1 INFO-VAX	Thu, 23 Sep 2004	Volume 2004 : Issue 528       Contents:> Re: 1920x1200 resolution with VMS and the HP L2335 LCD monitor) Re: AlphaServer DS10L - IDE disk question , Re: Can YOU save the Hubble Space Telescope?, Re: Can YOU save the Hubble Space Telescope?, Re: Can YOU save the Hubble Space Telescope?, Re: Can YOU save the Hubble Space Telescope?, Re: Can YOU save the Hubble Space Telescope?, Re: Can YOU save the Hubble Space Telescope?, Re: Can YOU save the Hubble Space Telescope?, Re: Can YOU save the Hubble Space Telescope?$ Re: HOw to convert seq to ascii file$ Re: HOw to convert seq to ascii file Re: Login using UAF? Help  looking for console firmware  Re: looking for console firmware  Re: looking for console firmware Re: McNeally understands Re: Mixed Cluster certification  OT: Spam protection software Re: OT: Sun's fighting chance  Re: OT: Sun's fighting chance  Re: OT: Sun's fighting chance  Re: OT: Sun's fighting chance  Re: OT: Sun's fighting chance , Re: Readable error log while process running Re: RMS and threads + Re: search txt file for a specified string? % Re: Sniffer Can't see cluster traffic % Re: Sniffer Can't see cluster traffic % Re: Sniffer Can't see cluster traffic % Re: Sniffer Can't see cluster traffic % Re: Sniffer Can't see cluster traffic  Re: VMS on IBM's Itanium Re: VMS on IBM's Itanium5 Re: Windoze not rebooted monthly shuts down airports! 5 Re: Windoze not rebooted monthly shuts down airports! : Re: ZIP "-V" v. UNIX, et al.: Problem, possible solutions.< Re: [DCL REQUEST] New ignore keyword for DIFFERENCES command  F ----------------------------------------------------------------------  # Date: Wed, 22 Sep 2004 17:40:29 GMT 4 From: "Fred Kleinsorge" <fred.nospam@nospam.dec.com>G Subject: Re: 1920x1200 resolution with VMS and the HP L2335 LCD monitor 3 Message-ID: <hOi4d.11453$e86.5790@news.cpqcorp.net>   @ "Vance Haemmerle" <vance@toyvax.Glendale.CA.US> wrote in message7 news:Ml94d.18230$QJ3.4384@newssvr21.news.prodigy.com...  > Fred Kleinsorge wrote:J > > We (VMS) have 2 of these flat panels on order to test them on a Radeon 7500, K > > so I can't tell you much about the panel yet.  I do have several 2035's I > > (1600x1200), and 1825's (1280x1024), which appear to work fine (but I  haveI > > not tried a P350).  The 2335 which is fairly new, was never tested or G > > qualified on the P350 (or any of the cards yet) which stopped being  built . > > and sold long before this FP was designed. > H >    I look forward to see what you think.  Especially as I read there's2 > no VMS support for 1920x1200 on the Radeon 7500. >   H The Radeon driver table shows 1920 x 1200 at 60 and 70Hz in 8, 16 and 32 (24) bit depths.    K > >  The P350 only has an analog output.  I know that the 2335 has a native K > > display of 1920x1200 @ 60Hz -- at least in digital (DVI) mode.  I would H > > expect it to be able to sync on an analog signal of that rate -- but until I I > > get one in, I won't be able to tell.  It may be that the timing being 6 > > generated is outside of what the panel can handle. > H >    With an LCD, native is native, for analog or digital.  For DVI modeJ > it has to be a reduced blanking 1920x1200 since the regular DVI standard$ > only supports up though 1600x1200. > J >    The 2335 has several inputs, Digital (DVI), Analog (DVI) (the DVI canF > be used as either), Analog (D-SUB), S-Video, Composite and ComponentB > Video.  I have the XP1000 hooked up to Analog (D-SUB) and my DECG > 3000/900 hooked up to the Analog (DVI).  Like I mentioned previously, C > the XP1000 is set to 1920x1200.  The 3000/900 has a ZLX-E2 set to G > 1280x1024@72Hz.  I also have an HDTV tuner hooked up to the component 6 > inputs and an watch it either full screen or as PIP. > I >    The monitor syncs fine to both signals.  It recognizes the 1280x1024 F > correctly and displays it correct.  Though the mode display calls itG > 73Hz.  It isn't even one of the three presets for the monitor at this E > resolution which are 60, 75 and 85Hz.  The monitor has three custom H > scaling options:  Fill to Screen, Fill to Aspect Ratio and One to One.D > All three of these settings works as expected for this resolution. > J >    For the P350 the monitor sees 1920x1200x60 as 1600x1200x60.  It seemsF > that whatever the P350 is putting out in 1920x1200 mode, the monitorE > doesn't see it as a 1920x1200 standard.  This means black side bars G > in the two latter scaling modes.  For the Fill to screen mode I don't C > think it is measuring 1920 samples across the screen but 1600 and @ > stretching them across the screen.  When I type vertical bars,G > ||||||||||||||||||||||, some are wider than others, though I guess it ( > could be some sort of aliasing effect. > K >    In the 1280x1024 mode, One to One, the vertical bars are all the same. E > In Fill to Screen or Fill to Aspect Ratio they are uneven because I D > believe they're being sampled at 1024 and then stretched digitally > to 1920 or 1500. >  > J > > The P350 (or any of our drivers except perhaps the AlphaBook 1) driver isn't K > > smart enough to generate a 1600x1200 output when set to 1920x1200 - nor  doB > > we inquire the monitor capabilities (which can be done by some
 signallingF > > techniques on most video cards).  It is entirely possible that the monitor K > > however is doing something funky when seeing an input it can't handle -  the ( > > FP displays are quite sophisticated. > C >    I guess my question is what is really the difference between a H > 1920x1200 and 1600x1200 analog signal and is the P350s video differentI > between the two modes?  At a vertical refresh rate of 60Hz that means a I > horizontal rate of 72KHz, not counting the vertical interval size.  The J > monitor presets say 1600x1200 has a horizontal frequency of 75.00kHz butG > the 1920x1200 has 74.56kHz.  If it was a regular CRT it would stretch H > each line across the whole screen and wouldn't matter.  The LCD treats7 > the modes differently depending what it thinks it is.  >   = The P350 does 1920x1200 @ 24 bits at 60Hz with a dot clock of K 94.727479MHz  - while it has a 75Hz mode, this exceeds the RAMDAC specs --  E so it isn't really supported at 75Hz reliably (the dot clock would be  243.409349).  A 1600x1200 yields 161.795626MHz at 60Hz, and 202.500215MHz at 75Hz   L The R7500 on the other hand has a dot clock of 193.16MHz at 1920x1200 @60Hz.  L [video timings are strange things.  i think the p350 timings pre-date a VESAC standard.  in fact, i think the radeon timings were from a proposed 	 standard]   L Check the output in DECW$SERVER_0_ERROR.LOG to verify that the server/driver9 set the mode you requested for 60Hz - and don't use 75Hz.    ------------------------------  % Date: Wed, 22 Sep 2004 14:48:55 -0400 " From: Glenn Everhart <gce@gce.com>2 Subject: Re: AlphaServer DS10L - IDE disk question& Message-ID: <4151C917.6090500@gce.com>   Gremlin wrote: > Thanks for that  > 3 > I guess that an 80-120Gb drive should be OK then? ? > "Paul A. Jacobi" <Paul.Jacobi@nospam.hp.com> wrote in message ( > news:4150a691$1@usenet01.boi.hp.com... > 0 >>"Gremlin" <not-here@all.mate> wrote in message+ >>news:415017ae$1@duster.adelaide.on.net...  >>= >>>Can a AlphaServer DS10L use a large IDE drive - say 200Gb?  >>J >>The SRM console and the VMS DQDRIVER has a 137Gb limit on IDE drives.  I1 >>don't know any plans to remove this limitation.  >>J >>Be aware that even Windows XP requires SP1 to support drives larger than< >>137Gb.  See http://news.zdnet.com/2100-9584_22-946929.html >> >> >> >>Paul A. Jacobi >>HP OpenVMS Systems Group >>Nashua, NH >> >> >  >  > I The issue is that IDE has a limit at 137GB. A newer IDE standard has been I made to expand this, not very long ago, but lots of older IDE controllers  know only that size or less.  K If you want a larger IDE disk on VMS, it is said that an Acard IDE <-> SCSI O adapter will handle the new IDE standard. It will hit the 1TB or 2TB (depending D on sign bit in block number; don't count on over 1TB) limit in SCSI.  I Drives that will exceed this size seem not too far off. ODS2 or ODS5 have  limits at this size also.   J Note that negative block numbers will hit various problems in dkdriver and
 elsewhere.   ------------------------------    Date: 22 Sep 2004 13:12:04 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) 5 Subject: Re: Can YOU save the Hubble Space Telescope? 3 Message-ID: <ZKb1KDvEpQZy@eisner.encompasserve.org>   q In article <QOgD8bx7+LOC@eisner.encompasserve.org>, koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:   J >    If you're thinking of responding on usenet, you're a little bit lost.  N And those who get lost are not prime candidates for programming a space ship ?   ------------------------------  % Date: Wed, 22 Sep 2004 11:08:28 -0700 # From: "Tom Linden" <tom@kednos.com> 5 Subject: Re: Can YOU save the Hubble Space Telescope? ( Message-ID: <opseqp0ea5zgicya@hyrrokkin>  I On 22 Sep 2004 13:12:04 -0500, Larry Kilgallen <Kilgallen@SpamCop.net> w=  rote:   I > In article <QOgD8bx7+LOC@eisner.encompasserve.org>, koehler@eisner.nos= + pam.encompasserve.org (Bob Koehler) writes:  > I >>    If you're thinking of responding on usenet, you're a little bit lo=  st.  > I > And those who get lost are not prime candidates for programming a spac=  e ship ? >  And they are ... Lost in Space?    ------------------------------  % Date: Wed, 22 Sep 2004 15:58:41 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 5 Subject: Re: Can YOU save the Hubble Space Telescope? , Message-ID: <4151D96E.22F0A87C@teksavvy.com>   Bob Koehler wrote:J >    Since the Senate just passed the funds, we are anticipating work willI >    start soon on a robotic servicing mission to extend the life of HST.   M What are the odds of the shuttle doing what it had been designed to do and go 4 to hubble, service it and eventually bring it back ?  L Can a robotic mission really "service" Hubble ? I was aware of an attempt toM robotically attach some propulsion system that would allow Hubble to de-orbit N over ocean and not break international agreements to never again allow a largeG object to fall back uncontrolled on earth, but there is a big different K between destructively attaching a new module (no attach points exists)  and ) robotically performing surgery on Hubble.    ------------------------------  % Date: Wed, 22 Sep 2004 15:49:45 -0400 # From: "John Smith" <a@nonymous.com> 5 Subject: Re: Can YOU save the Hubble Space Telescope? , Message-ID: <b66dnfSAnMrGSszcRVn-tQ@igs.net>   Bob Koehler wrote:H >    Unofficially here's your chance to save the Hubble Space Telescope. > E >    Since the Senate just passed the funds, we are anticipating work F >    will start soon on a robotic servicing mission to extend the life	 > of HST.  > ) >    I cannot announce jobs at this time: A >       1)  I'm not a spokesman for Computer Sciences Corporation & >       2)  I have no hiring authority& >       3)  I'm not in human resources& >       4)  we don't have the work yetG >    When we do start seeking candidates, an offical spokesman will add * >    the usual legaleze about EEOC et. al. > H >    But we do plan to start seeking candidates soon.  When we do, we'll$ >    be looking for programmers who: > ? >       can explain why Windows is not a deterministic platform C >       program in more than one of C, C++, Ada, Java, Fortran, ... 2 >       program in more than one assembly languageG >       program on more than one of VMS, UNIX, MacOS, Windows, MVS, ... 4 >       program on one or more real-time embedded OS@ >       debug programs with logic analyzers, Jtags, and o-scopes2 >       have a little space cadet deep inside themH >       know physics, mathematics, mechanical, or electrical engineeringH >       do a lot of things that are not stricly programming, but require >       software expertise >  >    Location:  Greenbelt, MD.+ >    US Citizenship will likely be required     E Guess you won't be fielding a mission if you want the Canadians at MD K Robotics, ( http://www.mdrobotics.ca/wwdframe.html who will be suppying the ) robot system) to take out US citizenship.   G I heard that the MD Robotics SPDM proof-of-concept tests on Hubble-like  repairs went flawlessly.   ------------------------------    Date: 22 Sep 2004 16:04:39 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 5 Subject: Re: Can YOU save the Hubble Space Telescope? 3 Message-ID: <NiJFeakjvQgi@eisner.encompasserve.org>   R In article <b66dnfSAnMrGSszcRVn-tQ@igs.net>, "John Smith" <a@nonymous.com> writes: > G > Guess you won't be fielding a mission if you want the Canadians at MD M > Robotics, ( http://www.mdrobotics.ca/wwdframe.html who will be suppying the + > robot system) to take out US citizenship.   J    If the customer does place citizensship requirements I'm sure it won't G    be on work done in Canada and they will deal with making exceptions.    ------------------------------    Date: 22 Sep 2004 16:10:18 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 5 Subject: Re: Can YOU save the Hubble Space Telescope? 3 Message-ID: <alkfXHN+L0XL@eisner.encompasserve.org>   \ In article <4151D96E.22F0A87C@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes: > Bob Koehler wrote:K >>    Since the Senate just passed the funds, we are anticipating work will J >>    start soon on a robotic servicing mission to extend the life of HST. > O > What are the odds of the shuttle doing what it had been designed to do and go 6 > to hubble, service it and eventually bring it back ?  D    Zero.  The shuttle cannot be made safe for flight to HST prior toG    the expected end of life without some other kind of servicing in the     meantime.  N > Can a robotic mission really "service" Hubble ? I was aware of an attempt toO > robotically attach some propulsion system that would allow Hubble to de-orbit P > over ocean and not break international agreements to never again allow a largeI > object to fall back uncontrolled on earth, but there is a big different M > between destructively attaching a new module (no attach points exists)  and + > robotically performing surgery on Hubble.   D    Attach points do exist, and have been used by the shuttle severalN    times already.  NASA has at a minumum a requirement to grab HST and safely K    dispose of it.  Technology critical to doing a lot more is already well      demonstrated.  I've seen it.    ------------------------------  % Date: Wed, 22 Sep 2004 17:32:16 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 5 Subject: Re: Can YOU save the Hubble Space Telescope? , Message-ID: <4151EF56.573A3107@teksavvy.com>   John Smith wrote: G > Guess you won't be fielding a mission if you want the Canadians at MD M > Robotics, ( http://www.mdrobotics.ca/wwdframe.html who will be suppying the + > robot system) to take out US citizenship.   L There is HUGE difference between a contractor supplying a product delieveredG to the customer and a person working on site as essentially an employee - (whether on a contract, payroll or whatever).    ------------------------------  % Date: Wed, 22 Sep 2004 17:46:45 -0700 # From: "Tom Linden" <tom@kednos.com> 5 Subject: Re: Can YOU save the Hubble Space Telescope? ( Message-ID: <opseq8f7qrzgicya@hyrrokkin>  I On 22 Sep 2004 16:10:18 -0500, Bob Koehler <koehler@eisner.nospam.encomp=  asserve.org> wrote:   I > In article <4151D96E.22F0A87C@teksavvy.com>, JF Mezei <jfmezei.spamnot=  @teksavvy.com> writes: >> Bob Koehler wrote: I >>>    Since the Senate just passed the funds, we are anticipating work =  willI >>>    start soon on a robotic servicing mission to extend the life of H=  ST.  >>I >> What are the odds of the shuttle doing what it had been designed to d=  o and go7 >> to hubble, service it and eventually bring it back ?  > F >    Zero.  The shuttle cannot be made safe for flight to HST prior toI >    the expected end of life without some other kind of servicing in th=  e  >    meantime. > I >> Can a robotic mission really "service" Hubble ? I was aware of an att=  empt to I >> robotically attach some propulsion system that would allow Hubble to =  de-orbitI >> over ocean and not break international agreements to never again allo= 	 w a large I >> object to fall back uncontrolled on earth, but there is a big differe=  ntI >> between destructively attaching a new module (no attach points exists=  )  and, >> robotically performing surgery on Hubble. > F >    Attach points do exist, and have been used by the shuttle severalI >    times already.  NASA has at a minumum a requirement to grab HST and=   safely I >    dispose of it.  Technology critical to doing a lot more is already =  well! >    demonstrated.  I've seen it.   " Has anyone contacted Bruce Willis?   ------------------------------  % Date: Wed, 22 Sep 2004 16:04:33 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> - Subject: Re: HOw to convert seq to ascii file , Message-ID: <4151DACD.1C8FF728@teksavvy.com>   Chuck Aaron wrote: > E > I have a sequential file of userid's. How can I convert the file to 6 > ascii to where when I send it it does not look like: >  > useriduseriduseriduserid...  >  > but looks liks >  > userid > userid    + You need to provide us with with following:    DIR/FULL of the file,  and  DUMP/BLOCK=COUNT:1  N You need to find out what VMS thinks the attributes of the file are (DIR/FULL)O and then look at the file contents to see what the record separators are (DUMP)   4 Once you have that information, you can use SET FILEM file/ATTRIB=(RFM=xxx,RAT=yy) command to tell VMS what the real file structure M is so that it will properly interpret the text records/line delimitors inside 	 the file.    ------------------------------    Date: 22 Sep 2004 15:56:42 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) - Subject: Re: HOw to convert seq to ascii file 3 Message-ID: <ogyfV6TW$+qd@eisner.encompasserve.org>   [ In article <4151A099.40305@ceris.purdue.edu>, Chuck Aaron <caaron@ceris.purdue.edu> writes: F > I have a sequential file of userid's. How can I convert the file to 6 > ascii to where when I send it it does not look like: >  > useriduseriduseriduserid...  >  > but looks liks >  > userid > userid > userid  (    Gee, that all looks like ASCII to me.  B    You seem to be missing line delimiters.  Just how you got thereC    will tell us a lot about how to fix it.  Staring with a dir/full     may clue us in.   ------------------------------  % Date: Wed, 22 Sep 2004 15:52:19 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> " Subject: Re: Login using UAF? Help, Message-ID: <4151D7F0.94424AC8@teksavvy.com>   RT Carter wrote:F > are now 192.255.2.xxx (servers stayed the same). I know that we haveF > security set to only allow certain IP addresses into the VMS serversC > but I don't know where to change the access from 192.255.1.xxx to  > 192.255.2.xxx.  P This is done at the TCPIP stack level, well before it gets to the login process.  T If you are using the TCPIP Services stack, it is done from within the TCPIP utility.  N You can start with SHOW SERVICE TELNET/FULL to see if the blocks are in there.L After making changes, you need to: DISABLE SERVICE TELNET and ENABLE SERVICEO TELNET for changes to take effect. (not sure of impact on existing connections)    HELP SET SERVICE /ACCEPT and& HELP SET SERVICE/REJECT will help you.    D If it is not done at the TELNET service level, it can be done at the interface/communications level:    SHOW COMM/FULL   HELP SET COMM and  HELP SET CONF COMM    I (SET CONFIGURATION COMMUNICATIONS affects the permanent database for such @ settings, while SET COMMUNICATIONS affects only running system.)   ------------------------------  % Date: Wed, 22 Sep 2004 16:30:00 -0400 ) From: "Robert A. Healey" <healer@rpi.edu> % Subject: looking for console firmware + Message-ID: <2re5n1F198jr5U1@uni-berlin.de>    Hi.   I I'm running a VAX 8530 where the console HD recently died.  I've found a  G blank disk I can swap in, but I don't have any media to install on it.  G Would anyone know of a source for P/OS, the microcode for an 8530, and  - the VAX Console software to make it all work?    Thanks in advance.  
 Bob Healey healer@rpi.edu   ------------------------------  % Date: Wed, 22 Sep 2004 17:20:32 -0400 ) From: "Robert A. Healey" <healer@rpi.edu> ) Subject: Re: looking for console firmware + Message-ID: <2re8lqF19fkdrU1@uni-berlin.de>   D Just want to add that this is a hobby system, and as such, there is 8 nothing important on it that it needs to be ressurected.     Robert A. Healey wrote:  > Hi.  > K > I'm running a VAX 8530 where the console HD recently died.  I've found a  I > blank disk I can swap in, but I don't have any media to install on it.  I > Would anyone know of a source for P/OS, the microcode for an 8530, and  / > the VAX Console software to make it all work?  >  > Thanks in advance. >  > Bob Healey > healer@rpi.edu   ------------------------------   Date: 22 Sep 04 19:12:49 EDT) From: cook@wvnvms.wvnet.edu (George Cook) ) Subject: Re: looking for console firmware ! Message-ID: <l9vBQPm4pap9@wvnvms>   W In article <2re5n1F198jr5U1@uni-berlin.de>, "Robert A. Healey" <healer@rpi.edu> writes:  > Hi.  > K > I'm running a VAX 8530 where the console HD recently died.  I've found a  I > blank disk I can swap in, but I don't have any media to install on it.  I > Would anyone know of a source for P/OS, the microcode for an 8530, and  / > the VAX Console software to make it all work?   E I believe we donated the Pro (VAX Console) you currently have.  I did E some digging in our "archives" and found an RD52 disk drive which may H contain an older version (9.4 maybe) of the console software.  The driveE we sent with the Pro had version 10 if I recall correctly.  The drive ) is yours if you tell me where to send it.   A I still have the version 10 floppies somewhere, but it could take % several days of digging to find them.      George Cook  WVNET    ------------------------------  % Date: Wed, 22 Sep 2004 16:25:30 -0400 # From: "John Smith" <a@nonymous.com> ! Subject: Re: McNeally understands , Message-ID: <UYCdnZeEAI4hQszcRVn-iA@igs.net>   John Smith wrote:  > L http://story.news.yahoo.com/news?tmpl=story&cid=1620&ncid=1208&e=1&u=/sv/200& > 40921/tc_sv/suntriestowoowallstfirms > ! > Sun tries to woo Wall St. firms  > ' > By Therese Poletti and Dean Takahashi  > Mercury News > B > Sun Microsystems made some of its first sales in the early 1980sE > selling workstations to Wall Street. Now the struggling Santa Clara F > computer maker is returning to its roots with a campaign to win back2 > some of its former financial-industry customers. > D > President and Chief Operating Officer Jonathan Schwartz today willC > unveil a new line of Sun servers designed to woo Wall Street back A > into Sun's orbit. Schwartz hopes to convince these consumers of A > bleeding-edge technology that the new products are as cheap and B > powerful as the Linux-based servers that have stolen much of the? > company's Wall Street business over the past couple of years.  > D > After its founding in 1982, Sun scored its first customers on WallF > Street, where companies bought its networked workstations that couldG > crunch numbers and run financial software more cheaply than mainframe  > computers. > E > Sun began losing its grip on the information-technology departments C > of the world's biggest investment-banking firms and other trading 8 > houses during the bust that followed the dot-com boom. > 7 > Faced with lower trading volumes and excess capacity, F > financial-services firms began buying lower-cost servers using IntelF > chips and running the open-source Linux operating system. Instead ofD > supplying these new machines to Wall Street, Sun tried to stem theC > tide by promoting computers running Sun's own chips and software.  >  > Linux deployed NowB > Morgan Stanley, Merrill Lynch and other firms are using Linux inG > parts of their operations. TowerGroup, a research and consulting firm @ > that focuses on the financial-services industry, reported in aF > January study that Linux was deployed on 9 percent of all servers inB > the North American securities industry. TowerGroup projects thatF > Linux will continue to grow at an annual rate of 22 percent in North > America through 2006.  > H > Sun Chief Executive Scott McNealy in an interview said the company has: > refocused on low-cost technology for its core customers. > G > "We aren't going to announce Sun TV sets, or Sun digital cameras or a G > patent for a new Sun inkjet cartridge," said McNealy, taking a jab at @ > Silicon Valley rival Hewlett-Packard. "This event will be very9 > focused on solving complex network computing problems."  > B > John Fowler, executive vice president of network systems at Sun,D > noted that financial firms that switched to competitors' computersF > remain Sun customers. He noted that Sun servers still handle trading4 > and other transactions for many Wall Street firms. > E > McNealy, who will join Schwartz in New York next week, said Sun now @ > offers a more appealing mix of products, including Linux-basedB > servers that use Advanced Micro Devices' low-cost Opteron chips.G > Those servers start at $8,495 for a server with four microprocessors.  > F > McNealy said Sun also plans to tout Solaris 10, its upcoming versionE > of its Solaris operating system. Solaris 10 will run on both SPARC, G > Sun's proprietary processors, and on servers designed around both AMD  > chips and Intel chips. > F > Sun also plans to announce a new simplified data-storage system thatB > it says will help Wall Street firms deal with the Sarbanes-Oxley" > financial reporting regulations. > B > Joel Scotkin, CEO of Random Walk, which resells both Sun and IBME > equipment to financial firms, said Wall Street customers are likely B > to welcome the features of Solaris 10. The operating system willG > offer a new file system and the ability to isolate parts of computers E > into "containers." These containers protect one part of the machine E > that supports a particular application from the rest of the machine  > when another program crashes.  > E > But Scotkin noted that IBM's Power5-based servers are well-received > > on Wall Street and that IBM appears to be ahead in so-calledA > virtualization technology, which resembles Sun's containers and 0 > enables a customer to maximize a server's use. > F > `Frequent disconnect' Some analysts said Sun's financial problems ofF > recent years and a perception that it has lost focus has contributed6 > to the company's diminished presence on Wall Street. > F > "There has been a frequent disconnect from what customers have heardC > from headquarters and what they have heard from the sales force," F > said Michael Dortch, a principal business analyst at consulting firm > the Robert Frances Group.  > C > Sun recently completed its annual sales force reorganization, but G > Dortch said it is not clear if that will help. "It doesn't matter how E > they reorganize, it matters how it plays out at the customer site," 
 > he said. > F > For its part, Sun said it hopes to recapture its leading position in" > the financial-services industry. > D > "It's an opportunity to re-establish ourselves," said McNealy. "We > put the story back together."     	 See also:   L http://story.news.yahoo.com/news?tmpl=story&cid=1738&ncid=1208&e=10&u=/zd/20 040921/tc_zd/135712     < Financial Industry Insiders Warm Up to Sun's New Initiatives   Tue Sep 21, 5:04 PM ET   Darryl K. Taft - eWEEK  F NEW YORK-In his inimitable, unflappable style, Sun Microsystems Inc.'sK president and chief operating officer, Jonathan Schwartz, Tuesday delivered I Sun's message to the financial community: Sun is back and it's making its 2 comeback by returning to its roots on Wall Street.  I Analysts and customers welcomed the news, especially over advances in the B Solaris architecture and the company's new grid computing program.  D During an hour-long presentation to more than 100 financial servicesI customers and financial analysts at Sun's quarterly Network Computing '04 K event here, Schwartz outlined Sun's new initiatives, including new hardware J and software; a preview of the next refresh of the Solaris OS, Version 10;K global support for Linux (news - web sites) and Unix; and a new pay-for-use - computing offer around Sun's N1 Grid program.   I "Wall Street is the swamp from which we spawned," Schwartz said. "Sun was K built on Wall Street. You are our spawning grounds and this is the place we , believe we need to return to grow," he said.  F However, Schwartz said Wall Street gave Sun a shopping list, which theI company largely ignored and led to some lean years for the systems maker.   E The shopping list included: multi-platform Solaris, industry standard ? hardware, choice and interoperability, and innovation to define J price/performance-all things that Sun is now delivering in spades with its new offerings, Schwartz said.   C "You came knocking and we didn't listen," he said. "You said Sun is K proprietary and expensive, and we declined year over year for a few years."   I But, "What we've been doing for the past few years is getting back to our G roots so we can deliver real systems innovation to you," Schwartz said.   B Now, Sun offers Solaris on x86 for 249 different systems, he said.  D Meanwhile, the company said Solaris 10, which will ship by year end,J features Dynamic Tracing, N1 Grid Containers and Trusted Solaris. SchwartzG said Solaris 10 offers "extreme performance, utilization/consolidation, J unparalleled security, relentless availability, and rationalizes Linux andG Unix (news - web sites) with native Red Hat and SuSE execution." Sun is K involved in a project called Project Janus to run Linux applications native  on Solaris.   H Meanwhile, Solaris x86 now has an ecosystem of more than 700 application5 software partners offering 1,100 solutions, Sun said.   H Peter Lankford, senior vice president and head of enterprise informationK systems at Reuters Information Technology LLC, Oak Brook, Ill., said at the J event, "We're quite excited by Solaris 10. We're looking at it with a real intense focus on innovation."   D Miriam Soza, senior vice president for system development at ThomsonL Financial, a New York-based division of Thomson Corp., Stamford, Conn., saidC Thomson Financial uses Solaris x86 "on a number of mission-critical K applications. We also started testing Solaris 10 in preliminary stages, but % we gained 35 percent in performance."   K "We know we can win on performance, this company was built on performance," B Schwartz said. "We are absolutely targeting Red Hat specifically."  J Indeed, "Linux isn't so free anymore; it's about $1,000 per CPU," SchwartzB said. "Sun is about $700 a CPU. Linux is now a farming ground-more  opportunity for us to go after."  K Jean Bozman, research vice president, global enterprise server solutions at K International Data Corp., in Mountain View, Calif., said "What Sun answered > here is how they might approach managing a mixed environment."  J Schwartz announced a deal offering 50 percent off the price of Solaris for5 customers moving to Solaris from Red Hat Inc.'s line.   F However, many attendees said perhaps the most interesting of the day'sH announcements was the pay-for-use computing announcement around Sun's N1H Grid program, where Schwartz said Sun would give customers access to its1 computing grids at a cost of $1 per CPU per hour.   I "If you want to re-invent the computer industry around service then you'd B want to be able to deliver computing as a service," Schwartz said.  I "This is what we think is an evolution and not hosting, but instead we're I saying 'here's our infrastructure and if you can map your workloads to it D will work," he said. "In the long run this is truly virtualization."  L Schwartz said the first evolution of the N1 Grid offering is a computational grid.   L "We don't want to be in the business of owning and operating datacenters, weH want to be in the business of expressing our computing as a service," heL said. "The N1 Grid for now is going to be a computational grid and over time> we'll look at the technical hurdles to get to a service grid."  L Mark Stahlman, managing director of equity research at Caris & Company Inc.,I of New York, said: "Sun has dramatically increased the number of customer J requests for high-level meetings, which have been building at a rapid rate over the last three quarters."  E Stahlman said the N1 Grid offer means "Sun will get another 100 majorhI accounts asking for meetings. At some point it becomes a liability if youc haven't wet with Sun."  K IDC's Bozman said, "They [Sun] realize they need a much bigger footprint inaL the marketplace. Right now Sun gets 35 percent of its revenue from services.+ Services are going to continue to develop."e  I Meanwhile, Schwartz said the relationship between Sun and Microsoft Corp.bL continues to blossom, with the first phase of the plan to cooperate focusingK on identity and Web services interoperability. "Identity is the fundamental C element behind commerce, that's why we're starting there," he said.e  H "We send hardware to Microsoft now and that's a bizarre thing," SchwartzF said. "But we have dialogue with them on how to make the hardware more performant."   ------------------------------  % Date: Wed, 22 Sep 2004 20:12:37 -0400g* From: Joel Loveless <joell@mindspring.com>( Subject: Re: Mixed Cluster certification8 Message-ID: <gi34l0lmhah5m07ds0lnkbnicdlcnharo9@4ax.com>  @ We did rolling upgrades on our cluster by taking one system to aE higher version first to work out any issues. Our first upgrade was to E go to 7.3-1 on one system with the remaining 11 on 7.2-1. It is ok toMF step up to 7.3-1 from 7.2-1, not sure if you can go directly to 7.3-2.? You will need to check the SPD to see if you can jump to 7.3-2,I otherwise do 2 upgrades. b  A The key to doing mixed upgrades in a cluster is to make sure thatFE software packages are compatable with the higher version. Examples of-E what we had to be concerned with were shadowing, host based RAID, RMS F global buffering and Multinet to name a few. We allways ran full patchD reports and applied all patches for each software package, includingE VMS, before making the upgrade. We also applied the latest patches toC' the new version of VMS at upgrade time.Q  D  We would then run this for several weeks before proceeding with the= remaining nodes. The last 2 upgrades took about 8-10 weeks toaB complete. We encountered no issues each time leaving it as a mixedF Alpha cluster, systems included GS140 and GS160 systems. You will alsoC need to upgrade the firmware for each system to support the versioni? OS, I recommend that be done on all systems before OS upgrades.   A In order to facilitate a quick return to the previous version, welA always made a copy of the system disk and upgraded the copy. ThisaB leaves an easy reboot to back out the upgrade if needed. In our 12E node cluster, we had 3 system disks and each system had a unique root1 identifier.   @ One word of advise, a GS160 with a shared galaxy license pak forE partitions does not support rolling reboots. I always had to shutdown2B all partitions and bring them up on the same VMS version. I am notA sure if this was a license issue or the fact that we had 2 systemn? disks during the upgrade. Maybe someone out here can correct myt
 procedure.  C It may seem like a daunting task but it's not. Just take the neededWE steps to assure compatibility and back everything up. Ours was smoothtF after the first disk, just document your steps and proceed. We had theA advantage of a development cluster to test our steps which helped  develop procedures.,   Good luck, t Joel.c  4 On Tue, 21 Sep 2004 17:43:37 +0000 (UTC), "Robert H"$ <robert.heyes@btinternet.com> wrote:  L >Can anyone direct me to a document which highlights what is possible for a ( >mixed cluster, or tell me if they know. > M >I am looking at upgrading one of our core systems but I cannot upgrade that @L >Alpha due to other software restrictions so I'm looking at getting another M >Alpha, which would be at the bare bones be 7.3 while my existing cluster is  N >7.2-1. I am guessing I would be safe clustering 7.2-1 with a 7.3 machine but L >anything above 7.3 would not be certified...? Ideally I would love to take A >the new machine to 7.3-2 but I dont think this will be possible.  >e >TVM,r >Rob r >c   ------------------------------  % Date: Thu, 23 Sep 2004 01:04:09 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>p% Subject: OT: Spam protection softwarer+ Message-ID: <41525945.B831F8E@teksavvy.com>0  6 SpamAssasin has switched to an APACHE type of licence.  " However, take a look at this page:/ http://spamassassin.apache.org/tests_3_0_x.htmln  N It contains the list of tests it performs by default.  It is a VERY LONG list,N and makes you wonder whether any of your own emails get filetered out or not !   ------------------------------  % Date: Wed, 22 Sep 2004 15:42:14 -0400|- From: JF Mezei <jfmezei.spamnot@teksavvy.com>h& Subject: Re: OT: Sun's fighting chance, Message-ID: <4151D593.ED239134@teksavvy.com>   Keith Parris wrote:eJ > This statement shows that Sun clearly doesn't understand the Open Source; > movement. And so their strategy is doomed from the start.t   Au contraire !  J Sun realise it could not attack the open source movement, so it started toK attack those profiting from it. And since corporations won't be doanloadingaL Linux binaries from the net, they will turn most likely to Red hat for sales and support of the product.b  H So Red Hat is the commercial vendor for a product called Linux. In otherN words, when a corporation considers Linux vs Solaris, it is Red Hat versus Sun who compete in all likelyhood.  M And yes, if Red Hat disapears, someone else will take over. But it takes timehG to build the image. So if Sun can work to keep those vendors "small" bysJ constantly weakening them, then Linix won't grow so fast in the enterpriseT sector because corporations won't be seeing a trustable vendor of the linux systems.   ------------------------------  % Date: Wed, 22 Sep 2004 15:52:04 -0400t# From: "John Smith" <a@nonymous.com>9& Subject: Re: OT: Sun's fighting chance, Message-ID: <mNCdnRua_bp7SszcRVn-uQ@igs.net>   Keith Parris wrote:l > Fabio Cardoso wrote:E >> "Sun Microsystems seeks to avoid oblivion by pursuing a simple buth >> powerful strategy.  >> Its plan? Attack Red Hato >oF >  From the text at the URL you pointed at: "Linux is like every other? > operating system; it's about the foibles, greed, mistakes andcH > engineering prowess (or lack thereof) of one vendor--in this case, Red@ > Hat. Step No. 1: Make the argument that Linux equals Red Hat." >cC > This statement shows that Sun clearly doesn't understand the OpengG > Source movement. And so their strategy is doomed from the start. What E > about SuSE? Mandrake? TurboLinux, Slackware, etc.? Red Hat may havetF > achieved majority market share of Linux distros in recent years, andA > it has made major engineering contributions, but it doesn't own-E > Linux, and even if Red Hat as a company fails, Linux will continue.v >aA > And even Sun could take down Linux, what about OpenVMS, NetBSD,n
 > FreeBSD?    ? Sun doesn't have to 'take down' OpenVMS. carly(tm) will do that  single-handedly.   ------------------------------   Date: 22 Sep 2004 20:00:40 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)& Subject: Re: OT: Sun's fighting chance+ Message-ID: <2re3v8F173dpcU1@uni-berlin.de>a  , In article <mNCdnRua_bp7SszcRVn-uQ@igs.net>,& 	"John Smith" <a@nonymous.com> writes: > Keith Parris wrote:0 >> Fabio Cardoso wrote:lF >>> "Sun Microsystems seeks to avoid oblivion by pursuing a simple but >>> powerful strategy. >>> Its plan? Attack Red Hat >>G >>  From the text at the URL you pointed at: "Linux is like every othere@ >> operating system; it's about the foibles, greed, mistakes andI >> engineering prowess (or lack thereof) of one vendor--in this case, Red A >> Hat. Step No. 1: Make the argument that Linux equals Red Hat."s >>D >> This statement shows that Sun clearly doesn't understand the OpenH >> Source movement. And so their strategy is doomed from the start. WhatF >> about SuSE? Mandrake? TurboLinux, Slackware, etc.? Red Hat may haveG >> achieved majority market share of Linux distros in recent years, andoB >> it has made major engineering contributions, but it doesn't ownF >> Linux, and even if Red Hat as a company fails, Linux will continue. >>B >> And even Sun could take down Linux, what about OpenVMS, NetBSD, >> FreeBSD?r >  > A > Sun doesn't have to 'take down' OpenVMS. carly(tm) will do that  > single-handedly.  i: I resisted the temptation to actually put this into words.   bill     -- SJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   y   ------------------------------   Date: 22 Sep 2004 21:44:15 GMT+ From: "Doc." <doc.cypher@openvms-rocks.com>-& Subject: Re: OT: Sun's fighting chance6 Message-ID: <Xns956CF1C33FEAdcovmsrox@212.100.160.123>  B %NEWS-I-NEWMSG, Bill Gunshannon wrote in news:2re3v8F173dpcU1@uni-	 berlin.deo  . > In article <mNCdnRua_bp7SszcRVn-uQ@igs.net>,, >      "John Smith" <a@nonymous.com> writes: >> Keith Parris wrote:  C >>> And even Sun could take down Linux, what about OpenVMS, NetBSD,  >>> FreeBSD?  B >> Sun doesn't have to 'take down' OpenVMS. carly(tm) will do that >> single-handedly.e  < > I resisted the temptation to actually put this into words.  G Point is, most of the people in this newsgroup are of the opinion that s? appropriate marketing will make the difference between OpenVMS eI marketshare growing, and the OS slowly declining to a point where it can j be axed.  H Where *is* the damn marketing?  Are we still waiting on two dozen focus 6 groups coming back with the right target demographics?     Doc. --  G OpenVMS:     Eight out of ten hackers prefer *other* operating systems.iG http://www.openvms-rocks.com    Deathrow Public-Access OpenVMS Cluster.e   ------------------------------    Date: 22 Sep 2004 15:34:12 -0700( From: bob@instantwhip.com (Bob Ceculski)& Subject: Re: OT: Sun's fighting chance= Message-ID: <d7791aa1.0409221434.25e5c8c2@posting.google.com>n  t Andrew Harrison <andrew_remove__s._harrison@s_u_n.com> wrote in message news:<cis6vr$dj2$1@new-usenet.uk.sun.com>... > D > Sun has no intention of attacking NetBSD or FreeBSD, RedHat is theC > target not them and OpenVMS despite having the Open bit tacked one! > the front is hardly OpenSource.g > 	 > Regardsb > Andrew Harrison-  = it certainly is open source ... you can port any c unix/linuxm8 garbage code onto it including libraries and it runs ...: and you can get the source if you want to ... but vms does9 not need to be tinkered with by people who think they cane/ code understand os's but don't (i.e. linux) ...p   ------------------------------  % Date: Wed, 22 Sep 2004 22:23:59 -0700r From: Z <z@no.spam>o5 Subject: Re: Readable error log while process running 0 Message-ID: <10l4nfbespbe3df@corp.supernews.com>   Paul Jones wrote:'E > Is there a trick to making a program's error log readable while theeN > executable is running when using the vms run statement in DCL? I'm using the	 > format:  >  > $       run program_xyz.exe-- >               /error=program_xyz_error.log-  >               /output=_nla0:-C >               /input=_nla0:a > D > I know the VMS SYS$INPUT is equivalent to C's  stdin descriptor 0,N > SYS$OUTPUT is equivalent to stdout descriptor 1, and SYS$ERROR is equivalentN > stderr descriptor 2. At DCL level, I would like to be able to open the errorH > log with a process write-lock and allow a world read privilege without( > having to do this at the C code level.   Submit it to a batch queue.e  2 The LOG file can be read while the job is running.   ------------------------------  # Date: Thu, 23 Sep 2004 02:50:36 GMTa( From: Jonathan Cronin <jtc@theworld.com> Subject: Re: RMS and threads> Message-ID: <0001HW.BD77B14C002E6EC3F02845B0@news.verizon.net>   Consider the following models.   One master process.i Two slave processes.  Each slave process does 1000 IOs2 IOs take either 10ms or 20ms with 50% probability.  ( Time for an unco-ordinated slave to run:  0 	(.5 * 10ms) + (.5 * 20 ms) * 1000 = 15 seconds.  C Total time for two uncordinated slave processes is also 15 seconds.n   Now co-ordinated:t  D The time for a slave to complete an IO and start the next one is theG max of its IO time and the other slaves IO time.  Note that because the0J slaves are co-ordinated, the effective IO time for each slave is the same.   There are four possiblities:  ! Slave 1    10ms  10ms  20ms  20msJ! Slave 2    10ms  20ms  10ms  20ms "            -----------------------! Effective  10ms  20ms  20ms  20msu   so  1 (.25 * 10ms) + (.75 * 20ms) * 1000 = 17.5 seconds     B The co-ordinated system run slower even with the same IO times and distribution of IO times.s   ------------------------------  % Date: Wed, 22 Sep 2004 15:34:10 -0400t- From: JF Mezei <jfmezei.spamnot@teksavvy.com>u4 Subject: Re: search txt file for a specified string?, Message-ID: <4151D3AF.CC2009B7@teksavvy.com>   Bob Koehler wrote:I >    search, like all good DCL commands, sets $STATUS when it exits.  YouvG >    will find that there are different values for $STATUS which map toe7 >    file not found, string not found, or string found.   K And you can use /WINDOW=0 to essentially hide the results, or /OUTPUT=NLA0:vJ which does a similar thing, so that the procedure doesn't spew out lots of text as it performs searches.k   ------------------------------  % Date: Wed, 22 Sep 2004 16:11:49 -0400o- From: JF Mezei <jfmezei.spamnot@teksavvy.com>d. Subject: Re: Sniffer Can't see cluster traffic, Message-ID: <4151DC81.21E981D7@teksavvy.com>   James O'Shea wrote:   L > Our network guy, who has never monitored a VMS cluster previously, says he > is only see multicast packetsr  > 1- Please turn off HTML posting. comp.os.b,s is a text (ascii) newsgroup/mailing list.r  N 2- You have not mentioned how the machines are physically connected. Is it viaA thinwire ethernet, is it with CAT5 wiring to a hub, to a switch ?e    N There is a utility (probably on freeware) that runs on VMS called ETHERMON. ItN requires all migty privs to run on VMS (since it is promiscuous and will sniffK everyone's packets). It will give you all ethernet packets pass through theA wire connected to that VMS box.e  I If a switch is involved, remember that the switch (which is funtionally aqN bridge) builds a table of ethernet adresses connected to each of its ports andC then only passes packets that shoudl be received by each port. (and-( multicast/broadcasts) get sent to port).  J If you have an unmanaged switch, then you cannot use a packet sniffer. YouN need to have a managed switch where you can designate a port to be promiscuousN where your packet sniffer computer will then receive all traffice that is sent to the switch.   ------------------------------  % Date: Wed, 22 Sep 2004 16:27:10 -0400w= From: "John Gemignani, Jr." <john@nfw-invalid.cibtrikker.com>n. Subject: Re: Sniffer Can't see cluster traffic1 Message-ID: <RJ-dnX2WqtiCfczcRVn-iQ@adelphia.com>o  ; "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message w& news:4151DC81.21E981D7@teksavvy.com... > James O'Shea wrote:- >-K >> Our network guy, who has never monitored a VMS cluster previously, says @ >> hec  >> is only see multicast packets >u@ > 1- Please turn off HTML posting. comp.os.b,s is a text (ascii) > newsgroup/mailing list.0  A     JF - please move your right hand one key to the left, thanks.      >nM > 2- You have not mentioned how the machines are physically connected. Is it v > viawC > thinwire ethernet, is it with CAT5 wiring to a hub, to a switch ?h >u >,D > There is a utility (probably on freeware) that runs on VMS called  > ETHERMON. ItK > requires all migty privs to run on VMS (since it is promiscuous and will   > sniffaJ > everyone's packets). It will give you all ethernet packets pass through  > theb! > wire connected to that VMS box.  >nK > If a switch is involved, remember that the switch (which is funtionally ahM > bridge) builds a table of ethernet adresses connected to each of its ports t > andiE > then only passes packets that shoudl be received by each port. (andt* > multicast/broadcasts) get sent to port). > L > If you have an unmanaged switch, then you cannot use a packet sniffer. YouE > need to have a managed switch where you can designate a port to be l
 > promiscuousuL > where your packet sniffer computer will then receive all traffice that is  > sent > to the switch.     ------------------------------  % Date: Wed, 22 Sep 2004 20:30:33 -0400n- From: "John E. Malmberg" <wb8tyw@qsl.network>s. Subject: Re: Sniffer Can't see cluster traffic1 Message-ID: <Q4CdnTxosNW0hM_cRVn-vw@adelphia.com>d   Richard Brodie wrote: < >>"James O'Shea" <seamas_ose@ameritech.net> wrote in message< >>news:20040922161014.93547.qmail@web81102.mail.yahoo.com... >>> >>Forget to mention, he is also seeing LAT and DECnet packets.I >>Do you still think he should throw another non->vms box on the network?  > M > Both these protocols also use both multicast and direct packets, so you maya! > just be seeing more multicasts.a > G > If you are running DECnet (and sometimes when not), you will be usingeG > the DECnet hardware address (AA-00-04-xx-yy-zz) rather than the burntuJ > in one; likely (08-00-2B-aa-bb-cc) from machines of that era. You should9 > be seeing those as the source addresses anyways though.T  H NIC cards can support multiple MAC addresses at the same time.  The SCS C communications uses a different MAC address than the DECNET or the e& default hardware address for the card.  F I remember seeing the information documented in one of the cluster or C I/O manuals as to how to translate them back to the cluster member.l  K The SCS protocol is non-routable, so it should show up on all switch ports.-  F By using some managed hubs you can keep the SCS packets restricted to  the VMS systems.   -John  wb8tyw@qsl.network Personal Opinion Only    ------------------------------  % Date: Wed, 22 Sep 2004 21:34:00 -0400 - From: "John E. Malmberg" <wb8tyw@qsl.network>h. Subject: Re: Sniffer Can't see cluster traffic1 Message-ID: <rJGdncKepvyXtc_cRVn-rw@adelphia.com>v   JF Mezei wrote:  > "John E. Malmberg" wrote:  > M >>The SCS protocol is non-routable, so it should show up on all switch ports.  >  > Can you explain this ?  G As I remember from looking at the protocol traces, they were sent to a iI protocol and cluster specific multicast MAC address, from a protocol and f! node specific source MAC address.   M > I was under the impression that switches acted solely at the ethernet level2O > and did not examine contents of ethernet packets to determine protocol. BasediP > on that, i don't see how an ethernet packet containing an SCS packet would bneM > treated any differently by a switch than an ethernet packet containing somey& > virus infested microsoft SMB packet.  G There are bits in the packets that the switches use to determine if it 2, is a multi-cast packet or a directed packet.  E Typically the characteristics of a non-routable protocol are that it  H communicates with broadcast destination addresses so it will not depend F on keeping a mapping table of MAC addresses to the remote hosts.  The G upper level of the protocol handler then filters out what packets that T it cares about.S  I It has been a while since I have looked at the ethernet packet structure nG in detail for the specifics.  Unless you are directly working on a low h- level protocol, that knowledge is not needed.e  H Most of it is structured to allow a network handler to quickly divert a ) packet to an application that can use it.C   -John  wb8tyw@qsl.network Personal Opinion Onlyt   ------------------------------  % Date: Thu, 23 Sep 2004 00:06:29 -0400c- From: JF Mezei <jfmezei.spamnot@teksavvy.com>-. Subject: Re: Sniffer Can't see cluster traffic, Message-ID: <41524BC4.515A0D0C@teksavvy.com>   "John E. Malmberg" wrote:oH > As I remember from looking at the protocol traces, they were sent to aJ > protocol and cluster specific multicast MAC address, from a protocol and# > node specific source MAC address.e  H When I signed up to a cable ISP years ago and found out it theyr serviceM didn't work, they, like any good civil servant, blamed my setup.  I had to do H protocol traces to show to them that I was sending DHCP requests and notN getting any response from the modem, although I was seeing ARP requests coming from the modem.u  M In the process, I got to see many SCS packets that were specifically adressed.I to each of the 2 nodes on my lan. When BIKE has an MSCP request to a disk = served by VELO, there is no reason to broadcast that request.   H > There are bits in the packets that the switches use to determine if it. > is a multi-cast packet or a directed packet.  M Yep and that is solely at the ethernet packet level. There is actually a goodeT explanation of these in the good old VMS do set (grey wall) for the ethernet driver.  F > Typically the characteristics of a non-routable protocol are that itI > communicates with broadcast destination addresses so it will not dependEC > on keeping a mapping table of MAC addresses to the remote hosts. f  N My understanding and experience is that both LAT and SCS do have tables of MACM adresses to whom they talk. For instance, a terminal server talking to node A N will be sending packets to NODE A's  ethernet address with a protocol field ofL LAT. The ethernet card on Node A, upon seeing the protocol field, then feeds that packet to the LAT driver.  L Similarly, when Node A wants to send a packet to a terminal, the packet willK be adresses specifically to the terminal server in charge of that terminal. M Where there are broadcasts is the regular service announcements. (in terms ofoL LAT). And there would also be broadcasts for SCS when a node needs to send a) packet to all other nodes in the network.o   > The H > upper level of the protocol handler then filters out what packets that > it cares about.s  N nop. The ethernet card filters out packets that are not supposed to be seen byN that ethernet address (packets adressed to specifically to another MAC addressL for instance). For those packets that this nodes is sopposed to see, it thenK looks at the protocol field and then delivers that packet to the approriate + driver (decnet, lat, scs, tcpip, whatever).&  M A standard (non vlan) switch only deals with MAC addresses. It has a table ofuK which MAC adresses live on which port of the switch. And only sends to that L port packets that should be seen by that any of the MAC adresses that resideG on that PORT. And this includes multicast and broadcasts since they are / designed to be seen by everyone on an ethernet.s  F VLANS complicate things because more sophisticated ones can filter outG protocols. However, they still do not route nor do they look inside theaM ehthernet packet contents. They only look at the ethernet header, and as such  as protocol independant.   ------------------------------  % Date: Wed, 22 Sep 2004 15:37:08 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>h! Subject: Re: VMS on IBM's Itaniuml, Message-ID: <4151D461.3705CC99@teksavvy.com>   Bob Ceculski wrote: G > Your recent e-mail query to Sam Palmisano regarding an article in thee@ > Inquirer.net has been forwarded to me for review and response.  C > most enterprise customers are now looking for more open operatingpE > environments. As noted in the article, Compaq, and now HP, have noteH > invested to keep up with the requirements of the customer base and the4 > result is an inevitable decline in installed base.  N Well, bob, I think you did a great job in contacting IBM, even if the responseN was brutal and true. Just make sure nobody at HP sees it because they will use/ it as further justification to deemphasize VMS.u   ------------------------------  % Date: Wed, 22 Sep 2004 15:56:18 -0400e# From: "John Smith" <a@nonymous.com>r! Subject: Re: VMS on IBM's Itanium , Message-ID: <b8-dnbg8kYp5RczcRVn-og@igs.net>   JF Mezei wrote:  > Bob Ceculski wrote: H >> Your recent e-mail query to Sam Palmisano regarding an article in theA >> Inquirer.net has been forwarded to me for review and response.i >sD >> most enterprise customers are now looking for more open operatingF >> environments. As noted in the article, Compaq, and now HP, have notE >> invested to keep up with the requirements of the customer base and 9 >> the result is an inevitable decline in installed base.  >cG > Well, bob, I think you did a great job in contacting IBM, even if thelC > response was brutal and true. Just make sure nobody at HP sees it G > because they will use it as further justification to deemphasize VMS.n     JF,a  I Shhhhh! Don't you know that carly(tm) and the HP Bored of Directing trollaL comp.os.vms looking for exactly these sorts of ideas?  How else could a sane person explain HP's policies?e   ;-)e   ------------------------------  % Date: Wed, 22 Sep 2004 15:32:46 -0400a- From: JF Mezei <jfmezei.spamnot@teksavvy.com>S> Subject: Re: Windoze not rebooted monthly shuts down airports!+ Message-ID: <4151D35C.2E462B5@teksavvy.com>i   Bob Koehler wrote:* >    Each radar sweep takes a few seconds.  M However, doesn't the "system" keep track of estimated position of an aircraft L based on its alttude, speed and heading calculated from previous sweeps ? It: could then be updating aircraft positions more frequently.   ------------------------------    Date: 22 Sep 2004 16:11:09 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)r> Subject: Re: Windoze not rebooted monthly shuts down airports!3 Message-ID: <S2tAerxcp0V2@eisner.encompasserve.org>.  [ In article <4151D35C.2E462B5@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes:l > Bob Koehler wrote:+ >>    Each radar sweep takes a few seconds.t > O > However, doesn't the "system" keep track of estimated position of an aircraftIN > based on its alttude, speed and heading calculated from previous sweeps ? It< > could then be updating aircraft positions more frequently.  D    Not to my knowledge.  There's no telling when a pilot is going to    twist a control.s   ------------------------------  + Date: Wed, 22 Sep 2004 17:41:16 -0500 (CDT)r From: sms@antinode.orgC Subject: Re: ZIP "-V" v. UNIX, et al.: Problem, possible solutions.i) Message-ID: <04092217411658@antinode.org>   ( From: David Froble <davef@tsoft-inc.com>   > [...]oP > Ok, you're implying that there ia already a provision for 'custom' data in an O > archive.  I'll go with that, which does make me a bit of a moron for talking n9 > about something without knowing what I'm talking about.e  F    May I suggest, then, that you give serious consideration to greaterC use of a particular keyboard entity?  On my LK411, it's a "Shift-/"  (ASCII 63, I believe).   >   However, the sources I  Q > have indicate that ZIP is implemented in C, which I refuse to even read, which DE > is a definite sign of intellegence.  On average that puts me ahead.a  1    Determined ignorance can be tough to overcome.   H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode-orgs    Saint Paul  MN  55105-2547a   ------------------------------    Date: 22 Sep 2004 17:00:45 -0700. From: spamsink2001@yahoo.com (Alan E. Feldman)E Subject: Re: [DCL REQUEST] New ignore keyword for DIFFERENCES command.= Message-ID: <b096a4ee.0409221600.359370c7@posting.google.com>r  v koehler@eisner.nospam.encompasserve.org (Bob Koehler) wrote in message news:<C9SItGJVewFQ@eisner.encompasserve.org>...g > In article <Xns956C9B42E6078dcovmsrox@212.100.160.123>, "Doc." <doc.cypher@openvms-rocks.com> writes:n( > > %NEWS-I-NEWMSG, Bob Koehler wrote in/ > > news:WFAGstJ320P6@eisner.encompasserve.org i > > B > >> In article <b096a4ee.0409211611.455028c7@posting.google.com>,6 > >> spamsink2001@yahoo.com (Alan E. Feldman) writes: 4 > >>> I'd like to see /IGNORE=LEADING_DOLLAR_SIGNS.  > >> o1 > >>    diff /ignore=comments/comment_delimiter=$c > > J > > This will cause all the DCL code he's presumably trying to diff to be  > > ignored. > I >    I think instead of /IGNORE=LEADING_DOLLAR_SIGNS, he'd want somethingt- >    like /comment_delimiter="!", but he saide" >    /IGNORE=LEADING_DOLLAR_SIGNS.  & Yes, and I said it was to be used with/ /IGNORE=(COMMENTS,TRAILING_SPACES,BLANK_LINES).o  E Even so, ignoring leading dollar signs IS NOT the same as the $ beingnD a comment character. For example, ignoring form feeds and spacing is= not the same as /COMMENT=(either of those). Let's not confuset  /COMMENT_DELIMITER with /IGNORE.  D I want all four at once, but I can only use the three that have been implemented by VMS Engineers.   $ Explicitly, I request the following:  C     $ DIFFERENCES file1,file2 /IGNORE=(LEAD,COMM,TRAI,BLAN)/COMM=EXm  8 which would be great for DIFFing DCL command procedures.   [EOP]y   ------------------------------   End of INFO-VAX 2004.528 ************************mance-all things that Sun is now delivering in spades with its new offerings, Schwartz said.   C "You came knocking and we didn't listen," he said. "You said Sun is K proprietary and expensive, and we declined year over year for a few years."   I But, "What we've been doing for the past few years is getting back to our G roots so we can deliver real systems innovation to you," Schwartz said.   B Now, Sun offers Solaris on x86 for 249 different systems, he said.  D Meanwhile, the company said ools/distn/gets.mar (1906 bytes) started.5: >>> 226 Transfer completed.  1108 (8) bytes transferred. <<< PORT 83,31,156,101,135,1335 >>> 200 Port 135.13 at Host 83.31.156.101 accepted.  <<< RETR grep.fd_ >>> 150 IMAGE retrieve of /disk$misc/decus/vax82b/lbltools/distn/grep.f (5252 bytes) started.t: >>> 226 Transfer completed.  4789 (8) bytes transferred. <<< PORT 83,31,156,101,135,14H5 >>> 200 Port 135.14 at Host 83.31.156.101 accepted.> <<< RETR gtftok.mar c >>> 150 IMAGE retrieve of /disk$misc/decus/vax82b/lbltools/distn/gtftok.mar (1260 bytes) started. 9 >>> 226 Transfer completed.  744 (8) bytes transferred.1 <<< PORT 83,31,156,101,135,15 5 >>> 200 Port 135.15 at Host 83.31.156.101 accepted.  <<< RETR hsh.f_ >>> 150 IMAGE retrieve of /disk$misc/decus/vax82b/lbltools/distn/hsh.f (77970 bytes) started.d; >>> 226 Transfer completed.  73859 (8) bytes transferred.1 <<< PORT 83,31,156,101,135,16 5 >>> 200 Port 135.16 at Host 83.31.156.101 accepted.  <<< RETR impath.mardc >>> 150 IMAGE retrieve of /disk$misc/decus/vax82b/lbltools/distn/impath.mar (1338 bytes) started. 9 >>> 226 Transfer completed.  667 (8) bytes transferred.1 <<< PORT 83,31,156,101,135,17 5 >>> 200 Port 135.17 at Host 83.31.156.101 accepted.M <<< RETR incl.fs_ >>> 150 IMAGE retrieve of /disk$misc/decus/vax82b/lbltools/distn/incl.f (2146 bytes) started.7: >>> 226 Transfer completed.  1848 (8) bytes transferred. <<< PORT 83,31,156,101,135,18 5 >>> 200 Port 135.18 at Host 83.31.156.101 accepted.  <<< RETR index.marb >>> 150 IMAGE retrieve of /disk$misc/decus/vax82b/lbltools/distn/index.mar (1398 bytes) started.9 >>> 226 Transfer completed.  605 (8) bytes transferred.1 <<< PORT 83,31,156,101,135,19H5 >>> 200 Port 135.19 at Host 83.31.156.101 accepted.  <<< RETR intro.f` >>> 150 IMAGE retrieve of /disk$misc/decus/vax82b/lbltools/distn/intro.f (6894 bytes) started.: >>> 226 Transfer completed.  6183 (8) bytes transferred. <<< PORT 83,31,156,101,135,20H5 >>> 200 Port 135.20 at Host 83.31.156.101 accepted.> <<< RETR isam.fe_ >>> 150 IMAGE retrieve of /disk$misc/decus/vax82b/lbltools/distn/isam.f (3964 bytes) started.m: >>> 226 Transfer completed.  3027 (8) bytes transferred. <<< PORT 83,31,156,101,135,21 5 >>> 200 Port 135.21 at Host 83.31.156.101 accepted. <<< RETR kill.fe^ >>> 150 IMAGE retrieve of /disk$misc/decus/vax82b/lbltools/distn/kill.f (578 bytes) started.9 >>> 226 Transfer completed.  421 (8) bytes transferred.6 <<< PORT 83,31,156,101,135,2225 >>> 200 Port 135.22 at Host 83.31.156.101 accepted.  <<< RETR kwic.fr_ >>> 150 IMAGE retrieve of /disk$misc/decus/vax82b/lbltools/distn/kwic.f (2226 bytes) started.c: >>> 226 Transfer completed.  1769 (8) bytes transferred. <<< PORT 83,31,156,101,135,2325 >>> 200 Port 135.23 at Host 83.31.156.101 accepted.  <<< RETR lam.f^ >>> 150 IMAGE retrieve of /disk$misc/decus/vax82b/lbltools/distn/lam.f (3482 bytes) started.: >>> 226 Transfer completed.  2522 (8) bytes transferred. <<< 