1 INFO-VAX	Thu, 11 Sep 2003	Volume 2003 : Issue 503       Contents:+ Re: "HP breakup on the way" - Merrill Lynch + Re: "HP breakup on the way" - Merrill Lynch + RE: "HP breakup on the way" - Merrill Lynch + Re: "HP breakup on the way" - Merrill Lynch + Re: "HP breakup on the way" - Merrill Lynch + Re: Backup foreground/background operations  Block size in show dev?  Re: Block size in show dev?  Re: Block size in show dev?  Re: Block size in show dev?  Re: Block size in show dev?  Re: Block size in show dev? % Re: Choosing between ASTs and Threads % Re: Choosing between ASTs and Threads % Re: Choosing between ASTs and Threads % Re: Choosing between ASTs and Threads % Re: Choosing between ASTs and Threads  Re: Do we need a DCL debugger? Re: Image tools for VMS # Re: Image tools for VMS - follow up & Re: Mac OS X -> Pathworks/Mac problems Re: MESSAGE problem  RE: MESSAGE problem ' Re: Moving a Tk70 from 3400 to 4000/600 + Need better SMTP RBL blocking error message  Re: Need HP Reseller in Germany ' Re: New linker ECOs for V7.3 and V7.3-1 L Re: OpenVMS clusters (Alpha) - standalone workstation joined to the cluster?P Re: OpenVMS clusters (Alpha) - standalone workstation joined to the cluster? cluP Re: OpenVMS clusters (Alpha) - standalone workstation joined to the cluster? clu OpenVMS I64 V8.0 test drive  Re: OpenVMS I64 V8.0 test drive = Re: Runtime linking of Java to AWT graphics libraries problem , Re: Solaris Security (was: OpenVMS Security) Re: VMS Upgrade Needed VMS Upgrade Needed Re: VMS Upgrade Needed Re: VMS Upgrade Needed Re: VMS Upgrade Needed Re: VMS website updated. Re: VMS website updated. Re: VMS website updated. Re: VMS website updated. Re: VMS website updated.  F ----------------------------------------------------------------------  # Date: Wed, 10 Sep 2003 17:50:33 GMT & From: Rick Jones <foo@bar.baz.invalid>4 Subject: Re: "HP breakup on the way" - Merrill Lynch2 Message-ID: <JvJ7b.4401$JE7.3582@news.cpqcorp.net>  P Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote:? > The bug is a problem for non IP network traffic that does not  > do checksumming on blocks.  F I take it then the data corruption happens after the GbE chip does theB Internet checksum offload?  Or is Sun not taking advantage of that" feature of the Broadcom GbE chips?  7 > On an affected V240 (not all the servers manufactured < > use the batch of NICs with the manufacturing defect) using7 > IP this results in ~0.3% retransmission rate which is  > pretty much background noise.   C Yes and no.  For bulk data transfer with a "large enough" window to C still get fast RTX and such, 0.3% is not going to bring things to a D screaching halt.  However, for stuff where the windows are not largeF enough to get fast RTX, or the connection latency is large (high-speed> WAN where you need to keep cwnd large) then 0.3% can indeed be plusungood.   C Some old data showing the effect on bulk transfer of having to wait 6 for RTX timeouts at various RTX rates can be found at:  = ftp://ftp.cup.hp.com/dist/networking/briefs/SE_372_2_99.ps.gz   < Which showed that for a 0.01% loss rate the timeout-only TCPE connection lost 20% performance, going to a loss of about 65% at 0.1%  and 95% by 1%.  @ When the connection was able to fast RTX, a 0.1% throughput held: steady (with a 32KB window), and was down about 55% at 1%.  D Also, while 0.3% might be background noise for an Internet connectedD server, for something in the department or intranet that is going to$ be quite large comapred with normal.  C Lets assume we have a synchronous request/reply application going - D say very much like single instance of netperf TCP_RR :).  A UP J5000E (440 MHz PA-8500) can do roughly 13100 single-byte TCP_RR's through a F J6825A GbE NIC (Broadcom like the V2[14]0, but 5701 rather than 5704).6 That suggests an RTT of 1/13000 or ~0.08 milliseconds.  E If there were a 0.3% retransmission rate, then that RTT would become:      (.997 * 0.08) + (.003 * 500)    C or ~1.57 milliseconds per transaction, which would be a transaction  rate of ~635 or a loss of 95%.  F (If one is using the typical tcp_rexmit_interval_min. I'm ass-u-me-ingA that Solaris doesn't use the 1 second default the RFC's suggest - ! HP-UX uses 500 ms as the default)   C Now, it is perhaps unreasonable to expect real applications to have C 0.08 ms response time, and I've no idea if a V240 can achieve 13000 E TCP_RR transactions per second through one of its GbE interfaces with B a single instance of netperf, packet losses or no, so lets instead@ assume that the application RTT is 1 millisecond when the NIC is< functioning correctly.  That suggests it would be doing 1000B transactions per second without loss.  With that 0.3% loss rate we instead have     (.997 * 1) + (.003 * 500)   E or 2.497 ms response time, which means a transaction rate of ~400 per $ second or a 60% loss of performance.  9 > Of course if you are using a non IP based protocol with , > no checksumming then 0.3% is unacceptable.  C Not all IP-based protocols do retransmissions - UDP leaves it up to D the application.  NFS, DNS etc would have to do their retransmission5 things, and both are rather request/reply in nature.    
 rick jones --  = portable adj, code that compiles under more than one compiler F these opinions are mine, all mine; HP might not want them anyway... :)A feel free to post, OR email to raj in cup.hp.com  but NOT BOTH...    ------------------------------  # Date: Wed, 10 Sep 2003 18:15:50 GMT # From: "John Smith" <a@nonymous.com> 4 Subject: Re: "HP breakup on the way" - Merrill LynchI Message-ID: <qTJ7b.212756$_V.153410@news04.bloor.is.net.cable.rogers.com>   8 "Rob Young" <young_r@encompasserve.org> wrote in message- news:zvpVnRu+zgtl@eisner.encompasserve.org...  > = > It wouldn't be too much of a stretch to imagine that on the < > next go round of upgrades, Sun made the short list but IBM< > was and is being chosen.  After all, how could IBM Unix be@ > flourishing and Sun Unix floundering (considerable Sun revenueF > decline - 20% year over year)?  Afer all, CIOs read mags like Forbes > .....   G This is a Eureka moment for Rob. "CIOs read mags like Forbes", said he.   K Presumably those same CIO's are the ones who would possibly purchase VMS if K they knew anything about it. Think about it...in today's corporations there L are many CIO's who have only used Windows in their shops, or grew up on unixK at university. Maybe they ought to be told about something that works for a  change.   9 So why isn't HP advertising VMS in magazines like Forbes?    It's a simple question. & Do you have a simple answer Mr. Young?F If you don't, then perhaps some of the other HP employees who frequentK c.o.v. could convince Bob Blatz or Mark Gorham or Rich Marcello to deign us L with their presence here and offer a rational explaination as to why this is not so.   L Perhaps this is one of the reasons why the Merrill Lynch analayst wrote what= he did in his report (bringing the discussion back on-topic).    ------------------------------    Date: 10 Sep 2003 13:24:05 -0500+ From: young_r@encompasserve.org (Rob Young) 4 Subject: RE: "HP breakup on the way" - Merrill Lynch3 Message-ID: <lC36pjFfI2dS@eisner.encompasserve.org>   _ In article <CIEJLCMNHNNDLLOOGNJICECMHPAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> writes:  >  >  >>-----Original Message-----D >>From: Bob Koehler [mailto:koehler@eisner.nospam.encompasserve.org]- >>Sent: Wednesday, September 10, 2003 7:55 AM  >>To: Info-VAX@Mvb.Saic.Com 6 >>Subject: Re: "HP breakup on the way" - Merrill Lynch >> >>B >>In article <bjkomh$iha$1@new-usenet.uk.sun.com>, Andrew HarrisonC >>SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes:  >>; >>> VMS relys on Alpha and is migrating to Itanium it isn't B >>> if you include OpenVMS's share of developing and manufacturing> >>> those platforms in your cost base I very much doubt if you >>> are remotely profitable. >>H >>   Nonsense.  Alphas have never made as much profit as VMS.  The money >>   is in software, not chips.  > I > In fact, I think you might argue that Alpha in part was responsible for J > putting Digital out of business.  Taking this argument one step further,J > it would have been better to port VMS to x86 rather than Itanium.  After > all,K > the most successful OS vendor doesn't make HW, just uses it as a delivery F > vehicle.  Same problem, of course, confronts Sun, and I would not beI > surprised to see them abandon Sparc in favour volume produce gear (e.g.  > Dell)     E 	Vehicle yes.  But wrong vehicle. If it was just a vehicle, in theory ? 	Dell wouldn't be able to make any money but as we know Dell is D 	quite content to average 16% margins, keeping prices low and makingE 	tons of money due to the high volume and low overhead (Walmart model @ 	as recently discussed/highlighted).  Sun's "problem" among manyB 	is that they are very dependent on hardware profits.  Since theirC 	higher-end margins are >40% (read that others talk of >50% margins ? 	in high-end kit, assumption is Sun is similar), in theory they E 	should be making a lot more than Dell with less volumes.  Problem is @ 	they have tons of overhead and a slide in sales translates into: 	losing money.   CPU design, engineering design, OS designB 	and verification, manufacturing, etc.  Contrast Dell - a lot less
 	overhead.  A 	The best model would be to drop overall margins to 25% ? and cut C 	accordingly.  But won't happen for roll your own, HP, Sun , IBM as G 	their overhead (by nature is higher) and Dell will continue to punish.   G 	That's the bad news.  The good news is that HP is doing a mini-version 	 	of Dell  E 	putting all their wood behind one arrow (IA64) on the high-end.  CPU @ 	design costs go away and a good deal of overlapping engineeringA 	design goes away (one would suspect that HP/UX IA64 hardware and D 	VMS IA64 hardware is VERY similar - if not the exact box.  Same forD 	high-end Windows boxes.)    I'd almost bet that Mr. TangleWood putsB 	that memory controller on CPU further reducing engineering design= 	costs - BUT realworldtech.com techies disagree and they have D 	strong arguments as to why NOT to put memory controller on CPU.  We? 	shall see.  But the danger is to assume Intel won't be able to F 	end-to-end *most* of the engineering in the system design and ship a C 	board that gets put in a box and that box has superior reliabilty.    					Rob   ------------------------------  % Date: Thu, 11 Sep 2003 04:51:40 +0800 , From: Paul Repacholi <prep@prep.synonet.com>4 Subject: Re: "HP breakup on the way" - Merrill Lynch- Message-ID: <87k78gicyr.fsf@prep.synonet.com>   = koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:    > In article <bjkomh$iha$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes:  A >> VMS relys on Alpha and is migrating to Itanium it isn't if you @ >> include OpenVMS's share of developing and manufacturing thoseD >> platforms in your cost base I very much doubt if you are remotely >> profitable.  B >    Nonsense.  Alphas have never made as much profit as VMS.  The% >    money is in software, not chips.      E No, the real money is in the people. The tin and SW are just a way of D giving you a big leg up over all the others. Plus you can brag about> having the bigest hammer on the block, you generally can't say, anything about what you do at the customers.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Wed, 10 Sep 2003 16:56:00 -0700 * From: "Jack Peacock" <peacock@simconv.com>4 Subject: Re: "HP breakup on the way" - Merrill Lynch2 Message-ID: <-i2dnXmsOYGMJ8KiXTWJiA@mpowercom.net>  4 "Carl Perkins" <carl@gerg.tamu.edu> wrote in message& news:9SEP200322020301@gerg.tamu.edu...? > "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> writes...  > 0 > Which is a radio station. Thus, it is not PBS. >  > PBS is TV. >  > NPR is radio.  > J I often make the same mistake.  It's understandable...the same Ministry of, Truth supplies the content for both outlets.    Jack Peacock    ------------------------------  % Date: Wed, 10 Sep 2003 22:29:36 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> 4 Subject: Re: Backup foreground/background operations' Message-ID: <3F5FEC20.7C3AA26B@fsi.net>    "Keith A. Lewis" wrote:  >  > "David J. Dachtera" <djesys.nospam@fsi.net> writes in article <3F5E9522.AC0B34EC@fsi.net> dated Tue, 09 Sep 2003 22:06:10 -0500: > >Bart Zorn wrote:  > >>* > >> See HELP BACKUP_command /RELEASE_TAPE > > K > >Problem with that is all it does is release (deallocate) the tape drive. H > >There's no way to trigger another process to take over the drive just > >released. > 
 > $tape_loop:  > $ wait 0:0:5' > $ on severe_error then goto tape_loop  > $ backup .... /release_tape  > I > Could be smarter, but in most cases it will do what the original poster  > needs.  H ...except that he still lacks a mechanism for kicking off another backupH process using the drive just released while the current process executes the /RECORD.  D Oh, yeah: /RELEASE will DISMOUNT/UNLOAD the tape. Maybe not what the user wants.   @ > >I wrote some DCL code once to use DFU to set the backup dates >  > DFU?  F See HELP SET /BACKUP_DATE in DFU. DFU is freeware from DEC. Find it on? the freeware CD and in my VMS freeware area for older versions.   H I'll be moving to temporary quarters this weekend. So, my VMS machine is5 down for now. Else, I'd cut/paste the help text here.    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------    Date: 10 Sep 2003 12:34:22 -0700% From: tadamsmar@yahoo.com (Tom Adams)   Subject: Block size in show dev?= Message-ID: <ea44f5a1.0309101134.753ae8f5@posting.google.com>   + SHOW DEV DKA0 shows 17,773,524 total blocks   8 are these 512 byte blocks?  What is the block size here?   ------------------------------  % Date: Wed, 10 Sep 2003 14:51:04 -0500 ( From: brandon@dalsemi.com (John Brandon)$ Subject: Re: Block size in show dev?1 Message-ID: <03091014510447@dscis6-0.dalsemi.com>   - > SHOW DEV DKA0 shows 17,773,524 total blocks  > : > are these 512 byte blocks?  What is the block size here?   Block size = 512 bytes.     ( Roughly 17,773,524 / 2,000,000 = 8.88 GB  F The value of 2,000,000 is not exact but it is close enough for my use. Isn't it like 209,000,00?        J*o*h*n B*r*a*n*d*o*n  VMS Systems Administrator * firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  % Date: Wed, 10 Sep 2003 13:43:38 -0600 4 From: Norman Lastovica <norman.lastovica@oracle.com>$ Subject: Re: Block size in show dev?* Message-ID: <3F5F7EEA.9FD50BA2@oracle.com>  = as far as I can figure out, blocks are always 512 bytes long. : the term "page" can be less strictly defined (depending on8 where and how it is used along with what platform you're running on).   Tom Adams wrote: > - > SHOW DEV DKA0 shows 17,773,524 total blocks  > : > are these 512 byte blocks?  What is the block size here?   ------------------------------   Date: 10 Sep 2003 22:38:17 GMT) From: Tony Arnold <tony.arnold@man.ac.uk> $ Subject: Re: Block size in show dev?9 Message-ID: <3f5fa7d9$0$247$cc9e4d1f@news.dial.pipex.com>   ) John Brandon <brandon@dalsemi.com> wrote: / > > SHOW DEV DKA0 shows 17,773,524 total blocks  > > < > > are these 512 byte blocks?  What is the block size here?   > Block size = 512 bytes.     * > Roughly 17,773,524 / 2,000,000 = 8.88 GB  H > The value of 2,000,000 is not exact but it is close enough for my use. > Isn't it like 209,000,00?   F Isn't 2,000,000 actually exact? If a block is 512 bytes, then 1Gb, hasC to have 2,000,000 blocks exactly? Or has my maths gone rusty? (I am D assuming that 1Gb is 1024 megabytes, which in turn is 1024 kilobytes etc.)    Tony.  --  F Tony Arnold, Deputy to the Head of COS Division, Manchester Computing,: University of Manchester, Oxford Road, Manchester M13 9PL.F T: +44 (0)161 275 6093, F: +44 (0)870 136 1004, M: +44 (0)773 330 0039E E-mail: tony.arnold@man.ac.uk, Home: http://www.man.ac.uk/Tony.Arnold    ------------------------------  # Date: Wed, 10 Sep 2003 22:49:19 GMT # From: hoff@hp.nospam (Hoff Hoffman) $ Subject: Re: Block size in show dev?1 Message-ID: <PTN7b.4431$Rc.4248@news.cpqcorp.net>   e In article <ea44f5a1.0309101134.753ae8f5@posting.google.com>, tadamsmar@yahoo.com (Tom Adams) writes: , :SHOW DEV DKA0 shows 17,773,524 total blocks : 9 :are these 512 byte blocks?  What is the block size here?   E   Please review the OpenVMS Frequently Asked Questions (FAQ) document C   for the answer to this and to the many variants of this question, F   and for details of the various common units used to measure "stuff".  $     http://www.hp.com/go/openvms/faq    N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq N  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com    ------------------------------  % Date: Wed, 10 Sep 2003 18:39:09 -0500 ( From: brandon@dalsemi.com (John Brandon)$ Subject: Re: Block size in show dev?1 Message-ID: <03091018390986@dscis6-0.dalsemi.com>   + > John Brandon <brandon@dalsemi.com> wrote: 1 > > > SHOW DEV DKA0 shows 17,773,524 total blocks > > > > are these 512 byte blocks?  What is the block size here? > > Block size = 512 bytes.   , > > Roughly 17,773,524 / 2,000,000 = 8.88 GBJ > > The value of 2,000,000 is not exact but it is close enough for my use. > > Isn't it like 209,000,00?  >  Tony Arnold wrote:H > Isn't 2,000,000 actually exact? If a block is 512 bytes, then 1Gb, hasE > to have 2,000,000 blocks exactly? Or has my maths gone rusty? (I am F > assuming that 1Gb is 1024 megabytes, which in turn is 1024 kilobytes > etc.)  > Tony.    Your math is fine.  H Two definitions (check the definition by IBM - ooo was that a bad word?)  A http://whatis.techtarget.com/definition/0,,sid9_gci212542,00.html     A If you assume that 1-GB is 1,024 mega-bytes then you are correct.   %      17,773,524 * 512 = 9,100,044,288 )      9,100,044,288 / 1,024,000,000 = 8.88            -or-"      17,773,524 / 2,000,000 = 8.88    ; However if you assume that 1-GB is 1,000 mega-bytes then... 7 mega means 1,000,000 bytes and giga means 1,000,000,000 3 Therefore 1-GB is 1,000 mega-bytes or 1,000,000,000   %      17,773,524 * 512 = 9,100,044,288 3      9,100,044,288 / 1,000,000.00 = 9,100 megabytes 3      9,100,044,288 / 1,000,000,000 = 9.10 gigabytes   L I have seen bytes represented both ways - and since the division by 2,000,00% works for me... well it works for me.   N And I bet the disk is labeled completely differently... need to check that one out!    P So that still does not explain where I came up with the 2,090,000 - oh bother...   Then what Hoff said : A http://h71000.www7.hp.com/wizard/faq/vmsfaq_018.html#index_x_1092          J*o*h*n B*r*a*n*d*o*n  VMS Systems Administrator * firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  # Date: Wed, 10 Sep 2003 20:28:09 GMT " From:   VAXman-  @SendSpamHere.ORG. Subject: Re: Choosing between ASTs and Threads0 Message-ID: <00A25B40.FB994E54@SendSpamHere.ORG>  c In article <bKEDl5oxbIDY@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes: b >In article <6skulvo84khbte1rdeo4ucu0esmoi9u2jc@4ax.com>, jlsue <jefflsxxxz@sbcglobal.net> writes:H >> On 8 Sep 2003 21:32:40 -0500, Kilgallen@SpamCop.net (Larry Kilgallen)	 >> wrote:  > J >>>>> Not having done any thread programming, I'll ask what may be a sillyM >>>>> question:  Can you use the lock manager to synchronize threads for some O >>>>> activities?  For example, when pulling work off the queue, can one thread M >>>>> lock the queue to prevent multiple threads from stepping on each other?  >>>>> K >>>>    Generally, no.  Either your process has the lock or it doesn't (VMS M >>>>    DLM locks).  There's no mechanism to distinguish what thread in your   >>>>    process has it.  >>> N >>>Generally yes you _can_, since which thread has the lock need only be known$ >>>by the thread which has the lock. >>>  >>  N >> That's the question... what if multiple threads in the same process need toB >> synchronize.... it appears that lock manager won't work in that >> circumstance. > B >I don't know for certain, but what makes you think it would not ? > - >In what fashion do you think it would fail ?  > E >Individual threads can "stall" and "wake up" based on IO completion. ; >Do you think they cannot do so based on Lock Acquisition ?  > A >Obviously the Lock Value Block must be in thread-specific memory  >if not on the stack.   H The lock value block is located in the RSB which is carved out of system0 space (S2 space for the latest versions of VMS).   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM              5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  % Date: Wed, 10 Sep 2003 21:40:45 +0100 ( From: "John Travell" <john@jomatech.com>. Subject: Re: Choosing between ASTs and Threads9 Message-ID: <bjo2kv$kr94l$2@ID-120847.news.uni-berlin.de>   3 "jlsue" <jefflsxxxz@sbcglobal.net> wrote in message 2 news:6skulvo84khbte1rdeo4ucu0esmoi9u2jc@4ax.com...G > On 8 Sep 2003 21:32:40 -0500, Kilgallen@SpamCop.net (Larry Kilgallen)  > wrote: > 6 > >In article <nRhQIvm$3yv9@eisner.encompasserve.org>,= koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes: C > >> In article <jv9plv4s1kdgdra6h07c6aorljrcqsk1et@4ax.com>, jlsue " <jefflsxxxz@sbcglobal.net> writes:K > >>> On 8 Sep 2003 08:06:43 -0500, koehler@eisner.nospam.encompasserve.org  (Bob > >>> Koehler) wrote:  > >>>  > >>>  > >>>>K > >>>>   You use threads when there's significant parallel work that can be G > >>>>   done and the synchronization techniques provided by the thread  > >>>>   package suffice.  > >>> J > >>> Not having done any thread programming, I'll ask what may be a sillyH > >>> question:  Can you use the lock manager to synchronize threads for someH > >>> activities?  For example, when pulling work off the queue, can one threadF > >>> lock the queue to prevent multiple threads from stepping on each other? > >>> K > >>    Generally, no.  Either your process has the lock or it doesn't (VMS L > >>    DLM locks).  There's no mechanism to distinguish what thread in your > >>    process has it.  > > H > >Generally yes you _can_, since which thread has the lock need only be known $ > >by the thread which has the lock. > >  > J > That's the question... what if multiple threads in the same process need toA > synchronize.... it appears that lock manager won't work in that  > circumstance.  >  > = > >>    Thread packages provide their own locking mechanisms.  > > F > >Which likely will be more efficient if you are certain the resourceA > >will always be in use by another thread from the same process.  > L > Well, if you have mulitple threads (in the same process) pulling items offI > a queue, as each thread goes to do the "pop" activity, it needs to lock  the L > queue (head) so that nobody else gets the same item, or some indeterminate > state. > L > If each "thread" is actually in a separate process, then you could use the > lock mgr.  > L If you are talking about proper queues (absolute or self-relative, not thoseA batch or print thingies) the self-relative queues use interlocked L instructions, e.g INSQTI, etc. In which case use of other lock management is superfluous and wasteful.      -- John Travell" Independent VMS crashdump analyst. john- at - jomatech - dot - com  +44-(0)23-92552229 http://www.jomatech.com/       --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).A Version: 6.0.515 / Virus Database: 313 - Release Date: 01/09/2003    ------------------------------    Date: 10 Sep 2003 15:18:47 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) . Subject: Re: Choosing between ASTs and Threads3 Message-ID: <bKEDl5oxbIDY@eisner.encompasserve.org>   a In article <6skulvo84khbte1rdeo4ucu0esmoi9u2jc@4ax.com>, jlsue <jefflsxxxz@sbcglobal.net> writes: G > On 8 Sep 2003 21:32:40 -0500, Kilgallen@SpamCop.net (Larry Kilgallen)  > wrote:  I >>>> Not having done any thread programming, I'll ask what may be a silly L >>>> question:  Can you use the lock manager to synchronize threads for someN >>>> activities?  For example, when pulling work off the queue, can one threadL >>>> lock the queue to prevent multiple threads from stepping on each other? >>>>  J >>>    Generally, no.  Either your process has the lock or it doesn't (VMSL >>>    DLM locks).  There's no mechanism to distinguish what thread in your  >>>    process has it. >>M >>Generally yes you _can_, since which thread has the lock need only be known # >>by the thread which has the lock.  >> > M > That's the question... what if multiple threads in the same process need to A > synchronize.... it appears that lock manager won't work in that  > circumstance.   A I don't know for certain, but what makes you think it would not ?   , In what fashion do you think it would fail ?  D Individual threads can "stall" and "wake up" based on IO completion.: Do you think they cannot do so based on Lock Acquisition ?  @ Obviously the Lock Value Block must be in thread-specific memory if not on the stack.   ------------------------------  % Date: Wed, 10 Sep 2003 19:52:55 -0400   From: John Santos <JOHN@egh.com>. Subject: Re: Choosing between ASTs and Threads5 Message-ID: <1030910172703.3498A-100000@Ives.egh.com>   3 On Wed, 10 Sep 2003 VAXman-@SendSpamHere.ORG wrote:   e > In article <bKEDl5oxbIDY@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes: d > >In article <6skulvo84khbte1rdeo4ucu0esmoi9u2jc@4ax.com>, jlsue <jefflsxxxz@sbcglobal.net> writes:J > >> On 8 Sep 2003 21:32:40 -0500, Kilgallen@SpamCop.net (Larry Kilgallen) > >> wrote:  > > L > >>>>> Not having done any thread programming, I'll ask what may be a sillyO > >>>>> question:  Can you use the lock manager to synchronize threads for some Q > >>>>> activities?  For example, when pulling work off the queue, can one thread O > >>>>> lock the queue to prevent multiple threads from stepping on each other?  > >>>>> M > >>>>    Generally, no.  Either your process has the lock or it doesn't (VMS O > >>>>    DLM locks).  There's no mechanism to distinguish what thread in your   > >>>>    process has it.  > >>> P > >>>Generally yes you _can_, since which thread has the lock need only be known& > >>>by the thread which has the lock. > >>>  > >>  P > >> That's the question... what if multiple threads in the same process need toD > >> synchronize.... it appears that lock manager won't work in that > >> circumstance. > > D > >I don't know for certain, but what makes you think it would not ? > > / > >In what fashion do you think it would fail ?  > > G > >Individual threads can "stall" and "wake up" based on IO completion. = > >Do you think they cannot do so based on Lock Acquisition ?  > > C > >Obviously the Lock Value Block must be in thread-specific memory  > >if not on the stack.  > J > The lock value block is located in the RSB which is carved out of system2 > space (S2 space for the latest versions of VMS).  . I think Larry must mean the Lock Status Block.  F I haven't done any thread programming, but AFAIK, a single process can) have multiple locks on the same resource.   D For example, in BASIC, using RMS, you can open the same indexed fileD on multiple channels.  If you read (and lock) a record on channel 1,> and then attempt to read the same record on channel 2, you get@ %BAS-F-RECBUCLOC, Record/bucket locked, which is triggered by anH RMS error: -RMS-E-RLK, target record currently locked by another stream.  C If the lock manager was distinguishing locks only by process ID, it G would think this was the same lock, and accept it as a null conversion. B But when RMS first enqueues a lock for each channel, it is using aG separate Lock Status Block, which receives a distinct Lock ID for each  J lock.  It's the distinct Lock ID's (which must be used for input to later J lock conversions), that distinguish the locks.  (The Lock ID is stored in I the Lock Status Block, so if you keep separate LSB's for each thread, you  should be all set.)   J (The Lock Value Block (or a copy of it) is also stored in the Lock Status E Block.  I think this should actually be named a Resource Value Block, D since it is actually associated with the resource rather than with a! particular lock on the resource.)    --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------    Date: 10 Sep 2003 22:22:59 -07001 From: usenet_vms@lehrerfamily.com (Joshua Lehrer) . Subject: Re: Choosing between ASTs and Threads= Message-ID: <477e0934.0309102122.7dfc4fe7@posting.google.com>   ] John Santos <JOHN@egh.com> wrote in message news:<1030910172703.3498A-100000@Ives.egh.com>...  > H > I haven't done any thread programming, but AFAIK, a single process can+ > have multiple locks on the same resource.  >   F You need to be careful.  The lock manager knows nothing about threads.D  If thread A has an exclusive lock, and thread B queues an exclusiveB lock, it will return immediately with a deadlock error.  The queueE manager doesn't know that you are going to release the lock in thread < A at some point, at which point, thread B can have the lock.  B There is a flag that you can set on the queueing of the request to ignore deadlocks.   
 joshua lehrer  factset research systems NYSE:FDS   ------------------------------  % Date: Thu, 11 Sep 2003 00:55:44 +0100 6 From: Chris Townley <news@townleyc.nospam.demon.co.uk>' Subject: Re: Do we need a DCL debugger? 8 Message-ID: <0aevlvoagru4du9rsspcjg3nmf40gok8o2@4ax.com>  @ On Wed, 03 Sep 2003 15:12:18 +0300, Guy Peleg <guy.peleg@hp.com> wrote:   > D >The DCL engineering team is currently evaluating new features to be. >added in future version of VMS (post V7.3-2). > H >We were wondering if folks think that developing a DCL debugger will be% >useful. It won't be as sophisticated H >as the VMS debugger, I was thinking about implementing step,examine and >set break.  > 8 >Please post a reply with your comments or send me mail. >  >Thanks in advance,  > 
 >Guy Peleg >OpenVMS Engineering > I >** Please note that this note is no commitment for actually developing a 2 >DCL debugger, this is only a survey that will not* >necessarily result in actual product.....  C Definitely very useful. As would a built in syntax checker, much as 1 DCL_CHECK is good, ksh -n does nicelu under Unix.        --  
 Chris Townley ) chris at townleyc dot demon dot co dot uk    ------------------------------  % Date: Thu, 11 Sep 2003 02:01:12 +0800 , From: Paul Repacholi <prep@prep.synonet.com>  Subject: Re: Image tools for VMS- Message-ID: <87n0dcjzfb.fsf@prep.synonet.com>   + Jack Fortune <jcfortune3@fedex.com> writes:   D > I am researching the range of available VMS tools that can be used8 > to scan, store, and display images of paper documents.  8 > Are there scanners that can be used directly with VMS?  C > What are the best methods to display images stored on a VMS host?   < > Are there software products designed to build this kind of > application?  B There was about 3 or 4 layered product on at least the Vax CONDISTA for this stuff. Or have they all been left to wither on the vine?    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------    Date: 10 Sep 2003 11:33:02 -07000 From: keith.cayemberg@conti.de (Keith Cayemberg), Subject: Re: Image tools for VMS - follow up= Message-ID: <3a65a5c8.0309101033.166049e3@posting.google.com>    Hi Jack,  D I found this Scanning Software that appears to support for OpenVMS. / Sorry, but I only found a Web-Site in German...r  2 EasyCopy/Scan (see the list of supported scanners)< http://www.h-p-eurosoft.de/scanner/easycopy-scan/scannen.htm  # It works with EasyCopy/X on OpenVMS 6 ftp://ftp.augrin.com/pub/OldDocumentation/ecxvmsus.pdf5 http://www.h-p-eurosoft.de/imaging/easycopy/index.htmw    - DEC did have a scanning software at one time.eI http://www.sysworks.com.au/disk$vaxdocdec952/decw$book/d3sjaaa1.decw$book       G Perhaps a short list of Document Management Systems for VMS would help?    Redwood - Outrun1 http://www.redwood.com/products/outrun/index.htmlh0 http://www.redwood.com/news/butlergrp/outrun.htm   Sec 1.01 - i-engineg" http://www.sec101.ch/default_e.htm* http://www.sec101.ch/deutsch/paradocs.html  	 Swiftbasek8 http://www.swiftbase.com/applications2/document_mgr.html5 http://www.swiftbase.com/utilities2/cd_archiving.html    hp VTX2 http://h18000.www1.hp.com/info/SP6405/SP6405PF.PDF  7 As stated earlier, I do not recommend, endorse or vouch 7 for any of these products in regard to their quality orl" usefulness for a specific purpose.   Cheers!n   Keith Cayembergd ICA GmbH - Hannover, Germany  i Jack Fortune <jcfortune3@fedex.com> wrote in message news:<t45ulv0p1pqgleu3cthvqgl0vk7vd539ts@4ax.com>...p2 > On Tue, 09 Sep 2003 09:37:52 -0400, Jack Fortune > <jcfortune3@fedex.com> wrote:b >  *SNIP* > F > Thanks to everyone for the replies. Perhaps I can clarify & expand a > little bit on our needs. > A > We currently run an application on our Alpha cluster to handle aD > customs brokerage processing for our customers. This is a somewhatG > vintage application written in COBOL which stores data in RMS files. -F > We are beginning to investigate how we might be able to stop storingA > all paper documents associated with our customs transactions byeG > developing some kind of image storage/retrieval system. This would bepD > used primarily by our employees working on PC desktop computers inF > remote offices connected to the Alpha data center over a frame relay > WAN. > F > These PC workstations are running terminal emulation software to useG > our VMS/COBOL application and they have web browsers, as well. I fear E > that reconfiguring all of these machines with X-window servers will @ > not be a viable option. This leads me to believe that the bestF > approach for image viewing is to utilize a VMS web server to display) > the image files to the users' desktops.r > G > From many of the replies, I gather that there are SCSI scanners stillaD > available that would connect to my Alpha's. Is there any way theseF > kind of scanners could be connected from a remote office? If not, weC > would need scanners located in several dozen remote offices. ThiscC > would require a PC front end to scan paper & send images onto the. > Alpha cluster. > G > We haven't given any thought yet to whether we would store image dataoG > as separate RMS files or as blob's inside a dbms. If we went the dbmst6 > route, we would probably be required to use Oracle.  > ( > once again thanks for all information, >  > Jack Fortune   ------------------------------   Date: 11 Sep 2003 02:14:01 GMT From: healyzh@NOaracnetSPAM.come/ Subject: Re: Mac OS X -> Pathworks/Mac problemsd, Message-ID: <bjolp902i8h@enews3.newsguy.com>  + Paul Anderson <paul.anderson@hp.com> wrote:kJ > In article <bjlja50657@enews1.newsguy.com>, <healyzh@aracnet.com> wrote:  H >> The moment that 'xemacs' tried to read the directory of the AppletalkH >> share, the Mac locked up SOLID.  After rebooting I decided to try andA >> repeat this, and unfortunatly was able to without any problem.0  @ > I just tried accessing and writing new versions of files to myI > PATHWORKS for OpenVMS (Macintosh) share from my Mac OS X system, and ittD > worked.  TextEdit and Excel could both read and write files to the > share.  (I'm running 10.2.6.)a  F > I recall having problems similar to yours in the past, and have alsoG > found that the performance is not stellar in doing this, so I usually  > don't use this method.  K I'm wondering if part of the problem is that I'm trying to access it from anH *Unix* application rather than a *Mac* application.  At the same time itG simply shouldn't lock the Mac up solid.  I couldn't even ssh in to kill J xemacs.  Accessing files from Finder, and bringing them up in TextEdit and Word didn't offer any problems.d  1 I'm running OpenVMS 7.2-1H1, and Mac OS X 10.2.6.s  C >> I'm guessing the correct solution is to access the data via NFS.   H > Or an FTP connection, which is what I usually use, although if I spentC > the time figuring out NFS it would probably be a better solution.o  K What I don't understand is why Apple is so anti-NFS...  I take that back, I L think I do, they want you to buy thier servers.  What is interesting is thatK I seem to recall reading something about them making FTP in 10.3 be able toe act about like a disk drive.   		Zane   ------------------------------  # Date: Wed, 10 Sep 2003 22:45:10 GMT # From: hoff@hp.nospam (Hoff Hoffman). Subject: Re: MESSAGE problem1 Message-ID: <WPN7b.4429$Rc.2089@news.cpqcorp.net>   h In article <xBE7b.11897$sY3.9843@news-binary.blueyonder.co.uk>, Edward Brocklesby <ejb@goth.net> writes:  H :I'm trying to use the MESSAGE utility to generate some message objects  :for an application.    B   Acquire GNM off the Freeware, if you want to start from a common?   source file for messages and error message and recovery text."B   (The following example assumes a newer version of GNM, while the?   foreign command for the existing Freeware version of GNM mustl@   reference a slightly different image filename with a different   command syntax.)       $ gnm :== $srh:[gnm]gnm.exe "     $ gnm 'src' 'trgmsg' 'trgsdml'  A   If you want to start with a message file, here are the commandsd0   that would be used within OpenVMS Engineering:  ,     $ MESSAGE/NOOBJECT/SDL='trgsdl' 'trgmsg'(     $ MESSAGE/OBJECT='trgmsgobj' 'trgmsg)     $ link /notrace/nodebug 'build_opt' -r8       /shareable='trgmsgexe' 'trgmsgobj', xyz.opt/optionC     $ MESSAGE/OBJECT='trgmsgobj' /FILE_NAME='trgsysmsgexe' 'trgmsg' 5     $ library/object/replace lib$:xyz.olb 'trgmsgobj's  B   To make full use of the undocumented and unsupported MESSAGE/SDL@   tool, acquire the Freeware SDL tool, and invoke it as follows:       $ @srh:[sdl]sdl.comg&     $ SDL/ALPHA/LANG=C='trgh' 'trgsdl'  D   trgh is a C header file, and can be directly included in the code.E   SDL can generate definition files for various other languages, too.u  D   I expect a substationally updated version of GNM -- I re-wrote theC   tool, using lib$table_parse -- and I also expect a new version ofc'   SDL to appear on the next Freeware.  l    N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faqhN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com5   ------------------------------  % Date: Wed, 10 Sep 2003 16:16:54 -0700o# From: "Tom Linden" <tom@kednos.com>D Subject: RE: MESSAGE problem9 Message-ID: <CIEJLCMNHNNDLLOOGNJICEDNHPAA.tom@kednos.com>m   >-----Original Message-----@+ >From: Hoff Hoffman [mailto:hoff@hp.nospam] , >Sent: Wednesday, September 10, 2003 3:45 PM >To: Info-VAX@Mvb.Saic.Com >Subject: Re: MESSAGE problemp >b >1@ >In article <xBE7b.11897$sY3.9843@news-binary.blueyonder.co.uk>,) >Edward Brocklesby <ejb@goth.net> writes:> >>H >:I'm trying to use the MESSAGE utility to generate some message objects >:for an application.H >iC >  Acquire GNM off the Freeware, if you want to start from a commoni@ >  source file for messages and error message and recovery text.C >  (The following example assumes a newer version of GNM, while the @ >  foreign command for the existing Freeware version of GNM mustA >  reference a slightly different image filename with a different  >  command syntax.)n >c  >    $ gnm :== $srh:[gnm]gnm.exe# >    $ gnm 'src' 'trgmsg' 'trgsdml'e >nB >  If you want to start with a message file, here are the commands1 >  that would be used within OpenVMS Engineering:p > - >    $ MESSAGE/NOOBJECT/SDL='trgsdl' 'trgmsg' ) >    $ MESSAGE/OBJECT='trgmsgobj' 'trgmsge* >    $ link /notrace/nodebug 'build_opt' -9 >      /shareable='trgmsgexe' 'trgmsgobj', xyz.opt/optioneD >    $ MESSAGE/OBJECT='trgmsgobj' /FILE_NAME='trgsysmsgexe' 'trgmsg'6 >    $ library/object/replace lib$:xyz.olb 'trgmsgobj' >iC >  To make full use of the undocumented and unsupported MESSAGE/SDLmA >  tool, acquire the Freeware SDL tool, and invoke it as follows:, >e >    $ @srh:[sdl]sdl.com' >    $ SDL/ALPHA/LANG=C='trgh' 'trgsdl'n   Or in the case postedo(     $ SDL/ALPHA/LANG=PLI='trgh' 'trgsdl'' which produces an include file trgh.plie     >eE >  trgh is a C header file, and can be directly included in the code. F >  SDL can generate definition files for various other languages, too. >iE >  I expect a substationally updated version of GNM -- I re-wrote theeD >  tool, using lib$table_parse -- and I also expect a new version of& >  SDL to appear on the next Freeware. >e > 1 > ---------------------------- #include <rtfaq.h>r >-----------------------------L >    For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faq3 > --------------------------- pure personal opinionn >---------------------------F >        Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com >n >---' >Incoming mail is certified Virus Free.o; >Checked by AVG anti-virus system (http://www.grisoft.com).iA >Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003d >g ---t& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.512 / Virus Database: 309 - Release Date: 8/19/2003   ------------------------------  % Date: Thu, 11 Sep 2003 02:55:55 +0800M, From: Paul Repacholi <prep@prep.synonet.com>0 Subject: Re: Moving a Tk70 from 3400 to 4000/600- Message-ID: <8765k0jww4.fsf@prep.synonet.com>y   tony.arnold@man.ac.uk writes:h  B > Seeing Adian Stapley's post recently about his TK-70 reminded meC > that I have a 3400 with a TK-70 and I want to move the TK-70 to aiD > MicroVAX 4000/600. Should this be possible and is there anything I > need to be careful about?e  E If there are no other MU controllers, and the RH slot is available it D is straight forward to do. You may have fun finding a 5 pin to 4 pin@ power cable to go from the DSSI drive cradle to the TK70 though.   -- 8< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.i@                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Wed, 10 Sep 2003 15:27:09 -0400r# From: sol gongola <sol@adldata.com>n4 Subject: Need better SMTP RBL blocking error message+ Message-ID: <3F5F7B0D.3636969C@adldata.com>   ?   Compaq TCP/IP Services for OpenVMS Alpha Version V5.1 - ECO 3a4   on a AlphaServer 2100 4/275 running OpenVMS V7.2-1  B When a message is blocked due to RBL, the OPER logged message has:G 	TCPIP-W-SMTP_CLNTINRBL, client IP address nn.nn.nn.nn matched RBL list = but the default message returned to the email sender only has6" 	Your IP address is in my RBL list  F Is there a way to get more information returned to the sender such as A the offending IP address and the RBL facility where it was found?   	 thank you  sol gongola  adl data systems inc.t dobbs ferry, nyu   ------------------------------   Date: 10 Sep 2003 16:12:35 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)J( Subject: Re: Need HP Reseller in Germany0 Message-ID: <bjnihj$nj8$1@n.ruf.uni-freiburg.de>  d In article <s%F7b.3185$z5.248042@news20.bellglobal.com>, "Scott Greig" <jsgreig@geminaq.com> writes:4 >Sorry for this off-topic request - I can't think of >another area to post this.z >d4 >We are an HP reseller in Canada, and have ordered a6 >DS10 system to be shipped to our client's location in7 >Canada for application setup, and our client will thenl. >ship the system to their location in Germany. >t5 >For all other (Canadian) systems, I can purchase the-3 >3 year warranty uplift to 4-hour hardware/software00 >response (Canadian part #'s FM-A01BE-36 for the4 >system, and FP-T14EC-36 for the external DLT tape). >s5 >Is there a Germany based HP-reseller who can provide  >me with  i) Part numbersd( >             ii) Contacts for ordering.   http://www.pdv-systeme.de/ http://www.technidata.de/e   Regards,    Christoph Gartmannn   --  E  Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452t  ImmunbiologieA  Postfach 1169                 Internet: gartmann@immunbio.mpg.deo  D-79011  Freiburg, Germanya9                http://www.immunbio.mpg.de/home/menue.html    ------------------------------    Date: 10 Sep 2003 11:51:29 -0700) From: ejohnson@factset.com (Eric Johnson)u0 Subject: Re: New linker ECOs for V7.3 and V7.3-1= Message-ID: <ef79676b.0309101051.4f277f9f@posting.google.com>i  | "Martin Vorlaender" <martin.vorlaender@pdv-systeme.de> wrote in message news:<bjmfij$kpd47$1@ID-56200.news.uni-berlin.de>...  F > INSTALL /HEADER_RESIDENT makes VMS remember the file header, so thatE > when the image is activated, VMS doesn't have to go out to the disk  > and find the file.  D I would think the value of /header_resident would be obviated by theC advent of XFC.  Of course, I'm assuming that the image was recentlytF activated and had no reason to be kicked out of the cache.  Given that+ operational environment, aren't they equal?-  G > With INSTALL /RESIDENT, on the other hand, "The Install utility movesiJ > certain portions of resident images into a granularity hint region (GHR)C > in system space; there, they function as a large single page witha< > granularity hints set, which provides better performance."  F What kind of performance gains should one see?  Earlier up the thread,E I see someone attributing oracle.exe to getting a 30% bang out of the @ /resident qualifier.  I find that hard to believe.  What type ofE executive environments benefit the most from this?  Big images?  Lots F of images?  Lots of processes running through all of the images?  Lots6 of processes running through a small set of the image?  C > A prerequisite is to link the image with the /SECTION_BINDING andcG > /NOTRACEBACK qualifiers. With /SECTION_BINDING, certain optimizations06 > are suppressed which would inhibit resident loading.  D The /notraceback restriction is a real bummer.  I see no good reasonE for it.  I also think its a bummer that one can't deinstall /residenteF images.  Last time I checked, that required a node reboot.  Huh?  What is this, Windows?   
 -eric johnsons   ------------------------------  # Date: Wed, 10 Sep 2003 19:32:07 GMT?3 From: hammond@not@peek.ssr.hp.com (Charlie Hammond)rU Subject: Re: OpenVMS clusters (Alpha) - standalone workstation joined to the cluster?t0 Message-ID: <X_K7b.4418$FP7.62@news.cpqcorp.net>  1 In article <3F5E7BAA.3010604@thisnewsgroup.com>,  5 Jeffrey Green <pleasereply@thisnewsgroup.com> writes:t  C >I'm clearing the cobwebs from my VMS cluster knowledge and have a hJ >question I hope someone can help with.   Ages ago (Vax 4000 server, 3100 > >  workstations clustered) we had a requirement for couple of G >workstations that had to be standalone but tied to the cluster.  I.e.  J >They don't boot from the cluster and there is no boot root for them, yet J >they participate in the cluster and use the same authorize files.   Back J >when I was the admin for that system we had a DEC contractor come in and  >set it up for us.  H If you don't understand how to do this, hiring a contractor is not a badF idea -- you can get some {ahem} interesting situation if you do things "not quite right".  E What you need to do is to install OpenVMS on each of the workstationsaB that are to have their own system disks.  Then configure them intoL the cluster, and assign logicals to point the the main cluster authorizationD files.  You will probably want these worstations to have zero votes,< so that they can not boot without the cluster being present.  F I think that a lot of this is in the Cluster Systems and the Guide to E Cluster Configuration (something like that) manuals.  I advise you to  get them and read them.r   -- fJ       Charlie Hammond -- Hewlett-Packard Company -- Ft Lauderdale  FL  USAF           (hammond@not@peek.ssr.hp.com -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------  % Date: Thu, 11 Sep 2003 02:46:53 +0800s, From: Paul Repacholi <prep@prep.synonet.com>Y Subject: Re: OpenVMS clusters (Alpha) - standalone workstation joined to the cluster? clut- Message-ID: <87ad9cjxb6.fsf@prep.synonet.com>o  5 Jeffrey Green <pleasereply@thisnewsgroup.com> writes:u  C > Needless to say that was almost 10 years ago and I can't find any-C > documentation on how they did it, which brings me to my point.  Ii> > have group at my work that wants to stand up a very small (4> > alphastation workstations will grow to 10) VMS OpenVMS AlphaB > cluster, but they want all the workstations to act the same as IE > previosly discussed. Standalone with the OS and page/swap local but C > still tied to the cluster.  I've been searching the Web and everyuE > VMS FAQ that I can find and I haven't seen anything that looks like0@ > the answer.  I've pleaded to set it up booting off the server,E > something I still know fairly well, but the answer was a firm "no".y  F > I know it can be done because that DEC contractor did it, although I$ > do realize that was VMS/Vax (5.2).  6 I'm not sure why they want to do it that way, however.  F Problem is, you need to have some files in *1* place, SYSUAF and such,E and you do not have any shared storage, so the drives are all tied to A a machine. No node, no disk, no SYSUAF :( How ever, if you set uprF several drives into a shadow set, you can then point SYLOGICALS at the3 shadow DSA unit and run with that *very carefully*..  C Problem is, if you take down some nodes, change stuff on the shadowoF set, then drop those nodes and reboot the ones that have been down you end up with the old contents :(s  / Doing this without a CI/DSSI or a HS is a pain.,   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  # Date: Thu, 11 Sep 2003 03:14:16 GMT03 From: Jeffrey Green <pleasereply@thisnewsgroup.com>eY Subject: Re: OpenVMS clusters (Alpha) - standalone workstation joined to the cluster? clu 0 Message-ID: <3F5FE88C.3080401@thisnewsgroup.com>  
 Hello all,  H Took all of your suggestions, along with some web surfing and (woohoo!) D I got my OpenVMS Docset pulled out of the company long-term storage.  H I have the cluster master set up (boot server, etc), and 3 workstations E loaded with OpenVMS.  I changed the disk label on all the satellites dH after I figured out that was why they kept dumping out after attempting  to join the cluster.  I Thank you all for the help, information and timely responses.  I did get sH this project handed to me when the person that started it admitted they C didn't know what they were doing.  This was confirmed today when I t= rebooted two of the workstations and they joined the cluster  @ successfully, only to note that you can't log into them anymore.  E After a "boot -flags 0,1" and a sysboot p1-min setting on one of the oI cluster members, I noticed something peculiar after successfully logging uF in.  A little warning message tells me that somebody forgot to buy an E OpenVMS cluster license...  A cursory check of all four workstations eH shows just OpenVMS, VMS Additional User and Net-app-sup-200 are loaded. G   Well at least they DID join the cluster successfully as confirmed by h show cluster/continuous =)  A I hope to have this completed soon, but I'm going to keep a safe iF distance when I tell the project leader that they need to spend a few 1 thousand more $$$ to get what they were promised.>  
 Thanks again,i
 Jeffrey Green.   Jeffrey Green wrote: > Hello there, > D > I'm clearing the cobwebs from my VMS cluster knowledge and have a K > question I hope someone can help with.   Ages ago (Vax 4000 server, 3100 tK >  workstations clustered) we had a requirement for couple of workstations iK > that had to be standalone but tied to the cluster.  I.e. They don't boot k@ > from the cluster and there is no boot root for them, yet they K > participate in the cluster and use the same authorize files.   Back when rJ > I was the admin for that system we had a DEC contractor come in and set  > it up for us.i > D > Needless to say that was almost 10 years ago and I can't find any I > documentation on how they did it, which brings me to my point.  I have TG > group at my work that wants to stand up a very small (4 alphastation sI > workstations will grow to 10) VMS OpenVMS Alpha cluster, but they want 0A > all the workstations to act the same as I previosly discussed. CC > Standalone with the OS and page/swap local but still tied to the 4J > cluster.  I've been searching the Web and every VMS FAQ that I can find K > and I haven't seen anything that looks like the answer.  I've pleaded to :H > set it up booting off the server, something I still know fairly well, ! > but the answer was a firm "no".s > J > I know it can be done because that DEC contractor did it, although I do ! > realize that was VMS/Vax (5.2).o > F > Anyone have a reference or a link to a favorite page with an answer? > I > Thanks in advance, I'm really struggling with this one, and I'm off to  ' > do some more search engine yahtzee =)  >  > Jeffrey Greena >    ------------------------------  # Date: Wed, 10 Sep 2003 18:04:49 GMT ( From: Mark Schafer <mark.schafer@hp.com>$ Subject: OpenVMS I64 V8.0 test drive2 Message-ID: <5JJ7b.4404$oq7.2457@news.cpqcorp.net>   http://www.testdrive.hp.com/  G The HP Test Drive Program is pleased to offer an OpenVMS Itanium cross oA development environment. Two servers are used in this testdrive. ,H 192.233.54.184 is a AlphaServer DS20 running OpenVMS V7.3-1 and has the F cross tools installed on it. 192.233.54.183 is an HP Integrity rx2600 I running OpenVMS I64 V8.0. To get started, use your testdrive account for nF OpenVMS to get the IPF$DOCS:README.TXT file from 192.233.54.184 using % ftp. Sign up and give it a try today!    ------------------------------  # Date: Wed, 10 Sep 2003 20:21:15 GMTl" From:   VAXman-  @SendSpamHere.ORG( Subject: Re: OpenVMS I64 V8.0 test drive0 Message-ID: <00A25B40.074D8BD5@SendSpamHere.ORG>  ] In article <5JJ7b.4404$oq7.2457@news.cpqcorp.net>, Mark Schafer <mark.schafer@hp.com> writes:i >http://www.testdrive.hp.com/  > H >The HP Test Drive Program is pleased to offer an OpenVMS Itanium cross B >development environment. Two servers are used in this testdrive. I >192.233.54.184 is a AlphaServer DS20 running OpenVMS V7.3-1 and has the fG >cross tools installed on it. 192.233.54.183 is an HP Integrity rx2600 cJ >running OpenVMS I64 V8.0. To get started, use your testdrive account for G >OpenVMS to get the IPF$DOCS:README.TXT file from 192.233.54.184 using o& >ftp. Sign up and give it a try today! >   O Signed up for TestDrive account and supplied username and password do not work!a --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMN             5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  # Date: Wed, 10 Sep 2003 19:43:50 GMT ' From: Don Sykes <anonymous@pacbell.net>2F Subject: Re: Runtime linking of Java to AWT graphics libraries problem* Message-ID: <3F5F7F6B.B0F916B@pacbell.net>  G Are you sure you're running the correct Java setup file before startingR this?D> I've had similar problems when I had multiple versions of Java
 installed.   -- 5   Have VMS, Will Travelo Wire paladin, San Franciscoj   (paladinATalphaseDOTcom)   Theo Platt wrote:g >  > Hi there,D > E > We're having some runtime problems with java (1.4.1) linking to the2 > AWT graphics libraries.x >  > Here's the test code - > & > import java.awt.GraphicsEnvironment; >  > public class awttest { > . >     public static void main(String[] args) { > & >         System.out.println("Start"); > 1 >         System.out.println("Setting headless");d: >         System.setProperty("java.awt.headless", "true"); > = >         System.out.println("Getting graphics environment");i" >         GraphicsEnvironment ge =4 > GraphicsEnvironment.getLocalGraphicsEnvironment(); > $ >         System.out.println("End"); >     }s > }h > " > and the results when we run it - >  > Start  > Setting headless > Getting graphics environmentE > %IMGACT-F-SYMVECMIS, shareable image's symbol vector table mismatcha@ > -IMGACT-F-FIXUPERR, error when JAVA$FONTMANAGER_SHR referenced > JAVA$MAWT_SHRnA > java.lang.UnsatisfiedLinkError: Last error code from dlsym was:m > 273517796w> > %LIB-E-ACTIMAGE, error activating image JAVA$FONTMANAGER_SHR > D >         at java.lang.ClassLoader$NativeLibrary.load(Native Method)F >         at java.lang.ClassLoader.loadLibrary0(ClassLoader.java:1504)E >         at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1400) = >         at java.lang.Runtime.loadLibrary0(Runtime.java:788)o: >         at java.lang.System.loadLibrary(System.java:832)Q >         at sun.security.action.LoadLibraryAction.run(LoadLibraryAction.java:50):G >         at java.security.AccessController.doPrivileged(Native Method)1O >         at sun.awt.font.NativeFontWrapper.<clinit>(NativeFontWrapper.java:42) U >         at sun.java2d.SunGraphicsEnvironment$1.run(SunGraphicsEnvironment.java:109)lG >         at java.security.AccessController.doPrivileged(Native Method) U >         at sun.java2d.SunGraphicsEnvironment.<init>(SunGraphicsEnvironment.java:79)SS >         at sun.awt.X11GraphicsEnvironment.<init>(X11GraphicsEnvironment.java:154)p? >         at java.lang.Class.native_newInstance0(Native Method)y8 >         at java.lang.Class.newInstance(Class.java:261)b >         at java.awt.GraphicsEnvironment.getLocalGraphicsEnvironment(GraphicsEnvironment.java:62)* >         at awttest.main(awttest.java:13) > 4 > The same test code runs fine on Linux/Windows etc. >   > Any help would be appreciated. >  > Thanks >  > Theo   ------------------------------  # Date: Wed, 10 Sep 2003 22:13:34 GMT # From: hoff@hp.nospam (Hoff Hoffman)>5 Subject: Re: Solaris Security (was: OpenVMS Security)c1 Message-ID: <imN7b.4427$Rc.1670@news.cpqcorp.net>t  E   As a comparsion to the five year old CERT reports cited, as well asrH   the the Bind 8 bug found in early TCP/IP Services V5.3 releases though.   not in older releases nor in current ECOs...  F   91 matches for Solaris among the CERT Advisories -- not VulerabilityD   Notes, or other areas -- just advisories.   Twenty breaches in theE   first page of the display, from the month of July 2003 -- the other 7   five reports on the first page are newer than July.     7   The next twenty five look to be heavily in July, too.a  (   Bind bugs reported in CA-1999-14, too.    K CERT Advisory CA-2001-15 Buffer Overflow In Sun Solaris in.lpd Print Daemon n .. CERT  Advisory CA-2001-15 Buffer Overflow In Sun Solaris in.lpd Print Daemon ... Solaris 2.6 for SPARC ...G http://www.cert.org/advisories/CA-2001-15.html - 14.8KB - solaris: 21 	i	 28 Jul 03n  I CERT Advisory CA-2002-34 Buffer Overflow in Solaris X Window Font Servicem .. CERT  Advisory CA-2002-34 Buffer Overflow in Solaris X Window Font Service ... Sun Microsystems Solaris 2.5.1 (Sparc/Intel) ...nG http://www.cert.org/advisories/CA-2002-34.html - 17.8KB - solaris: 21 	d	 28 Jul 03   = CERT Advisory CA-1993-19 Solaris System Startup Vulnerabilitya .. CERT  Advisory CA-1993-19 Solaris System Startup Vulnerability ... the system startup scripts on Solaris 2.x and Solaris x86 systems. The changes described ...rG http://www.cert.org/advisories/CA-1993-19.html - 10.8KB - solaris: 17 	B	 02 Sep 03   1 CERT Advisory CA-1995-09 Solaris ps Vulnerabilityh .. CERT  Advisory CA-1995-09 Solaris ps Vulnerability ... A vulnerability exists in Solaris systems that allows a race condition to be exploited to gain root access. The essential problem is that the ps(1 ... G http://www.cert.org/advisories/CA-1995-09.html - 18.8KB - solaris: 17 	 	 28 Jul 03h   Find SimilarC CERT Advisory CA-1996-15 Vulnerability in Solaris 2.5 KCMS programs  .. CERT  Advisory CA-1996-15 Vulnerability in Solaris 2.5 KCMS programs ... contact information has been replaced with CERT/CC contact information. ... Note that this vulnerability also affects Solaris 2.5.1. ... G http://www.cert.org/advisories/CA-1996-15.html - 13.1KB - solaris: 17 	s	 28 Jul 03   H http://www.cert.org/advisories/ CA-93.09a.SunOS.expreserve.vulnerability .. 1, 1993 REVISION NOTICE: SunOS/Solaris /usr/lib/expreserve Vulnerability ... 4.1.2, 4.1.3, Solaris 2.0 (SunOS 5.0), Solaris 2.1 (SunOS 5.1 ...e` http://www.cert.org/advisories/ CA-93.09a.SunOS.expreserve.vulnerability - 4.8KB - solaris: 17 		 16 Mar 01   ; CERT Advisory CA-1996-16 Vulnerability in Solaris admintool  .. CERT  Advisory CA-1996-16 Vulnerability in Solaris admintool ... has received a report of a vulnerability in the Sun Microsystems Solaris 2.x distribution involving the program admintool. This program is used ...G http://www.cert.org/advisories/CA-1996-16.html - 13.7KB - solaris: 13 	h	 28 Jul 03a  6 CERT Advisory CA-1996-17 Vulnerability in Solaris vold .. known to be present in Solaris 2.4 and Solaris 2.5. Solaris distributions prior ... 2.4 and Solaris 2.5. Solaris distributions prior to Solaris 2.4 are also expected ...G http://www.cert.org/advisories/CA-1996-17.html - 15.1KB - solaris: 13 	s	 28 Jul 03e  , CERT Advisory CA-2001-11 sadmind/IIS Worm...  ... Systems running unpatched versions of Solaris up to, and including, Solaris 7 ... worm exploits a vulnerability in Solaris systems and subsequently installs software ...E http://www.cert.org/advisories/CA-2001-11.html - 16.6KB  - solaris: 9   M ---------------------------- #include <rtfaq.h> -----------------------------gK     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faqeN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.come   ------------------------------  % Date: Wed, 10 Sep 2003 21:55:21 +0100 ( From: "John Travell" <john@jomatech.com> Subject: Re: VMS Upgrade Needede9 Message-ID: <bjo33t$l54u4$1@ID-120847.news.uni-berlin.de>   4 "Don Sykes" <anonymous@pacbell.net> wrote in message% news:3F5F8C36.D28312B2@pacbell.net...l > Can anyone help?0 > I need to upgrade from my VMS 7.2 to VMS7.2-2.C > I've been all over the HP site looking for a downloadable file toyA > accomplish this, but can't seem to find one. I see all kinds offC > downloads in ftp://ftp.service.digital.com/public/vms/axp/v7.2-2/sF > like .CHECKSUM .README & layered products, but where's the file that > will do the VMS upgrade?  L You didn't say if you are running VAX or Alpha. I presume Alpha as V7.2-2 is' Alpha only, and does not exist for VAX.d  G There is NO single file to do this. It is a full upgrade which replaced H hundreds, possibly thousands, of files. Usually done from CD these days.     -- John Travell" Independent VMS crashdump analyst. john- at - jomatech - dot - comi +44-(0)23-92552229 http://www.jomatech.com/       --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).A Version: 6.0.515 / Virus Database: 313 - Release Date: 01/09/2003,   ------------------------------  # Date: Wed, 10 Sep 2003 20:38:24 GMTa' From: Don Sykes <anonymous@pacbell.net>e Subject: VMS Upgrade Neededl+ Message-ID: <3F5F8C36.D28312B2@pacbell.net>e   Can anyone help?. I need to upgrade from my VMS 7.2 to VMS7.2-2.A I've been all over the HP site looking for a downloadable file toD? accomplish this, but can't seem to find one. I see all kinds of A downloads in ftp://ftp.service.digital.com/public/vms/axp/v7.2-2/SD like .CHECKSUM .README & layered products, but where's the file that will do the VMS upgrade?   -- r   Have VMS, Will Travelf Wire paladin, San Franciscoa   (paladinATalphaseDOTcom)   ------------------------------  % Date: Wed, 10 Sep 2003 22:48:40 +0200f9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com>  Subject: Re: VMS Upgrade Needede' Message-ID: <3F5F8E28.CD5E7951@aaa.com>u  . I don't think there are any complete VMS kits. Why do think there are ?  	 Jan-Erik.M   Don Sykes wrote: >  > Can anyone help?0 > I need to upgrade from my VMS 7.2 to VMS7.2-2.C > I've been all over the HP site looking for a downloadable file to A > accomplish this, but can't seem to find one. I see all kinds ofwC > downloads in ftp://ftp.service.digital.com/public/vms/axp/v7.2-2/cF > like .CHECKSUM .README & layered products, but where's the file that > will do the VMS upgrade? >  > -- >  > Have VMS, Will Travel/ > Wire paladin, San Franciscoe >  > (paladinATalphaseDOTcom)   ------------------------------  # Date: Wed, 10 Sep 2003 22:58:13 GMTl# From: hoff@hp.nospam (Hoff Hoffman)s Subject: Re: VMS Upgrade Needed-1 Message-ID: <90O7b.4434$Rc.3438@news.cpqcorp.net>x  d In article <bjo33t$l54u4$1@ID-120847.news.uni-berlin.de>, "John Travell" <john@jomatech.com> writes: :25 :"Don Sykes" <anonymous@pacbell.net> wrote in message & :news:3F5F8C36.D28312B2@pacbell.net... :> Can anyone help? 1 :> I need to upgrade from my VMS 7.2 to VMS7.2-2..D :> I've been all over the HP site looking for a downloadable file toB :> accomplish this, but can't seem to find one. I see all kinds ofD :> downloads in ftp://ftp.service.digital.com/public/vms/axp/v7.2-2/G :> like .CHECKSUM .README & layered products, but where's the file thats :> will do the VMS upgrade?   E   Those are patches and patch kits, and not product kits.  That's them2   old OpenVMS Alpha patch FTP server address, too.  3     ftp://ftp.support.compaq.com/public/vms/axp/...t  8   The new (but not yet entirely populated) patch URL is:  3     ftp://ftp.itrc.hp.com/openvms_patches/alpha/...   D :You didn't say if you are running VAX or Alpha. I presume Alpha as 2 :V7.2-2 is Alpha only, and does not exist for VAX.  C   A reasonable assumption, and one which the reference to the Alpha '   patch site URL would tend to confirm.b  H :There is NO single file to do this. It is a full upgrade which replacedI :hundreds, possibly thousands, of files. Usually done from CD these days.n  F   OpenVMS Alpha is only installed over the network (eg: InfoServer, orF   cluster-based operations), or from disk (or most commonly CD) media.  F   The most common and often the most efficient distribution is via CD,E   but contract customers should now be able to download kits directly C   off an FTP server.  (Contract customers should have received some.E   details of this kit availability, and hopefully I have not let thiseD   out prematurely. :-)  Once the FTP installation kit is downloaded,7   it must be unpacked onto a spare disk, and installed.     N  ---------------------------- #include <rtfaq.h> -----------------------------K     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faqrN  --------------------------- pure personal opinion ---------------------------E         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.comw   ------------------------------  # Date: Wed, 10 Sep 2003 23:33:36 GMTm' From: Don Sykes <anonymous@pacbell.net>r Subject: Re: VMS Upgrade NeededA+ Message-ID: <3F5FB547.4D851CB3@pacbell.net>    Hoff Hoffman wrote:e > f > In article <bjo33t$l54u4$1@ID-120847.news.uni-berlin.de>, "John Travell" <john@jomatech.com> writes: > :w7 > :"Don Sykes" <anonymous@pacbell.net> wrote in messageM( > :news:3F5F8C36.D28312B2@pacbell.net... > :> Can anyone help?T3 > :> I need to upgrade from my VMS 7.2 to VMS7.2-2.tF > :> I've been all over the HP site looking for a downloadable file toD > :> accomplish this, but can't seem to find one. I see all kinds ofF > :> downloads in ftp://ftp.service.digital.com/public/vms/axp/v7.2-2/I > :> like .CHECKSUM .README & layered products, but where's the file thatn > :> will do the VMS upgrade?n > G >   Those are patches and patch kits, and not product kits.  That's thes4 >   old OpenVMS Alpha patch FTP server address, too.  E Ouch. I thought I just needed a patch to bring it to the -2 level. It G sounds like I may as well upgrade to 7.3. I didn't want to do this now.s: I'm a firm believer in "if it ain't broke, don't fix it". ' And all this just to install Java 1.3. hH AND the only reason I need to install Java 1.3 is because the Java based! James SMTP facility requires it. uE AND the only reason to use James instead of HP/Compaq's TCPIP/SMTP iseB because I CAN'T MAKE THAT FACILITY SAY "NO SUCH USER" AND DROP THEG CONNECTION, during the connect phase, when there really IS no such usera anymore!   GGGRRRRRRRR...   > 5 >     ftp://ftp.support.compaq.com/public/vms/axp/...- > : >   The new (but not yet entirely populated) patch URL is: > 5 >     ftp://ftp.itrc.hp.com/openvms_patches/alpha/...e > E > :You didn't say if you are running VAX or Alpha. I presume Alpha as$4 > :V7.2-2 is Alpha only, and does not exist for VAX. > E >   A reasonable assumption, and one which the reference to the Alphae) >   patch site URL would tend to confirm.t > J > :There is NO single file to do this. It is a full upgrade which replacedK > :hundreds, possibly thousands, of files. Usually done from CD these days.i > H >   OpenVMS Alpha is only installed over the network (eg: InfoServer, orH >   cluster-based operations), or from disk (or most commonly CD) media. > H >   The most common and often the most efficient distribution is via CD,G >   but contract customers should now be able to download kits directlywE >   off an FTP server.  (Contract customers should have received some G >   details of this kit availability, and hopefully I have not let this F >   out prematurely. :-)  Once the FTP installation kit is downloaded,9 >   it must be unpacked onto a spare disk, and installed./ > P >  ---------------------------- #include <rtfaq.h> -----------------------------M >     For additional, please see the OpenVMS FAQ -- www.hp.com/go/openvms/faqoP >  --------------------------- pure personal opinion ---------------------------G >         Hoff (Stephen) Hoffman   OpenVMS Engineering   hoff[at]hp.com'   -- t   Have VMS, Will Travele Wire paladin, San Franciscoy   (paladinATalphaseDOTcom)   ------------------------------  % Date: Thu, 11 Sep 2003 02:16:35 +0800x, From: Paul Repacholi <prep@prep.synonet.com>! Subject: Re: VMS website updated.$- Message-ID: <87ekyojypo.fsf@prep.synonet.com>o  * "John Travell" <john@jomatech.com> writes:  M > If anyone who has any of those other browsers can bother to look at my site-) > with them, I would appreciate feedback.S  @ Works fine with Opera 7.11 with js enabled. Why not use ordinary> links? Touting the security of VMS then using Javascript seems& somewhat contarian to say the least :)   -- h< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.D@                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Thu, 11 Sep 2003 02:10:48 +08008, From: Paul Repacholi <prep@prep.synonet.com>! Subject: Re: VMS website updated.t- Message-ID: <87iso0jyzb.fsf@prep.synonet.com>t  * "John Travell" <john@jomatech.com> writes:  E > If anyone who has any of those other browsers can bother to look atr1 > my site with them, I would appreciate feedback.   , The page refs don't work with Lynx on linux.   Kill the javashite!    -- f< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.m@                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Wed, 10 Sep 2003 21:34:50 +0100 ( From: "John Travell" <john@jomatech.com>! Subject: Re: VMS website updated.l9 Message-ID: <bjo2kv$kr94l$1@ID-120847.news.uni-berlin.de>t  9 "Paul Repacholi" <prep@prep.synonet.com> wrote in messages' news:87ekyojypo.fsf@prep.synonet.com...e, > "John Travell" <john@jomatech.com> writes: >lJ > > If anyone who has any of those other browsers can bother to look at my site+ > > with them, I would appreciate feedback.  > B > Works fine with Opera 7.11 with js enabled. Why not use ordinary@ > links? Touting the security of VMS then using Javascript seems( > somewhat contarian to say the least :) >e- Why? MY JavaScript is not a security problem!l  C The page is substantially standards compliant, and unlike a massiveoK proportion of the internet it does not demand the use of non-cross-platform J plugins. It is fully operational on a browser (Mozilla) running on VMS andJ nearly every other computing environment available. I acknowledge the lack on VAX.-L The overwhelming majority of users are running browsers that see the page inA it's full glory, and most of the few that don't see an acceptable- alternative.G I grant that I can do more to support the trivial minority of text-only I browser users, and may be able to also better support machine readabilityIJ for braille readers. I am now learning about those tiny minority of users,F and how they view the internet, and while I must aim to satify 100% ofK potential viewers, it is an impossible goal. I do not speak any non-englisheK languages, and do not have the funds to pay people to translate the contentsG for me, so already I have lost everyone who doesn't understand english.-H A significant part of the problem is that many people used to seeing theH flashy stuff see good old fashioned basic html as intolerably boring andE promptly move on. Perhaps they associate the amount of time and moneyhD spent^h^h^h^h^hwasted on the flashy crap as a measure of a companies interest in their customers.H Note that MY site works better in Mozilla than does much of HP's site...   -- John Travell" Independent VMS crashdump analyst. john- at - jomatech - dot - come +44-(0)23-92552229 http://www.jomatech.com/       ---r& Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).A Version: 6.0.515 / Virus Database: 313 - Release Date: 01/09/2003    ------------------------------  + Date: Wed, 10 Sep 2003 16:58:14 -0500 (CDT)m From: sms@antinode.org! Subject: Re: VMS website updated.u) Message-ID: <03091016581486@antinode.org>p  ( From: "John Travell" <john@jomatech.com>  D > > Works fine with Opera 7.11 with js enabled. Why not use ordinaryB > > links? Touting the security of VMS then using Javascript seems* > > somewhat contarian to say the least :) > >a/ > Why? MY JavaScript is not a security problem!i  C    Please explain how I can set my browser to enable JavaScript forpC trustworthy sites only.  If I have to switch it manually, it's morep trouble than it's worth.  H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode.org     Saint Paul  MN  55105-2547l   ------------------------------   Date: 11 Sep 03 07:47:24 +0200) From: p_sture@elias.decus.ch (Paul Sture)h! Subject: Re: VMS website updated.t) Message-ID: <XUiQVCU3BEqk@elias.decus.ch>n  d In article <bjo2kv$kr94l$1@ID-120847.news.uni-berlin.de>, "John Travell" <john@jomatech.com> writes: > ; > "Paul Repacholi" <prep@prep.synonet.com> wrote in messaget) > news:87ekyojypo.fsf@prep.synonet.com...e- >> "John Travell" <john@jomatech.com> writes:n >>K >> > If anyone who has any of those other browsers can bother to look at my  > site, >> > with them, I would appreciate feedback. >>C >> Works fine with Opera 7.11 with js enabled. Why not use ordinaryWA >> links? Touting the security of VMS then using Javascript seems ) >> somewhat contarian to say the least :)n >>/ > Why? MY JavaScript is not a security problem!  >i  ( Oh yes it is. The very fact that you say  H "I (John Travell, john@jomatech.com) personally CERTIFY that there is NOH malicious code in this document or anywhere on this OpenVMS hosted site,* wholly and exclusively maintained by me."   K means that you acknowledge that other sites may have malicious code, and bydI persuading folks to switch it on, you leave them open to attack from such  sites.      sE > The page is substantially standards compliant, and unlike a massivenM > proportion of the internet it does not demand the use of non-cross-platform5L > plugins. It is fully operational on a browser (Mozilla) running on VMS andL > nearly every other computing environment available. I acknowledge the lack	 > on VAX.nN > The overwhelming majority of users are running browsers that see the page inC > it's full glory, and most of the few that don't see an acceptableC > alternative.  2 You idea of its full glory may not be everybody's.  I > I grant that I can do more to support the trivial minority of text-onlyoK > browser users, and may be able to also better support machine readabilityLL > for braille readers. I am now learning about those tiny minority of users,H > and how they view the internet, and while I must aim to satify 100% ofM > potential viewers, it is an impossible goal. I do not speak any non-englishdM > languages, and do not have the funds to pay people to translate the contentnI > for me, so already I have lost everyone who doesn't understand english.i  D Try www.apple.com. As I pointed out here recently, it is possible toI navigate their site and determine product prices using Netscape 3.03 witho Javascript and Java disabled.)  D Now, please try viewing your site with Netscape 3.03 with Javascript8 enabled. Not exactly anybody's idea of its "full glory".  @ Given that until recently this was the only browser which workedI satisfactorily under VMS, and still is for VAX users, please ask yourself  which market you are targeting.d  J > A significant part of the problem is that many people used to seeing theJ > flashy stuff see good old fashioned basic html as intolerably boring andG > promptly move on. Perhaps they associate the amount of time and moneyhF > spent^h^h^h^h^hwasted on the flashy crap as a measure of a companies > interest in their customers.  I But you are trying to force that flashy stuff onto those who have alreadyu4 made the decision that they prefer to do without it.  J > Note that MY site works better in Mozilla than does much of HP's site... >  So what?  = Another point to consider is that Mozilla uses so much memoryqH that at work I run it from a server with 4 GB RAM, rather then my XP900. Not everyone has that luxury.    ------------------------------   End of INFO-VAX 2003.503 ************************)rU Subject: Re: OpenVMS clusters (Alpha) - standalone workstation joined to the cluster?t0 Message-ID: <X_K7b.4418$FP7.62@news.cpqcorp.m.17izښ6v,hsm[6:B3&1k}̠NW<e'*Vuz,r6j&R~E$Zm Zô_p˽'^5yChZ92y=΍،[{jZ	Ǻ`.է|Miw{d]Jil:%	96j3<Hq+Nim۶m۶m۶m۶m۶m=3}y'J}T*gUV=J}BwA@CoOsƓF^5|4R^]SRR+G3&Ɨ+'M@9͖z2"
4axf/R7rčzxTv~}#E7SVoT;}.{}~M܁[K[2HڵZb͆t
JujKI%/NQLέ<CJi&#USc̼+F7xwOLE}Ll_5bCwkaIGYe˝~.BzkN
Vt5cDE1\֦D[5m,lӢcmsE"<V+v
@KwZ8F؇fa?]uMUaV>NE`V~Xk.m>]/e\&eX->Wf!4.ܖ	/ueU6vgl̼؋Cg|;w]!TDF%;M

̍N#Mw+`.N<BUH|yڮ[eԻR{. U

~7 9O֊Be9s|;9y]0.pqtI!oq^GOT0R5yk|t+`N >=7Y!sl3cO,;65$01ִ+T
̢
*nҫc{Fwi