > optimistic hope that I can plug all of this together and make it work. >  > Q  >    ------------------------------  % Date: Sat, 16 Apr 2005 20:45:13 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> B Subject: Communications between DECterm and DECW session manager ?B Message-ID: <1113698694.1815b49e443e68c324151b2727f35d4a@teranews>  ? Are there any  hooks to allow an interactive decterm session to 1 communicate with the DECwindows session manager ?   B For instance, to tell the decwindows session manager to change itsH default directory when it opens an application.; Thisd way for instance,H if I start XV from the session manager, it would already present me with' file selection of my current directory.   B It would be also neat if from DCL in a decterm, you could tell theF session manager to start an editor session with file x.y.z.  This way,H the edtor session would show up in the work in progress and you wouldn't6 have some subprocess tied to your current DCL process.  F Similarly, it would be neat if you could "register" a DCL process withH the session manager so that certain custom menuy options would send dataA to the decterm session after having done standard session manager H prompting etc. (think of it as glorified dynamic PFkey definitions where> the session manager would send keystrokes to the dcl process).   ------------------------------    Date: 16 Apr 2005 20:52:36 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) F Subject: Re: Communications between DECterm and DECW session manager ?3 Message-ID: <OL7T$rpIbnS4@eisner.encompasserve.org>   r In article <1113698694.1815b49e443e68c324151b2727f35d4a@teranews>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes:A > Are there any  hooks to allow an interactive decterm session to 3 > communicate with the DECwindows session manager ?   F    Some of what you want could be achieved by starting application via    FileView instead of DECterm.    ------------------------------    Date: 16 Apr 2005 21:22:55 -0700% From: "Leo" <Leo.Rosegrave@gmail.com> Y Subject: Re: Do You Want To Know For Sure That You Are Going To Heaven? The reason some p C Message-ID: <1113711775.779748.144350@g14g2000cwa.googlegroups.com>   A didn't mean to post that 3 times, but if it annoys someone, well, F that's how i feel about the Christy Trolls polluting the google boards incessantly.   ------------------------------    Date: 16 Apr 2005 21:18:55 -0700% From: "Leo" <Leo.Rosegrave@gmail.com> Y Subject: Re: Do You Want To Know For Sure That You Are Going To Heaven? The reason some p C Message-ID: <1113711535.075773.227300@l41g2000cwc.googlegroups.com>   8 i'm a civil libertarian and free speech advocate, but...  > i'm sorry. this mad-dog X-ian proselytizing is just incrediblyG offensive and totally OT for this group. is someone here unable or just * afraid to ban this sanctimonious garbage??  G some of us do not need someone to tell us how to think and behave. save G the "good news" for someone who cares. or share some actual good news - " like, the war in Iraq is over.....E ....or, perish forbid, something about the (ostensible) point of this  group: DC Comics?   
 Leo Rosegrave    ------------------------------    Date: 16 Apr 2005 21:20:10 -0700% From: "Leo" <Leo.Rosegrave@gmail.com> Y Subject: Re: Do You Want To Know For Sure That You Are Going To Heaven? The reason some p C Message-ID: <1113711610.552162.247980@f14g2000cwb.googlegroups.com>    i agree about the Pope though.   ------------------------------    Date: 16 Apr 2005 21:20:19 -0700% From: "Leo" <Leo.Rosegrave@gmail.com> Y Subject: Re: Do You Want To Know For Sure That You Are Going To Heaven? The reason some p C Message-ID: <1113711619.682154.170490@g14g2000cwa.googlegroups.com>    i agree about the Pope though.   ------------------------------    Date: 16 Apr 2005 21:20:24 -0700% From: "Leo" <Leo.Rosegrave@gmail.com> Y Subject: Re: Do You Want To Know For Sure That You Are Going To Heaven? The reason some p C Message-ID: <1113711624.851514.260680@l41g2000cwc.googlegroups.com>    i agree about the Pope though.   ------------------------------  % Date: Sat, 16 Apr 2005 20:02:55 -0400 " From: Glenn Everhart <gce@gce.com> Subject: Re: FA: VMS SW kits. , Message-ID: <0qKdnQwnEI6oNPzfRVn-gA@rcn.net>   John Vottero wrote:  > [snip] >  > K >>>I am not a lawyer.  When you buy the media, you own that copy.  You can   >>>sell F >>>it or give it away.  You can't install that software on a computer 
 >>>becauseI >>>that is copying and all rights to copy are controlled by the copyright 
 >>>holder.G >>>If you want to copy it to your computer, you need a license from the  >>>copyright holder. >>G >>I'm not a lawyer either, but I thought (I do not know as I have never H >>been the primary in the transaction as I get all my in house VMS stuffG >>from the datacebter) you don't actually "buy" a condist.  You license F >>it and one of the requirements of the license is that you destroy itF >>when your done with it.  (Still hoping someone from HP would jump in >>here with the real story!) >> >  > N > Well that would certainly muddy the legal waters.  I suppose you can't sell  > something you don't own. > L > I checked a recent condist and I couldn't find any information either way. >  > I Seems to me there was a case against Adobe in which the courts ruled that N in spite of the "license" rhetoric, if it looked like a sale it was a sale andJ the doctrine of first sale applied. This had to do with packages one wouldH buy, possibly from a retailer, but which claimed to be selling licenses.  G If that precedent applies, the license restrictions may be unsupported.   K There are publishers who would be delighted to sell licenses for books too, I so they couldn't be resold, and one might well imagine the same done for, N say, automobiles. However the courts have at times held that doctrine of firstK sale (copyright holder can control only first sale of an item, in brief and S if my understanding is right: IANAL) cannot be so easily avoided. It was refreshing 8 to see it move, finally, into an explicit software case.   ------------------------------  + Date: Sun, 17 Apr 2005 01:11:05 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: FA: VMS SW kits. $ Message-ID: <d3sd39$hoo$1@online.de>  M > There are publishers who would be delighted to sell licenses for books too, K > so they couldn't be resold, and one might well imagine the same done for, P > say, automobiles. However the courts have at times held that doctrine of firstM > sale (copyright holder can control only first sale of an item, in brief and U > if my understanding is right: IANAL) cannot be so easily avoided. It was refreshing   * ???  I'm not up to date on my MTTLAs.  :-)  : > to see it move, finally, into an explicit software case.  B Note that the discussion here has been revolving around CONDIST.  I However, it is possible to just buy the media CDs as a one-time thing (I  < have done this).  The rules might be different in this case.   ------------------------------    Date: 16 Apr 2005 19:52:49 +01006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)1 Subject: Re: How to disable outgoing FTP service? , Message-ID: <42616d11$1@NEWS.LANGSTOEGER.AT>  c In article <oECVpaeXY2Dl@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes: f >In article <42610c85$1@NEWS.LANGSTOEGER.AT>, peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) writes:N >> Better forget the idea. You can make it more difficult, but not prevent it.% >> What are to trying to accomplish ?  > I >It sounds like an attempt to achieve DECnet levels of security using IP. L >That is not viable in a protocol stack which defaults to no authentication.  - What DECnet security ? The NETMBX privilege ?   H If I think over it, there might be a way to disable access to FTP_CLIENTL How about restricting access to the BG0: device (eg. with ACLs) for all I/O?   LikeG $ SET SECU/OBJ=DEV/ACL=((ID=IP_ALLOW,ACC=R,W,L,P),(ID=*,ACC=NONE)) BG0:   L And then granting the IP_ALLOW identifier to users allowed to do IP, any IP. Would that work ?   * btw: Is NETMBX required for use of TCPIP ?2 If not, how about restricting TCPIP to its usage ?   --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------    Date: 16 Apr 2005 19:59:54 +01006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)1 Subject: Re: How to disable outgoing FTP service? , Message-ID: <42616eba$1@NEWS.LANGSTOEGER.AT>  s In article <1113671128.3bb80dcee0fca9da3bc4050b02ced673@teranews>, JF Mezei <jfmezei.spamnot@vaxination.ca> writes: A >Another simpler way to disable outgoing FTP is simply to set the E >security of the executable file in SYS$SYSTEM so that "world" cannot  >read or execute the file.    < No. It would prevent access to the FTP client of TCPIP only.C There are a lot more ways to run an FTP client on the machine then. @ CSWB, FETCH_HTTP (?), LYNX, MOSAIC, MOZILLA, NETSCAPE, WGET, ...  F >Also, after you have use the configuration procedure to shut down the6 >FTP client, you need to go to the INSTALL utility and > 0 >INSTALL> DELETE SYS$SYSTEM:UCX$FTP_CLIENT.EXE   > E >(of whatever the exact name of the executable is for your version of H >software). This will instantly remove the entry (but not the file). And  >you then do not need to reboot.  J The TCPIP$FTP_CLIENT_SHUTDOWN.COM which is run by TCPIP$CONFIG.COM in such< a case (and in case of a shutdown) does deinstall the image.  So, no need to do it yourself...  I >(note that the UCX configuration utility may have done this for you. You 	 >can use:  > H >INSTALL> LIST SYS$SYSTEM:UCX$FTP_CLIENT.EXE (or whatever the exact file. >name is) to see if it is still a known image.  D As the image is only installed /OPEN/HEADER/SHARED, I see no need to8 deinstall it for (the idea of) a security reason at all.   --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------    Date: 16 Apr 2005 15:44:29 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) 1 Subject: Re: How to disable outgoing FTP service? 3 Message-ID: <F5uBwLRw9IiR@eisner.encompasserve.org>   e In article <42616d11$1@NEWS.LANGSTOEGER.AT>, peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) writes: e > In article <oECVpaeXY2Dl@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes: g >>In article <42610c85$1@NEWS.LANGSTOEGER.AT>, peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) writes: O >>> Better forget the idea. You can make it more difficult, but not prevent it. & >>> What are to trying to accomplish ? >>J >>It sounds like an attempt to achieve DECnet levels of security using IP.M >>That is not viable in a protocol stack which defaults to no authentication.  > / > What DECnet security ? The NETMBX privilege ?   3 That is the one particularly relevant to this case.   B Another important one is the requirement for privilege to set up aD service that can accept incoming connections without authentication.  J > If I think over it, there might be a way to disable access to FTP_CLIENTN > How about restricting access to the BG0: device (eg. with ACLs) for all I/O? >  > LikeI > $ SET SECU/OBJ=DEV/ACL=((ID=IP_ALLOW,ACC=R,W,L,P),(ID=*,ACC=NONE)) BG0:  > N > And then granting the IP_ALLOW identifier to users allowed to do IP, any IP. > Would that work ?  > , > btw: Is NETMBX required for use of TCPIP ?4 > If not, how about restricting TCPIP to its usage ?   ------------------------------  % Date: Sat, 16 Apr 2005 14:27:42 -0400 ' From: "Main, Kerry" <kerry.main@hp.com> ' Subject: RE: Itanium ovms 8.2 migration R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB5EC9BD@tayexc19.americas.cpqcorp.net>   > -----Original Message-----B > From: Peter 'EPLAN' LANGSTOEGER [mailto:peter@langstoeger.at]=20 > Sent: April 16, 2005 4:55 AM > To: Info-VAX@Mvb.Saic.Com ) > Subject: Re: Itanium ovms 8.2 migration  >=20 > In article=20 = > <1113574696.425304.268200@o13g2000cwo.googlegroups.com>,=20 ( > "ugex" <edgar_ulloa@yahoo.com> writes:G > >I need to migrate all aplications in vax vms 7.1 to Itanium ovms 8.2  > >(all builded in cobol). >=20+ > COBOL (V2.8) is already available on I64.  >=20 > >The questions are.. > > = > >1.-- Can I do it direcly .. vax - vms  Itanium..? or is=20  > necesary Vax - > >alpha -itanium..? >=20E > I tend to think so, though I don't know your specific requirements. @ > The problem in porting usually lies in implicit assumptions=20 > of the platform C > (like 32/64 bit, [endianess, which isn't here an issue], platform @ > specific features of the compiler, memory layout) and other=20 > bad codings.< > If good coded, there should be no real problem in porting. > What remains, is testing...  >=20E > >2.- Some one knows what itanium licences I need for start ovms 8.2 
 > >itanium..? 	 > >- ovms  > >- vms users > >- tcp/ip 
 > >- dvnetend 	 > >-cobol  >=20 > FOE and COBOL 0 > http://www.compaq.com/info/XAV12X/XAV12XPF.PDF0 > http://www.compaq.com/info/XAV12Z/XAV12ZPF.PDF0 > http://www.compaq.com/info/SP4592/SP4593PF.PDF >=20G > >3.- some one knows what point missing me to know for this proyect..?  >=20H > Assistance for porting to Itanic is available at HP's porting centers.A > Assistance for porting to Alpha can be found here, but maybe=20  > no longer by HP.= > Cheap Itanic development systems are available by HP (DSPP) 7 > Cheap Alpha develepment systems can be found at EBAY.  >=20 > --=20      All,  ? Porting assistance on all HP platforms is available from HP at: 1 http://h18011.www1.hp.com/iqcenter/expertise.html      Regards   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  $ "OpenVMS has always had integrity .. Now, Integrity has OpenVMS .."   ------------------------------  % Date: Sat, 16 Apr 2005 14:19:51 -0400 ' From: "Main, Kerry" <kerry.main@hp.com> 3 Subject: RE: Microsoft Joins the Battle for Itanium R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB5EC9BC@tayexc19.americas.cpqcorp.net>   > -----Original Message------ > From: John Smith [mailto:a@nonymous.com]=20  > Sent: April 15, 2005 9:53 PM > To: Info-VAX@Mvb.Saic.Com 5 > Subject: Re: Microsoft Joins the Battle for Itanium  >=20  
 [snip ...]  @ > > ... With the full backing of Intel, HP and now Microsoft,=20
 > Itanium has ) > > a real shot of owning the enterprise.  > > 2 > > Stay tuned. This is about to get interesting." >=20 >=20. > There was a version of NT for the Alpha too. >=206 > Those who forget history are condemned to repeat it. >=20   John,   A One big difference is that Digital/Compaq did all of the porting, G debugging, testing, QA'ing, marketing and packaging for the OS versions F of Alpha Windows NT. Microsoft primary work wrt to the server versionsH of Alpha NT was to gather up the license revenues. [Microsoft did do the' SQL/Exchange/SMS versions of Alpha NT.]   & This will not be the case for Itanium.   Regards   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  $ "OpenVMS has always had integrity .. Now, Integrity has OpenVMS .."   ------------------------------  % Date: Sat, 16 Apr 2005 14:43:30 -0400 ' From: "Main, Kerry" <kerry.main@hp.com> - Subject: RE: Operating System Security Survey R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB5EC9BF@tayexc19.americas.cpqcorp.net>   > -----Original Message-----, > From: John Santos [mailto:john@egh.com]=20 > Sent: April 16, 2005 1:27 AM > To: Info-VAX@Mvb.Saic.Com / > Subject: Re: Operating System Security Survey  >=20 > Larry Kilgallen wrote:? > > In article <13cxl2mzy5dpk.dlg@wagmail.com>, Mitch Wagner=20  > <mitch@wagmail.com> writes:  > >=20 > >=20B > >>Likewise, our Voting Booth polls are designed to get people=20 > thinking andF > >>generate discussion. Which is what this one has certainly done.=20 > >=20 > >=20A > > Please do not be under the illusion that participants here=20  > are otherwise " > > reticent to discuss things :-) >=20= > True.  Unfortunately, people here are discussing polling=20  > methods ratherD > than OS security.  Though, Larry, you could explain why you always- > disable cookies to get us back on track :-)  >=20 > --=20 
 > John Santos  > Evans Griffiths & Hart, Inc. > 781-861-0670 ext 539 >=20   John,   ? A few examples of why cookies are potentially not a good thing:   , "Are P2P networks leaking military secrets?"/ http://news.zdnet.com/2100-1009_22-5285918.html   ) "FTD.com hole leaks personal information" ) http://news.com.com/2100-1017-984585.html   , "Doubleclick sued over alleged cookie abuse"H http://www.theregister.co.uk/2000/01/28/doubleclick_sued_over_alleged_co okie/   > "Cookies" targeted as Congress, advocates address Net privacy"J http://news.com.com/2100-1023-236800.html?legacy=3Dcnet&tag=3Dst.ne.ron.l= thd  1005-200-1547443   Regards   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  $ "OpenVMS has always had integrity .. Now, Integrity has OpenVMS .."   ------------------------------  % Date: Sat, 16 Apr 2005 19:22:54 -0400 . From: JF Mezei <jfmezei.spamnot@vaxination.ca>- Subject: Re: Operating System Security Survey B Message-ID: <1113693766.b0f5acce5eac77d9b9abe68e84b7c9e4@teranews>   "Main, Kerry" wrote:A > A few examples of why cookies are potentially not a good thing:   K I dsagree. Good quality freshly baked chocolate chip cookies are excellent.   D A browser however should only send cookies that belong to the domainE which set the cookies in the first place (and also only allow cookies 3 belonging to the server whih is setting the cookie.     A The fact that microsoft browsers may have been ill copnfigured by C default did open up a vast array of abuses, but it doesn't make the 9 cookie concept wrong just because Microsoft did it wrong.    ------------------------------  % Date: Sat, 16 Apr 2005 21:20:00 -0400 ' From: Dave Froble <davef@tsoft-inc.com> - Subject: Re: Operating System Security Survey 0 Message-ID: <1163ee76gjrncd5@corp.supernews.com>   Main, Kerry wrote:  A > A few examples of why cookies are potentially not a good thing:  > . > "Are P2P networks leaking military secrets?"1 > http://news.zdnet.com/2100-1009_22-5285918.html  > + > "FTD.com hole leaks personal information" + > http://news.com.com/2100-1017-984585.html  > . > "Doubleclick sued over alleged cookie abuse"J > http://www.theregister.co.uk/2000/01/28/doubleclick_sued_over_alleged_co > okie/  > @ > "Cookies" targeted as Congress, advocates address Net privacy"J > http://news.com.com/2100-1023-236800.html?legacy=cnet&tag=st.ne.ron.lthd > 1005-200-1547443  E Kerry, the above all appear to be bad implementations.  Some of this  B comes from people who blindly use Microsoft tools and never think H through the entire application.  Also, when these people go looking for H a vendor, what is more important to them, security, or price?  They get  what they paid for.   C Seriously, if someone with sensitive data shares his data over the  0 internet, don't blame the tool, blame the idiot.  G There can be occations when the intent is to share everything.  That's   what the tool is for.    Dave   ------------------------------  % Date: Sat, 16 Apr 2005 16:07:22 -0700 0 From: glen herrmannsfeldt <gah@ugcs.caltech.edu>> Subject: Re: TCPIP: why can an MX record not point to a CNAME?0 Message-ID: <v8KdnU-X8ogHB_zfRVn-vw@comcast.com>  / Phillip Helbig---remove CLOTHES to reply wrote:   I > This has nothing to do with VMS per se, but perhaps someone here knows  I > the reasoning behind this restriction.  An MX record is not allowed to  H > point to a CNAME record, but only to an A-record or to a numerical IP ) > address.  What's the logic behind this?    (snip)  " A record, not a numerical address.  @ The reason has to do with secondary mail servers and forwarding.  ; If you have only one MX there is no technical reason why it H can't work with CNAME, though I believe the restriction is now enforced 	 by named.   C If you have more than one, and the mail is received by a secondary  E server, that server needs to know which it can forward to. It does an G MX lookup and removed all lower preference (equal or higher number) MX  G entries than itself.  To do this, it must match up the MX name with its 3 own name, which it can't do if the name is a CNAME.    -- glen    ------------------------------  + Date: Sat, 16 Apr 2005 21:59:45 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)% Subject: two questions about MINICOPY $ Message-ID: <d3s1sh$6eu$1@online.de>   Make that three questions.  H First question:  If I dismount one member of a shadow set in a non-cleanH fashion (for instance, if there are files open on the shadow set (it is H well known that one cannot dismount a member of a "live" shadow set and D expect to get a "clean" copy) and/or if the only node with a direct E connection to the shadow set fails), can I expect that a) a minicopy  B will take place and b) the files which were open will be restored ( correctly to the member when it returns?  5 Second question:  Is there any reason not to specify  F POLICY=MINICOPY=OPTIONAL when initially mounting a disk?  This is one D way to enable minicopy, the other being to specify it on DISMOUNT.  G However, if the dismount is unplanned (see first question), then there  I is no chance to specify a proper DISMOUNT command.  The only advantage I  H can see to specifying it on DISMOUNT as opposed to MOUNT is that in the F former case, fewer resources will be used.  Surely the maintenance of H the write bitmap can't be that resource-intensive.  (Another reason, of E course, would be to specify it on DISMOUNT of a shadow set which had  I previously been mounted before one learned about minicopy, but I find it  H hard to believe that that is the reason it is possible to specify it on  DISMOUNT as well as MOUNT.)   G Third question:  It is recommended to set SHADOW_MAX_COPY on VAXes to 0 G to avoid the chance that a VAX might be assigned a copy, resulting in a G full copy when no minicopy is possible.  However, imagine the following H situation: in a mixed cluster, there is a shadow set mounted only on VAXE nodes.  Hence, no minicopy is possible anyway.  It would be better to H have a VAX with a local connection to at least one of the members do theB full copy, as opposed to an ALPHA doing it over the network (afterG mounting it from an ALPHA after all, of course).  Or imagine a VAX with A a shadowed system disk.  Even if it is mounted by an ALPHA in the C cluster, the ALPHA can't do the INITIAL mount.  If a shadow copy is E needed (for example, if a member temporarily fails and returns), then H again it would make more sense to have a VAX with a direct connection doH the full copy than have an ALPHA do it over the network.  (There doesn'tF seem to be a situation in which specifying POLICY=MINICOPY on DISMOUNTA would make sense in the case of a system disk.)  If the action is E planned, then presumably one could temporarily define SHADOW_MAX_COPY H (it is a dynamic parameter) to 1 on a VAX, do the copy, then redefine it@ to 0 after the copy is done.  However, for unplanned, unassistedH scenarios, either no copy is done, or an ALPHA does it over the network.   ------------------------------  + Date: Sat, 16 Apr 2005 22:39:54 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)) Subject: Re: two questions about MINICOPY $ Message-ID: <d3s47q$8dn$1@online.de>  D In article <d3s1sh$6eu$1@online.de>, helbig@astro.multiCLOTHESvax.de3 (Phillip Helbig---remove CLOTHES to reply) writes:    I > Third question:  It is recommended to set SHADOW_MAX_COPY on VAXes to 0 I > to avoid the chance that a VAX might be assigned a copy, resulting in a + > full copy when no minicopy is possible.     B I meant "when no minicopy is possible via the VAX".  The importantC thing, though, is that a minicopy IS possible, if done by an ALPHA.   ; I decided to give it a try.  Unfortunately, I forgot to set D SHADOW_MAX_COPY to 0 on the VAXes.  From an ALPHA, I dismounted the B member of a shadow set connected to a VAX, the other member being G connected to the ALPHA.  I then remounted it with POLICY=MINICOPY (the  @ default is POLICY=MINICOPY=OPTIONAL if, as in this case, it was F dismounted with POLICY=MINICOPY).  A message came up saying that such B and such a write bitmap was being used.  This looked good.  Then, F however, a message FROM A VAX (the VAX hosting the dismounted member) E came, saying that a shadow copy had been initiated---and since a VAX  I initiated it, it was a full copy.  (Fortunately, I chose a small disk to   test this out.)   I This makes me wonder what determines which node in a cluster handles the  H shadow-copy operation.  Is this documented somewhere?  Can one override D the default behaviour (other than by setting SHADOW_MAX_COPY to all 3 nodes other than the one which should do the copy)?    ------------------------------  + Date: Sat, 16 Apr 2005 19:56:58 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)+ Subject: why no new AUDSRV patch for 7.3-2? $ Message-ID: <d3rqma$t28$1@online.de>  D The newest AUDSRV patch for 7.3-1 says that the next version of VMS H with this functionality is 8.2.  Why is there not a corresponding patch I for 7.3-2, either as a standalone patch or part of another patch?  I can  I understand that sometimes fixes are not back-ported to OLDER versions of  : VMS, but why fix something on 7.3-1 and 8.2 but not 7.3-2?  F On a related note, what actually happens to executables and sharables D during an UPGRADE?  Does (in this respect) the same thing happen as  during a fresh install?   D In other words, if I install the newest AUDSRV patch on 7.3-1, then C later upgrade to 7.3-2 and apply all available 7.3-2 patches, am I  E correct in assuming that the AUDSRV fix won't be available to me, at  3 least in any official/supported fashion, until 8.2?   G I suppose it is at least theoretically possible that the upgrade would  C behave otherwise, but I tend to doubt it, for a variety of reasons.    ------------------------------   End of INFO-VAX 2005.213 ************************                                                                                                                                                                                                                                                        c{|GUgL_(>Wa\_%7	z
/MeWl>uObβ %G)^QvF gI~	:/x25gKc0C'v.b>Wa}ZE{	;g.\ki./=+~S 9.4g6\j؋z>jr#uT#sd,\;[C 繁QS9oT|h6*̕?>[XD96t=ɏِّ~|ׇl@:ѳv['&o?ofɲ?c<Թ(!4݊==^˩9'1`z$ύX7}5}Y//T/S~
ОyX3>"i~_#}4Vސ,l=_pe|+cR¿cݳkIw_K gIٗ\!꺿!B
;Tgyq|߭X1om2^Ayxa߲+`%I/p=O}zo_xɸ;+R>aGǋlϙ%?H~7߇?5׼C~? xƓX>$~Km%$W'>=UCO&3ClowWx)j?^7"{{6".k~&|d]jnNbS1rRX^?Z|P
v`8i'|ݏމu'7?DaLoX<h~C#?<BgoD!h&wt`	RHVm=Stn޲seꇔ<w@NƼw^py}vÖ.W~Wگx>QI;W9~>fe@cmREzkry
RH%~A$^sFxOmgmOgt<ezHrI{_0=i:s[C_!c,\B隩gg}8S?ܛ:?!qwy?*p^:"_X"[Tm

	CO3d|}&	>/r?xWZJעyUF͓rSJ2~%H򼱘=2DEd>j?)|??Jޏ$h&BCMd/KNKySДw
Ϥg}F~V姷U~eni~R>%$8qhG(yF>~9OZk[Dȑ%<oMyP,y.v1On$왎'Ox?Y\#?9'!YSѣ\{O.^SHyBiۣӶJ>"wpaq=R@xJ2|35>>cm\|s[盕QI2[._LF={̯dRFb?/+zQ7<.ߤoieσ߱O\	>/ӿ秕8?34GW?&Qk^]oyffv$fu`:9k#ӱ{q
Nǧ2UdvSsRCXn{)rFL.qöqAiam"+7N)Į>g杙d[./Bμ=xQfϞg[+Tٙ"ydm˞gle۔|Ty(mR|?ǕIoWf_b