1 INFO-VAX	Sat, 20 Mar 2004	Volume 2004 : Issue 156       Contents:@ Re: AMD OpteronT Processor Powers New Family of Sun Fire Servers+ Re: ANN: bison 1.35 and flex 2.5.4a for GNV  Re: Athlon 64 vs Opteron: diff? ' Re: Check out these new VT flat panels! & Re: Cluster hang -- Getting Crash Dump/ Re: Contract VMS programming position available  Re: Hang Time: 15 minutes  Re: Intel releases 64bit Xeon  Re: Intel releases 64bit Xeon  Re: Intel releases 64bit Xeon  Re: ODS-5 after five years Re: ODS-5 after five yearsH ODS-5 after five years (was Re: ANN: bison 1.35 and flex 2.5.4a for GNV)G Re: OpenVMS virus proof while other disks get zapped by latest viruses! G Re: OpenVMS virus proof while other disks get zapped by latest viruses! D Re: Professional Agitator - Harrison, Andrew. Confimed on sun.com!!! Quorum Disk Question Re: Quorum Disk Question Re: Quorum Disk Question Re: Quorum Disk Question Re: scanning the PCI bus
 SIMH problems  Re: SIMH problems  Re: SIMH problems 5 Re: Sun nearing bankruptcy ... better buy VMS Andrew! 5 Re: Sun nearing bankruptcy ... better buy VMS Andrew! 5 Re: Sun nearing bankruptcy ... better buy VMS Andrew!  Re: Sun On The Run?  Re: Sun On The Run?  Re: Sun On The Run? 3 Re: VMS 5.5-2 disk size limitations on VAX 4000-200 3 Re: VMS 5.5-2 disk size limitations on VAX 4000-200 3 Re: VMS 5.5-2 disk size limitations on VAX 4000-200 - Re: What causes backups to produce split IOs? - RE: [Simh] Newbie has network compile problem - Re: [Simh] Newbie has network compile problem   F ----------------------------------------------------------------------  % Date: Fri, 19 Mar 2004 18:58:50 +0000 9 From: Andrew Harrison <andrew_._remove_harrison@su_n.com> I Subject: Re: AMD OpteronT Processor Powers New Family of Sun Fire Servers 0 Message-ID: <c3ffta$369$2@new-usenet.uk.sun.com>   Fred Kleinsorge wrote:  0 > "John Smith" <a@nonymous.com> wrote in messageE > news:t7v4c.21572$1gU1.15168@news04.bloor.is.net.cable.rogers.com...  >  >>>  >>>You mean Sparc is dead? >> >>J >>Reading between the lines, probably yes. Not this generation or the next- >>ones due, but most probably the ones after.  >>L >>Yet Sun has not said it in so many words, and not gone out of their way toK >>tell their customers that they are killing a proven architecture in favor  >  > of > L >>one that has yet to prove itself (funny, do I hear an echo of the Big Bang	 >>here?).  >> >  > N > As Lou Reed says - stick a fork in it, it's done.  Of course, if I was a SunC > customer I'd want to know about the future of Sparc for planning.  >    UltraSPARC IV+ UltraSPARC V Niagara  Rock  6 All currently under development, some in first silicon   Regards  Andrew Harrison  >  >  >    ------------------------------    Date: 19 Mar 2004 15:23:39 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 4 Subject: Re: ANN: bison 1.35 and flex 2.5.4a for GNV3 Message-ID: <3zmOST6m$$3g@eisner.encompasserve.org>   O In article <405AFDDC.EC81F04D@hp.com>, Joshua Cope <Joshua.Cope@hp.com> writes: H >> Bernard, do you think that I'll convert ODS especialy for using FLEX? > K >   On the other hand, should we still be asking developers to add special  J > filename handling code to every UNIX port? ODS-5 has been available for  > something like 5 years now.   F    ODS-5 doesn't handle everything UNIX does.  There are many examplesF    of ODS-2 compatable fix-the-name code.  Nobody has to re-invent the@    wheel, but it would be nice if they picked up ODS-2 vs. ODS-5    sensative code.   ------------------------------  % Date: Fri, 19 Mar 2004 18:55:39 +0000 9 From: Andrew Harrison <andrew_._remove_harrison@su_n.com> ( Subject: Re: Athlon 64 vs Opteron: diff?0 Message-ID: <c3ffnc$369$1@new-usenet.uk.sun.com>   Didier Morandi wrote:   K > Could someone please share some light on the differences between the AMD  ' > Athlon-64 and the Opteron processors?  >   @ Athlon-64 is for single CPU systems, Opteron is for SMP systems.  > The Athlon-64 has a single HyperTransport interface which only3 allows it to be configured in uniprocessor systems.   > Athlon-64 currently has a 1 MB onchip L2 cache same as Opteron9 however AMD are about to introduce a lower cost Athlon-64  with 512K of onchip L2 cache.   8 Athlon-64 also has a number of variants aimed at Mobile,. standard desktop and high performance desktop.   Regards  Andrew Harrison    ------------------------------  % Date: Fri, 19 Mar 2004 20:57:42 +0000 9 From: Andrew Harrison <andrew_._remove_harrison@su_n.com> 0 Subject: Re: Check out these new VT flat panels!0 Message-ID: <c3fms6$5dv$1@new-usenet.uk.sun.com>   Bob Koehler wrote:  n > In article <c2j0ch$6tu$1@new-usenet.uk.sun.com>, Andrew Harrison <andrew_._remove_harrison@su_n.com> writes: > 8 >>I still have one bridge, second hand one carefull lady >>owner. >>. >>Please apply at london-bridge-auctions.co.uk >  > G >    Andrew sure can find some odd ways to be wrong.  The London Bridge D >    was sold a long time ago, and is alive and well at its new home >    in Arizona. >   3 Ohh dear just when you though it was safe to go out  you mess up again.  1 The old London Bridge is safe and well and in the 1 US, but hey what do you think they did when faced 3 with two roads one on either side of the Thames and  no bridge between them ?  1 Guess what they built a new one which I walk over  quite regularly.  1 I don't know how often you intend to imply that I 3 regularly get things wrong but I would suggest that 5 on the basis of our last 20-30 exchanges your chances 5 supporting that claim are very poor and getting worse  with every post.   Regards  Andrew Harrison    ------------------------------  + Date: Fri, 19 Mar 2004 23:11:09 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney) / Subject: Re: Cluster hang -- Getting Crash Dump ( Message-ID: <c3fumd$993$2@pcls4.std.com>  = koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:   F >> EXPECTED_VOTES should always be set to the total of all VOTES (plus( >> QDSKVOTES if present) in the cluster.  D >   With VOTES = 1 each and EXPECTED_VOTES = 1, either system can beE >   brought up alone and operate as a one node cluster.  Since quorum F >   will be 2 as soon as the other node joins, there is no possibility' >   of operating a fragemented cluster.   F There is a possibility of multiple clusters - boot both nodes with the  LAN between them broken somehow.  H You can get away with this if the cluster isn't much more than two nodesD on a network (but "closer").  But if you have shared devices (sharedJ SCSI bus or something, or shadowing) you WILL scrozzle disks, and quickly, too.  It's not pretty.  6 >   I do this at home, and yes, I would do it at work.  B >   EXPECTED_VOTES should be 2 if a critical application won't run >   without both nodes present.    ugh!     --   -Mike    ------------------------------  # Date: Sat, 20 Mar 2004 01:31:13 GMT " From: tutor_removespam_@cfl.rr.com8 Subject: Re: Contract VMS programming position available8 Message-ID: <7c7n50tj6n2u4un52n2p9t1k368493a9vl@4ax.com>   I'll go with your guess.  . But now that we are moving down memory lane...7 The wine company was clearly visiable from the highway.   That's why Vineland rang a bell.  E It makes more sense to me that a "smaller" company then Dofasco would  keep "older" equipment.   ? After all - a typical vax shop took (up to) 10 people to run it ? (including a manager, data entry, hardware support, and maybe 2 
 programmers).   E Vax programs rarely needed maintenance - even the y2k fiasco would be B a snap on vax programming - just because vax programmers tended toC think far into the future. Something about "if you fix it, then you * don't have to be on call at midnight" ....  C Also, I've heard, vax (now alpha etc), is not suspect to microsofts F frequent upgrades and virus attacks. In other words, the only fear for9 a cobol program is a cobol programmer. (insert laughter).   % Thanks for the trip down memory lane. ( me? University of Western Ontario. 1979.% Old? yeah. but it was a good life....       ( On Fri, 19 Mar 2004 14:38:48 GMT, wrote:   > 0 ><tutor_removespam_@cfl.rr.com> wrote in message3 >news:50bi509j2k3neuhah01de2gfp5n3bssvio@4ax.com... 2 >> Darn, I used to know the name of the company... >>  >> I used to work in Toronto....G >> And if memory served me better, I would know the name of the Winery. I >> They were indeed a vax/vms shop. Heck, back as far as 1979 - they were ( >> probably a DEC 10 (20?) shop as well. >>F >> For some reason, the name "Jordan Wines" pops up. But that's wrong.B >> All I remember for "sure" is that they are in or near Vineland,? >> Ontario (aptly named - the "wine belt" of the Niagara penn.)  >  > I >Jordan has been incorporation into a larger entity named Vincor. You got % >your geography more or less correct.  > I >My guess is that it is Dofasco or Stelco, both are steel companies. Both K >used to use VMS and if my guess is correct at least one of them still does  >to some extent. >    ------------------------------  # Date: Fri, 19 Mar 2004 22:54:31 GMT & From: Don Sykes <paladin@mydomain.com>" Subject: Re: Hang Time: 15 minutes> Message-ID: <GSK6c.40586$%A5.22611@newssvr25.news.prodigy.com>   Keith Parris wrote: l > Don Sykes <paladin@mydomain.com> wrote in message news:<78t4c.9471$Qx2.6534@newssvr27.news.prodigy.com>... > 1 >>During that time one person was able to log on,  >  > 5 > So it appears SYSUAF and RIGHTSLIST weren't locked.  >  > ! >>but then could do nothing else.  >  > C > Perhaps the Audit Server SUSPending processes due to insufficient + > system disk space to log security events?   F After listening to all of the wise folks at cov, I decided to look at G the sizes of other system files and found that the ACCOUNTNG.DAT files  A totaled 1.3M blocks and hadn't been recycled since 1997, so your  F observation makes sense to me. They have since been recycled, as have  the OPERATOR logs.F I think they may have taken the idea of "lights out" administration a  little too far.    > @ > I second the recommendation to use console management softwareG > (ConsoleWorks, Unicenter Console Management for OpenVMS, etc.) to log  > all console output.  > A > I also second the recommendation to run Availability Manager or F > DECamds. This allows you to examine the cluster while it's hung, andH > also detects and logs things like lock contention (user holding a lockE > for an excessive length of time) that can cause apparent hangs, and B > even allows you to clear many problems on-the-fly with immediate > fixes.  H Unlikely they will be implementing any new measures now, as the systems C are scheduled to be replaced soon by some shiney new Unix boxes! ;)      --     Have VMS, Will Travel  Wire paladin, San Francisco    (paladinATalphaseDOTcom)   ------------------------------  % Date: Fri, 19 Mar 2004 19:09:06 +0000 9 From: Andrew Harrison <andrew_._remove_harrison@su_n.com> & Subject: Re: Intel releases 64bit Xeon0 Message-ID: <c3fggj$3dm$1@new-usenet.uk.sun.com>   Rick Jones wrote:   < > Andrew Harrison <andrew_._remove_harrison@su_n.com> wrote: > < >>Perhaps you have now worked out how Sun calculated its I/OC >>throughput numbers for the F25K ? Rather more scientific than the 0 >>I/O bridge peak throughpu x 16 as in the Dome. >  > ; > Well if taking the sums of the peak throughput of the PCI 8 > busses and multiplying by .8 is more scientific, sure. >   1 Well you seem confused again, because that wasn't 0 how we calculated the sustained I/O bandwidth of the system.    > F >>Perhaps its also time to draw a line under what seems to be becoming? >>an increasing unhappy discussion as far as you are concerned.  >  > B > Why should I be unhappy?  I'm learing more about my competition. >   > Actually, you appear to be learning from your competition, its3 a subtle distinction but one that could be crucial.    Regards  Andrew HArrison    ------------------------------  # Date: Fri, 19 Mar 2004 20:28:02 GMT & From: Rick Jones <foo@bar.baz.invalid>& Subject: Re: Intel releases 64bit Xeon1 Message-ID: <mJI6c.1302$Nx4.882@news.cpqcorp.net>   : Andrew Harrison <andrew_._remove_harrison@su_n.com> wrote: > Rick Jones wrote:   C >> Well if taking the sums of the peak throughput of the PCI busses 2 >> and multiplying by .8 is more scientific, sure.  E > Well you seem confused again, because that wasn't how we calculated , > the sustained I/O bandwidth of the system.  C And how did Sun calcuate the sustained I/O bandwidth of the system?   ? In Message-ID: <c25257$gmo$1@new-usenet.uk.sun.com> you stated:   	 <excerpt> < A fully configured F25K has 54 x 66Mhz 64 bit PCI busses and 18 x 33 Mhz 64 bit PCI busses.  8 The peak bandwidth of a 33 Mhz 64bit PCI bus is 264 MB/s8 The peak bandwidth of a 66 Mhz 64bit PCI bus is 528 MB/s  9 At this point you should have realised that your maths is : duff but I since your contribution to this thread has been( dubious I will do the math to rub it in.   18 x 264 = 4752 MB/s 54 x 528 = 28512 MB/s    28512 + 4752 = 33264 MB/s 
 </excerpt>  C Those being "SI" units rather than "CS" units (powers of ten rather D than powers of two). And would be 256 MB/s and 512 MB/s respectively@ in CS units, which would be 4608 MB/s + 27648 MB/s or 32256 MB/s  9 http://www.sun.com/servers/highend/sunfire_e25k/specs.xml   F says "25.2 GB/s sustained."  32256 * 0.8 is, as it happens, 25.2 GB/s.E If that is indeed just a coincidence, I would be keen to hear the way  it was done.  
 rick jones --  . a wide gulf separates "what if" from "if only"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: Fri, 19 Mar 2004 21:00:38 GMT & From: Rick Jones <foo@bar.baz.invalid>& Subject: Re: Intel releases 64bit Xeon1 Message-ID: <WbJ6c.1305$tA4.521@news.cpqcorp.net>   ' Rick Jones <foo@bar.baz.invalid> wrote: E > Those being "SI" units rather than "CS" units (powers of ten rather F > than powers of two). And would be 256 MB/s and 512 MB/s respectivelyB > in CS units, which would be 4608 MB/s + 27648 MB/s or 32256 MB/s  B Actually, I got the CS units wrong, trusting memory rather than myD 16C.  It is still an interesting coincidence to have had the numbersD come-out that way.  I'm not sure when in the mists of time I got 256 and 512 MB/s stuck in my mind.  A I'm still curious to know just how Sun arrived at their 25.2 GB/s  figure though.  
 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: Fri, 19 Mar 2004 21:20:03 -0500 - From: "John E. Malmberg" <wb8tyw@qsl.network> # Subject: Re: ODS-5 after five years 1 Message-ID: <yI2dncwdJd_JN8bdRVn-vw@adelphia.com>    Craig A. Berry wrote:  >   J > 1.)  Perl was ported to VMS about 10 years ago, well before ODS-5 was a F > glimmer in the eye of its designers, so there's a lot of home-grown H > file parsing that needs a thorough rewrite, plus the addition of long F > name support with NAML, honoring some of the settings of the DECC$*  > feature logical names, etc.   H Apparently to build a component of Mozilla, a version of PERL is needed H that when run under BASH on an ODS-5, keep the filenames in UNIX format  including preserving case.  D I have a patch procedure that should build such a version, but that H patch hardcodes the behavior instead of making it selectable outside of  the script.   I > 3.)  VMS versions of ZIP either don't preserve case in filenames or do  G > so poorly.  The version in the latest GNV distribution goes half-way  G > toward preserving case but leaves in some case sensitive comparisons  J > that make the whole thing rather flaky.  For example, if your directory D > ends in ".DIR;1" rather than ".dir;1" you bomb out with the error F > "directory not version 1".  Also, the caret escape character is not H > removed when placing files in archives, so when you unpack on another G > platform you end up with "foo.bar^.baz" instead of "foo.bar.baz".  I  E > have a patch for these issues but don't know who to send it to.  I  J > haven't looked at the UNZIP sources yet, but it needs an option so that 7 > it doesn't always munge names into ODS-2 safe format.   F Post on comp.os.vms with a subject indicating about your patch, and a G maintainer may show up.  Hunter Goatley seems to be doing some work in  H that area, if not the patches themselves, assisting in the coordination  of them.   -John  wb8tyw@qsl.network Personal Opinion Only    ------------------------------  % Date: Fri, 19 Mar 2004 23:42:19 -0500 - From: "John E. Malmberg" <wb8tyw@qsl.network> # Subject: Re: ODS-5 after five years 1 Message-ID: <urmdnVj0HZYxVsbdRVn-gg@adelphia.com>   G As a followup, if a program needs to detect if it needs to operate in a E traditional VMS mode, but also operate in "UNIX" mode when run under  H BASH, there is a way to detect that it is running in the BASH under GNV.  D BASH supplied with GNV sets the environment variable "SHELL" to the  string "bash".  G It also sets the environment variable "BASH" to the UNIX format of the  
 path to BASH.    -John  wb8tyw@qsl.network Personal Opinion Only    ------------------------------  # Date: Fri, 19 Mar 2004 23:52:53 GMT 6 From: "Craig A. Berry" <craigberry@mac.com.spamfooler>Q Subject: ODS-5 after five years (was Re: ANN: bison 1.35 and flex 2.5.4a for GNV) @ Message-ID: <f95f9f628b5f6535ae9e2d0d9bf84d61@news.teranews.com>  3 In article <3zmOST6m$$3g@eisner.encompasserve.org>, =  koehler@eisner.nospam.encompasserve.org (Bob Koehler) wrote:   J > In article <405AFDDC.EC81F04D@hp.com>, Joshua Cope <Joshua.Cope@hp.com> 	 > writes: J > >> Bernard, do you think that I'll convert ODS especialy for using FLEX? > > M > >   On the other hand, should we still be asking developers to add special  L > > filename handling code to every UNIX port? ODS-5 has been available for  > > something like 5 years now.  > H >    ODS-5 doesn't handle everything UNIX does.  There are many examplesH >    of ODS-2 compatable fix-the-name code.  Nobody has to re-invent theB >    wheel, but it would be nice if they picked up ODS-2 vs. ODS-5 >    sensative code.  B I'm quite sympathetic to folks requiring GNV and/or ODS-5 for new D ports. That's what these things are there for, to ease porting, and G sometimes they ease deployment as well.  However, there are still many  G native utilities and older ports that either don't handle ODS-5 at all  F or don't handle it very well.  That makes it tough for folks who need E to keep things working that only work on ODS-2 but are interested in  F having the new ports too.  A short list of tools that need ODS-5 work  that I'm aware of:  H 1.)  Perl was ported to VMS about 10 years ago, well before ODS-5 was a D glimmer in the eye of its designers, so there's a lot of home-grown F file parsing that needs a thorough rewrite, plus the addition of long D name support with NAML, honoring some of the settings of the DECC$*  feature logical names, etc.   E 2.) MMS is buggy with mixed-case file names.  A user-defined rule of  H .FOO.BAR may or may not go missing when MMS encounters files with names H ending in .Foo and .Bar.  No, I don't have a small reproducer, but I've E fixed it every time by renaming all the files to have all upper case   names.  G 3.)  VMS versions of ZIP either don't preserve case in filenames or do  E so poorly.  The version in the latest GNV distribution goes half-way  E toward preserving case but leaves in some case sensitive comparisons  H that make the whole thing rather flaky.  For example, if your directory B ends in ".DIR;1" rather than ".dir;1" you bomb out with the error D "directory not version 1".  Also, the caret escape character is not F removed when placing files in archives, so when you unpack on another E platform you end up with "foo.bar^.baz" instead of "foo.bar.baz".  I  C have a patch for these issues but don't know who to send it to.  I  H haven't looked at the UNZIP sources yet, but it needs an option so that 5 it doesn't always munge names into ODS-2 safe format.   F 4.)   I don't remember all the details, but I've seen cases where the F librarian defaulted to a lower or mixed-case module name derived from H an ODS-5 filename, but the linker then couldn't find the module because > it was looking for an all upper case version.  Existing build E procedures can break simply by converting a disk from ODS-2 to ODS-5.    ------------------------------  % Date: Fri, 19 Mar 2004 22:19:08 +0000 9 From: Andrew Harrison <andrew_._remove_harrison@su_n.com> P Subject: Re: OpenVMS virus proof while other disks get zapped by latest viruses!0 Message-ID: <c3frkt$6qr$1@new-usenet.uk.sun.com>   Bill Gunshannon wrote:/ > In article <c2k1ec$285t$1@news.cybercity.dk>, % > 	"Dr. Dweeb" <dr@dweeb.com> writes:  > J >>No, there are some that run some version of the PocketPC OS (not windows >>last time I looked).   >  > ? > http://www.epinions.com/cmd-review-15CA-7B5915-394AB4AC-prod3 = >   "The PocketPC OS is basically WinCe with still more added * >     features, without most of the bugs." > G > Considering the source, I will leave others to decide on the validity C > of the last phrase.  The important thing is, it most definitel is 
 > Windows. >   4 Not only is it Windows based but it also has Viruses& and requires Firewalls/Virus scanners.  5 I wonder if Bob has anything like that on his systems ) assuming they are running PocketPC/WINCE.    Regards  Andrew Harrison  > I >>                     Excuse me for not knowing the name of this OS this N >>week, but it is the one MS is flogging for mobile devices, PDAs and at least$ >>for a while (still?) thin clients. >  > H > As stated above, it's called WindowsCE warmed over and with a new nameG > most likely intended to lull the users into a flase sense of security ) > that their "new" OS is actually better.  >  > 3 >>I have no clue if Bob uses these, but they exist.  >  > G > Bob's thin clients run Powerterm and IE.  If it walks like a duck and  > quacks like a duck.....  >  > bill >    ------------------------------  % Date: Fri, 19 Mar 2004 17:50:04 -0400 * From: JF Mezei <jfmezei.spamnot@istop.com>P Subject: Re: OpenVMS virus proof while other disks get zapped by latest viruses!) Message-ID: <405B6AFA.D0F2EA91@istop.com>   A > > http://www.epinions.com/cmd-review-15CA-7B5915-394AB4AC-prod3 ? > >   "The PocketPC OS is basically WinCe with still more added , > >     features, without most of the bugs."  $ Ok, that article is not worth much.   N One needs to look at the year it was written: 2000. The writer mentions he hasP tried all PDAs, but fails to mention PSIONs which were still alive at that time.  K In fact, when Microsoft developped WinCE in 1998, it was to counter PSIONs, M not PALMS. And so much so, that PSION released its PSION 5 (first with the 32 J bit EPOC OS) before it was actually ready because it wanted to be first to6 market since Microsoft had promised so much for WinCE.  I Turns out that the first couple of versions of WinCE were total duds, and I PSION's product, with all its flaws because it was released before it was L finished was still far superior. PALM at that time still had not "taken over the world" (but had begun to).  K Now, by 2000, Palm had become the "de facto standard" PDA in north america, M and Microsoft was quick to get Compaq to produce a new WINCE, this time aimed I at Palm instead of PSION (which by that time, had pulled out of the north F american market, retrenching to Europe). It was rebranded as PocketPC.  J Meanwhile, PSION had spun off its EPOC32 OS to Symbian and had gotten manyM mobile phone nmanufacturers to jump in, despite strong efforts from Microsoft K to lure them to WinCE. But WinCE just wasn't capable of handling phones (no N real time capability for instance) and was full of bug, and very bloated (moreI expensive hardware to run it). Microsoft never did manage to make serious I inroads into the mobile market, although it has managed to ruin cars like 3 Mercedes by bribing its way in luxury car cockpits.   N Meanwhile, the little PDA that could did conquer the world. And you know why ?H because they had rock solid synchronisation with computers (both mac andF windows). (Of course, it is easier when you don't have to convert word' processing, spreadsheet and databases).     K Palm is to PDAs what SUN is to servers. It rode the .com boom and everybody I got one because it was so trendy. But now, they aren't so healthy because L people don't buy anymore either because they doN't need one, or they already	 have one.    ------------------------------  % Date: Fri, 19 Mar 2004 20:26:37 +0000 9 From: Andrew Harrison <andrew_._remove_harrison@su_n.com> M Subject: Re: Professional Agitator - Harrison, Andrew. Confimed on sun.com!!! 0 Message-ID: <c3fl1u$4sv$1@new-usenet.uk.sun.com>   David Svensson wrote:   ] > JF Mezei <jfmezei.spamnot@istop.com> wrote in message news:<404CD136.84C8780A@istop.com>...  >  >>David Svensson wrote:  >>; >>>to react to it like you do. Those who still use VMS need D >>>encouragement, not a bully person setting them straight. I cannot@ >>>understand how you can take posts from Bob Ceculski seriosly. >>N >>Hey, don't blast Bob.  While his posts are overly enthousiastic, he seems to+ >>be defending VMS, not HP's bad decisions.  >  > F > I don't mind Bob, it is OK with me, and it is also funny that AndrewD > feels that he needs to answer :), but I don't find the discussions > that follow so interesting.     / So you are happy about the image Bob portrays ?   ! Is he the classic OpenVMS admin ?   , Lets hope not but some of his postings would+ make horribly usefull collateral for anyone - trying intent on showing that OpenVMS is just / a marginal OS for people with very very strange  ideas.  - I like Bob as well, his posts are normally so 1 full of BS that correcting them is a pleasure and 2 one that doesn't require me to break into a sweat.   Regards  Andrew Harrison    ------------------------------    Date: 19 Mar 2004 14:08:22 -08000 From: dave.baxter@bannerhealth.com (Dave Baxter) Subject: Quorum Disk Question = Message-ID: <a3c44af1.0403191408.2c60dfee@posting.google.com>   F Thanks to everyone who responded to my note, you all help a lot.   New@ Question.     The reason I am running a 2-node cluster without aC quorum disk is that I expected to have a third node in by now!    I = have now decided to go ahead and create a quorum disk anyway.   ,      I have carried out the following steps.  E 1.    enabled each node as a Quorum Watcher by running cluster_config 1 on each cluster node and defined the quorum disk. = 2.    I checked modparams.dat to make sure that the necessary ! information is included, (it is). = 3.    I initialized the new disk ($1$dga1014:) with the label  "QUORUM$DISK".D 4.    My cluster storage is SAN (hence the dga device) so both nodes can see the disk directly.  :    My question is this.    Will the system mount this diskE automatically on the next boot, or will I have to mount it explicitly  (say in SYPAGSWPFILES.com)?    thanks   ------------------------------  # Date: Fri, 19 Mar 2004 23:35:45 GMT 4 From: brad@.gateway.2wire.net (Bradford J. Hamilton)! Subject: Re: Quorum Disk Question - Message-ID: <ltL6c.44210$SR1.51985@attbi_s04>   p In article <a3c44af1.0403191408.2c60dfee@posting.google.com>, dave.baxter@bannerhealth.com (Dave Baxter) writes:G !Thanks to everyone who responded to my note, you all help a lot.   New A !Question.     The reason I am running a 2-node cluster without a D !quorum disk is that I expected to have a third node in by now!    I> !have now decided to go ahead and create a quorum disk anyway. ! - !     I have carried out the following steps.  ! F !1.    enabled each node as a Quorum Watcher by running cluster_config2 !on each cluster node and defined the quorum disk.> !2.    I checked modparams.dat to make sure that the necessary" !information is included, (it is).> !3.    I initialized the new disk ($1$dga1014:) with the label !"QUORUM$DISK". E !4.    My cluster storage is SAN (hence the dga device) so both nodes  !can see the disk directly.  ! ; !   My question is this.    Will the system mount this disk F !automatically on the next boot, or will I have to mount it explicitly !(say in SYPAGSWPFILES.com)?  I It must be mounted explicitly - the only cluster I currently have manager E access to mounts the disk in SYLOGICALS.  ISTR the mount can exist in > SYPAGSWPFILES, as well, but my recollection could be faulty...   !  !thanks   J __________________________________________________________________________A Bradford J. Hamilton                    "All opinions are my own" K bMradAhamiPltSon-at-coMmcAast.nPeSt     "Lose the MAPS, and replace '-at-'  0                                          with @"   ------------------------------  + Date: Fri, 19 Mar 2004 23:55:53 +0000 (UTC) % From: "Basil Fong" <reply@usenet.com> ! Subject: Re: Quorum Disk Question 0 Message-ID: <c3g1a8$h4d$1@sparta.btinternet.com>  J We have a 2 AXP (SAN) node cluster with the system disk also acting as the quorum disk too.  L This seems to work quite well, is anybody else doing this or are we the only ones ?      A "Bradford J. Hamilton" <brad@.gateway.2wire.net> wrote in message ' news:ltL6c.44210$SR1.51985@attbi_s04... ? > In article <a3c44af1.0403191408.2c60dfee@posting.google.com>, 2 dave.baxter@bannerhealth.com (Dave Baxter) writes:I > !Thanks to everyone who responded to my note, you all help a lot.   New C > !Question.     The reason I am running a 2-node cluster without a F > !quorum disk is that I expected to have a third node in by now!    I@ > !have now decided to go ahead and create a quorum disk anyway. > ! / > !     I have carried out the following steps.  > ! H > !1.    enabled each node as a Quorum Watcher by running cluster_config4 > !on each cluster node and defined the quorum disk.@ > !2.    I checked modparams.dat to make sure that the necessary$ > !information is included, (it is).@ > !3.    I initialized the new disk ($1$dga1014:) with the label > !"QUORUM$DISK". G > !4.    My cluster storage is SAN (hence the dga device) so both nodes  > !can see the disk directly.  > ! = > !   My question is this.    Will the system mount this disk H > !automatically on the next boot, or will I have to mount it explicitly > !(say in SYPAGSWPFILES.com)? > K > It must be mounted explicitly - the only cluster I currently have manager G > access to mounts the disk in SYLOGICALS.  ISTR the mount can exist in @ > SYPAGSWPFILES, as well, but my recollection could be faulty... >  > ! 	 > !thanks  > L > __________________________________________________________________________C > Bradford J. Hamilton                    "All opinions are my own" L > bMradAhamiPltSon-at-coMmcAast.nPeSt     "Lose the MAPS, and replace '-at-'2 >                                          with @"   ------------------------------  # Date: Sat, 20 Mar 2004 04:47:55 GMT % From: "Mike Naime" <mnaime@kc.rr.com> ! Subject: Re: Quorum Disk Question 9 Message-ID: <%1Q6c.24870$4B1.20602@twister.rdc-kc.rr.com>   . Basil Fong <reply@usenet.com> wrote in message* news:c3g1a8$h4d$1@sparta.btinternet.com...L > We have a 2 AXP (SAN) node cluster with the system disk also acting as the > quorum disk too. > I > This seems to work quite well, is anybody else doing this or are we the  only > ones ? >   G We have over 60 2-node VMS clusters using the system disk as the quorum I disk.  Management did not want to waste space/spindles for a quorum disk. A Now that we are using EVA's for new clusters, I can make an extra > pagefile/quorum disk as an extra LUN without using unnecessary spindles/space.   
 Mike Naime   ------------------------------  + Date: Fri, 19 Mar 2004 20:08:34 +0000 (UTC) 6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)! Subject: Re: scanning the PCI bus 1 Message-ID: <newscache$anauuh$3oq1$1@news.sil.at>   X In article <iLo5c.895$vT.493@news.cpqcorp.net>, hoffman@hp.nospam (Hoff Hoffman) writes:\ >In article <Pine.LNX.4.44.0403101555520.21657-100000@localhost.localdomain>, Chip Coldwell ' ><coldwell@physics.harvard.edu> writes:  > D >:Long-term, I want to do this with an ICBM and IO AUTOCONFIGURE, of	 >:course.  > H >  I tend to prefer to use file-based configurations now, and not ICBMs.  F Can you explain a little bit what ICBMs (and their disadvantages) are,6 and why you decided to tend to prefer another method ?   TIA    -- : Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialistf E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Sat, 20 Mar 2004 02:14:20 +0100e  From: "Dr. Dweeb" <dr@dweeb.com> Subject: SIMH problems, Message-ID: <c3g5td$dak$1@news.cybercity.dk>   Hi,e  * Is there a place where SIMH is discussed ?  I In my instance I am trying to set up a VAX using SIMH on a Win2K machine.   J I would like then to connect and X-terminal to the SIMH, if that is indeedG possible.  My machine is a 512MB 3.2Gz PIV, so I am assuming I have the ' power to actually do this successfully.   + Anyhow, any direction would be appreciated.   
 Dr. Dweeb.  : The logfile is below (I have cleaned out some whitespace).   Logging to file "vax.log"o do> load -r ka655.binm do> attach nvr ka655.nvr do> set cpu 64mu do> set lpt disableu do> set tq disable do> set rl disable do> set ts disable do> set rq0 ra92 do> set rq1 ra92 do> set rq2 ra92 do> set rq3 cdromu do> att rq0 vms073.dsk do> att rq1 vmsdata.dsk- do> att rq2 tmpdata.dskn do> att -r rq3 oraclerdbvax.isok
 do> dep bdr 0n do> set cpu simhalte  do> set xq mac=08-00-2B-AA-BB-CC do> attach xq eth0> Eth: opened \Device\NPF_{6185B870-7EBC-4C1D-95C3-6ABD274FE847}   do> boot cpu [cKA655-B V5.3, VMB 2.7   Performing normal system tests.   @ 40..39..38..37..36..35..34..33..32..31..30..29..28..27..26..25..@ 24..23..22..21..20..19..18..17..16..15..14..13..12..11..10..09.. 08..07..06..05..04..03..   Tests completed. Loading system software. (BOOT/R5:0 ETH0F       2..n   ?41 DEVASSIGN, ETH. HALT instruction, PC: 00000C1A (MOVL (R11),SP)	 sim> exitg Goodbye > Eth: closed \Device\NPF_{6185B870-7EBC-4C1D-95C3-6ABD274FE847} Log file closed@   ------------------------------  % Date: Sat, 20 Mar 2004 02:23:34 +0100a5 From: "Philip Lewis" <philipdotjdotlewis@writeme.com>N Subject: Re: SIMH problems, Message-ID: <c3g6en$e0l$1@news.cybercity.dk>  # See below (Sorry, I am a moron :-()s  I However, I still will need some help figuring out how to X-Term into thishI from the same Win2K machine, coz Networking is a black art from my chair.s  
 Dr. Dweeb.   Dr. Dweeb wrote: > Hi,  >o, > Is there a place where SIMH is discussed ? >uB > In my instance I am trying to set up a VAX using SIMH on a Win2K
 > machine. >IE > I would like then to connect and X-terminal to the SIMH, if that isAG > indeed possible.  My machine is a 512MB 3.2Gz PIV, so I am assuming Is2 > have the power to actually do this successfully. > - > Anyhow, any direction would be appreciated.e >9 > Dr. Dweeb. >9< > The logfile is below (I have cleaned out some whitespace). >r > Logging to file "vax.log". > do> load -r ka655.binV > do> attach nvr ka655.nvr > do> set cpu 64me > do> set lpt disable0 > do> set tq disable > do> set rl disable > do> set ts disable > do> set rq0 ra92 > do> set rq1 ra92 > do> set rq2 ra92 > do> set rq3 cdrom  > do> att rq0 vms073.dsk > do> att rq1 vmsdata.dskp > do> att rq2 tmpdata.dske! > do> att -r rq3 oraclerdbvax.isor > do> dep bdr 0    Whoops - autoboot enabled !!   > do> set cpu simhalt5" > do> set xq mac=08-00-2B-AA-BB-CC > do> attach xq eth0@ > Eth: opened \Device\NPF_{6185B870-7EBC-4C1D-95C3-6ABD274FE847} >e > do> boot cpu > [cKA655-B V5.3, VMB 2.7 >p! > Performing normal system tests.i >cB > 40..39..38..37..36..35..34..33..32..31..30..29..28..27..26..25..B > 24..23..22..21..20..19..18..17..16..15..14..13..12..11..10..09.. > 08..07..06..05..04..03.. >  > Tests completed. > Loading system software. > (BOOT/R5:0 ETH0  >i >M >   2..o >  > ?41 DEVASSIGN, ETH0 > HALT instruction, PC: 00000C1A (MOVL (R11),SP) > sim> exitt	 > Goodbye @ > Eth: closed \Device\NPF_{6185B870-7EBC-4C1D-95C3-6ABD274FE847} > Log file closedM   ------------------------------  % Date: Sat, 20 Mar 2004 05:06:30 +0100y2 From: martin@radiogaga.harz.de (Martin Vorlaender) Subject: Re: SIMH problems; Message-ID: <405bc346.524144494f47414741@radiogaga.harz.de>u   Dr. Dweeb (dr@dweeb.com) wrote:6L > I would like then to connect and X-terminal to the SIMH, if that is indeedI > possible.  My machine is a 512MB 3.2Gz PIV, so I am assuming I have the ) > power to actually do this successfully.-  @ The problem with using the pcap library is that it can't monitorA outgoing packets; so a TCP/IP connection to the SIMH VAX from thetF same machine it runs on can only be accomplished by using two ethernet cards.   cu,e   Martin -- nD One OS to rule them all       | Martin Vorlaender  |  OpenVMS rules!7 One OS to find them           | work: mv@pdv-systeme.deoJ One OS to bring them all      |   http://www.pdv-systeme.de/users/martinv/> And in the Darkness bind them.| home: martin@radiogaga.harz.de   ------------------------------  % Date: Fri, 19 Mar 2004 20:49:05 +0000,9 From: Andrew Harrison <andrew_._remove_harrison@su_n.com> > Subject: Re: Sun nearing bankruptcy ... better buy VMS Andrew!0 Message-ID: <c3fmc1$59l$1@new-usenet.uk.sun.com>   Fred Kleinsorge wrote:  7 > "Jack Peacock" <peacock@simconv.com> wrote in messageo >  > I >>This "revelation" comes from a race tout^W^W stock analyst.  In generaluH >>their credibility ranks somewhat below that of astrologer.  However, I >  > will > G >>consult the I Ching as soon as possible to see if it agrees with thist >>prognostication of doom. >  > L > Well...  While that might apply to commentary in the Inquirer - the rating< > of debt has a little more structure and reality around it. > G > As I've said in the past, Sun is in a lot of respects an echo of whatoJ > happened at DEC.  I have contended (while tipping a pint) that DEC mightL > have recovered and rebounded had it "stayed the course" - it had somethingN > like $4 billion in cash, almost no debt, and was losing money at a rate that? > it could have gone many years before it ran out of cash.  TheiG > "restructuring" cost more money than we were losing.  Sun hasn't beensN > profitable in 2-3 years, and it is showing the strain - it's stock price andH > debt ratings reflect it.  The question is if they can adapt to the new9 > environment quickly enough to turn the business around.p >  Umm not exactly.  < Digital leapt into the arms of Compaq at a time when most of; the rest of the computer business was doing rather well andhD where Digitals main markets were actually expanding not contracting.  : Sun's lack of profitability has coincided with a worldwide8 recession and a contraction in the main markets that Sun operates in.  ; And Sun's debt rating is interesting but not that relevant,T6 after all we have 5.2 billion dollars in cash or other6 liquid assets rather more than the 1.3 billion we have3 in debt. Nor does Sun have a strategy that involves 3 borrowing money, this debt has remained very staticm# while our cash reserves have grown.n  9 Ammusingly this thread has been started by Bob's trolling8; with a Register article, the same organ that recently wrotew. that HP are rethinking their Itanium strategy.  5 Ammusing because you work for the division that looksn2 increasingly like a roadkill because of the latest8 AMD and Intel developments. You also work for a division2 that is entirely dependant on the margins on toner3 and to a much lesser extent to a services business.e  5 Over the same 2-3 year period that Sun has lost moneya< the two hardware businesses in HP have lost rather more than; Sun a fact that has been masked by the change in the way HPe7 reports BU P&L. Someone has to pay for the ~4 billion a"8 year HP spends on R&D, it used to be the indevidual BU's its now the printing division.  K > They are moving quickly to try and adopt AMD as a quick fix to the costly N > Sparc strategy.  Can they move quickly enough?  One thing that helps them isL > that nobody seems to want to try and buy them out for their customer base.N > I'd guess it's a case of perhaps anti-trust, and perhaps the assumption thatH > they will impode on their own and the customers will leave anyway, and= > taking over Sun isn't a guarantee of keeping the customers.x >   C How ammusing since that would appear to be exactly HP's strategy as  well.   A You have after all spent rather more on Itanium and the migrationaD to it than Sun has on Sparc. And now that Itanium is looking decidly5 dodgy you have rolled out a range of AMD based boxes.o  1 Stones glass houses, when will you HP guys learn.W  8 Oh and thanks to Bob for getting you to post that tripe.   Regardst Andrew Harrisonr   ------------------------------    Date: 19 Mar 2004 15:22:03 -0600+ From: young_r@encompasserve.org (Rob Young)n> Subject: Re: Sun nearing bankruptcy ... better buy VMS Andrew!3 Message-ID: <w9crbvuH9Jkq@eisner.encompasserve.org>a  l In article <c3fmc1$59l$1@new-usenet.uk.sun.com>, Andrew Harrison <andrew_._remove_harrison@su_n.com> writes:   > < > Sun's lack of profitability has coincided with a worldwide: > recession and a contraction in the main markets that Sun > operates in. >   ; 	Parts of it - sure.  But news from the last 3, 4  quarters% 	can't be explained away:d    IBM steals server sales from Sun+ Last modified: August 28, 2003, 9:01 PM PDTs By Stephen Shankland 2 Staff Writer, CNET News.comc            >I IBM extended its lead in the server market in the second quarter of 2003,YK stealing a sizable slice of the worldwide market for the powerful computerse/ away from Sun Microsystems, a new study shows.    L Sales of IBM servers increased 10.1 percent to $3.2 billion, giving Big BlueN 30.4 percent of the $10.6 billion market, according to new figures released byJ market share researcher IDC. Third-ranked Sun Microsystems, meanwhile, saw+ revenue fall 18.7 percent to $1.4 billion. ,   	And  - http://zdnet.com.com/2100-1103_2-5165213.htmle  % IBM rises, Sun sinks in server marketg    By Stephen Shankland a
 CNET News.com  February 25, 2004, 4:19 PM PT   t  M IBM widened its lead in the worldwide server market in 2003 at the expense ofaH Sun Microsystems, making particular gains in the Unix server market, new figures show.   J Server revenue for IBM grew 10 percent to $14.8 billion from 2002 to 2003,M outpacing the overall market's 5 percent growth, market research firm Gartner N said Wednesday. IBM increased its share 1.7 percentage points to 32 percent of  the $46.1 billion total market.   O In contrast, No. 3 server seller Sun saw its revenue decline 15 percent to $5.4iN billion and its share drop 2.8 percentage points to 11.8 percent. No. 2 seller1 Hewlett-Packard grew 5 percent to $12.5 billion. o    2 	Sun is a company in decline while others expand.   ; 	What's that I hear running?  The washing machine.  I thinkh 	it is time for the spin cycle.-   				Rob    ------------------------------  % Date: Fri, 19 Mar 2004 17:08:57 -0400e* From: JF Mezei <jfmezei.spamnot@istop.com>> Subject: Re: Sun nearing bankruptcy ... better buy VMS Andrew!) Message-ID: <405B6159.95DEE868@istop.com>r   Andrew Harrison wrote:> > Digital leapt into the arms of Compaq at a time when most of= > the rest of the computer business was doing rather well andaF > where Digitals main markets were actually expanding not contracting.  
 Not quite.  D When Olsen was kicked out, Digital was without aim, fat, bloated andJ uncompetitive, partly for having tried to emulate IBM instead of competingM agianst Sun, Apollo and even then the nascent Windows in server market (early K 1990s). Note that IBM was also bleeding profusely and about to take radicalm actions to divest itself.e  K Well, Palmer did to Digital what Gerstner prevented IBM from doing. DigitalbM divested itself without fixing any of the problems, while Gerstner managed totE change a long established and too well rooted corporate mentality andt transformed IBM big time.e  F Meanwhile, a couple years into his massacre, Palmer started to talk toN Pfeiffer who gave Palmer the list of things that Compaq didn't want to buy, soL Palmer continued his divestiture until Digital was small enough to be bought
 by Compaq.  I (Did Digital even have a single profitable year during Palmer's tenure ?)W  L Palmer admitted that he had begun discussions with Compaq 3 years before theJ merger was announced. That would have been in the 1995/1996 timeframe, andN Palmer had already done lots of damage to Digital. In fairness, the Digital he inherited wasn't healthy.e  E Had Olsen hired Gerstner instead of Palmer, perhaps Compaq would haveoB purchased IBM and Digital would be the big computer company today.  M At the time of the .com boom, Digital was already out of the picture, without R a sales force, and with resellers not interested in selling non-marketed products.    < > Sun's lack of profitability has coincided with a worldwide: > recession and a contraction in the main markets that Sun > operates in.  K However, like other .com companies such as Nortel, Sun made the decision to4M expand like nuts to benefit from .com balloon, and it didn't contract back toS, its original size the day the boom was over.  K Anyone with a brain knew that companies like DrKoop.com had no hope of ever J being profitable as soon as they would run out of the Wall Street Casino's money, they'd close shop.s  M So yes, Sun is hurting from its above-normal glory days, but the big questioneM is where does Sun go from here. Do proprietary Unix system provide sufficienthC value added to justify the costs versus the apparently free Linux ?   K Just like the Windows craze where companmiues decided to migrate to windowsnM because that was the fad and it was initially seen as much cheaper, companies  are now migrating to Linux.-  M Will Linux prove to be as expensive to own/operate as Solaris or AIX or HP-UX>N ?  If its costs remain significantly lower, then one has the right to ask whatM sort of future lies for a one-trick pony like Sun.  HP lives on printers. IBMi lives on much more than AIX. r  N One also has to ask where do Linux customers come from ? Lets say Linux were aL company. Would it be seeking to steal customers from HP/IBM/Sun, or would it be targetting Windows users ?o  J If Linux steals mostlty Windows users, then HP/IBM/Sun can survive. But ifT Linux steals customers from the proprietary Unixes, then Sun is in a lot of trouble.   ------------------------------  % Date: Fri, 19 Mar 2004 19:15:01 +0000t9 From: Andrew Harrison <andrew_._remove_harrison@su_n.com>k Subject: Re: Sun On The Run?0 Message-ID: <c3fgrl$3dm$4@new-usenet.uk.sun.com>   David Svensson wrote:l   > Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<c2i1p2$pob$1@new-usenet.uk.sun.com>...- >  >>Rob Young wrote: >> >>>In article <c2hpjr$mtb$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes:n >>>d >>>a >>>>Daryl Jones wrote: >>>e >>>  >>>>>Second: >>>>>oI >>>>>What is being done to SUN by LINUX it is the same thing SUN has beeneJ >>>>>doing to HP, IBM, Compaq, and others and vise versa. Take over a DataI >>>>>Center by initially cutting its prices and then later boosting them.L' >>>>>Business as usual in the IT world.C >>>>>  >>>>  >>>>Provide examples if you can. >>>> >>>  >>>p> >>>	Really.  That claim is misleading.  I personally don't see; >>>	Linux getting a whole lot more expensive and Intel kit lA >>>	has been stable or declining for years in price.  I seriouslydD >>>	doubt HP boosts their Intel kit prices - Dell is always lurking! >>>i >> >>C >>>http://www.forbes.com/technology/2004/03/03/cz_dl_0303linux.htmlo >>>rQ >>>Same goes for Jorge Borbolla, chief information officer at AutoTradeCenter, aniP >>>online auto wholesaler in Menlo Park, Calif.--Sun let him go without a fight.N >>>Borbolla just replaced all of his Sun computers with HP Intel-based serversP >>>running Linux. "It was the money," Borbolla says. "The Linux systems were 40%" >>>the cost of the Sun machines."  >>>a >>@ >>SPARC boxes all include Solaris right to use licenses with the
 >>systems. >>G >>HP x86 boxes don't include a commecrial Linux license and if you wantoG >>to run almost any kind of commercial app you have to buy a commerciale2 >>Linux release which ends up being mainly Redhat. >>B >>RedHat AS which is what most customers end up with costs between; >>$1499 and $2499 per year so on a cheap HP you add anothert
 >>$4.5-$7.5K.u >>? >>This means that low end x86 boxes actually end up costing ~2xo >>the cost of a low end Sun. >>< >>Of course if you only consider the hardware costs then the8 >>Sun's may be more expensive but then you have to run a: >>community version of Linux and you cannot do that if you! >>want supported commercial apps.h >>@ >>Of course if you are smart you add up the cost of the hardware >>................. and the OS.b >>	 >>regardsa >>Andrew Harrisono >  > D > Are you saying that the world is screwed and don't understand it'sF > best? I am seeing lots of sites that have been and are replacing SunA > Sparc servers with x86 Linux servers. Most of this replacing iss > because of costs.     ' Absolutely correct do you want a prize.i  / Replacing old Sun's with new PC's running Linux 1 may well save money, but then replacing old Sun's % with new Sun's would also save money.O  . Chosing new Server running Linux as opposed to3 new Sun's running Solaris is unlikely to save money 3 unless you use free Linux and you can't if you want5! supported apps or a supported OS.o   Regardsf Andrew HarrisonV   ------------------------------  % Date: Fri, 19 Mar 2004 19:11:24 +0000a9 From: Andrew Harrison <andrew_._remove_harrison@su_n.com>e Subject: Re: Sun On The Run?0 Message-ID: <c3fgks$3dm$2@new-usenet.uk.sun.com>   John Smith wrote:u  H > "Andrew Harrison" <andrew_._remove_harrison@su_n.com> wrote in message, > news:c2lbv7$32i$1@new-usenet.uk.sun.com... >  >>Bob Koehler wrote: >> >>C >>>In article <c2ivjq$6o1$2@new-usenet.uk.sun.com>, Andrew Harrison  > - > <andrew_._remove_harrison@su_n.com> writes:o >  >>>>Bob Koehler wrote: >>>>C >>>>Because if you get audited by them (which their contract allowsn@ >>>>them to do) you end up paying a premium for the unsubscribed >>>>systems. >>>>+ >>>>That and its software piracy of course.e >>>O >>>vB >>>   Piracy to use free software multiple times?  What Red Hat isH >>>   generally selling is a pre-configured copy of free software.  TheyG >>>   may also be selling is support contracts.  Generally one does notwI >>>   need multiple support contracts unless there is some possibility ofdD >>>   not demonstrating some problem on systems which are supported. >>>u >>4 >>You really do need to read the contract because it5 >>entirely refutes the theory you have just advanced.r >>" >>Fred has kindly provided the URL >> >  >  > As I read it:o > E > One Machine or part thereof, one License, one support contract. Two ! > machines, double it, and so on.n > 0 > ..from A. General Terms and Conditions sectionH > .....The term "Installed Systems" means the number of Systems on whichI > Customer installs or executes the Software. The term "System" means anyeD > hardware on which the Software is installed, which may be, withoutE > limitation, a server, a work station, a virtual machine, a blade, aw( > partition or an engine, as applicable. > F > 4. REPORTING AND AUDIT. If Customer wishes to increase the number ofG > Installed System, then Customer will purchase from Red Hat additionalP0 > Services for each additional Installed System. >  >   4 Quite so Bob's theory could well land him in trouble with RedHats lawyers.n   Regardsl Andrew Harrisone >    ------------------------------  % Date: Fri, 19 Mar 2004 19:12:19 +0000 9 From: Andrew Harrison <andrew_._remove_harrison@su_n.com>o Subject: Re: Sun On The Run?0 Message-ID: <c3fgmk$3dm$3@new-usenet.uk.sun.com>  * Alan Winston - SSRL Admin Cmptg Mgr wrote:  s > In article <eXc3zm5tKXNf@eisner.encompasserve.org>, koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:e > n >>In article <umn3c.504$hu1.107@news.cpqcorp.net>, "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> writes: >>F >>>RedHat doesn't actually sell you "Linux" so much as they sell you aI >>>subscription service which includes the OS distribution as part of the O >>>contract.  When you agree to the contract, you are agreeing to use what they5N >>>provide *only* on the number of systems for which you have paid them to use
 >>>it on.  >>D >>  Sounds liek RedHat has found a way to deploy IBM's old mainframe- >>  licensing scheme on a Linux distribution.- >>F >>  Now I have good reason not to bother looking at RedHat next time I >>  want a Linux distribution. >  > L > As I understand it, there is an unencumbered fork of Redhat called Fedora.% > No maintenance charges; no support.e >   ? And probably more important very little commercial ISV support.5   Regardsb Andrew Harrison"	 > -- Alan< >    ------------------------------  % Date: Fri, 19 Mar 2004 14:47:24 -0500o+ From: "DanO" <oz55555.nospam@bellsouth.net>.< Subject: Re: VMS 5.5-2 disk size limitations on VAX 4000-200: Message-ID: <h%H6c.24930$zP2.24293@bignews5.bellsouth.net>  / Frank Gribbin <fjg@ing.iac.es> wrote in message ! news:4059c8ae$1@news.ll.iac.es...oI >     We have Vaxes running VMS 5.5-2. We can't upgrade because we depend H > on third party software that was never upgraded beyond 5.5-2. I recallC > that one can't address SCSI disks that are bigger than 8Gb. We'veVJ > recently had a 4Gb disk fail. Fortunately we had disk shadowing and haveJ > a couple more disks on-line. However we don't have any spares to replace > the failed 4Gb disk. >sI >     Does anyone know if its possible to use larger disks but initialiseSH > so it looks like it's 8Gb or less? It seems like it's difficult to get7 > SCSI disks less than 18Gb. Suggestions for suppliers?n >s	 > Thanks,v > Frankt  8 If you are using the DEC/Compaq/HP StorageWorks systems,@ you can mix sizes of drives into the same mirror set and it willD pretend they are all the size of the smallest dive.  So you can plug> an 18GB drive into an existing mirror set with 1 remaining 4GB> drive, and the mirror will act like both drives are 4GB units., This wastes some space, but works just fine.  : Also StorageWorks will let you build a large container andA then carve smaller containers out of it... IIRC older controllerse> would let you build 4 LUNS out of a single container and newer' controllers would let you build 8 LUNS.w  < So for example you could take 4 18GB drives and build a 54GB: Raid5, then partition it into 7 units just under 8GB each.6 You could build a second 4x18GB raid5, then VMS shadow( units between the two and be good to go.      DanO0   ------------------------------  % Date: Fri, 19 Mar 2004 18:48:36 -0500h( From: David Froble <davef@tsoft-inc.com>< Subject: Re: VMS 5.5-2 disk size limitations on VAX 4000-200* Message-ID: <405B86D4.10704@tsoft-inc.com>   Barry Treahy, Jr. wrote:   > Nic Clews wrote: >  >> Frank Gribbin wrote:a >>   >>J >>>    We have Vaxes running VMS 5.5-2. We can't upgrade because we dependJ >>> on third party software that was never upgraded beyond 5.5-2. I recallE >>> that one can't address SCSI disks that are bigger than 8Gb. We'veaL >>> recently had a 4Gb disk fail. Fortunately we had disk shadowing and haveL >>> a couple more disks on-line. However we don't have any spares to replace >>> the failed 4Gb disk. >>>DJ >>>    Does anyone know if its possible to use larger disks but initialiseJ >>> so it looks like it's 8Gb or less? It seems like it's difficult to get9 >>> SCSI disks less than 18Gb. Suggestions for suppliers?u >>>    >> >> >> Three points. >>K >> 1. Have you actually tried the application on higher versions of VMS? V4 H >> VMS compiled code works perfectly well on 7.3 and all points between, >> this is VMS you know.   >>G > That is not entirely true, if there are any executive or kernel code f= > hooks or references to data structures, you are in trouble.s > Barrya >   O Yes, you're correct.  I've seen some.  I think I can count them on the fingers kO of one hand, maybe not.  Point is, it sure doesn't hurt to try a later version xP of VMS.  VMS people usually try very hard to be accomodating with such testing. 7   Temporary PAKs and such are available for the asking.s  O Unless the user knows that there is a hard issue, the smart move is to do some ' testing.   Dave   -- r4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com 170 Grimplin Roada Vanderbilt, PA  15486a   ------------------------------  % Date: Fri, 19 Mar 2004 20:20:59 -0700-+ From: "Barry Treahy, Jr." <Treahy@MMaz.com> < Subject: Re: VMS 5.5-2 disk size limitations on VAX 4000-200' Message-ID: <405BB89B.3030903@MMaz.com>e   David Froble wrote:i   > Barry Treahy, Jr. wrote: >! >> Nic Clews wrote:1 >> >>> Frank Gribbin wrote: >>>  i >>>nK >>>>    We have Vaxes running VMS 5.5-2. We can't upgrade because we depend<K >>>> on third party software that was never upgraded beyond 5.5-2. I recallcF >>>> that one can't address SCSI disks that are bigger than 8Gb. We'veI >>>> recently had a 4Gb disk fail. Fortunately we had disk shadowing and  	 >>>> havefF >>>> a couple more disks on-line. However we don't have any spares to  >>>> replace >>>> the failed 4Gb disk.> >>>>K >>>>    Does anyone know if its possible to use larger disks but initialisetK >>>> so it looks like it's 8Gb or less? It seems like it's difficult to getu: >>>> SCSI disks less than 18Gb. Suggestions for suppliers? >>>>   o >>>a >>>p >>>o >>> Three points.r >>> E >>> 1. Have you actually tried the application on higher versions of i >>> VMS? V4uI >>> VMS compiled code works perfectly well on 7.3 and all points between,t >>> this is VMS you know.  >>H >> That is not entirely true, if there are any executive or kernel code > >> hooks or references to data structures, you are in trouble. >> Barry >> >hI > Yes, you're correct.  I've seen some.  I think I can count them on the  I > fingers of one hand, maybe not.  Point is, it sure doesn't hurt to try  B > a later version of VMS.  VMS people usually try very hard to be I > accomodating with such testing.  Temporary PAKs and such are available h > for the asking.o >sI > Unless the user knows that there is a hard issue, the smart move is to   > do some testing.  H No argument here, presuming they either know their apps very well, have = a test system to try it on, or have a solid backup system to  C 'fall-back.'  In my case, I had a lot of personal code, as well as bA commercial software and utilities that would have failed without c upgrades or recoding...d   Regards,   Barry.   -- h  > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                        v   ------------------------------  % Date: Fri, 19 Mar 2004 14:13:48 -0500 + From: "DanO" <oz55555.nospam@bellsouth.net>l6 Subject: Re: What causes backups to produce split IOs?: Message-ID: <nxH6c.63850$p77.31006@bignews3.bellsouth.net>  + AM <aileenm79@hotmail.com> wrote in message57 news:4034601c.0403190944.47ba2910@posting.google.com...lE > Could anyone provide some insight on why backups produce split IOs?5 >5	 > Thanks!   ) It depends upon how you are using backup,p4 but the basic cause for split IO is probably becauseA you are doing backups file by file and your files are fragmented.      Dan Ol   ------------------------------  % Date: Fri, 19 Mar 2004 22:29:45 +0000 % From: Alan Fay <alan@fay.demon.co.uk>h6 Subject: RE: [Simh] Newbie has network compile problem4 Message-ID: <c3fs9f$3j0$1$8302bc10@news.demon.co.uk>  ; Bob Supnik's SIMH for VAX runs great. Here is my SIMH VAX:-    $ sh sysH VAX/VMS V5.5  on node MVAX  19-MAR-2004 21:30:00.81   Uptime  0 03:01:23M   Pid    Process Name    State  Pri      I/O       CPU       Page flts Ph.MemyP 00000021 SWAPPER         HIB     16        0   0 00:00:45.89         0      0   P 00000025 CONFIGURE       HIB      8        7   0 00:00:00.04        98    159   P 00000026 IPCACP          HIB     10        9   0 00:00:00.04        96    156   P 00000028 ERRFMT          HIB      8      206   0 00:00:00.29       102    102   P 00000029 OPCOM           HIB      8      110   0 00:00:00.24       249    125   P 0000002A AUDIT_SERVER    HIB     10       47   0 00:00:00.25      1359    423   P 0000002B JOB_CONTROL     HIB     10      120   0 00:00:00.20       125    209   P 0000002C TP_SERVER       HIB     10     1033   0 00:00:00.47       162    268   P 0000002D UCX$INET_ACP    HIB      9      205   0 00:00:00.34       331    372   P 0000002F ALAN            CUR      7    75430   0 00:32:55.47   4877696    258    $   ) I have 8 RA92 disks plus 4 CD's mounted:-   
 $ sh dev dP Device                  Device           Error    Volume         Free  Trans MntP  Name                   Status           Count     Label        Blocks Count CntP MVAX$DUA0:              Mounted alloc        0  VAXVMSV073     1430037     1   1P MVAX$DUA1:              Mounted              0  VAXVMSV055     1499511   162   1P MVAX$DUA2:              Mounted alloc        0  DUA2           1662207     1   1P MVAX$DUA3:              Mounted alloc        0  DUA3           2940585     1   1P MVAX$DUB4:              Mounted alloc        0  DUB4            430863     1   1P MVAX$DUB5:              Mounted alloc        0  DUB5           2940585     1   1P MVAX$DUB6:              Mounted alloc        0  I64081         1524915     1   1P MVAX$DUB7:              Mounted alloc        0  DUB7           2940585     1   1P MVAX$DUC8:              Mounted alloc        0  VAXBINDEC033    183264     1   1(                         wrtlck          P MVAX$DUC9:              Mounted alloc        0  FREEWARE50_2     56934     1   1(                         wrtlck          P MVAX$DUC10:             Mounted alloc        0  VAXVMS073        16137     1   1(                         wrtlck          P MVAX$DUC11:             Mounted alloc        0  VAXVMS062       134463     1   1(                         wrtlck           $    Networking also works:-g   $ ucx sh host mvax        n      LOCAL databasey    Host address    Host namet  . 192.168.1.57    mvax, MVAX $ 
 $ ucx sh servtL Service             Port  Proto    Process          Address            StateO BPCD               13782  TCP      BPCD             0.0.0.0             EnabledtO FTP                   21  TCP      UCX$FTPD         0.0.0.0             EnabledeO REXEC                512  TCP      UCX$REXECD       0.0.0.0             Enabled O RLOGIN               513  TCP      not defined      0.0.0.0             Enabled_O RSH                  514  TCP      UCX$RSHD         0.0.0.0             EnablednO TELNET                23  TCP      not defined      0.0.0.0             Enabled  $   2 eth1 is actually configured as (not the same ip):-  ! [root@enigma simh]# ifconfig eth1a9 eth1      Link encap:Ethernet  HWaddr 00:0C:6E:7A:03:0B  .I           inet addr:192.168.0.22  Bcast:192.168.0.255  Mask:255.255.255.02<           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1A           RX packets:610912 errors:0 dropped:0 overruns:0 frame:0hC           TX packets:421887 errors:0 dropped:0 overruns:0 carrier:0e'           collisions:0 txqueuelen:1000  F           RX bytes:586141351 (558.9 Mb)  TX bytes:548798118 (523.3 Mb))           Interrupt:10 Memory:feaf4000-0 r   [root@enigma simh]#   ! This is running on RH Linux 7.2:-    [root@enigma simh]# uname -aA Linux enigma 2.4.23 #13 Sat Feb 28 18:48:34 GMT 2004 i686 unknown  [root@enigma simh]#   * Now list the libraries SIMH VAX is using:-   [root@enigma simh]# ldd vaxe. 	libm.so.6 => /lib/i686/libm.so.6 (0x40033000); 	libpcap.so.0.6.2 => /usr/lib/libpcap.so.0.6.2 (0x40056000)e. 	libc.so.6 => /lib/i686/libc.so.6 (0x40071000)6 	/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000) [root@enigma simh]#   0 Notice /usr/lib/libpcap.so.0.6.2 for networking.   Compile with:-  . [root@enigma simh]# make USE_NETWORK=1 BIN/vax   Here is my vax.ini file:-0   ;o ; Load CPU microcode load -r ka655.bino ;e# ; Attach non-volatile RAM to a file  attach nvr nvram.bin ; % ; This virtual machine has 64M memory  set cpu 64m  ; ? ; Define disk drive types. RA92 is largest-supported VAX drive.o ;  set rqb enable set rqc enable ;t ; DUA0: DUA1: DUA2: DUA3:d set rq0 ra92 set rq1 ra92 set rq2 ra92 set rq3 ra92 attach rq0 dua0.dsk  attach rq1 dua1.dskN attach rq2 dua2.dskr attach rq3 dua3.dsks ; DUB4: DUB5: DUB6: DUB7:a
 set rqb0 ra92-
 set rqb1 ra92S
 set rqb2 ra92 
 set rqb3 ra92s attach rqb0 dub4.dsk attach rqb1 dub5.dsk attach rqb2 i64081.dsk attach rqb3 dub7.dsk ; DUC8: DUC9: DUC10: DUC11:@ set rqc0 cdrom set rqc1 cdrom set rqc2 cdrom set rqc3 cdrom attach -r rqc0 vaxvmsspl02.iso attach -r rqc1 freeware52.isou attach -r rqc2 vaxvms073.iso attach -r rqc3 vms062.isog ; tk50's ; MUA0: MUA1: MUA2: MUA3:  set tq0 tk70 set tq1 tk70 set tq2 tk70 set tq3 tk70 attach tq0 vms0621.tap attach tq1 vms0622.tap attach tq2 vms551.tapg ;hK ; Disable unused devices. It's also possible to disable individual devices,o9 ; using a construction like "set rq2 disable" if desired.( ;a set rl disable set ts disable ; - ; Attach Ethernet eth1 to a network interfacei set xq mac=08-00-2B-AA-BB-CC attach xq eth1 ;o ; Now start the emulator boot cpu   Here is my boot sequence:-   [root@enigma simh]# ./vax2   VAX simulator V2.10-4e NVR: buffering file in memory  RQC: unit is read only RQC: unit is read only RQC: unit is read only RQC: unit is read only Invalid argument Invalid argument Invalid argument Invalid argument     KA655-B V5.3, VMB 2.7  Performing normal system tests.8@ 40..39..38..37..36..35..34..33..32..31..30..29..28..27..26..25..@ 24..23..22..21..20..19..18..17..16..15..14..13..12..11..10..09.. 08..07..06..05..04..03.. Tests completed. >>>boot. (BOOT/R5:0 DUA1)         2... -DUA1.   1..0..    E    VAX/VMS Version V5.5     Major version id = 1 Minor version id = 0f  K $!  Copyright (c) 1991 Digital Equipment Corporation.  All rights reserved.C  > %STDRV-I-STARTUP, VMS startup begun at 19-MAR-2004 22:13:13.55H %SET-I-NEWAUDSRV, identification of new audit server process is 0000002A8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:13:42.36  %%%%%%%%%%%6 Operator _MVAX$OPA0: has been enabled, username SYSTEM  8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:13:42.36  %%%%%%%%%%%( Operator status for operator _MVAX$OPA0:K CENTRAL, PRINTER, TAPES, DISKS, DEVICES, CARDS, NETWORK, CLUSTER, SECURITY, O LICENSE, OPER1, OPER2, OPER3, OPER4, OPER5, OPER6, OPER7, OPER8, OPER9, OPER10,f OPER11, OPER12  8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:13:45.48  %%%%%%%%%%%4 Logfile has been initialized by operator _MVAX$OPA0:4 Logfile is MVAX::SYS$SYSROOT:[SYSMGR]OPERATOR.LOG;55  8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:13:45.49  %%%%%%%%%%%F Operator status for operator MVAX::SYS$SYSROOT:[SYSMGR]OPERATOR.LOG;55K CENTRAL, PRINTER, TAPES, DISKS, DEVICES, CARDS, NETWORK, CLUSTER, SECURITY, O LICENSE, OPER1, OPER2, OPER3, OPER4, OPER5, OPER6, OPER7, OPER8, OPER9, OPER10,o OPER11, OPER12  8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:13:47.71  %%%%%%%%%%%  Message from user SYSTEM on MVAXG Warning: DECdtm log file not found (SYS$JOURNAL:SYSTEM$MVAX.LM$JOURNAL)0"         %RMS-E-FNF, file not found!         TP server process waitingo    = %RUN-S-PROC_ID, identification of created process is 0000002Dq8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:51.73  %%%%%%%%%%%" Message from user INTERnet on MVAX INTERnet Loadeda  5 The Internet driver and ACP were successfully loaded.b1 %UCX-I-SETLOCAL, Setting domain and/or local host0, %UCX-I-SETPROTP, Setting protocol parameters( %UCX-I-STARTCOMM, Starting communication8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:54.93  %%%%%%%%%%%" Message from user INTERnet on MVAX INTERnet Started  # %UCX-I-DEFINTE, Defining interfaces.8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:56.12  %%%%%%%%%%%" Message from user INTERnet on MVAX, INTERnet ACP Created INTERnet interface: QE0  " %UCX-I-ENABSERV, Enabling services8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:56.22  %%%%%%%%%%%" Message from user INTERnet on MVAX! INTERnet ACP Activate FTP Server d  8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:56.24  %%%%%%%%%%%" Message from user INTERnet on MVAX# INTERnet ACP Activate REXEC Server    8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:56.26  %%%%%%%%%%%" Message from user INTERnet on MVAX4 INTERnet ACP Remote Terminal Services START - RLOGIN  8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:56.26  %%%%%%%%%%%" Message from user INTERnet on MVAX$ INTERnet ACP Activate RLOGIN Server   8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:56.28  %%%%%%%%%%%" Message from user INTERnet on MVAX! INTERnet ACP Activate RSH Server    8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:56.30  %%%%%%%%%%%" Message from user INTERnet on MVAX4 INTERnet ACP Remote Terminal Services START - TELNET  8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:56.31  %%%%%%%%%%%" Message from user INTERnet on MVAX$ INTERnet ACP Activate TELNET Server   ' %UCX-I-STARTNAME, Starting name_service 5 %UCX-I-LOADSERV, Loading UCX Server proxy informationa> %UCX-I-SERVLOADED, UCX Aux. Server loaded with 1 proxy records7 -UCX-I-SERVSKIP,  Skipped 0 communication proxy records-0 -UCX-I-SERVTOTAL,  Total of 1 proxy records read9 %DECW-W-NODEVICE, No graphics device found on this systeme> -DECW-I-NODECW, DECwindows graphics drivers will not be loadedJ %SET-I-INTSET, login interactive limit = 64, current interactive value = 0   19-MAR-2004 22:15:238   SYSTEM       job terminated at 19-MAR-2004 22:15:24.16     Accounting information:oL   Buffered I/O count:            1856         Peak working set size:     806L   Direct I/O count:              1141         Peak page file size:     11393L   Page faults:                  30643         Mounted volumes:             0M   Charged CPU time:           0 00:00:30.26   Elapsed time:     0 00:02:10.82i    = %RUN-S-PROC_ID, identification of created process is 0000002D 8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:51.73  %%%%%%%%%%%" Message from user INTERnet on MVAX INTERnet Loaded   5 The Internet driver and ACP were successfully loaded.a1 %UCX-I-SETLOCAL, Setting domain and/or local hoste, %UCX-I-SETPROTP, Setting protocol parameters( %UCX-I-STARTCOMM, Starting communication8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:54.93  %%%%%%%%%%%" Message from user INTERnet on MVAX INTERnet Started  # %UCX-I-DEFINTE, Defining interfacess8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:56.12  %%%%%%%%%%%" Message from user INTERnet on MVAX, INTERnet ACP Created INTERnet interface: QE0  " %UCX-I-ENABSERV, Enabling services8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:56.22  %%%%%%%%%%%" Message from user INTERnet on MVAX! INTERnet ACP Activate FTP Server e  8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:56.24  %%%%%%%%%%%" Message from user INTERnet on MVAX# INTERnet ACP Activate REXEC Server    8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:56.26  %%%%%%%%%%%" Message from user INTERnet on MVAX4 INTERnet ACP Remote Terminal Services START - RLOGIN  8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:56.26  %%%%%%%%%%%" Message from user INTERnet on MVAX$ INTERnet ACP Activate RLOGIN Server   8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:56.28  %%%%%%%%%%%" Message from user INTERnet on MVAX! INTERnet ACP Activate RSH Server    8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:56.30  %%%%%%%%%%%" Message from user INTERnet on MVAX4 INTERnet ACP Remote Terminal Services START - TELNET  8 %%%%%%%%%%%  OPCOM  19-MAR-2004 22:14:56.31  %%%%%%%%%%%" Message from user INTERnet on MVAX$ INTERnet ACP Activate TELNET Server   ' %UCX-I-STARTNAME, Starting name_service 5 %UCX-I-LOADSERV, Loading UCX Server proxy informationi> %UCX-I-SERVLOADED, UCX Aux. Server loaded with 1 proxy records7 -UCX-I-SERVSKIP,  Skipped 0 communication proxy recordsr0 -UCX-I-SERVTOTAL,  Total of 1 proxy records read9 %DECW-W-NODEVICE, No graphics device found on this systema> -DECW-I-NODECW, DECwindows graphics drivers will not be loadedJ %SET-I-INTSET, login interactive limit = 64, current interactive value = 0   19-MAR-2004 22:15:238   SYSTEM       job terminated at 19-MAR-2004 22:15:24.16     Accounting information: L   Buffered I/O count:            1856         Peak working set size:     806L   Direct I/O count:              1141         Peak page file size:     11393L   Page faults:                  30643         Mounted volumes:             0M   Charged CPU time:           0 00:00:30.26   Elapsed time:     0 00:02:10.82l   Username: ALAN
 Password: '         Welcome to VAX/VMS version V5.5s7     Last interactive login on Friday, 19-MAR-2004 18:30e=     Last non-interactive login on Saturday, 13-MAR-2004 18:21r $      Alan Fay       > Date: 2004-03-18 11:30:47 PSTh+ > From: Steve Smith (ssmith94132@yahoo.com)d4 > Subject: [Simh] Newbie has network compile problem( > Newsgroups: comp.os.vms, alt.sys.pdp11 > 5 > I didn't see anything on the newsgroups about this.u4 > I'm sure it's simple, but I'm not having any luck." > Using "make USE_NETWORK=1". TIA. >   > SIMH V3.1-0 updated 31-Dec-03  >  > Linux version 2.4.18-35 > (bhcompile@stripples.devel.redhat.com) (gcc versionr > 2.96( > 20000731 (Red Hat Linux 7.3 2.96-110)) >  > libpcap 0.8.1  > * > /tmp/cc3ffVSZ.o: In function `eth_open':6 > /tmp/cc3ffVSZ.o(.text+0x4fa): undefined reference to > `pcap_setnonblock', > /tmp/cc3ffVSZ.o: In function `eth_filter':6 > /tmp/cc3ffVSZ.o(.text+0xd8b): undefined reference to > `pcap_setnonblock'- > /tmp/cc3ffVSZ.o: In function `eth_devices': 6 > /tmp/cc3ffVSZ.o(.text+0xdd3): undefined reference to > `pcap_findalldevs'6 > /tmp/cc3ffVSZ.o(.text+0xf01): undefined reference to > `pcap_freealldevs'% > collect2: ld returned 1 exit status  > make: *** [BIN/pdp11] Error 1t >  >    ------------------------------  % Date: Sat, 20 Mar 2004 05:18:16 +0100o2 From: martin@radiogaga.harz.de (Martin Vorlaender)6 Subject: Re: [Simh] Newbie has network compile problem; Message-ID: <405bc608.524144494f47414741@radiogaga.harz.de>l  & Alan Fay (alan@fay.demon.co.uk) wrote:# > This is running on RH Linux 7.2:-  >  > [root@enigma simh]# uname -aC > Linux enigma 2.4.23 #13 Sat Feb 28 18:48:34 GMT 2004 i686 unknown  > [root@enigma simh]#  > , > Now list the libraries SIMH VAX is using:- >  > [root@enigma simh]# ldd vaxr0 > 	libm.so.6 => /lib/i686/libm.so.6 (0x40033000)= > 	libpcap.so.0.6.2 => /usr/lib/libpcap.so.0.6.2 (0x40056000)i0 > 	libc.so.6 => /lib/i686/libc.so.6 (0x40071000)8 > 	/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000) > [root@enigma simh]#  >u2 > Notice /usr/lib/libpcap.so.0.6.2 for networking.  E I just built SIMH without problems using libpcap-0.8.1 (under SuSE8):r   notemv:/opt/simh # uname -arC Linux notemv 2.4.19-4GB #1 Fri Feb 7 15:37:36 UTC 2003 i686 unknowni notemv:/opt/simh # ldd vax0         libm.so.6 => /lib/libm.so.6 (0x40026000)B         libpcap.so.0.8.1 => /usr/lib/libpcap.so.0.8.1 (0x40049000)0         libc.so.6 => /lib/libc.so.6 (0x4006b000)=         /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)  notemv:/opt/simh # i   cu,p   Martin  H P.S.: Before, I used xterm as a SIMH console. I now found that PuTTY for Linux works pretty well, too.  -- p;    OpenVMS @ 25      | Martin Vorlaender  |  OpenVMS rules!a.                      | work: mv@pdv-systeme.deA    Still exceeding   |   http://www.pdv-systeme.de/users/martinv/c5    expectations      | home: martin@radiogaga.harz.deE   ------------------------------   End of INFO-VAX 2004.156 ************************