1 INFO-VAX	Sat, 28 Feb 2004	Volume 2004 : Issue 117       Contents:P Re: 2-node LAVC cluster with quorum disk - network disappears - which node CLUEXP Re: 2-node LAVC cluster with quorum disk - network disappears - which node CLUEXP Re: 2-node LAVC cluster with quorum disk - network disappears - which node CLUEXP Re: 2-node LAVC cluster with quorum disk - network disappears - which node CLUEX Re: Alpha Powered logo- Re: Educating HP employees on the nettiquette - Re: Educating HP employees on the nettiquette - Re: Educating HP employees on the nettiquette - Re: Educating HP employees on the nettiquette - Re: Educating HP employees on the nettiquette - Re: Educating HP employees on the nettiquette - Re: Educating HP employees on the nettiquette  Re: Hobbyist license for VMS Re: Identify graphics on XP1000  Re: Intel releases 64bit Xeon  Re: Intel releases 64bit Xeon : Re: It is almost certain now, INTEL will have 64bit x86 !!: Re: It is almost certain now, INTEL will have 64bit x86 !!: Re: It is almost certain now, INTEL will have 64bit x86 !!/ Re: KGPSA-C and EMC DS-24 M2 SAN Switch Problem  RE: Power Button on PWS + Re: vms 552 + timeline software experience? ' vms 552 + timeline software experience? + Re: VMS Hobbyist License program abolished? + Re: VMS Hobbyist License program abolished? + Re: VMS Hobbyist License program abolished? + Re: VMS Hobbyist License program abolished? + Re: VMS Hobbyist License program abolished? @ Re: VMS upgrade changed program behavior, what could cause this?@ Re: VMS upgrade changed program behavior, what could cause this?@ Re: VMS upgrade changed program behavior, what could cause this?2 Re: Why does smtp mail between local systems fail?3 Re: [GAME] you've been appointed VMS Mkt Mgr, so...   F ----------------------------------------------------------------------    Date: 28 Feb 2004 01:00:27 -08007 From: jones.computer.srv@worldnet.att.net (Daryl Jones) Y Subject: Re: 2-node LAVC cluster with quorum disk - network disappears - which node CLUEX = Message-ID: <8a646952.0402280100.110b3763@posting.google.com>    Dear Roy Omond:   A I am sorry for not clearly reading your question. Anyway, here is  another try.  E According to VAXcluster Principle, Preliminary Edition - Fall 1992 us  DECUS, page 7-11:   F Since the cluster system manager has given them equal number of votes,D they are considered to be of "equal importance". Thus, it is more orC less random which node is forced out of the cluster. Whichever node F first proposes a new configuration will remain in the cluster, and theC other node will BUGCHECK. If you want a one node to be favored over D the other then change the vote on the favored to be greater than the" other node. Generally by one vote.   From page 7-12:   C The Connection Manager "logically" considers all possible connected B subcluster. For each of these subclusters it computes a "figure of merit" using the formula  )                 256+V+N="fiqure of merit"   D Where V is the total number of votes in the subcluster, and N is the' total of VAX systems in the subcluster.   A The Connection Manager then picks as the "optimal subcluster" the E totally connected subcluster with the highest "figure of merits", and   discards the other possiblities.   From Page 7-13:   < "When multiple subclusters qualify for being optimal, one isC effectively chosen at random, and the others are discarded. In such C situations, the optimal subcluster will generally be the subcluster 6 containing the system that first noticed the problem."   I hope this helps.   Regards: Daryl Jones   o Roy Omond <Roy.Omond@BlueBubble.UK.Com> wrote in message news:<c1kefd$1jqibc$1@ID-225674.news.uni-berlin.de>...  > Daryl Jones wrote: > F > > Years ago I ran across a problem where the cluster information wasF > > going over the ethernet and not over the CI connection. Therefore,J > > when the ethernet connection was lost there went the cluster until theA > > traffic restablished over the CI. Check to see if the cluster D > > connection was lost over the SCSI or ethernet. There is a sysgenI > > parameter that controls which path is used. I don't recall the sysgen A > > parameter because it was over 30 years ago. Sorry about that.  > A > Cluster communications do *not* go over SCSI (an all too common  > misapprehension).  > D > As stated before, the single and only path for SCS traffic (namely! > the Ethernet) was disconnected.  >  > Roy Omond  > Blue Bubble Ltd.   ------------------------------  + Date: Sat, 28 Feb 2004 18:29:16 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney) Y Subject: Re: 2-node LAVC cluster with quorum disk - network disappears - which node CLUEX ( Message-ID: <c1qmls$11g$1@pcls4.std.com>  ! Roy Omond <Roy@Omond.net> writes:    >> From page 7-12: >>  F >> The Connection Manager "logically" considers all possible connectedE >> subcluster. For each of these subclusters it computes a "figure of  >> merit" using the formula  >>  , >>                 256+V+N="fiqure of merit"    That's really     256*V+N.   D >Yes, all well and good.  But as far as I can tell, this all assumesF >that the nodes can still communicate with one another.  The situationF >as described precludes any such communication, both nodes have quorumH >(due to the shared quorum disk), so I'm impressed that one of the nodesD >voluntarily exits.  IMHO it is a tribute to the design fundamentals: >of VMS clusters that this situation is handled correctly.  F They are still aware of each other.  As far as each node is concerned,D if there were no other nodes other than itself, it would expect thatE whenever it reads the quorum disk, it will read exactly what it wrote G last.  But the other node is writing to it as well, so the first 'sees' D the second (and vice versa) since what it writes must be informationI about itself (CSID etc.) and when it reads it doesn't match what it wrote  when the other node wrote last.     H >By the way, this cluster is totally symmetric in the sense that neitherG >one nor the other is considered to be more "important" (as others have F >suggested granting more votes to one or other, or playing around withG >LOCKDIRWT).  I'm pretty convinced that the SCSSYSTEMID is the deciding D >factor, as Mike Moroney has suggested from inspection of the actual >sources (thanks Mike !).   H Note that it'n not strictly a case of the lowest SCSSYSTEMID wins, sinceJ the CSIDs are compared with a CMPC3, and the little-endianism of VMS makes  this different from a CMPW/CMPL.  .   The thing that impresses me most is the factA >that both nodes must be aware that the other one is still alive.    Through the quorum file.   --   -Mike    ------------------------------  % Date: Sat, 28 Feb 2004 18:31:57 +0000 - From: John Laird <nospam@laird-towers.org.uk> Y Subject: Re: 2-node LAVC cluster with quorum disk - network disappears - which node CLUEX 8 Message-ID: <jqm140lpom2fqn9t7vvhcfkuj051qjcq5a@4ax.com>  D On Sat, 28 Feb 2004 10:02:09 +0000, Roy Omond <Roy@Omond.net> wrote: > D >Yes, all well and good.  But as far as I can tell, this all assumesF >that the nodes can still communicate with one another.  The situationF >as described precludes any such communication, both nodes have quorumH >(due to the shared quorum disk), so I'm impressed that one of the nodesD >voluntarily exits.  IMHO it is a tribute to the design fundamentals: >of VMS clusters that this situation is handled correctly. > H >By the way, this cluster is totally symmetric in the sense that neitherG >one nor the other is considered to be more "important" (as others have F >suggested granting more votes to one or other, or playing around withG >LOCKDIRWT).  I'm pretty convinced that the SCSSYSTEMID is the deciding D >factor, as Mike Moroney has suggested from inspection of the actualG >sources (thanks Mike !).  The thing that impresses me most is the fact A >that both nodes must be aware that the other one is still alive.   J I think you will find that both machines must be quorum disk watchers, andL that in fact they are not simply watching but actively writing to QUORUM.DATJ - it would be a disaster if they were not.  I don't have an active clusterF with a quorum disk to check, but I would guess the file records activeJ watchers and cluster incarnation information.  Once one node has decide toL continue with the cluster, it would be sensible for it to update the file soH that on the off-chance the other node was still up and resumed watching,B that it would notice the cluster had moved on and commit the usual
 hara-kiri.  A >I consider this to be the end of this thread - thanks to all who 
 >contributed.   - CLUEXIT, thread voluntarily exiting Usenet...    --  : Diplomacy: the art of letting someone else have your way.    Mail john rather than nospam...    ------------------------------  % Date: Sat, 28 Feb 2004 10:02:09 +0000  From: Roy Omond <Roy@Omond.net> Y Subject: Re: 2-node LAVC cluster with quorum disk - network disappears - which node CLUEX 4 Message-ID: <c1pov1$e6r$1$830fa7a5@news.demon.co.uk>   Daryl Jones wrote:  - > Dear Roy Omond (no need to be so formal :-)  > C > I am sorry for not clearly reading your question. Anyway, here is  > another try. > G > According to VAXcluster Principle, Preliminary Edition - Fall 1992 us  > DECUS, page 7-11:  > H > Since the cluster system manager has given them equal number of votes,F > they are considered to be of "equal importance". Thus, it is more orE > less random which node is forced out of the cluster. Whichever node H > first proposes a new configuration will remain in the cluster, and theE > other node will BUGCHECK. If you want a one node to be favored over F > the other then change the vote on the favored to be greater than the$ > other node. Generally by one vote. >  > From page 7-12:  > E > The Connection Manager "logically" considers all possible connected D > subcluster. For each of these subclusters it computes a "figure of > merit" using the formula > + >                 256+V+N="fiqure of merit"  > F > Where V is the total number of votes in the subcluster, and N is the) > total of VAX systems in the subcluster.  > C > The Connection Manager then picks as the "optimal subcluster" the G > totally connected subcluster with the highest "figure of merits", and " > discards the other possiblities. >  > From Page 7-13:  > > > "When multiple subclusters qualify for being optimal, one isE > effectively chosen at random, and the others are discarded. In such E > situations, the optimal subcluster will generally be the subcluster 8 > containing the system that first noticed the problem."  C Yes, all well and good.  But as far as I can tell, this all assumes E that the nodes can still communicate with one another.  The situation E as described precludes any such communication, both nodes have quorum G (due to the shared quorum disk), so I'm impressed that one of the nodes C voluntarily exits.  IMHO it is a tribute to the design fundamentals 9 of VMS clusters that this situation is handled correctly.   G By the way, this cluster is totally symmetric in the sense that neither F one nor the other is considered to be more "important" (as others haveE suggested granting more votes to one or other, or playing around with F LOCKDIRWT).  I'm pretty convinced that the SCSSYSTEMID is the decidingC factor, as Mike Moroney has suggested from inspection of the actual F sources (thanks Mike !).  The thing that impresses me most is the fact@ that both nodes must be aware that the other one is still alive.  @ I consider this to be the end of this thread - thanks to all who contributed.  4 As ever, VMS - when only the best is good enough :-)  ? I just wonder how other so-called "clusters" would handle this   particular situation.   	 Roy Omond  Blue Bubble Ltd.   ------------------------------    Date: 28 Feb 2004 03:38:02 -0800. From: alexdaniels@themail.co.uk (Alex Daniels) Subject: Re: Alpha Powered logo < Message-ID: <9f7f13a8.0402280338.6a58d2c@posting.google.com>  h jealousxmp@aol.com (jealous xmp) wrote in message news:<20040226154808.26340.00000416@mb-m20.aol.com>...O > >The site's HTTP responses only mention Apache. No hint of which OS they run.  > F > Netcraft says it's Apache on Linux, but that is 6 months old figure. > P > I know with passive fingerprinting you can verify that it's linux, but i don'tL > know how to verify it's on alpha.  I know some servers also show the wrong7 > thing (like IIS on Linux) due to load balancers, etc.  > 	 > Michael   7 Telnet'ing to the SSH port(22) reveals a little more...   + SSH-2.0-OpenSSH_3.6.1p2 Debian 1:3.6.1p2-11    Regards    Alex   ------------------------------  % Date: Sat, 28 Feb 2004 10:24:01 +0100 * From: Paul Sture <nospam@sture.homeip.net>6 Subject: Re: Educating HP employees on the nettiquette0 Message-ID: <40406C41.176882AE@sture.homeip.net>   R. D. Davis wrote: > 2 > In article <403E96A4.17B13CBA@sture.homeip.net>,6 >         Paul Sture <nospam@sture.homeip.net> writes: > 7 > > And no, swapping cables around is not the answer...  > E > Perhaps you're right.  One should just insert a FreeBSD CD-ROM into A > the CD-ROM drive of the PC and begin the installation, deleting ? > Windoze partitions or just running fdisk on them. :-) :-) :-)  >   H And then find that you can't run that timesheet program, which means you don't get paid :-)  E But your previous suggestion of connecting your own laptop would have F had you chucked out by security at any place I have worked in the last< 10 or so years. At some places you could end up in jail too.   --  
 Paul Sture   ------------------------------  # Date: Fri, 27 Feb 2004 17:11:08 GMT " From:   VAXman-  @SendSpamHere.ORG6 Subject: Re: Educating HP employees on the nettiquette0 Message-ID: <00A2E0B3.3F120B27@SendSpamHere.ORG>  U In article <403F4517.1090006@MMaz.com>, "Barry Treahy, Jr." <Treahy@MMaz.com> writes:   >VAXman-@SendSpamHere.ORG wrote: > ^ >>In article <403E9B97.BC9E2C8@sture.homeip.net>, Paul Sture <nospam@sture.homeip.net> writes: >>{...snip...} >>   >>C >>>But the ***very real*** problem there is that 99% of the html or 3 >>>quoted/printable emails I receive are pure spam.  >>>      >>>  >>M >>Any email that is received here as Content-Type: */html is considered to be   >>fertilizer for the bit bucket. >>L >>I'd toss that rediculously stupid "quoted-printable" formatting (oh, to beK >>Freed of that idiocy ;) but that would mean I'd never hear from DEComHPaq 5 >>again.  Why does PLAIN OLD TEXT need to be encoded?  >>   >>H >Because the SMTP RFC is based on 7-bit text streams and if you include E >anything odd like a British pound symbol, or the Euro, or any other  J >language punctuation that is not English based, those characters require E >an 8-bit encoding at which point the MTA MUST encode the message as  G >quoted-printable.  So even if there isn't HTML, embedded graphics, MS  D >crap, there is STILL the need for folks to get over themselves and ; >except that pure, raw, 7-bit ASCII text e-mail is history!    Duh!  6 Euro and British pound symbols are NOT PLAIN OLD TEXT!   --B http://www.legacy-2000.com  for the *best* OpenVMS system securityC                             solutions that others only claim to be.  --  K VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM              5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  % Date: Sat, 28 Feb 2004 10:39:33 +0100 * From: Paul Sture <nospam@sture.homeip.net>6 Subject: Re: Educating HP employees on the nettiquette/ Message-ID: <40406FE5.FE4D84F@sture.homeip.net>   ! VAXman-, @SendSpamHere.ORG wrote:  > W > In article <403F4517.1090006@MMaz.com>, "Barry Treahy, Jr." <Treahy@MMaz.com> writes: " > >VAXman-@SendSpamHere.ORG wrote: > > ` > >>In article <403E9B97.BC9E2C8@sture.homeip.net>, Paul Sture <nospam@sture.homeip.net> writes: > >>{...snip...} > >> > >>E > >>>But the ***very real*** problem there is that 99% of the html or 5 > >>>quoted/printable emails I receive are pure spam.  > >>>  > >>>  > >>O > >>Any email that is received here as Content-Type: */html is considered to be " > >>fertilizer for the bit bucket. > >>N > >>I'd toss that rediculously stupid "quoted-printable" formatting (oh, to beM > >>Freed of that idiocy ;) but that would mean I'd never hear from DEComHPaq 7 > >>again.  Why does PLAIN OLD TEXT need to be encoded?  > >> > >>I > >Because the SMTP RFC is based on 7-bit text streams and if you include F > >anything odd like a British pound symbol, or the Euro, or any otherK > >language punctuation that is not English based, those characters require F > >an 8-bit encoding at which point the MTA MUST encode the message asH > >quoted-printable.  So even if there isn't HTML, embedded graphics, MSE > >crap, there is STILL the need for folks to get over themselves and = > >except that pure, raw, 7-bit ASCII text e-mail is history!  >  > Duh! > 8 > Euro and British pound symbols are NOT PLAIN OLD TEXT! >   C But the British pound can be plain old text if you use a proper VT. G Somewhere in the setup you can toggle the display between # and the GBP  symbol, certainly on UK models.   F When it comes to accented characters, Barry does have a point, as then: even mail between 2 VMS systems can go "quoted-printable".  H But that's still _no excuse_ for munging mails which don't require 8 bit characters.    --  
 Paul Sture   ------------------------------  % Date: Sat, 28 Feb 2004 10:51:43 +0100 * From: Paul Sture <nospam@sture.homeip.net>6 Subject: Re: Educating HP employees on the nettiquette0 Message-ID: <404072BF.5A725BDC@sture.homeip.net>   PhilThayer wrote:  > ] > JF Mezei <jfmezei.spamnot@istop.com> wrote in message news:<403E508B.D48324E9@istop.com>...    <snip>   D > Formatted email is valuable to many people.  Whether the format isF > HTML or a MS-Whatever attachment.  Whether people want to believe itF > or not MS is an industry leader in computers.  I would bet that at aG > minimum 75% of office systems are running on MS based platforms.  So, F > that means that 25% of the industry is using a non standard platform= > for their office management environments.  It't the anti-MS H > hackeristas that cause the excessive viri and worms that are currentlyH > hitting systems anc costing the industry billions in lost productivityF > not to mention profits due to the purchasing of expensive Anti-VirusG > software.  How in the world can you blame Viri and Worms on a company ) > that is setting the industry standards.  >   $ You have a fine sense of humour Sir. :-)    --  
 Paul Sture   ------------------------------  % Date: Sat, 28 Feb 2004 10:12:59 -0700 + From: "Barry Treahy, Jr." <Treahy@MMaz.com> 6 Subject: Re: Educating HP employees on the nettiquette' Message-ID: <4040CC1B.9010604@MMaz.com>    R. D. Davis wrote:  = >In article <5ee1d1b7.0402271217.8bffb98@posting.google.com>, 1 >	phillip_thayer@hotmail.com (PhilThayer) writes:  >    > F >>Doing things acording to company policy has EVERTHING to do with theA >>"best interest" of me getting a paycheck to support my family.  C >>Whether I like it or not is beside the point.  I work for my best H >>interest and not for the companies best interest.  If my input happens >>     >> > B >Then it's also in your best interest to avoid having to work withE >tools that are broken and annoying to use, like Windoze on a "locked G >down" PC that one can't configure to one's liking.  It's so refreshing C >to nuke windoze and install something useful, like FreeBSD, on the + >PC that one's employer puts on one's desk.  >  >    > F >>to help sway the company one way or another in making it's decisionsA >>then that's great.  But rocking the boat and falling out is not 5 >>something I enjoy. (Speaking from experience here.)  >>     >> > E >"Rocking the boat" does make sense if the policies and rules make no F >sense---and many do not make sense in many companies.  It does no one? >any good to have to bother with following insipid policies and F >procedures, particularly if they make your work more difficult to do. >    > E You sound like a person that has not worked for someone in some time  F because your statements are utterly ignorant!  In a world of ISO 9000 G certification and compliance, procedure and policy are mandates as are  F policies for site security clearances, let alone policy developed and D dictated for the protection of intellectual properties.  I can only F deduce from your gibberish that you are simply clueless or looking to H just agitate those on this list with topics that really have no purpose   or place on a technical group...   Barry    --    > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                            ------------------------------  % Date: Sat, 28 Feb 2004 10:16:28 -0700 + From: "Barry Treahy, Jr." <Treahy@MMaz.com> 6 Subject: Re: Educating HP employees on the nettiquette' Message-ID: <4040CCEC.7050706@MMaz.com>    VAXman-@SendSpamHere.ORG wrote:   V >In article <403F4517.1090006@MMaz.com>, "Barry Treahy, Jr." <Treahy@MMaz.com> writes: >    > ! >>VAXman-@SendSpamHere.ORG wrote:  >> >>     >>_ >>>In article <403E9B97.BC9E2C8@sture.homeip.net>, Paul Sture <nospam@sture.homeip.net> writes:  >>>{...snip...}  >>>  >>> 	 >>>        >>> D >>>>But the ***very real*** problem there is that 99% of the html or4 >>>>quoted/printable emails I receive are pure spam. >>>>     >>>> >>>>         >>>>N >>>Any email that is received here as Content-Type: */html is considered to be! >>>fertilizer for the bit bucket.  >>> M >>>I'd toss that rediculously stupid "quoted-printable" formatting (oh, to be L >>>Freed of that idiocy ;) but that would mean I'd never hear from DEComHPaq6 >>>again.  Why does PLAIN OLD TEXT need to be encoded? >>>  >>> 	 >>>        >>> I >>Because the SMTP RFC is based on 7-bit text streams and if you include  F >>anything odd like a British pound symbol, or the Euro, or any other K >>language punctuation that is not English based, those characters require  F >>an 8-bit encoding at which point the MTA MUST encode the message as H >>quoted-printable.  So even if there isn't HTML, embedded graphics, MS E >>crap, there is STILL the need for folks to get over themselves and  < >>except that pure, raw, 7-bit ASCII text e-mail is history! >>     >> >  >Duh!  > 7 >Euro and British pound symbols are NOT PLAIN OLD TEXT!  >    > G Well tell that to all of our European friends, also on this list, also  E running VMS, that their currency symbol is not a valid or legitimate  8 text character, you are thinking on this matter is very B narrow-minded...  I suspect that you could still find manual type G writers in England with old currency symbols though I would yield that  2 you would most likely never find one for the Euro.   Barry    --    > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                            ------------------------------  % Date: Sat, 28 Feb 2004 10:20:27 -0700 + From: "Barry Treahy, Jr." <Treahy@MMaz.com> 6 Subject: Re: Educating HP employees on the nettiquette' Message-ID: <4040CDDB.4090308@MMaz.com>    Paul Sture wrote:   " >VAXman-, @SendSpamHere.ORG wrote: >    > W >>In article <403F4517.1090006@MMaz.com>, "Barry Treahy, Jr." <Treahy@MMaz.com> writes:  >>     >>" >>>VAXman-@SendSpamHere.ORG wrote: >>> 	 >>>        >>> ` >>>>In article <403E9B97.BC9E2C8@sture.homeip.net>, Paul Sture <nospam@sture.homeip.net> writes: >>>>{...snip...} >>>> >>>> >>>>         >>>>E >>>>>But the ***very real*** problem there is that 99% of the html or 5 >>>>>quoted/printable emails I receive are pure spam.  >>>>>  >>>>>  >>>>>            >>>>> O >>>>Any email that is received here as Content-Type: */html is considered to be " >>>>fertilizer for the bit bucket. >>>>N >>>>I'd toss that rediculously stupid "quoted-printable" formatting (oh, to beM >>>>Freed of that idiocy ;) but that would mean I'd never hear from DEComHPaq 7 >>>>again.  Why does PLAIN OLD TEXT need to be encoded?  >>>> >>>> >>>>         >>>>I >>>Because the SMTP RFC is based on 7-bit text streams and if you include F >>>anything odd like a British pound symbol, or the Euro, or any otherK >>>language punctuation that is not English based, those characters require F >>>an 8-bit encoding at which point the MTA MUST encode the message asH >>>quoted-printable.  So even if there isn't HTML, embedded graphics, MSE >>>crap, there is STILL the need for folks to get over themselves and = >>>except that pure, raw, 7-bit ASCII text e-mail is history! 	 >>>        >>>  >>Duh! >>8 >>Euro and British pound symbols are NOT PLAIN OLD TEXT! >> >>     >> > D >But the British pound can be plain old text if you use a proper VT.H >Somewhere in the setup you can toggle the display between # and the GBP  >symbol, certainly on UK models. > G >When it comes to accented characters, Barry does have a point, as then ; >even mail between 2 VMS systems can go "quoted-printable".  > I >But that's still _no excuse_ for munging mails which don't require 8 bit 
 >characters.   >  >    > G I would agree, no MTA should quote for just the hell of it, and in the  I case of our systems, our MTA's are Process's SMTP on VMS and Sendmail on  D Linux systems and we have never, ever, had a message mangled for no G cause, it is ALWAYS because a character exists in the body of the text  I that is outside the standard 7-bit ASCII character definition, once that  E ASCII character reaches 128 (decimal) or greater it must be encoded,  # that is standard in the SMTP RFC...      Barry    --    > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                            ------------------------------  % Date: Sat, 28 Feb 2004 18:33:59 +0100 " From: Didier Morandi <no@spam.com>% Subject: Re: Hobbyist license for VMS 4 Message-ID: <4040d12b$0$21676$636a15ce@news.free.fr>   BobbydaMong wrote: > Howdy there... > I > Have joined the DECUS (Encompass) group a few weeks ago, with a view to ' > getting a VAX/VMS Hobbyist license...  > N > Received my number OK, but when trying to log in to www.montagar.com, I justM > get a message saying that it couldn't find my Encompass user / reference...  > L > Is anyone else having these sort of probs - does anyone know how I can getN > round this. Have tried mailing the contact mail address 3 times now (as well4 > as Encompass twice), but have yet had no response.   Welcome to the Club!  Q Montagar is supposed to receive mew membership numbers every month or so, except  N when local DECUS organizations do not know about that process, as it occurred  for France a few months ago.   You have three options:    a) ask Montagar to trust you  F b) ask your local DECUS Board to send to Montagar their montly records   c) wait.   D. --  2 VAXUS - Your new helpful friend in the DEC Family!2 EHQ: 19 chemin de la Butte, 31400 Toulouse, France/       Phone: +336 7983 6418 Fax: +335 6154 1928 $                 http://www.vaxus.org   ------------------------------  % Date: Fri, 27 Feb 2004 09:23:03 -0800 3 From: Alan Frisbie <Usenet01REMOVE@Flying-Disk.com> ( Subject: Re: Identify graphics on XP1000. Message-ID: <403F7CF7.8050608@Flying-Disk.com>   Vance R. Haemmerle wrote: < >  How does one identify which graphics card is in an XP1000= > without taking it out and seeing what part number is on the  > card?   = Some of them have a part number *lightly* stamped next to the ; video connector.   Sometimes there is a paper lable next to : the connector.   For instance, mine has a paper lable that/ says "SN-PBXGD-AD".   This is a PowerStorm 300.   > If you can take off the cover of the system (three thumbscrews= is all it takes), you can tell a lot by the size of the card. < The PowerStorm cards are full length (12+ inches), while the ATI and ELSA cards are short.   : Also, when the system is first powered up, some video BIOS< code displays the card information on the screen for a brief6 moment.   This is where a camera comes in handy.   :-)7 Just remember that this is the OEM information, not the ; DEC/Compaq/HP card info.   For instance, the PowerStorm 300 	 displays:         VGA BIOS Version 2.20B 0      Copyright (C) 1990-1998 Elpin Systems, Inc.      All rights reserved. 5        EVANS & SUTHERLAND   REALimage 2100/IBM RAMDAC    Followed a moment later by:         BIOS Emulation V1.15A:      Copyright (c) 1993-1995 Digital Equipment Corporation      All Rights Reserved      Patent Pending    Alan   ------------------------------  % Date: Sat, 28 Feb 2004 12:20:22 -0500 * From: "Bill Todd" <billtodd@metrocast.net>& Subject: Re: Intel releases 64bit Xeon2 Message-ID: <Mtedneup5uEoUN3d4p2dnA@metrocast.net>  8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:k$SWacO8XQd8@eisner.encompasserve.org... @ > In article <AoKdnTfh5v8rAaLd4p2dnA@metrocast.net>, "Bill Todd"  <billtodd@metrocast.net> writes: > >  > A > >> I still don't believe that the balance of the data out there F > >> sufficiently supports the claim that the ratios would hold in the > >> lesser-end systems. > > H > > Well, the best 32-processor Itanic TPC-C result is only a *bit* over halfH > > that of the new 32-processor POWER4+ result.  And at about twice the $/tpmCD > > at that (you actually have to shell out more absolute dollars to purchaseF > > the 577K tpmC Itanic system than you have to for the 1025K POWER4+ system). > >  >  >  > So?   K 'So', it's hardly any wonder that Itanic is still wallowing in the shallows K with minimal market share, Rob:  it just doesn't compete well in some areas : at all, and at best only equals its competition in others.  =   In a current situation, I can't help but sit and grin.  One = > quote will be several hundred thousand cheaper than another   K When you're talking $6 million systems, Rob, a few hundred thousand may not  matter all that much.    .  We E > will see a lot of posturing but at the end of the day, one solution F > clearly wins the other clearly loses.  Biggest factor?  Cost.  Money > is important.   K That's why getting nearly twice the performance for somewhat *less* capital E outlay (in the above example) matters, Rob.  Or, in the case of the 1 E million+ tpmC systems, why people might care that the IBM system that . out-performs HP's costs about $1 million less.   > A > When/if Monticeto further pulls alongside (better tpmC/CPU with  > less memory)  I Even your most optimistic projections didn't seem to claim that Montecito K would anywhere nearly *double* Madison's performance, Rob - and that's just I comparing against today's POWER4+, rather than against POWER5 (which will H appear before Montecito does).  Just what bodily orifice are you pulling! this ludicrous suggestion out of?    - bill   ------------------------------  % Date: Sat, 28 Feb 2004 12:24:33 -0500 * From: "Bill Todd" <billtodd@metrocast.net>& Subject: Re: Intel releases 64bit Xeon2 Message-ID: <YrGdnajkW90tU93dRVn-uQ@metrocast.net>  3 "Rick Jones" <foo@bar.baz.invalid> wrote in message + news:9uP%b.1266$PF1.725@news.cpqcorp.net... + > Bill Todd <billtodd@metrocast.net> wrote: H > > Well, the best 32-processor Itanic TPC-C result is only a *bit* over5 > > half that of the new 32-processor POWER4+ result.  > & > Keep going down the processor count.  K Last time I looked, IBM hadn't submitted TPC-C systems with lower processor K counts.  Though (for reasons that I've already described elsewhere) one can J reasonably infer that a 4-CPU top-end POWER4+ configuration, if submitted,J would handily beat the best 4-CPU Itanic platform, though perhaps not by a factor approaching 2.    - bill   ------------------------------  % Date: Fri, 27 Feb 2004 15:46:02 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> C Subject: Re: It is almost certain now, INTEL will have 64bit x86 !! 0 Message-ID: <c1nont$ra3$2@new-usenet.uk.sun.com>   Bob Koehler wrote:[ > In article <wDs%b.1184$CG.424@news.cpqcorp.net>, Rick Jones <foo@bar.baz.invalid> writes:  > ! >>Dirk Munk <munk@home.nl> wrote:  >>: >>>Another example? Tru64 is the *only* real 64 bit Unix.  >>, >>What makes something a "real" 64 bit Unix? >  > F >    Mostly the marketing department.  There are, of course, technicalB >    facts to back it up.  Tru64 uses a flat 64 bit address space. >     & And No other UNIX does ???????????????   Regardsc Andrew Harrison    ------------------------------  % Date: Fri, 27 Feb 2004 15:42:45 +0000XO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>XC Subject: Re: It is almost certain now, INTEL will have 64bit x86 !!t0 Message-ID: <c1nohp$ra3$1@new-usenet.uk.sun.com>   Rick Jones wrote:p< > Andrew Harrison <andrew_._remove_harrison@su_n.com> wrote: >  >>Rick Jones wrote:n >>D >>>The tradeoffs between CPU/Memory/IO sound rather like the SunFire >>>systems.  >  > = >>No they are not, A SunFire system can have 72 CPU's, 576 GBS7 >>or RAM and 18 I/O controllers without any tradeoffs. b >  > ? > Perhaps we are talking about different variants of tradeoffs.N > G > What about the "MaxCPU" stuff - if I've read the stuff I've been ableiG > to find thusfar correctly, those go into IO slots.  Doesn't that mean H > that if one wants Maximum CPU count on the 15K one has to give-up I/O? >   ; That allows you to have more than 72 CPU's in a F15K if youw> want to lose I/O slots. You seem to have missed the difference> between this an 8400 which was that you had CPU boards, memory: boards and I/O boards. If you wanted the maximum number of9 CPU's you could not have the maximum amount of memory etc  or I/O  / In the F25K the maxcpu option no longer exists.t   > > >>>If they can actually then drive that I/O out the PCI slots. >  > @ >>Of course but then as you would also know Sun has demonstratedA >>12.5 GB/s on a table scan from an Oracle DBMS sustained throughn6 >>the older I/O subsystem for the F15K what can you do >  > G > The E15K specs sheet on www.sun.com claims the IO is "up to 21.6 GB/s8 > sustained" >   > As it is and as you will also notice this is not the number of9 PCI slots multipled by indevidual bandwidth of the slots.d  : So what sort of I/O rate can you sustain through an Oracle DBMS on a SuperDome ?O  4 You seem to have spent an inordinate ammount of time6 not answering the question, the fruits of which I have cut    <snip>  F > So, no, I am not intimately familiar with the 8400, but the high-endD > SunFire servers certainly seem to have their share of tradeoffs.   >   D Again this is incorrect. The E25K and E20K do not have any tradeoffsC and the tradeoffs in the F12K and F15K are not like the ones in then 8400.i   Regardsl Andrew Harrisonl   ------------------------------  % Date: Sat, 28 Feb 2004 12:27:19 -0500r* From: "Bill Todd" <billtodd@metrocast.net>C Subject: Re: It is almost certain now, INTEL will have 64bit x86 !!.2 Message-ID: <6vOdnQK9l9_LUt3dRVn-ug@metrocast.net>  5 "Jack Peacock" <peacock@simconv.com> wrote in message , news:SMKdnc1AZpfkfaLdRVn-iQ@mpowercom.net...I > > > In article <vzq%b.127388$jk2.539633@attbi_s53>, glen herrmannsfeldt   > <gah@ugcs.caltech.edu> writes:= > > > But according to Gartner - 64-bit extended x86 won't be = > > > around forever.  Which is big 64-bit environment in thep
 > > > future?r > > > J > Just out of curiosity, where did Gartner predict Itanium would be today,J > from a perspective of 1, 2, or 5 years ago?  How many systems installed?G > How many vendors shipping Itanium systems in larger than single digithL > quantities?  Does anyone bother to check on Gartner predictions to see how > accurate they are over time? >gK > In the casino business if a handicapper's predictions don't work out over:D > time then they get shown the door.  It seems that in the IT pundit businessJ > it's more like a race tout than a real handicapper...make up a differentK > story for every client then hope a profitable percentage come through, or C > hope the client forgets about the report they paid big bucks for.   I Well, I don't have time to dredge it up right now, but my recollection is E that back around Y2K IDC projected that Itanic systems would be a $20nK billion market in 2004.  Shouldn't be that difficult to find a reference toh that.r   - bill   ------------------------------  # Date: Fri, 27 Feb 2004 13:41:52 GMTe/ From: "Jeff Goodwin" <jgoodwin@maine.rrr-r.com> 8 Subject: Re: KGPSA-C and EMC DS-24 M2 SAN Switch Problem8 Message-ID: <AOH%b.94397$%72.55016@twister.nyroc.rr.com>  0 "Mike Naime" <mnaime@kc.rr.com> wrote in message3 news:9Wy%b.28020$LD5.24776@twister.rdc-kc.rr.com...  >., > Hey Phil.   Hello...   Long time no see... >2E > You might have the cables swapped.  Some Fibre cables come from the K > manufacturer in a "twisted" fashion.  You may need to physically swap oner= > pair of the cables in order to get the link lights to work.s  L Good point Phil.  I've purchased some non-HP cables that were swapped and itF was just about impossible to reverse them due to the connector design.H Subsequent purchases have been made only for cables that have individual connectors I could reverse..   -Jeff    ------------------------------  % Date: Sat, 28 Feb 2004 09:07:25 -0800s# From: "Tom Linden" <tom@kednos.com>   Subject: RE: Power Button on PWS9 Message-ID: <NDEMLKKEBOIFBMJLCECIAEEOCOAA.tom@kednos.com>a  J It appears to be a momentary switch.  I looked at the wires and I couldn'tD conclude anything from that ther were 6 with a carbon choke on them.  E I tried holding  the switch down, but that didn't work which is why I  concluded that it was a momentary.i  G Not sure why anyone would make a switch like that.  I see no functional  reason( for it, but I am sure there must be one.     -----Original Message-----5   From: Keith A. Lewis [mailto:lewis@mazda.mitre.org] *   Sent: Friday, February 27, 2004 12:40 PM   To: Info-VAX@Mvb.Saic.Coml"   Subject: Re: Power Button on PWS    1   "Tom Linden" <tom@kednos.com> writes in articlenA   <NDEMLKKEBOIFBMJLCECIIEDHCOAA.tom@kednos.com> dated Fri, 27 Febi   2004 08:54:41 -0800:B   >I have three of these, 2 running VMS and 1 running Tru64.  Upon9   >reestablishing power after outage the one running Unixr   automaticvally comes?   >up, the other two I have to manually depress the switch.  Isr   there perhaps a    >BiosC   >setting to correct this?d  L   There are 2 kinds of computer power switches.  One is a double throw whichL   actually cuts off the power to what's behind it, and the other is a singleC   throw which merely sends a signal to the mobo when it is pressed.     It shouldJ   be easy to tell by looking at the size of the wires going to the switch.  H   If your PWS remembers whether it was on or off when power returns, the@   switch is probably DT.  Otherwise it's ST.  My PWS 500au boots   back up intoC   VMS after a power outage, with no modification to the default SRM1   variables    except AUTO_ACTION.p  J   Is there a breaker switch on the back?  If you decide you don't need theA   functionality of the switch on the front.  You might be able to9
   wedge it incJ   the "depressed" position, or open up the box and bypass it electrically.  2   --Keith Lewis              klewis {at} mitre.org@   The above may not (yet) represent the opinions of my employer.     ---e(   Incoming mail is certified Virus Free.<   Checked by AVG anti-virus system (http://www.grisoft.com).B   Version: 6.0.593 / Virus Database: 376 - Release Date: 2/20/2004   ---a& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.593 / Virus Database: 376 - Release Date: 2/20/2004   ------------------------------  % Date: Sat, 28 Feb 2004 10:04:38 +0100r, From: "Hans Vlems" <hvlems.dotweg@zonnet.nl>4 Subject: Re: vms 552 + timeline software experience?: Message-ID: <c1plif$1l3kma$1@ID-143435.news.uni-berlin.de>  , "sal" <s1leone@yahoo.com> schreef in bericht7 news:a9d652c2.0402270921.6bf1fe7e@posting.google.com...MH > I have been tasked with a disaster recovery procedure and am restoringB > timeline software BCK to a VMS 552 Vaxstation 4000/90. TechnicalG > documentation regarding this ancient financial software (Timeline fort > VMS) is nill to be found.u >cA > It may be a long shot, but I am wondering: does anyone have any < > experience or recollection of the timeline software setup?H > I am running @tllogical.com, @tlsymbol.com and "menu" but getting sent2 > back to a prompt when the menu attempts to load. >n >w > sal   K The way you describe it, your only solution seems reverse engineering. Do anA SET HOST/LOG 0, log on again and SET VERIFY. Next run the command 3 procedures. Log off and study the SETHOST.LOG file. I My guess would be that some component is not in INSTALL. If there are any H TL*.EXE files in SYS$SHARE or SYS$SYSTEM you might try and INSTALL them, /HEADER/OPEN/PRIV.H Assuming that you have the VAXstation all for yourself. You might either; upset others or give them an, err, interesting opportunity.    Hans   ------------------------------    Date: 27 Feb 2004 09:21:55 -0800 From: s1leone@yahoo.com (sal)e0 Subject: vms 552 + timeline software experience?= Message-ID: <a9d652c2.0402270921.6bf1fe7e@posting.google.com>   F I have been tasked with a disaster recovery procedure and am restoring@ timeline software BCK to a VMS 552 Vaxstation 4000/90. TechnicalE documentation regarding this ancient financial software (Timeline for2 VMS) is nill to be found.   ? It may be a long shot, but I am wondering: does anyone have anyo: experience or recollection of the timeline software setup?F I am running @tllogical.com, @tlsymbol.com and "menu" but getting sent0 back to a prompt when the menu attempts to load.     salw   ------------------------------  % Date: Fri, 27 Feb 2004 10:27:09 +0100 " From: Didier Morandi <no@spam.com>4 Subject: Re: VMS Hobbyist License program abolished?4 Message-ID: <403f0d8f$0$22391$626a14ce@news.free.fr>  J To me, they got confused and understood that there was a way to go from a P commercial license to a free one via a so-called "hobbyist" process. I bet this I person does not know about the Hobbyist program which has, of course, no f/ relationship whatsoever with HP licenses sales.t   D.   Alan Frisbie wrote:s   > DAVID TURNER wrote:e > 0 >> Sounds to me that they are completely wrong ! >  > = > I hope so, but it's awfully hard to argue with the officialt > HP spokesperson. >  > Alan >    --  2 VAXUS - Your new helpful friend in the DEC Family!2 EHQ: 19 chemin de la Butte, 31400 Toulouse, France/       Phone: +336 7983 6418 Fax: +335 6154 1928 $                 http://www.vaxus.org   ------------------------------    Date: 27 Feb 2004 06:39:18 -0800. From: al5vf03p02@sneakemail.com (William Webb)4 Subject: Re: VMS Hobbyist License program abolished?= Message-ID: <d5ce4b06.0402270639.5b5152b1@posting.google.com>l  g Alan Frisbie <Usenet01REMOVE@Flying-Disk.com> wrote in message news:<403E7B05.20403@Flying-Disk.com>...a< > I just had an e-mail exchange with HP's licensing contact,2 > Madi Braje, as listed on the licensing web page: > 7 >    http://licensing.hp.com/swl/view.slm?page=contactst >  >    Madi Braje - West Regiony >    Telephone: 847-781-6756 >    Email: Madi.Braje@hp.comn > = > I was trying to see if I could get the original VMS licenset? > for the XP1000 I bought transferred to me.   Since I am goinge? > to be using it to play with Apache, etc., I mentioned that ift: > the license couldn't be transferred I would just use the? > hobbyist license.   However, it now appears that the hobbyist!3 > license program has been abolished.   Madi wrote:t > C >  > Digital/Compaq/HP has never distinguished between business andrD >  > hobbyist, so the license that transfers with the system, is theF >  > only type available for that particular system.  There is no such& >  > thing as a free hobbyist license. >  > R.I.P. >  > Alan    
 R.I.P.?  NOT!n  > I asked Mr. Gorham's permission to post this before doing so;   B I took the initiative of removing the email and phone information / to prevent it from being harvested by spambots.   < And, for those of you who, despite extensive evidence to the? contrary, still maintain the irrational position that HP isn't 0? listening or doesn't care about VMS, please take note that the t? elapsed time between my email and the response which corrected t, the misinformation was less than 15 minutes.   WWWebb   -----Original Message-----$ From: Gorham, Mark [mailto:REDACTED]' Sent: Friday, February 27, 2004 8:25 AM G To: Webb, William W - Raleigh, NC; Skonetski, Susan; Stallard, Scott J;L Smith, John (VMS - Govt)" Cc: Braje, Madi; Vazquez, MaryJaneF Subject: RE: This HP *licensing* point of contact does not know of the. VMS Hobbyist license program - please correct.     Thanks for the mail William.  G Madi, MJ can help you with the hobbyist program info.  We definitely doiA distinguish between business and hobbyist use and have a thrivings hobbyist program.7   Regardse Mark   Mark Gorham   Vice President   OpenVMS Systems Division REDACTED   -----End Original Message-----   ------------------------------  # Date: Fri, 27 Feb 2004 15:29:17 GMTu( From: "konabear" <maurert@ameritech.net>4 Subject: Re: VMS Hobbyist License program abolished?? Message-ID: <hnJ%b.6260$t16.4328904@newssvr28.news.prodigy.com>l  E Didn't Shakespeare write a play about this... Much Ado about Nothing?   * You beat me to the E-mail chained William.   Todd; "William Webb" <al5vf03p02@sneakemail.com> wrote in messagei7 news:d5ce4b06.0402270639.5b5152b1@posting.google.com... @ > Alan Frisbie <Usenet01REMOVE@Flying-Disk.com> wrote in message( news:<403E7B05.20403@Flying-Disk.com>...> > > I just had an e-mail exchange with HP's licensing contact,4 > > Madi Braje, as listed on the licensing web page: > >m9 > >    http://licensing.hp.com/swl/view.slm?page=contactsm > >e > >    Madi Braje - West Region  > >    Telephone: 847-781-6756 > >    Email: Madi.Braje@hp.com0 > >a? > > I was trying to see if I could get the original VMS licenseSA > > for the XP1000 I bought transferred to me.   Since I am goingmA > > to be using it to play with Apache, etc., I mentioned that if.< > > the license couldn't be transferred I would just use theA > > hobbyist license.   However, it now appears that the hobbyist 5 > > license program has been abolished.   Madi wrote:d > > E > >  > Digital/Compaq/HP has never distinguished between business andbF > >  > hobbyist, so the license that transfers with the system, is theH > >  > only type available for that particular system.  There is no such( > >  > thing as a free hobbyist license. > >O
 > > R.I.P. > >e > > Alan >> >  > R.I.P.?  NOT!S > ? > I asked Mr. Gorham's permission to post this before doing so;  > C > I took the initiative of removing the email and phone informationu1 > to prevent it from being harvested by spambots.r >e> > And, for those of you who, despite extensive evidence to the@ > contrary, still maintain the irrational position that HP isn't@ > listening or doesn't care about VMS, please take note that the@ > elapsed time between my email and the response which corrected. > the misinformation was less than 15 minutes. >p > WWWebb >a > -----Original Message-----& > From: Gorham, Mark [mailto:REDACTED]) > Sent: Friday, February 27, 2004 8:25 AM I > To: Webb, William W - Raleigh, NC; Skonetski, Susan; Stallard, Scott J;  > Smith, John (VMS - Govt)$ > Cc: Braje, Madi; Vazquez, MaryJaneH > Subject: RE: This HP *licensing* point of contact does not know of the0 > VMS Hobbyist license program - please correct. >: >3 > Thanks for the mail William. >sI > Madi, MJ can help you with the hobbyist program info.  We definitely doeC > distinguish between business and hobbyist use and have a thrivinge > hobbyist program.  >a	 > RegardsF > Mark >6 > Mark Gorham   Vice President > OpenVMS Systems Division
 > REDACTED >S  > -----End Original Message-----   ------------------------------  % Date: Fri, 27 Feb 2004 07:46:51 -0800.3 From: Alan Frisbie <Usenet01REMOVE@Flying-Disk.com>t4 Subject: Re: VMS Hobbyist License program abolished?. Message-ID: <403F666B.6050205@Flying-Disk.com>   William Webb wrote:   @ > I asked Mr. Gorham's permission to post this before doing so;   & > From: Gorham, Mark [mailto:REDACTED]  H > Subject: RE: This HP *licensing* point of contact does not know of the0 > VMS Hobbyist license program - please correct.   > Thanks for the mail William.  I > Madi, MJ can help you with the hobbyist program info.  We definitely docC > distinguish between business and hobbyist use and have a thrivingr > hobbyist program.d  ? Well, *THAT* is a relief.   Now I wonder if the other things heyA told me about license transfers were equally wrong: That I cannota? get the original license transferred unless I have the originalT= PAKs and a transfer form signed by the original owner.   Thisi? would be rather difficult, as it appears the original owner wasdC a victim of the dot com crash.   To make it even worse, the surplusi@ outlet I bought it from had installed Tru64 to wipe the disk andA prove that the system worked, so I couldn't even get the originaln" VMS PAK information from the disk.  A I really wanted to get it transferred so that I could sign up foreC VMS update service and get real support, as opposed to just postingo@ in comp.os.vms and hoping for the best.   Although, I am awfully" impressed with the responses here.   Alan   ------------------------------  # Date: Fri, 27 Feb 2004 15:03:39 GMTe# From: "John Smith" <a@nonymous.com>u4 Subject: Re: VMS Hobbyist License program abolished?H Message-ID: <f%I%b.35612$Qg7.25426@news04.bloor.is.net.cable.rogers.com>   William Webb wrote:m >y > I promise you this:  >s! > Although I *don't* work for HP-m >.& > 1. Sue Skonetski shall hear of this.$ > 2. Mark Gorham shall hear of this.* > 3. Scott Stallard(1) shall hear of this.& > 4. John Smith(2) shall hear of this. >lF > And I have complete faith that most, if not all of the people I justD > mentioned will inform this person of the existence, as well as the> > continued viability of the OpenVMS Hobbyist License Program. >f > WWWebb >r" > Not sticking my ususal tag here.$ > firstnamelastname at juno dot com. > 6 > (1) He *is* supportive of VMS- don't get me started!    K Re: your (1) above, as the senior or next to senior guy in the VMS specificvI group at HP, I just don't get it as to how 'supportive' doesn't translateyG into proactive, visible and sustained promotion of VMS to all manner ofaH potential customers. I don't know Mark other than through e-mail we haveI exchanged with one another from time-to-time, and I appreciate that he is C but a cog (a reasonable sized one) in a machine which has policies.   L I hear of his one-on-one efforts with current customers and the *occasional*J new customer face-to-face, and his  'keep the faith' messages, but it getsJ pretty thin when there is no advertising & promotion other than .pdf files from HP.  L I can only imagine that Mark's keeping the local Zantac and Tagamet suppliesL low in area pharmacies as his stomach lining devours itself at the stupidityC of HP's non-advertising/non-marketing policies with respect to VMS.n   ------------------------------  % Date: Sat, 28 Feb 2004 10:00:07 +0100t* From: Paul Sture <nospam@sture.homeip.net>I Subject: Re: VMS upgrade changed program behavior, what could cause this? 0 Message-ID: <404066A6.5C66EF29@sture.homeip.net>   Lawrence Bleau wrote:s > K > Okay, folks, I have more info.  First, toss aside all the arguments aboutoL > using extended precesion, or even about summing values; I produced the odd  > behvior with a simple program. > G > Here's the short of it: It seems that on umace (my older system still>G > runining VMS 7.1-2 and therefore an older LIBRTL) rounding of a valuehF > ending with .5 is down upwards except when the value is in the rangeG > 100.5-999.5, in which case it's rounded down.  On umtof (the upgraded 3 > system running 7.3-1) rounding is always upwards.a >  > Here's the test I did: > L > I wrote a Fortran program (ROUND2) to do the roundoff test.  You give it aR > starting value (integer) and number of iterations.  It loops that many times andQ > in each iteration generates the number   x = start + i + .5   so we get a largeeQ > range of values, all at .5.  It then does an internal WRITE of the value twice:OP > once with F15.5 and once with F10.0.  It then compares the first 10 charactersP > (the integer part) of each resultant string value.  If they are the same, thenQ > the LIBRTL routine rounded down; if different, it rounded up.  It outputs thoseiM > cases that round down.  (Btw, an internal WRITE calls the same conversation N > routine as an external WRITE, so we're getting the same behavior as if every > line was written.) > J > I ran ROUND2 over a large range of positive values.  Here's the results: > = >         value of x              UMACE (old)     UMTOF (new) < > in the range 0.5 - 99.5         rounded down    rounded up< > in the range 100.5 - 999.5      rounded up      rounded up< > in the range 1000.5 and higher  rounded down    rounded up > O > I didn't check values over 1E6; 1E6 itself (that is, 1000000.5) rounds up.  A@M > spot check of negative values shows they behave the same as their poisitive@Q > counterparts if you replace "down" with "towards zero" and "up" with "away fromp > zero" in the above table.  > P > So it seems there's a flaw in the 7.1-2 rounding routine that treats the rangeI > 100.5-999.5 differently than other values wrt rounding of .5, while thehC > 7.3-1 rounding routine is consistent in that it always rounds up.d > / > I *do* wish this had been documented, though.  > ; > If you want to test this yourself, here's my source code:o >  > C ROUND2.FORH > C Program rounds off values that always end in .5, detects those cases. > C where rounding is down down instead of up.$ >         CHARACTER STR1*15, STR2*10 > C 0 > 10      PRINT *,'Enter start value, # to test'" >         READ(*,*,END=90) START,N >         IF (N.LE.0) STOP >         DO I=1,N" >           VALUE = START + I + .5! >           WRITE(STR1,101) VALUEi! >           WRITE(STR2,102) VALUEt > 101       FORMAT(F15.5)r > 102       FORMAT(F10.0)e= >           IF (STR1(1:10) .EQ. STR2(1:10)) PRINT *,STR1,STR2e >         ENDDOc >         GOTO 10t > 90      STOP
 >         ENDe > I > If I had to theorize, I'd say that HP noticed a flaw in the floating to L > character conversion routines and corrected it, but forgot to tell anyone. >  > Can anyone confirm this? >   G Can you give me an example input to the program please? It compiles andb; links fine here, but simply loops around the input prompts.   H I'm probably doing something dumb here. Sorry - it's been too many years: since I used FORTRAN, but I am willing to test it for you.   M > Did anyone read anything in the release notes for 7.2, 7.2-x, 7.3, or 7.3-1cI > that mentioned a change to LIBRTL or to rounding behavior?  I only readtK > through the release notes for 7.3-1, since I upgraded directly there from. > 7.1-2. >         --  
 Paul Sture   ------------------------------  % Date: Fri, 27 Feb 2004 15:04:56 -0000 * From: "Richard Brodie" <R.Brodie@rl.ac.uk>I Subject: Re: VMS upgrade changed program behavior, what could cause this? + Message-ID: <c1nmap$dv6@newton.cc.rl.ac.uk>   ` "John Reagan" <john.reagan@hp.com> wrote in message news:NII%b.1226$nf1.1018@news.cpqcorp.net...  H > You'll want more than that.  Since LIBRTL is an installed known image,J > you want a semi-colon on the filename to ensure the image activator uses) > your side LIBRTL, not the installed oney >eB > $ DEFINE/USER LIBRTL disk:[dir]LIBRTL.EXE;  ! The semi is needed  I It shouldn't matter unless you are (unwisely) dropping it in SYS$LIBRARY.uG Unless something has changed recently, the image activator checks pathspK first, before preferring an installed image. Good habit to get into though.e   ------------------------------  % Date: Sat, 28 Feb 2004 18:35:15 +0000p- From: John Laird <nospam@laird-towers.org.uk>>I Subject: Re: VMS upgrade changed program behavior, what could cause this?e8 Message-ID: <bmn1401t59rrbuo87gbc4f634febc1kq45@4ax.com>  H On Sat, 28 Feb 2004 10:00:07 +0100, Paul Sture <nospam@sture.homeip.net> wrote:  H >Can you give me an example input to the program please? It compiles and< >links fine here, but simply loops around the input prompts.  J That is expected behaviour, Paul.  You will only get output if the rtl has; started incorrectly rounding down (as Lawrence discovered).m    A rather nasty little bug, imho.   --  ? You're growing old when your knees buckle and your belt won't.     Mail john rather than nospam...    ------------------------------  % Date: Sat, 28 Feb 2004 11:35:47 +0100i" From: <nicovanderboom@hotmail.com>; Subject: Re: Why does smtp mail between local systems fail?e> Message-ID: <40406e73$0$12846$3a628fcd@reader1.nntp.hccnet.nl>  7 "Lawrence Bleau" <bleau@umtof.umd.edu> wrote in messageu* news:c1gd1a$das$1@grapevine.wam.umd.edu...G > I'm having a problem with my TCPIP package, and I suspect it's due top aVG > recent upgrade.  I'm running OpenVMS AXP V7.3-1, TCPIP V5.3 on systemG? > umtof.umd.edu .  I have three test systems with the followingt > configurations:y >e% >                 VMS      TCPIP(UCX)t > umtof2.umd.edu  V7.3-2   V5.4-% > umace.umd.edu   V7.1-2   V4.2 ECO 1 % > uleis.umd.edu   V7.1-2   V4.2 ECO 1e >r> > What I'm observing is that email from the two test systems I
 administerE > that haven't been upgraded that is sent to the problem system usinge smtpF > seems to get stuck in the sending system's mail queue.  It goes into: > Processing state, stays there a while, then is requeued. > G > These two systems are next to each other on the same LAN!  A UCX PING7 from7 > each to the other returns immediately with no errors.e >eH > Email sent from umtof2.umd.edu, however, which runs the same or higherF > version of TCPIP, works nicely; it gets delivered within a minute or so.m >:D > No problem sending smtp email between the test systems, btw, works > immediately. >e <snip.>C  @ We noticed some time-out problems too, after upgrading to TCPIP. (eg. telnet)H it was due to the fact that de default configuration of TCPIP is waiting% for a reverse DNS lookup to complete.:D TELNET does this to record the remote system by name in the logfile.? during a DNS problem we experienced total failure of LAN telneti	 sessions.U  ) Maybe incoming SMTP tries to do the same.iG Can you check if all the systems are registered ok in DNS or local hostr table?  G The telnet problem can be solved with logical name definitions. (log ipt adress instead of host name)  
 good luck. Nico van der Booms   ------------------------------  % Date: Sat, 28 Feb 2004 10:52:02 +0100 , From: "Hans Vlems" <hvlems.dotweg@zonnet.nl>< Subject: Re: [GAME] you've been appointed VMS Mkt Mgr, so...: Message-ID: <c1pobb$1loahg$1@ID-143435.news.uni-berlin.de>  1 "Didier Morandi" <no@spam.com> schreef in berichtn. news:403f1255$0$22407$626a14ce@news.free.fr... > P*o*e*t*sw > Let's relax a bit. >tK > You have been recently appointed HP OpenVMS Marketing Manager and you arek inK > charge, together with MaryJane V., Sue S. and Hoff H. to promote the Nr 1  > Operating System worldwide.. >- > Congratulations. >l( > Now, what would be your first actions? >n) > [ ] retire and put Sue in your armchair ! > [ ] change the name back to VMS4( > [ ] ask Ogilvy to build an ad campaign > [ ] cancel i64 development' > [ ] restore the Alpha chip production % > [ ] start VMS/Athlon-64 engineeringOL > [ ] check if this is true that there are still PDP's running in production> > [ ] go and visit Customers with a 64 slides PPT presentation? > [ ] restore DECworld for VMS and build the greatest show everiH >      (if you choose this option, please select Cannes as the location. Thanks)mL > [ ] suddently wake up and think that all this was just a(nother) nightmare > [ ] other (please comment) >    Didier  G this is a very simple task: start the ad campaign first. Wait until thehI money comes in and then work down the list. At that point I'd skip item 1r= though. I like Sue but rather would want this job myself ....e   Hans   ------------------------------   End of INFO-VAX 2004.117 ************************