1 INFO-VAX	Fri, 29 Apr 2005	Volume 2005 : Issue 238       Contents:= Re: 64 bit addressing in FORTRAN with (character) descriptors = Re: 64 bit addressing in FORTRAN with (character) descriptors ! Check Apple's X server versus VMS % Re: Check Apple's X server versus VMS % Re: Check Apple's X server versus VMS % Re: Check Apple's X server versus VMS % Re: Check Apple's X server versus VMS % Re: Check Apple's X server versus VMS % Re: Check Apple's X server versus VMS % Re: Check Apple's X server versus VMS % Re: Check Apple's X server versus VMS % Re: Check Apple's X server versus VMS  Re: DEC/MMJ connectors Re: DEC/MMJ connectors Re: DEC/MMJ connectors Re: DECsystem 5500 Re: DECsystem 5500! FTP oddity for case sensitiveness % Re: FTP oddity for case sensitiveness % Re: FTP oddity for case sensitiveness $ Re: Kill multiple processes by name?8 Re: Maybe HP should get out of the hardware business....8 Re: Maybe HP should get out of the hardware business....8 Re: Maybe HP should get out of the hardware business....8 Re: Maybe HP should get out of the hardware business....8 Re: Maybe HP should get out of the hardware business....8 Re: Maybe HP should get out of the hardware business....8 Re: Maybe HP should get out of the hardware business....3 Re: mysql #1030 - Got error 127 from storage engine 3 Re: mysql #1030 - Got error 127 from storage engine 3 Re: mysql #1030 - Got error 127 from storage engine 3 Re: mysql #1030 - Got error 127 from storage engine  Processor scheduling Replacing Shadowed Boot Drive ! Re: Replacing Shadowed Boot Drive ! Re: Replacing Shadowed Boot Drive ! Re: Replacing Shadowed Boot Drive $ Re: Sandia says alpha the best chip!  Sandia says alpha the best chip! Setting up an NTP server Re: Setting up an NTP server Re: Setting up an NTP server Re: Setting up an NTP server Re: Setting up an NTP server Re: Setting up an NTP server Re: Setting up an NTP server Re: Setting up an NTP server Re: Setting up an NTP server Re: Setting up an NTP server Re: Setting up an NTP server Re: Slow Filesystem I/O  Re: Slow Filesystem I/O  Syntax Highlighting  Re: Syntax Highlighting  Re: Syntax Highlighting  Re: Syntax Highlighting  Re: Syntax Highlighting  Re: TCPIP V5.5 on OpenVMS V8.2 Re: TK505 Re: VMS equivalent of % xset fp+ directory_with_fonts + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + RE: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS?   F ----------------------------------------------------------------------  % Date: Fri, 29 Apr 2005 12:34:40 +0200 , From: Albrecht Schlosser <ajs856@tiscali.de>F Subject: Re: 64 bit addressing in FORTRAN with (character) descriptors+ Message-ID: <bk2t4d.ac.ln@news.hus-soft.de>    Steve Lionel wrote: L > On Thu, 28 Apr 2005 18:45:26 +0200, Albrecht Schlosser <ajs856@tiscali.de> > wrote: >  > M >>BTW, I found another "problem" with subroutines and descriptor types. If a  P >>subroutine includes an interface definition for itself, then it doesn't apply 6 >>the 64-bit attribute to its own formal parameter(s). >  > O > No surprise - that's not legal Fortran.  Or rather, the SHOW you are defining @ > in the interface is not the same one as the routine you're in. > N > You cannot define an "interface to self".  Within the routine, you can applyP > attributes to the dummy arguments, but you can't create an interface block for > yourself.   M Thanks for clarification. I'm still interested in a command line switch (see  N post of John Reagan), but if this isn't possible, I know what I have to do :-(   Albrecht   ------------------------------  # Date: Fri, 29 Apr 2005 15:15:57 GMT & From: John Reagan <john.reagan@hp.com>F Subject: Re: 64 bit addressing in FORTRAN with (character) descriptors2 Message-ID: <Ncsce.4715$AN2.1833@news.cpqcorp.net>   Albrecht Schlosser wrote:   J > Thanks for clarification. I'm still interested in a command line switch G > (see post of John Reagan), but if this isn't possible, I know what I   > have to do :-(  H I've passed along the suggestion to the right folks here in HP.  I have E no idea on the size of the effort and how that might fit into future   plans and schedules.   --   John Reagan / HP Pascal/{A|I}MACRO for OpenVMS Project Leader  Hewlett-Packard Company    ------------------------------  % Date: Fri, 29 Apr 2005 02:40:42 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> * Subject: Check Apple's X server versus VMSB Message-ID: <1114756809.6d26458acf93077bc14338c13f5fd8c4@teranews>  & http://www.apple.com/ca/server/macosx/  H Look at all the stuff that comes with Apple's server version of Max-OS X (the new one coming out today).    $599 canadian for 10 users.  $1200 for unlimited licence.    ? Even if those prices were in USD, it would still put VMS in the  "extremely expensive" category.   N And of course, Apple has a desktop version of the OS that is cheap to upgrade.    H MACOS comes with the Postfix SMTP server which is far more sophisticated, than the TCPIP Services vaniall SMTP server.  H While VMS engineers are stuck working on a port to an unwanted platform,C Apple manages to put in real new functionality to its OS. Note that H Apple also struggled for many years with a transition fron its old MacOSH to a new true operating systems with multitasking etc. But now, they are flying.   E Once the VMS engineers have completed the port to the 8086, hopefully E they will be able to start flying too. The earlier they are done with ! the port to the 8086, the better.    ------------------------------  # Date: Fri, 29 Apr 2005 11:36:50 GMT ! From: Nigel Barker <nigel@hp.com> . Subject: Re: Check Apple's X server versus VMS8 Message-ID: <gs647150mfntpdho5pts267rjh2bideaeo@4ax.com>  K On Fri, 29 Apr 2005 02:40:42 -0400, JF Mezei <jfmezei.spamnot@teksavvy.com>  wrote:  ' >http://www.apple.com/ca/server/macosx/  > I >Look at all the stuff that comes with Apple's server version of Max-OS X   >(the new one coming out today). >  >$599 canadian for 10 users. >$1200 for unlimited licence.  >  > @ >Even if those prices were in USD, it would still put VMS in the  >"extremely expensive" category.  O No. That's about the same price for an unlimited user license for OpenVMS on an  Integrity server.    -- Nigel Barker Live from the sunny Cote d'Azur    ------------------------------  % Date: Fri, 29 Apr 2005 08:27:45 -0400 # From: "John Smith" <a@nonymous.com> . Subject: Re: Check Apple's X server versus VMS, Message-ID: <G8udnfmygLTete_fRVn-sA@igs.net>   Nigel Barker wrote: . > On Fri, 29 Apr 2005 02:40:42 -0400, JF Mezei' > <jfmezei.spamnot@teksavvy.com> wrote:  > ) >> http://www.apple.com/ca/server/macosx/  >>B >> Look at all the stuff that comes with Apple's server version of+ >> Max-OS X (the new one coming out today).  >> >> $599 canadian for 10 users. >> $1200 for unlimited licence.  >> >>B >> Even if those prices were in USD, it would still put VMS in the" >> "extremely expensive" category. > C > No. That's about the same price for an unlimited user license for ! > OpenVMS on an Integrity server.     G Not having looked at the pricing in a while, where is an unlimited user H license priced at for an Alphaserver? It should be the same price as for
 Integrity.     --F OpenVMS - The never advertised operating system with the dwindling ISV base.    ------------------------------  % Date: Fri, 29 Apr 2005 07:09:18 -0600 . From: "Michael D. Ober" <mdo.@.wakeassoc..com>. Subject: Re: Check Apple's X server versus VMS. Message-ID: <3mqce.11$Fj4.122@news.uswest.net>  K Apple's OS X is based on Linux.  Based on my experiences with a Linux based L storage server, I'll stick with VMS and Windows Servers.  At least these two, OSs don't eat themselves on a regular basis.  
 Mike Ober.  : "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message< news:1114756809.6d26458acf93077bc14338c13f5fd8c4@teranews...( > http://www.apple.com/ca/server/macosx/ > J > Look at all the stuff that comes with Apple's server version of Max-OS X! > (the new one coming out today).  >  > $599 canadian for 10 users.  > $1200 for unlimited licence. >  > A > Even if those prices were in USD, it would still put VMS in the ! > "extremely expensive" category.  > G > And of course, Apple has a desktop version of the OS that is cheap to  upgrade. >  > J > MACOS comes with the Postfix SMTP server which is far more sophisticated. > than the TCPIP Services vaniall SMTP server. > J > While VMS engineers are stuck working on a port to an unwanted platform,E > Apple manages to put in real new functionality to its OS. Note that J > Apple also struggled for many years with a transition fron its old MacOSJ > to a new true operating systems with multitasking etc. But now, they are	 > flying.  > G > Once the VMS engineers have completed the port to the 8086, hopefully G > they will be able to start flying too. The earlier they are done with # > the port to the 8086, the better.  >    ------------------------------   Date: 29 Apr 2005 13:32:11 GMT( From: bill@cs.uofs.edu (Bill Gunshannon). Subject: Re: Check Apple's X server versus VMS, Message-ID: <3derarF6tnralU1@individual.net>  . In article <3mqce.11$Fj4.122@news.uswest.net>,1 	"Michael D. Ober" <mdo.@.wakeassoc..com> writes: " > Apple's OS X is based on Linux.   A OSX is not based on Linux.  It is the Mach Kernel and the FreeBSD 	 Userland.     bill   --  J 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>       ------------------------------  # Date: Fri, 29 Apr 2005 13:29:22 GMT ! From: Nigel Barker <nigel@hp.com> . Subject: Re: Check Apple's X server versus VMS8 Message-ID: <hkd471lmqpqve0jpmlfcc2ucmc5kb37c33@4ax.com>  H On Fri, 29 Apr 2005 08:27:45 -0400, "John Smith" <a@nonymous.com> wrote:   >Nigel Barker wrote:/ >> On Fri, 29 Apr 2005 02:40:42 -0400, JF Mezei ( >> <jfmezei.spamnot@teksavvy.com> wrote: >>* >>> http://www.apple.com/ca/server/macosx/ >>> C >>> Look at all the stuff that comes with Apple's server version of , >>> Max-OS X (the new one coming out today). >>>  >>> $599 canadian for 10 users.   >>> $1200 for unlimited licence. >>>  >>> C >>> Even if those prices were in USD, it would still put VMS in the # >>> "extremely expensive" category.  >>D >> No. That's about the same price for an unlimited user license for" >> OpenVMS on an Integrity server. >  > H >Not having looked at the pricing in a while, where is an unlimited userI >license priced at for an Alphaserver? It should be the same price as for  >Integrity.    Why?   -- Nigel Barker Live from the sunny Cote d'Azur    ------------------------------  % Date: Fri, 29 Apr 2005 07:07:22 -0700 # From: "Tom Linden" <tom@kednos.com> . Subject: Re: Check Apple's X server versus VMS( Message-ID: <opspzyukqczgicya@hyrrokkin>  F On 29 Apr 2005 13:32:11 GMT, Bill Gunshannon <bill@cs.uofs.edu> wrote:  0 > In article <3mqce.11$Fj4.122@news.uswest.net>,3 > 	"Michael D. Ober" <mdo.@.wakeassoc..com> writes: " >> Apple's OS X is based on Linux. > C > OSX is not based on Linux.  It is the Mach Kernel and the FreeBSD  > Userland.  > bill >  Wan't OSF-1 also based on Mach?    ------------------------------  % Date: Fri, 29 Apr 2005 10:16:25 -0400 2 From: "Stanley F. Quayle" <squayle@insight.rr.com>. Subject: Re: Check Apple's X server versus VMS. Message-ID: <42720979.12460.FE00028@localhost>  B On Fri, 29 Apr 2005 08:27:45 -0400, "John Smith" <a@nonymous.com>  wrote:  F > Not having looked at the pricing in a while, where is an unlimited >D > user license priced at for an Alphaserver? It should be the same > > price as for Integrity.     : The licensing model for Integrity is completely different.  E With Integrity, the number of users doesn't matter.  You can have an  ? infinite number of users (as long as they don't expect to *do*  
 anything).  D The Integrity licensing scheme is based on the number of processors B and on the type of "operating environment" in use.  The operating C environment determines which things (clustering, volume shadowing,   etc.) are available.  F So, your question doesn't make any sense unless you specify what your  definition of "unlimited" is. 
 --Stan Quayle  Quayle Consulting Inc.  
 ----------- Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363 3 8572 North Spring Ct., Pickerington, OH  43147  USA 0 stan-at-stanq-dot-com       http://www.stanq.com   ------------------------------   Date: 29 Apr 2005 14:27:56 GMT( From: bill@cs.uofs.edu (Bill Gunshannon). Subject: Re: Check Apple's X server versus VMS, Message-ID: <3deujcF6r23vmU1@individual.net>  ( In article <opspzyukqczgicya@hyrrokkin>,& 	"Tom Linden" <tom@kednos.com> writes:H > On 29 Apr 2005 13:32:11 GMT, Bill Gunshannon <bill@cs.uofs.edu> wrote: > 1 >> In article <3mqce.11$Fj4.122@news.uswest.net>, 4 >> 	"Michael D. Ober" <mdo.@.wakeassoc..com> writes:# >>> Apple's OS X is based on Linux.  >>D >> OSX is not based on Linux.  It is the Mach Kernel and the FreeBSD >> Userland. >> bill  >>! > Wan't OSF-1 also based on Mach?  >   E Couldn't say, I never used it (although I think I may actually have a + CD in one of my boxes of old DEC Software.)    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  # Date: Fri, 29 Apr 2005 15:17:18 GMT & From: John Reagan <john.reagan@hp.com>. Subject: Re: Check Apple's X server versus VMS2 Message-ID: <2esce.4717$AN2.4206@news.cpqcorp.net>   Tom Linden wrote:    >>! > Wan't OSF-1 also based on Mach?  >    Yes.   --   John Reagan / HP Pascal/{A|I}MACRO for OpenVMS Project Leader  Hewlett-Packard Company    ------------------------------  % Date: Fri, 29 Apr 2005 01:44:55 -0400 ' From: Dave Froble <davef@tsoft-inc.com>  Subject: Re: DEC/MMJ connectors 0 Message-ID: <1173iei1dle4n1f@corp.supernews.com>   GreyCloud wrote:< > Does anyone know where I can get a 6-pin DEC/MMJ to female > db25 cable??0 > I managed to acquire an unused VT420 terminal.  	 Location?    --  4 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 Road  Vanderbilt, PA  15486    ------------------------------  % Date: Fri, 29 Apr 2005 10:29:51 -0600 " From: GreyCloud <cumulus@mist.com> Subject: Re: DEC/MMJ connectors ' Message-ID: <427260FF.A831363@mist.com>    Dave Froble wrote: >  > GreyCloud wrote:> > > Does anyone know where I can get a 6-pin DEC/MMJ to female > > db25 cable??2 > > I managed to acquire an unused VT420 terminal. >  > Location?  >   < Montana.  But it is the cable I need to connect the terminal to the vlc with.   ------------------------------  % Date: Fri, 29 Apr 2005 10:29:07 -0600 " From: GreyCloud <cumulus@mist.com> Subject: Re: DEC/MMJ connectors ( Message-ID: <427260D3.9D359E01@mist.com>   David J Dachtera wrote:  >  > David J Dachtera wrote:  > >  > > GreyCloud wrote: > > > @ > > > Does anyone know where I can get a 6-pin DEC/MMJ to female > > > db25 cable??4 > > > I managed to acquire an unused VT420 terminal. > > H > > There's actually an adapter for that. H-857x, I think. Many variantsH > > depending upon the pinout (DCE or DTE, DB25M or DB25F, etc.). H-8575L > > sticks in my mind for some reason as a common one for use with terminals6 > > that don't have MMJ jacks. Should be MMJ to DB25F. > > J > > See also "MOD-TAP" and Black Box whilst Googling, searching eBay, etc. > > ...  > I > The one I use here says H8571-A on it. H8575-A had lots of filtering in @ > it for stray RF and such (Class B computing devices, I think).  9 I should've said a male 6-pin DEC/MMJ plug connected to a 1 cable that is connected to female db25 connector. ; The vaxstation vlc has the alternate port as a female 6-pin 1 mmj that I'm trying to connect to the vt420 with.    ------------------------------   Date: 29 Apr 2005 10:59:04 GMT( From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: DECsystem 5500 , Message-ID: <3deiboF6s2f5dU1@individual.net>  E In article <Pine.LNX.4.62.0504282240180.27715@localhost.localdomain>, + 	"P. Thompson" <no-spam@new.rr.com> writes: 5 > On Wed, 26 Apr 2005 tomarsin2015@comcast.net wrote:  >  >>F >> the version. I seen some where that Linux will run on a 5500, but IB >> would like Ultrix. If you have the binaries you can ftp them to >> 68.35.167.136 >  > I > Nope, definitely no Linux for the 5500.  A severe lack of QBUS drivers. J > No NetBSD even, because until recently not enough technical information E > was available.  Perhaps some day NetBSD will support the 5400/5500   > machines.   @ Can't have anything to do with QBUS as both Linux and NetBSD run on QBUS Vaxen.   bill   --  J 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>       ------------------------------  # Date: Fri, 29 Apr 2005 13:06:13 GMT ( From: "P. Thompson" <no-spam@new.rr.com> Subject: Re: DECsystem 5500 D Message-ID: <Pine.LNX.4.62.0504290755370.1083@localhost.localdomain>  + On Fri, 29 Apr 2005, Bill Gunshannon wrote:  > B > Can't have anything to do with QBUS as both Linux and NetBSD run > on QBUS Vaxen. >   E Interesting, I am a little behind the times it seems.  NetBSD I knew  I supported the QBUS VAX machines but I certainly was not aware that linux  E did.  Despite this, it still seems the MIPSfair type machines aren't   supported by either OS.   K I wonder if the reason is that the original or remaining installed base is  
 too small.  I I guess the QBUS made them attractive to those who had QBUS needs, but I  G never understood the thinking behind the 5400 which has DSSI disks but  F without clustering makes no use of their cool features.  This left it ; to using the disks like hard to find, expensive SCSI disks.    --- 6 Lord, protect me from those to whom you speak directlyE perl -e '#this program inspired by LookWAYup;print("hello world\n");'    ------------------------------  % Date: Fri, 29 Apr 2005 06:41:22 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> * Subject: FTP oddity for case sensitivenessB Message-ID: <1114771229.ef74daa7606101ce488817d1a442a5e4@teranews>   VAX VMS , TCPIP 5.3    FTP> CD "/" E %CLI-W-IVERB, unrecognized command verb - check validity ans spelling    FTP> cd "/"  250 Chdir okay   FTP> set def "/" 250 Chdir okay   FTP> SET DEF "/" 250 Chdir okay      G Come on guys, how come you have case sensitive for the "cd" command and F not the others, especially when the error message issued when enteringE an uppercase CD is a CLI error message and CLI is supposed to be case 
 insensitive ?   B Looks like some programmer actually added code in there to disable@ uppercase CD. Shouldn't the CLI command parser automatically andD transparently handle case insensitive stuff ? Or did the programmers fake the %CLI error message ?    ------------------------------  # Date: Fri, 29 Apr 2005 11:39:32 GMT * From: "FredK" <fred.nospam@nospam.dec.com>. Subject: Re: FTP oddity for case sensitiveness2 Message-ID: <U1pce.4695$rJ2.4138@news.cpqcorp.net>  8 I would assume that the FTP developers, in an attempt to7 make FTP both UNIX and VMS friendly - first look at the 6 command to see if it is a UNIX command (case sensitive7 of course - since UNIX itself would spit on CD).  If it 2 fails, then it try's to parse it as a VMS command.   So frankly it works.    There is a "cd" command for UNIX  3 There is a "set def" command for VMS, and VMS could  care less what the case is.   ) There is no "CD" command for UNIX or VMS.     : "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message< news:1114771229.ef74daa7606101ce488817d1a442a5e4@teranews... > VAX VMS , TCPIP 5.3  > 
 > FTP> CD "/" G > %CLI-W-IVERB, unrecognized command verb - check validity ans spelling  > 
 > FTP> cd "/"  > 250 Chdir okay >  > FTP> set def "/" > 250 Chdir okay >  > FTP> SET DEF "/" > 250 Chdir okay >  >  > I > Come on guys, how come you have case sensitive for the "cd" command and H > not the others, especially when the error message issued when enteringG > an uppercase CD is a CLI error message and CLI is supposed to be case  > insensitive ?  > D > Looks like some programmer actually added code in there to disableB > uppercase CD. Shouldn't the CLI command parser automatically andF > transparently handle case insensitive stuff ? Or did the programmers > fake the %CLI error message ?    ------------------------------  % Date: Fri, 29 Apr 2005 11:38:00 -0400 $ From: "Chris" <an.other@not_here.ca>. Subject: Re: FTP oddity for case sensitiveness: Message-ID: <txsce.13870$gA5.779911@news20.bellglobal.com>  % "CD" is a 'standard' FTP syntax error 9 "SET DEF" (or "set def") is translated as a VMS extension   I (yeah, they COULD fix it, or you could use TCPware, where this has worked  for several versions already)    CM  : "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message< news:1114771229.ef74daa7606101ce488817d1a442a5e4@teranews... > VAX VMS , TCPIP 5.3  > 
 > FTP> CD "/" G > %CLI-W-IVERB, unrecognized command verb - check validity ans spelling  > 
 > FTP> cd "/"  > 250 Chdir okay >  > FTP> set def "/" > 250 Chdir okay >  > FTP> SET DEF "/" > 250 Chdir okay >  >  > I > Come on guys, how come you have case sensitive for the "cd" command and H > not the others, especially when the error message issued when enteringG > an uppercase CD is a CLI error message and CLI is supposed to be case  > insensitive ?  > D > Looks like some programmer actually added code in there to disableB > uppercase CD. Shouldn't the CLI command parser automatically andF > transparently handle case insensitive stuff ? Or did the programmers > fake the %CLI error message ?    ------------------------------  # Date: Fri, 29 Apr 2005 15:19:09 GMT ( From: "Hein" <hein.nomail@hp.nomail.com>- Subject: Re: Kill multiple processes by name? 1 Message-ID: <Nfsce.4718$vS2.705@news.cpqcorp.net>   2 "DeanW" <dean.woodward@gmail.com> wrote in message0 news:3f119ada05042819031317a40@mail.gmail.com...E I've just had someone banging away, trying to break in via SSH. While E I shut it off at the firewall (after 980 failures- just a couple past F VMS setting intruder status, heh...), there are still a couple hundredD dropped SSH processes that I wish I could clear out. Is there a toolD that can kill a group of processes based on a wild-carded name, such as TCPIP$SS_*?  K perl -e "foreach (`show system`){($pid,$nam)=split; `stop/id=$pid` if ($nam  =~ /TCPIP\$SS_/)}"   hth, Hein.    ------------------------------    Date: 29 Apr 2005 01:33:24 -0700 From: icerq4a@spray.seA Subject: Re: Maybe HP should get out of the hardware business.... B Message-ID: <1114763604.522101.83830@f14g2000cwb.googlegroups.com>  * "Even the floating point performance beats	 Itanium."    Where have you seen that ?   ------------------------------    Date: 29 Apr 2005 02:01:58 -0700 From: icerq4a@spray.seA Subject: Re: Maybe HP should get out of the hardware business.... C Message-ID: <1114763397.934362.217010@o13g2000cwo.googlegroups.com>   B When the common platform between Xeon and Itanium is introduced inF 2007, system builders only have to build one system to support x86 and IA64.    ------------------------------    Date: 29 Apr 2005 02:18:39 -0700 From: icerq4a@spray.seA Subject: Re: Maybe HP should get out of the hardware business.... B Message-ID: <1114766319.206998.91870@z14g2000cwz.googlegroups.com>   " (also important A there, at least compared with the MP bus-strangulation which Xeon B experienced:  ever wonder why we never saw anything smaller than a* 4-processor Itanic submission in TPC-C?),"  G Well, atleast I have and it is still on the TPC-C list, with Madison 3M  1.3Ghz CPUs.   ------------------------------  % Date: Fri, 29 Apr 2005 06:36:35 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> A Subject: Re: Maybe HP should get out of the hardware business.... B Message-ID: <1114770942.75a495a80ff32a3c3891dc689b36751e@teranews>   icerq4a@spray.se wrote:  > D > When the common platform between Xeon and Itanium is introduced inH > 2007, system builders only have to build one system to support x86 and > IA64.   = I don't see why the IA64 supporters see this as a good thing.   H When the 8086 gains compatible system interfaces, it means that the 8086= will be able to scale to the same system types/sizes as IA64.   G Right now, IA64's claim to fame is that it can scale to systems such as C Superdome , whicn the 8086 can't really do yet. (at least Intel's).   E But once the 8086 gets the smae interfaces as IA64, what will prevent + the 8086 from being used on large systems ?   ? To me, the common interfaces means even more the death of IA64.    ------------------------------    Date: 29 Apr 2005 06:47:05 -0700* From: "Alan Greig" <greigaln@netscape.net>A Subject: Re: Maybe HP should get out of the hardware business.... C Message-ID: <1114782424.976562.180040@g14g2000cwa.googlegroups.com>    icer...@spray.se wrote: , > "Even the floating point performance beats > Itanium."  >  > Where have you seen that ?  G In the results for SPECfp_rate2000 for which Sun claim 138 at 2.2Ghz. I F was comparing it with the 1300Mhz Itanium 2. Admittedly if you compareD against the 1500Mhz HP system below it is approximately a tie. If weD were to use a 2.4Ghz dual core Opteron it would again take the lead.E Sandia/Cray have very interesting things to say about the Itanium 2's E actual "bursty" fp performance which helped them to rate it bottom in F their choice of supercomputer design. See my other posts in the thread on this subject.  ?  HP Integrity rx8620 (1500 MHz Itanium 2)  	8 cores, 8 chips, 1  core/chip  	126  	142   G In every other benchmark the 8 core Opteron completely trounces Itanium  2.  - "Opteron - the nuclear bomb designers choice"    --  
 Alan Greig   ------------------------------  % Date: Fri, 29 Apr 2005 12:16:39 -0400 ( From: Bill Todd <billtodd@metrocast.net>A Subject: Re: Maybe HP should get out of the hardware business.... = Message-ID: <p42dnRDSxt11wO_fRVn-vg@metrocastcablevision.com>    icerq4a@spray.se wrote:  > " (also important C > there, at least compared with the MP bus-strangulation which Xeon D > experienced:  ever wonder why we never saw anything smaller than a, > 4-processor Itanic submission in TPC-C?)," > I > Well, atleast I have and it is still on the TPC-C list, with Madison 3M  > 1.3Ghz CPUs.  D Why, so it is.  The list being so large, I usually don't notice new E submissions unless they fall *somewhere* near the top of their class.   I And this submission does not - by a lot.  I guess HP was more interested  I in providing a low-cost system than a high-performance one - though even  G there, HP's own dual-Xeon system submitted only a month later beats it  : on both $/tpmC *and* tpmC, so it's not exactly persuasive.   - bill   ------------------------------    Date: 29 Apr 2005 09:57:58 -0700 From: bob@instantwhip.com A Subject: Re: Maybe HP should get out of the hardware business.... C Message-ID: <1114793878.247629.231040@l41g2000cwc.googlegroups.com>   6 not when you run OpenVMS on it ... then the tpc winner is itanium/vms ...   ------------------------------  % Date: Fri, 29 Apr 2005 08:51:30 +0100 0 From: Chris Sharman <chris.sharman@sorry.nospam>< Subject: Re: mysql #1030 - Got error 127 from storage engine+ Message-ID: <4271E782.7090702@sorry.nospam>    Jean-Franois Pironne wrote:  > Chris Sharman wrote:  >> Jean-Franois Pironne wrote:9 >>>> Alpha VMS 7.3-1, Apache 2.0, php 4.3.2, mysql 4.1.9,  >>>> php_msql client 3.23.49 >>>>C >>>> Using the issue tracker 4.0.4 package, it broke with no error   >>>> indication.I >>>> Using phpmyadmin indicates #1030 - Got error 127 from storage engine  >>>> on the issues table.  >>> ' >>> Which storage engine have you used?  >>> D >>> Only Innodb work correctly, MyISAM is know to have some problem.I >>> Anyway ony Innodb support transaction, so I strongly suggest to only  I >>> use this storage engine, this is the default on OpenVMS and probably   >>> in future MySQL version. >>F >> Innodb is the default, right ? I didn't change it. [vms]my.cnf has F >> lines relating to innodb, and the rest are commented out. I didn't  >> make any changes. > F > Right, Innodb is the default, but you can specify when you create a < > table which storage engine you want to use for this table.
 > You can use + > mysql> show table status like 'my_table';  > ( > to check which storage engine is used.  F You've guessed it JF - every table in every db has engine MyISAM, not A innodb. That could be my problem, I guess. But I haven't created  C anything myself - I just ran the setup stuff I was told to run for  I mysql, and then for issuetracker - is there some way to check/change the  I default ? Some way to change (unload/drop/create/reload) existing tables  B & dbs ? Would it give the applications any problems, or would the  changes be transparent ?   Thanks,  Chris   I >> I wondered about skip-networking & safe-updates, but didn't turn them  D >> on, since I didn't really understand the implications. Should I ? >>H >> But the repair article I saw said it was only for MyISAM - which may 7 >> be why it didn't repair innodb that well, I suppose.  >>
 >> Thanks, >> Chris >>* >>>> I'm guessing it's a mysql corruption. >>>> Can anyone confirm ?  >>>> How can I fix it ? 4 >>>> Is there any known issue which could cause it ? >>>> >>>> Cross-posted to VAMP  >>>> >>>> Thanks,
 >>>> Chris >>>> >>>  >>> Jean-Franois    ------------------------------  % Date: Fri, 29 Apr 2005 08:54:03 +0100 0 From: Chris Sharman <chris.sharman@sorry.nospam>< Subject: Re: mysql #1030 - Got error 127 from storage engine4 Message-ID: <d4sp6r$n4f$2$8302bc10@news.demon.co.uk>   Jean-Franois Pironne wrote:  > Chris Sharman wrote:  >> Jean-Franois Pironne wrote:9 >>>> Alpha VMS 7.3-1, Apache 2.0, php 4.3.2, mysql 4.1.9,  >>>> php_msql client 3.23.49 >>>>C >>>> Using the issue tracker 4.0.4 package, it broke with no error   >>>> indication.I >>>> Using phpmyadmin indicates #1030 - Got error 127 from storage engine  >>>> on the issues table.  >>> ' >>> Which storage engine have you used?  >>> D >>> Only Innodb work correctly, MyISAM is know to have some problem.I >>> Anyway ony Innodb support transaction, so I strongly suggest to only  I >>> use this storage engine, this is the default on OpenVMS and probably   >>> in future MySQL version. >>F >> Innodb is the default, right ? I didn't change it. [vms]my.cnf has F >> lines relating to innodb, and the rest are commented out. I didn't  >> make any changes. > F > Right, Innodb is the default, but you can specify when you create a < > table which storage engine you want to use for this table.
 > You can use + > mysql> show table status like 'my_table';  > ( > to check which storage engine is used.  E You've guessed it JF - every table in every db has engine MyISAM, not @ innodb. That could be my problem, I guess. But I haven't createdB anything myself - I just ran the setup stuff I was told to run forH mysql, and then for issuetracker - is there some way to check/change theH default ? Some way to change (unload/drop/create/reload) existing tablesA & dbs ? Would it give the applications any problems, or would the  changes be transparent ?   Thanks,  Chris   I >> I wondered about skip-networking & safe-updates, but didn't turn them  D >> on, since I didn't really understand the implications. Should I ? >>H >> But the repair article I saw said it was only for MyISAM - which may 7 >> be why it didn't repair innodb that well, I suppose.  >>
 >> Thanks, >> Chris >>* >>>> I'm guessing it's a mysql corruption. >>>> Can anyone confirm ?  >>>> How can I fix it ? 4 >>>> Is there any known issue which could cause it ? >>>> >>>> Cross-posted to VAMP  >>>> >>>> Thanks,
 >>>> Chris >>>> >>>  >>> Jean-Franois    ------------------------------  % Date: Fri, 29 Apr 2005 11:55:50 +0100 # From: issinoho <issinoho@gmail.com> < Subject: Re: mysql #1030 - Got error 127 from storage engine0 Message-ID: <11744lo8ibfmje4@corp.supernews.com>   Chris Sharman wrote: > Jean-Franois Pironne wrote:v >  >> Chris Sharman wrote:  >>! >>> Jean-Franois Pironne wrote:  >>>v: >>>>> Alpha VMS 7.3-1, Apache 2.0, php 4.3.2, mysql 4.1.9, >>>>> php_msql client 3.23.49M >>>>> D >>>>> Using the issue tracker 4.0.4 package, it broke with no error  >>>>> indication.lJ >>>>> Using phpmyadmin indicates #1030 - Got error 127 from storage engine >>>>> on the issues table. >>>> >>>>( >>>> Which storage engine have you used? >>>>E >>>> Only Innodb work correctly, MyISAM is know to have some problem.sJ >>>> Anyway ony Innodb support transaction, so I strongly suggest to only J >>>> use this storage engine, this is the default on OpenVMS and probably  >>>> in future MySQL version.f >>>h >>>rG >>> Innodb is the default, right ? I didn't change it. [vms]my.cnf has  G >>> lines relating to innodb, and the rest are commented out. I didn't o >>> make any changes.w >> >>G >> Right, Innodb is the default, but you can specify when you create a a= >> table which storage engine you want to use for this table.1 >> You can use, >> mysql> show table status like 'my_table'; >>) >> to check which storage engine is used.3 >  > G > You've guessed it JF - every table in every db has engine MyISAM, notaB > innodb. That could be my problem, I guess. But I haven't createdD > anything myself - I just ran the setup stuff I was told to run forJ > mysql, and then for issuetracker - is there some way to check/change theJ > default ? Some way to change (unload/drop/create/reload) existing tablesC > & dbs ? Would it give the applications any problems, or would ther > changes be transparent ? > 	 > Thanks,  > Chris  > J >>> I wondered about skip-networking & safe-updates, but didn't turn them E >>> on, since I didn't really understand the implications. Should I ?e >>> I >>> But the repair article I saw said it was only for MyISAM - which may e8 >>> be why it didn't repair innodb that well, I suppose. >>>g >>> Thanks,e	 >>> Chrisg >>>t+ >>>>> I'm guessing it's a mysql corruption.0 >>>>> Can anyone confirm ? >>>>> How can I fix it ?5 >>>>> Is there any known issue which could cause it ?h >>>>>i >>>>> Cross-posted to VAMP >>>>> 
 >>>>> Thanks,n >>>>> ChrisA >>>>>S >>>> >>>> Jean-Franois >  >  >   C For the record, VAMP is using InnoDB, however some other databases i? within my MySQL installation are in fact MyISAM. Now I haven't AI explicitly set the engine type at any point, so I can only conclude that -: the SQL script that creates the database is specifying it.   ------------------------------  % Date: Fri, 29 Apr 2005 16:22:19 +0100a0 From: Chris Sharman <chris.sharman@sorry.nospam>< Subject: Re: mysql #1030 - Got error 127 from storage engine4 Message-ID: <d4tjfb$odu$1$8302bc10@news.demon.co.uk>   Jean-Franois Pironne wrote:h   > Chris Sharman wrote: >   >> Jean-Franois Pironne wrote:' >>> Which storage engine have you used?i >>>tD >>> Only Innodb work correctly, MyISAM is know to have some problem.I >>> Anyway ony Innodb support transaction, so I strongly suggest to only rI >>> use this storage engine, this is the default on OpenVMS and probably W >>> in future MySQL version. >>F >> Innodb is the default, right ? I didn't change it. [vms]my.cnf has F >> lines relating to innodb, and the rest are commented out. I didn't  >> make any changes. > F > Right, Innodb is the default, but you can specify when you create a < > table which storage engine you want to use for this table.
 > You can useo+ > mysql> show table status like 'my_table';  > ( > to check which storage engine is used.  H Seems like MyISAM is the default, and the only engine supported for the  system tables.G I've put "default-table-type=innodb" in the [mysqld] section of my.cnf )' and restarted - it wasn't there before..  B I've also done: "alter table <name> type=innodb;" to all my issue F tracker tables (but not to mysql or mysql_help). Doing that seemed to D lose some data too, but I'm hoping that that will be the end of the 
 suspect data.'  4 What are the known problems with the MyISAM engine ?? What engine are other people seeing used (show table status;) ?g) Has anyone got a default-table-type set ?    Thanks,n Chris    ------------------------------    Date: 29 Apr 2005 06:05:05 -0700' From: blake.batten@gmail.com (that_guy)  Subject: Processor scheduling4= Message-ID: <24c06f36.0504290505.19e87b8d@posting.google.com>c  D mabey im looking in the wrong place but i cant find much informationN could someone please give me a rundown (or a link) on VMS processor scheduling   ------------------------------    Date: 29 Apr 2005 06:27:31 -0700( From: "Shawn" <shawnm1964@sbcglobal.net>& Subject: Replacing Shadowed Boot DriveC Message-ID: <1114781251.750555.105530@g14g2000cwa.googlegroups.com>p   Hi All,h  G I have a 1000A 5/400 Alpha Server running OpenVMS 7.2-1 which currentlyvD utilizes Disk Shadowing on all the drives including the boot drives.  D The boot drive is  made up of  (2) 4.3GB shadowed disk drives.  I amD going to replace both of these drives with (2) 9.1GB Disk Drives and re-enable shadowing.  E I would like to list the steps I plan to follow and see if any of youtG have any suggestions or corrections my plan.  I have no extra drives toME copy the data from the system drives too, so I have to use a DLT TapehG Drive.  The backup command I am going to use is backup and then restore  is my main concern.y  B 1.  Comment any scripts called from startup which mount any disks.> 2.  Modify the modparams.dat file and set SHADOW_SYS_DISK to 0: 3.  Run the following command autogen savparams setparams. 4.  Shutdown the application.:5 5.  Show device/files to ensure all files are closed.5D 6.  Run analyze/disk/repair to fix anything on the drive which could cause a problem.( 7.  Dismount a member of the shadow set. 8.  Shutdown the systemr5 9.  Boot from an OpenVMS 7.2-1 CD and enter DCL mode.h@ ***** These next steps I would appreciate your comments on******9 10.  Mount one of the system drives:  mount/over=id dkb0:dD 11.  Backup the drive to tape as an image:  back/image/log/ign=label dkb0: mka500:dkb0.sav/saveF 12.  After the backup completes dismount the tape and disk drive: dism mka500: dism dkb0:0 13.  Shutdown the server and replace the drives.! 14.  Boot again off the CD Drive. = 15.  Mount the new drive to be restored too.  mount/for dkb0::8 16. Restore Data from tape:  back/image/ign=label/verify mka500:dkb0.sav dkb0:sB 17.  After the restore has finished dismount tape and disk drives.0 18. Shutdown the system and then boot from dkb0:B 19.  When server has come back up, modify startup scripts to mount drives as usual.: 20.  Modify modparams.dat file and set SHADOW_SYS_DISK = 1; 21.  Run the following command autogen savparams setparams.m 22.  Reboot the server.$G 23.  After the server comes back up and all drives are mounted completeE9 the final step of adding the 2nd drive to the shadow set.t6 mount/system/noassist DSA0:/shadow=($1$dkc0:) ALPHASYS  : Let me know if you see anything I need to change or modify   Thanks in advance  Shawn    ------------------------------  % Date: Fri, 29 Apr 2005 11:27:32 -0400e$ From: "Chris" <an.other@not_here.ca>* Subject: Re: Replacing Shadowed Boot Drive: Message-ID: <Ensce.13866$gA5.778961@news20.bellglobal.com>  B You're going to a lot more trouble than I have in past situations: 1.  Shutdown system normally/ 2.  Remove the secondary shadow (physical) units 3.  Replace with larger unit 4.  Boot from CD-ROM3 5.  Mount the primary shadow unit with /OVER=SHADOW C 6.  Perform image backup of primary shadow unit to new, larger unitu 7.  Shutdown systemv- 8.  Remove old (small) unit from primary slott= 9.  Replace with new (larger) unit just copied to in 5. above  10.  Leave secondary slot emptye5 11.  Reboot per normal from the primary physical unit0  I (an Analyze/Disk/Repair ahead of any of this is assumed -- the AUTOGEN isvA purely your call as to whether you think the system needs tuning)v  J In my experience, at boot, the Shadow Server will read the location of theL shadow unit from the primary (I think that's how it works), try to mount it,K fail because it isn't there, and continue with nothing more than a warning.oD Once the system is up, put your 2nd new drive in the secondary slot,F initialize it (to be cautious), then mount it into the boot shadowset.G Depending on how full the drive is, an hour or so later, it'll be up toP< snuff.  This has worked in at least 3 instances in the past.   CM3 "Shawn" <shawnm1964@sbcglobal.net> wrote in messagen= news:1114781251.750555.105530@g14g2000cwa.googlegroups.com...f	 > Hi All,r >iI > I have a 1000A 5/400 Alpha Server running OpenVMS 7.2-1 which currentlymF > utilizes Disk Shadowing on all the drives including the boot drives. >sF > The boot drive is  made up of  (2) 4.3GB shadowed disk drives.  I amF > going to replace both of these drives with (2) 9.1GB Disk Drives and > re-enable shadowing. >-G > I would like to list the steps I plan to follow and see if any of youcI > have any suggestions or corrections my plan.  I have no extra drives toSG > copy the data from the system drives too, so I have to use a DLT Tape4I > Drive.  The backup command I am going to use is backup and then restore  > is my main concern.  >LD > 1.  Comment any scripts called from startup which mount any disks.@ > 2.  Modify the modparams.dat file and set SHADOW_SYS_DISK to 0< > 3.  Run the following command autogen savparams setparams. > 4.  Shutdown the application.o7 > 5.  Show device/files to ensure all files are closed.nF > 6.  Run analyze/disk/repair to fix anything on the drive which could > cause a problem.* > 7.  Dismount a member of the shadow set. > 8.  Shutdown the systemc7 > 9.  Boot from an OpenVMS 7.2-1 CD and enter DCL mode.sB > ***** These next steps I would appreciate your comments on******; > 10.  Mount one of the system drives:  mount/over=id dkb0: F > 11.  Backup the drive to tape as an image:  back/image/log/ign=label > dkb0: mka500:dkb0.sav/saveH > 12.  After the backup completes dismount the tape and disk drive: dism > mka500: dism dkb0:2 > 13.  Shutdown the server and replace the drives.# > 14.  Boot again off the CD Drive.>? > 15.  Mount the new drive to be restored too.  mount/for dkb0:e: > 16. Restore Data from tape:  back/image/ign=label/verify > mka500:dkb0.sav dkb0: D > 17.  After the restore has finished dismount tape and disk drives.2 > 18. Shutdown the system and then boot from dkb0:D > 19.  When server has come back up, modify startup scripts to mount > drives as usual.< > 20.  Modify modparams.dat file and set SHADOW_SYS_DISK = 1= > 21.  Run the following command autogen savparams setparams.i > 22.  Reboot the server.uI > 23.  After the server comes back up and all drives are mounted complete5; > the final step of adding the 2nd drive to the shadow set.c8 > mount/system/noassist DSA0:/shadow=($1$dkc0:) ALPHASYS >u< > Let me know if you see anything I need to change or modify >n > Thanks in advance  > Shawn  >n   ------------------------------  % Date: Fri, 29 Apr 2005 11:29:29 -0400a$ From: "Chris" <an.other@not_here.ca>* Subject: Re: Replacing Shadowed Boot Drive: Message-ID: <upsce.13867$gA5.779067@news20.bellglobal.com>  ? OK, I can't count --- step 9 should read "6. above"   .....sigh-    / "Chris" <an.other@not_here.ca> wrote in message04 news:Ensce.13866$gA5.778961@news20.bellglobal.com...D > You're going to a lot more trouble than I have in past situations: > 1.  Shutdown system normally1 > 2.  Remove the secondary shadow (physical) unit@ > 3.  Replace with larger unit > 4.  Boot from CD-ROM5 > 5.  Mount the primary shadow unit with /OVER=SHADOWeE > 6.  Perform image backup of primary shadow unit to new, larger unitI > 7.  Shutdown systeme/ > 8.  Remove old (small) unit from primary slot.? > 9.  Replace with new (larger) unit just copied to in 5. aboveh! > 10.  Leave secondary slot empty(7 > 11.  Reboot per normal from the primary physical unit  >eK > (an Analyze/Disk/Repair ahead of any of this is assumed -- the AUTOGEN is C > purely your call as to whether you think the system needs tuning)m >sL > In my experience, at boot, the Shadow Server will read the location of theJ > shadow unit from the primary (I think that's how it works), try to mount it,uD > fail because it isn't there, and continue with nothing more than a warning.F > Once the system is up, put your 2nd new drive in the secondary slot,H > initialize it (to be cautious), then mount it into the boot shadowset.I > Depending on how full the drive is, an hour or so later, it'll be up toc> > snuff.  This has worked in at least 3 instances in the past. >e > CM5 > "Shawn" <shawnm1964@sbcglobal.net> wrote in messagea? > news:1114781251.750555.105530@g14g2000cwa.googlegroups.com...o > > Hi All,  > > K > > I have a 1000A 5/400 Alpha Server running OpenVMS 7.2-1 which currentlyrH > > utilizes Disk Shadowing on all the drives including the boot drives. > >sH > > The boot drive is  made up of  (2) 4.3GB shadowed disk drives.  I amH > > going to replace both of these drives with (2) 9.1GB Disk Drives and > > re-enable shadowing. > >sI > > I would like to list the steps I plan to follow and see if any of youcK > > have any suggestions or corrections my plan.  I have no extra drives to I > > copy the data from the system drives too, so I have to use a DLT Tape K > > Drive.  The backup command I am going to use is backup and then restorei > > is my main concern.t > >eF > > 1.  Comment any scripts called from startup which mount any disks.B > > 2.  Modify the modparams.dat file and set SHADOW_SYS_DISK to 0> > > 3.  Run the following command autogen savparams setparams.! > > 4.  Shutdown the application.19 > > 5.  Show device/files to ensure all files are closed. H > > 6.  Run analyze/disk/repair to fix anything on the drive which could > > cause a problem., > > 7.  Dismount a member of the shadow set. > > 8.  Shutdown the system-9 > > 9.  Boot from an OpenVMS 7.2-1 CD and enter DCL mode.FD > > ***** These next steps I would appreciate your comments on******= > > 10.  Mount one of the system drives:  mount/over=id dkb0: H > > 11.  Backup the drive to tape as an image:  back/image/log/ign=label > > dkb0: mka500:dkb0.sav/saveJ > > 12.  After the backup completes dismount the tape and disk drive: dism > > mka500: dism dkb0:4 > > 13.  Shutdown the server and replace the drives.% > > 14.  Boot again off the CD Drive.:A > > 15.  Mount the new drive to be restored too.  mount/for dkb0: < > > 16. Restore Data from tape:  back/image/ign=label/verify > > mka500:dkb0.sav dkb0: F > > 17.  After the restore has finished dismount tape and disk drives.4 > > 18. Shutdown the system and then boot from dkb0:F > > 19.  When server has come back up, modify startup scripts to mount > > drives as usual.> > > 20.  Modify modparams.dat file and set SHADOW_SYS_DISK = 1? > > 21.  Run the following command autogen savparams setparams.h > > 22.  Reboot the server. K > > 23.  After the server comes back up and all drives are mounted completer= > > the final step of adding the 2nd drive to the shadow set.r: > > mount/system/noassist DSA0:/shadow=($1$dkc0:) ALPHASYS > >w> > > Let me know if you see anything I need to change or modify > >  > > Thanks in advance 	 > > ShawnI > >g >a >p   ------------------------------    Date: 29 Apr 2005 10:35:42 -0700( From: "Shawn" <shawnm1964@sbcglobal.net>* Subject: Re: Replacing Shadowed Boot DriveC Message-ID: <1114796142.299316.141960@f14g2000cwb.googlegroups.com>o  F Wow, that is great thanks for the advice.  That does cut out a lot.  I really appreciate it.a   Shawnh   ------------------------------    Date: 29 Apr 2005 10:35:37 -0700 From: bob@instantwhip.coms- Subject: Re: Sandia says alpha the best chip! B Message-ID: <1114796137.637650.84320@g14g2000cwa.googlegroups.com>   continued from above ...  + per flops and the lowest page miss latency"    ------------------------------    Date: 29 Apr 2005 10:32:51 -0700 From: bob@instantwhip.com5) Subject: Sandia says alpha the best chip!tC Message-ID: <1114795971.045452.130850@f14g2000cwb.googlegroups.com>   G the following quotes are from this sandia pdf ... after looking at this4B it now makes now sense except either blatent stupidity or NIH (not< invented here) syndrome that HP chose itanium over alpha ...  B www.cs.sandia.gov/PublicationsPresentations/2004/CCPE03-Sandia.pdf   Page 15"  @ "In any case, as discussed in Section 3 below, these and similarG considerations led us to rank the processor choices for Red Storm basedwA on effectiveness for our design, ignoring cost for the moment, asa follows.   1. HP Alpha EV76 2. AMD Opteron! 3. Intel Pentium-4 and AMD Athlono 4. IBM Power-4 5. Intel Itanium IIi  E When we take into account performance relative to price, this rankingo changes somewhat:i   1. AMD Opteron! 2. Intel Pentium-4 and AMD Athlon/ 3. HP Alpha EV7s 4. IBM Power-4 5. Intel Itanium II"   Page 6  A "HP continues to sell the alpha ES-45 clusters formerly developedoD by Compaq.  However, they have not announced plans to build clusters6 based on the alpha EV7 processor, the Compaq-developedA replacement for the EV68, and one of the best-balanced processors 2 developed since the early Cray vector processors."  
 Page 14-15  D "alpha EV7 tops both the absolute and the relative performance list,4 since the EV7 has by far the highest memory bandwith   ------------------------------    Date: 29 Apr 2005 04:40:25 -0700' From: "tadamsmar" <tadamsmar@yahoo.com> ! Subject: Setting up an NTP server C Message-ID: <1114774825.582083.270220@z14g2000cwz.googlegroups.com>a  - Due to a network conversion, I have 6 Alpha'sn2 networked together, but temporarily without access to an NTP time server.  . One of the Alpha's has a clock that is running. about a minute a day fast.  But we need pretty+ accurate time to time stamp collected data.i  . I tried providing the IP address of one of the- Alphas as a PEER entry in the NTP config fileo, of another, but that did not appear to work.  * So how can I get one Alpha to serve NTP to another?  0 And, any ideas on how to fix a fast Alpha clock?   Thanks!i   ------------------------------  % Date: Fri, 29 Apr 2005 08:18:19 -0400 6 From: Brad Hamilton <braMdhamAiltPonS@comMcasAt.nPeSt>% Subject: Re: Setting up an NTP servery0 Message-ID: <56CdnU4gxdyRu-_fRVn-tQ@comcast.com>   tadamsmar wrote:/ > Due to a network conversion, I have 6 Alpha's 4 > networked together, but temporarily without access > to an NTP time server. > 0 > One of the Alpha's has a clock that is running0 > about a minute a day fast.  But we need pretty- > accurate time to time stamp collected data.r > 0 > I tried providing the IP address of one of the/ > Alphas as a PEER entry in the NTP config file . > of another, but that did not appear to work. >   F Did you remember to set a broadcasting address in the ntp.conf of the F Alpha that will serve time to the others?  The entry on my Alpha that F serves time to the other computers in my home network looks like this:  D broadcast 172.16.1.255 (my home network is 172.16.1/? - change your < broadcast entry to an appropriate address for your network).   <snip>   ------------------------------  % Date: Fri, 29 Apr 2005 13:55:33 +0100%* From: "Richard Brodie" <R.Brodie@rl.ac.uk>% Subject: Re: Setting up an NTP server:2 Message-ID: <d4tas5$1td$1@blackmamba.itd.rl.ac.uk>  2 "tadamsmar" <tadamsmar@yahoo.com> wrote in message= news:1114774825.582083.270220@z14g2000cwz.googlegroups.com...u  0 > I tried providing the IP address of one of the/ > Alphas as a PEER entry in the NTP config file . > of another, but that did not appear to work.  ! Should do. What does ntpq -p say?l   ------------------------------    Date: 29 Apr 2005 07:21:53 -0700' From: "tadamsmar" <tadamsmar@yahoo.com>r% Subject: Re: Setting up an NTP serversC Message-ID: <1114784513.418814.220490@o13g2000cwo.googlegroups.com>    Richard Brodie wrote:c4 > "tadamsmar" <tadamsmar@yahoo.com> wrote in message? > news:1114774825.582083.270220@z14g2000cwz.googlegroups.com...1 >82 > > I tried providing the IP address of one of the1 > > Alphas as a PEER entry in the NTP config filem0 > > of another, but that did not appear to work. > # > Should do. What does ntpq -p say?s   it shows the reach column as 0   ------------------------------    Date: 29 Apr 2005 07:25:31 -0700' From: "tadamsmar" <tadamsmar@yahoo.com> % Subject: Re: Setting up an NTP serverlC Message-ID: <1114784731.026756.234140@o13g2000cwo.googlegroups.com>a   Brad Hamilton wrote: > tadamsmar wrote:1 > > Due to a network conversion, I have 6 Alpha'sg6 > > networked together, but temporarily without access > > to an NTP time server. > > 2 > > One of the Alpha's has a clock that is running2 > > about a minute a day fast.  But we need pretty/ > > accurate time to time stamp collected data.- > >-2 > > I tried providing the IP address of one of the1 > > Alphas as a PEER entry in the NTP config fileo0 > > of another, but that did not appear to work. > >g >7G > Did you remember to set a broadcasting address in the ntp.conf of theS  G > Alpha that will serve time to the others?  The entry on my Alpha thatL  B > serves time to the other computers in my home network looks like this:o >aE > broadcast 172.16.1.255 (my home network is 172.16.1/? - change yours> > broadcast entry to an appropriate address for your network).  A I am not sure what address to use.  Should I put the address thato< I am broadcasting to, or the address I am broadcasting from?  < I tried to look up the broadcast command on the web, but the= ntp broacast command that I find does not take an IP address.o     >  > <snip>   ------------------------------    Date: 29 Apr 2005 08:19:53 -0700' From: "tadamsmar" <tadamsmar@yahoo.com>d% Subject: Re: Setting up an NTP serverCC Message-ID: <1114787993.134805.122190@l41g2000cwc.googlegroups.com>    tadamsmar wrote: > Brad Hamilton wrote: > > tadamsmar wrote:3 > > > Due to a network conversion, I have 6 Alpha'sr8 > > > networked together, but temporarily without access > > > to an NTP time server. > > >a4 > > > One of the Alpha's has a clock that is running4 > > > about a minute a day fast.  But we need pretty1 > > > accurate time to time stamp collected data.K > > >t4 > > > I tried providing the IP address of one of the3 > > > Alphas as a PEER entry in the NTP config filel2 > > > of another, but that did not appear to work. > > >. > >.E > > Did you remember to set a broadcasting address in the ntp.conf of  the  > D > > Alpha that will serve time to the others?  The entry on my Alpha that >nD > > serves time to the other computers in my home network looks like > this:s > >hG > > broadcast 172.16.1.255 (my home network is 172.16.1/? - change youro@ > > broadcast entry to an appropriate address for your network). >fC > I am not sure what address to use.  Should I put the address that~> > I am broadcasting to, or the address I am broadcasting from? > > > I tried to look up the broadcast command on the web, but the? > ntp broacast command that I find does not take an IP address.e  " Now I see, it is a broadcast mask.  ! I tried that and it did not work.h  = I need to get one of my alphas to be a server at the very topsB of the hierarchy, since I don't access to a real top level server. >  >  > > 
 > > <snip>   ------------------------------  # Date: Fri, 29 Apr 2005 15:39:30 GMT.% From: Rob Brown <mylastname@gmcl.com>7% Subject: Re: Setting up an NTP serverhE Message-ID: <Pine.LNX.4.61.0504290937330.10103@localhost.localdomain>   % On Fri, 29 Apr 2005, tadamsmar wrote:s  , > So how can I get one Alpha to serve NTP to
 > another?  C Some (early?) versions of ntp (maybe from UCX 4.x) had a MASTER or cD LOCAL MASTER command that you could put into your ntp.conf file.  I @ used this with VMS/VAX V6.2 while I was waiting for a real time  source to appear.r     -- s  B Rob Brown                        b r o w n a t g m c l d o t c o mA G. Michaels Consulting Ltd.      (866)438-2101 (voice) toll free!n6 Edmonton                         (780)438-9343 (voice)5                                   (780)437-3367 (FAX)C2                                   http://gmcl.com/   ------------------------------  + Date: Fri, 29 Apr 2005 15:51:34 +0000 (UTC)j% From: Dan Foster <usenet@evilphb.org> % Subject: Re: Setting up an NTP server16 Message-ID: <slrnd74m63.8r0.usenet@gaia.roc2.gblx.net>  K See if your version of NTP supports the 'fudge' keyword in the config file.    I.e.:u   fudge <IP> stratum 1   -Dan   ------------------------------    Date: 29 Apr 2005 09:59:24 -0700' From: "tadamsmar" <tadamsmar@yahoo.com> % Subject: Re: Setting up an NTP serverpB Message-ID: <1114793964.291081.71420@l41g2000cwc.googlegroups.com>   tadamsmar wrote:/ > Due to a network conversion, I have 6 Alpha'si4 > networked together, but temporarily without access > to an NTP time server. >t0 > One of the Alpha's has a clock that is running0 > about a minute a day fast.  But we need pretty- > accurate time to time stamp collected data.y >m0 > I tried providing the IP address of one of the/ > Alphas as a PEER entry in the NTP config file-. > of another, but that did not appear to work. >A, > So how can I get one Alpha to serve NTP to
 > another? >u2 > And, any ideas on how to fix a fast Alpha clock? >1	 > Thanks!7  7 I figured it out.  All I had to do was exactly what the.7 .conf file template said to do: comment in a server and % fudge command at the end of the file.   1 You guys were no help AT ALL.  Thanks for nothin'h   ;)   ------------------------------  % Date: Fri, 29 Apr 2005 12:54:49 -0400e' From: Dave Froble <davef@tsoft-inc.com>P% Subject: Re: Setting up an NTP server.0 Message-ID: <1174pmmgkjth727@corp.supernews.com>   tadamsmar wrote: > tadamsmar wrote: >  >>Brad Hamilton wrote: >> >>>tadamsmar wrote:  >>> 1 >>>>Due to a network conversion, I have 6 Alpha's 6 >>>>networked together, but temporarily without access >>>>to an NTP time server. >>>>2 >>>>One of the Alpha's has a clock that is running2 >>>>about a minute a day fast.  But we need pretty/ >>>>accurate time to time stamp collected data.e >>>>2 >>>>I tried providing the IP address of one of the1 >>>>Alphas as a PEER entry in the NTP config file 0 >>>>of another, but that did not appear to work. >>>> >>>6D >>>Did you remember to set a broadcasting address in the ntp.conf of >  > thed > C >>>Alpha that will serve time to the others?  The entry on my Alphav >  > that > C >>>serves time to the other computers in my home network looks liker >> >>this:o >>F >>>broadcast 172.16.1.255 (my home network is 172.16.1/? - change your? >>>broadcast entry to an appropriate address for your network).s >>C >>I am not sure what address to use.  Should I put the address that > >>I am broadcasting to, or the address I am broadcasting from? >>> >>I tried to look up the broadcast command on the web, but the? >>ntp broacast command that I find does not take an IP address.h >  > $ > Now I see, it is a broadcast mask. > # > I tried that and it did not work.w > ? > I need to get one of my alphas to be a server at the very top D > of the hierarchy, since I don't access to a real top level server. >  >>	 >>><snip>s >  > E Well, it's the broadcast command that you want.  For example, if all a2 your systems have IP addresses 192.168.7.???, then   broadcast 192.168.7.255H    should work for all the systems.  H If the first 3 bytes of the address aren't the same, possibly you could F use multiple broadcast commands.  Don't know.  Not an expert on this. 6 But I do have one VMS system feeding time to the rest.  F Now if I could only figure out how to get the Windows 2000 systems to  read the time from the server.   Dave   -- 74 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 Road0 Vanderbilt, PA  154869   ------------------------------  # Date: Fri, 29 Apr 2005 17:08:01 GMT4% From: Rob Brown <mylastname@gmcl.com>>% Subject: Re: Setting up an NTP serveriE Message-ID: <Pine.LNX.4.61.0504291106150.11016@localhost.localdomain>b  ' On Fri, 29 Apr 2005, Dave Froble wrote:o  E > Now if I could only figure out how to get the Windows 2000 systems m# > to read the time from the server.p   Off-topic now, but:t   Open a DOS window:   C\> NET TIME /HELP C\> NET TIME /SETSNTP mumble  , Seems to work on the W2K system I have here.     --    B Rob Brown                        b r o w n a t g m c l d o t c o mA G. Michaels Consulting Ltd.      (866)438-2101 (voice) toll free!I6 Edmonton                         (780)438-9343 (voice)5                                   (780)437-3367 (FAX)a2                                   http://gmcl.com/   ------------------------------    Date: 29 Apr 2005 07:47:11 -0700* From: "Alan Greig" <greigaln@netscape.net>  Subject: Re: Slow Filesystem I/OC Message-ID: <1114786031.385861.320210@o13g2000cwo.googlegroups.com>l   Rob Young wrote: >n? > 	the most part (smallness of buffers).  I'd guess a very hightC > 	percentage of VMS storage today is modern disk subsystems (HSG80y( > 	or better) as storage infrastructure. >  >n	 > 				Robd >/B > (1)  Lesson:  Careful what problem you are solving.  Those nasty hardware= >      folks might just come along and make you look foolish.h  G If you want to restrict *apparently" well performing VMS I/O to systems C coupled to an HSG80 or better then again you retreat from the loweru; end. Oops we already did that. For the university systems IfD administered we could not afford multiple HSG80s. In fact two HSC50sF were still in use when the last cluster was shut down a few years ago.F Spiralog allowed me to get high perforance (from a user point of view)> without breaking the bank. No I would not use Spiralog in thatB particular mode for a bank - or indeed even for the VMS university admin systems.  ; Typical example. Students coursework often involved lots of45 DCL/C/Fortran/Pascal/Cobol etc doing lots of I/O withsC edit/recompiles/links every other second it seemed. Students reallyeF noticed when this process suddenly became 10 times faster on our field test system.  F A VMSINSTAL rocketed along so quickly even I thought it had terminated- early first time I tried it as an experiment.> -- m
 Alan Greig   ------------------------------  % Date: Fri, 29 Apr 2005 11:58:45 -0400e( From: Bill Todd <billtodd@metrocast.net>  Subject: Re: Slow Filesystem I/O= Message-ID: <baidnZ_5rY0qxO_fRVn-vA@metrocastcablevision.com>    Rob Young wrote:   ...o  9    Newer hardware has masked the problem you describe forw( > 	the most part (smallness of buffers).  H If so, it would seem strange that people are still complaining about it H then.  Rather, it seems more likely that you're just typically clueless ; - a question which, fortunately, we can examine right here.c        I'd guess a very highC > 	percentage of VMS storage today is modern disk subsystems (HSG80 ( > 	or better) as storage infrastructure. > = > 	I'm seeing reads that hit disk cache and come back in lesscB > 	than a millisecond in some cases , which tells me it takes that" > 	long or less to write to cache.  B 1.  Read performance actually places only an upper bound on write F performance.  Reads come directly from a single cache (hopefully, the H closest, fastest one), whereas writes have to get mirrored (at least in G the level of system you're talking about) - and often mirrored between aE boxes or even between sites rather than within a single box - if you nK want durability guarantees comparable to that for the data on the platters.l  E 2.  A millisecond is still a millisecond.  Transfers to system cache  F take tens of microseconds (at least an order of magnitude faster than G your HSG cache read performance, which is itself at most only an order  D of magnitude faster than an actual access to a fast disk) - and the G available bandwidth to system cache is also orders of magnitude higher eC than that to a controller cache, so it's far less easy to saturate.u  F 3.  A surprisingly large amount of data written to Unix system caches E winds up being deleted before it ever needs to go anywhere else, and MG while Unix usually defers writing back data only for 30 seconds or so, -I VMS is certainly a sufficiently stable system to defer it for the full 5  D minutes which Jim Gray's 'rule' would suggest and thus achieve even I greater efficiency - since any application that's really concerned about  G making its writes persistent has to flush them explicitly anyway, even mG in current VMS environments.  By contrast, once you get below the file  G system (e.g., to the controller cache), you've not only already bought MF into a good deal more latency and bus utilization but there's no easy H way to let the controller cache know not to bother writing something it  already holds to disk.   ...>  G > 	(Funny thing there of course is the info is in the modest 128 MBytese? > 	of flash memory on the disk but isn't in the 4 Gigabytes of  ( > 	local memory?  I don't get that one).    You don't get rather a lot, Rob.   > G > 	It will only get better with MRAM and its *anticipated* use to mask pJ > 	filesystem or IO slowness (write to MRAM - destage to storage frame at  > 	your leisure).'  A Ah, selling futures yet again, I see - your stock in trade.  I'd  E actually tend to agree on this one, except that your track record of RF predictions is so lousy that it makes me hesitant to agree with *any*  prediction you make.  I So just wake me up if/when this actually happens, since I'm not aware of sF any current volume production of MRAM yet, or even of any target date 0 for volume availability at any reasonable price.  C (Of course, MRAM caches need to be mirrored just as current stable u caches do, by the way.)d   > @ > 	Cache is getting bigger, cpus faster, storage subsystems muchB > 	more powerful, interconnects much faster (2 Gigabit fibre cardsG > 	the norm today), the "problem" of a slow filesystem is a non-problemn > 	in my opinion.   I Well, I guess anyone who values your opinion might find that persuasive. s&   Can we have a show of hands, please?  9    In some ways (as much as I liked Spiralog), Spiralog's'E > 	timing was quite unfortunate as a few short years later the entire.D > 	infrastructure was so much faster and write-back cache the norm -. > 	Spiralog wouldn't have been a big win.  (1)  D Perhaps you're simply ignorant of the up-to-two-orders-of-magnitude ; difference between streaming sequential disk bandwidth and oI small-random-access disk bandwidth, then.  Though since I actually doubt h@ that, I guess your analysis was just as sloppy as it usually is.  G Or perhaps you have a budget which can tolerate buying and maintaining  I something like 5x - 20x as many high-performance disks (after taking all nC other potential optimizations for your proposed configuration into yC account - some of which, as in the case of increasing cache sizes,  G themselves add to system cost) as you'd otherwise need to support your - small, random-update workload.   ...5  L > (1)  Lesson:  Careful what problem you are solving.  Those nasty hardware = >      folks might just come along and make you look foolish.d  H There's a lesson here, Rob, but you've been exposed to it so many times H in the past that there's little hope you'll learn anything from it this  time.n   - bill   ------------------------------  % Date: Fri, 29 Apr 2005 09:05:25 +0200m: From: "Mark Vilstrup Svanesteen" <mark.svanesteen@mail.dk> Subject: Syntax Highlighting< Message-ID: <4271dcb1$0$73718$edfadb0f@dread14.news.tele.dk>  	 Hi group,>F Is there any way to get syntax highlighting, say for COBOL in OpenVMS?     --   Mark Vilstrup SvanesteenJ "...You don't have to be black or white, red or yellow, just be a person. J And if you practice it, you can learn it. Whatever 'it' is..." [B.B.King]    ------------------------------  # Date: Fri, 29 Apr 2005 11:39:13 GMT.! From: Nigel Barker <nigel@hp.com>l  Subject: Re: Syntax Highlighting8 Message-ID: <u57471p69u9s5gsv407gkh4kkpnlh0cfhf@4ax.com>  > On Fri, 29 Apr 2005 09:05:25 +0200, "Mark Vilstrup Svanesteen"  <mark.svanesteen@mail.dk> wrote:  
 >Hi group,G >Is there any way to get syntax highlighting, say for COBOL in OpenVMS?.  N Should be in Netbeans real soon now. Already there for Fortran, C, C++. Take aL look at http://h71000.www7.hp.com/openvms/products/ips/netbeans/modules.html   -- Nigel Barker Live from the sunny Cote d'Azur    ------------------------------  % Date: Fri, 29 Apr 2005 07:17:40 -0700t# From: "Tom Linden" <tom@kednos.com>(  Subject: Re: Syntax Highlighting( Message-ID: <opspzzbqaazgicya@hyrrokkin>  > On Fri, 29 Apr 2005 09:05:25 +0200, Mark Vilstrup Svanesteen    <mark.svanesteen@mail.dk> wrote:   > Hi group,tH > Is there any way to get syntax highlighting, say for COBOL in OpenVMS? >7 >@6 I believe there were packages for emacs that did that.   ------------------------------  % Date: Fri, 29 Apr 2005 18:49:33 +0200n3 From: "Dr. Dweeb" <NOSPAM_5msg0h202@sneakemail.com>-  Subject: Re: Syntax Highlighting= Message-ID: <4272659e$0$67262$157c6196@dreader2.cybercity.dk>>   Mark Vilstrup Svanesteen wrote:D > Hi group,-H > Is there any way to get syntax highlighting, say for COBOL in OpenVMS?   LSE ?    ------------------------------  # Date: Fri, 29 Apr 2005 17:18:09 GMTs& From: John Reagan <john.reagan@hp.com>  Subject: Re: Syntax Highlighting2 Message-ID: <l%tce.4738$FZ2.4462@news.cpqcorp.net>   Dr. Dweeb wrote:! > Mark Vilstrup Svanesteen wrote:9 >  >>Hi group,aH >>Is there any way to get syntax highlighting, say for COBOL in OpenVMS? >  >  > LSE ?  >  >   I No, LSE provides COBOL templates to help you write COBOL, but it doesn't yF know how to take existing COBOL, parse it, and highlight the language 	 keywords.k   -- a John Reagan / HP Pascal/{A|I}MACRO for OpenVMS Project Leadern Hewlett-Packard Company    ------------------------------  % Date: Fri, 29 Apr 2005 06:58:03 -0400s' From: Rich Whalen <whalenr@process.com>.' Subject: Re: TCPIP V5.5 on OpenVMS V8.2r8 Message-ID: <2l4471dhkc24spus3km08u8bb4u2uvnk3f@4ax.com>  3 On 28 Apr 2005 17:10:22 -0500, "Gareth V. Williams"n# <graff@cfa0.cfa.harvard.edu> wrote:t   >i ><rant>cJ >Why couldn't sftp have been designed to be backwards-compatible with ftp?F >Then in order to have encrypted transfer, all one would need to do to8 >make existing ftp-using command files work would be to: >$ FTP :== SFTPt >and voila!y= >The implementation of SSH has been very badly thought out...p ></rant>  @ Blame the problem with SFTP on the Unix people that designed the> initial protocol.  They specified it as binary transfers only.B Significant discussion has gotten text transfers included in later> revisions of the protocol (which is a full fledged file accessC protocol), but few implementations have been updated to support thec text functionality.i   ------------------------------    Date: 29 Apr 2005 09:57:41 -0400% From: kludge@panix.com (Scott Dorsey)t Subject: Re: TK50b+ Message-ID: <d4tegl$la9$1@panix2.panix.com>k  ) Bill Gunshannon <bill@cs.uofs.edu> wrote:a, >In article <d4lh85$jmi$1@panix2.panix.com>,) >	kludge@panix.com (Scott Dorsey) writes:i >> mG >> The only problem are the heads, which were made by Applied MagneticstG >> and are no longer available.  A couple head companies have said theytD >> would make replacements for me if I could order a couple hundred. >0C >I understand leaders are getting hard to find too and they tend todA >break with age as well.  And you have to admit, age is one thing-C >the TK50's have (meaning not many newer tapedrives like DAT lastedr
 >as long!)  ? You can use the leaders from DLT drives and make them work.  It-? can take some cutting and pasting but it's not that big a deal.g --scott      -- b6 "C'est un Nagra.  C'est suisse, et tres, tres precis."   ------------------------------  # Date: Fri, 29 Apr 2005 11:29:32 GMTi* From: "FredK" <fred.nospam@nospam.dec.com>> Subject: Re: VMS equivalent of % xset fp+ directory_with_fonts2 Message-ID: <wUoce.4694$BG2.3179@news.cpqcorp.net>  L "Phillip Helbig---remove CLOTHES to reply" <helbig@astro.multiCLOTHESvax.de>/ wrote in message news:d4rltb$6je$1@online.de...b< > In article <rhUbe.4631$oA1.2639@news.cpqcorp.net>, "FredK"& > <fred.nospam@nospam.dec.com> writes: >l >/* > > Sorry.  I meant DECW$FONT not SYS$FONT >t% > That seems to be undefined as well.w  -  $ sho log decw$font/table=decw$server0_tablem>    "DECW$FONT" = "DECW$SYSCOMMON:[SYSFONT.DECW.USER_CURSOR32]" (DECW$SERVER0_TAB  LE)t2         = "DECW$SYSCOMMON:[SYSFONT.DECW.CURSOR32]"7         = "DECW$SYSCOMMON:[SYSFONT.DECW.USER_CURSOR16]"t2         = "DECW$SYSCOMMON:[SYSFONT.DECW.CURSOR16]"5         = "DECW$SYSCOMMON:[SYSFONT.DECW.USER_100DPI]"e0         = "DECW$SYSCOMMON:[SYSFONT.DECW.100DPI]"4         = "DECW$SYSCOMMON:[SYSFONT.DECW.USER_75DPI]"/         = "DECW$SYSCOMMON:[SYSFONT.DECW.75DPI]"w5         = "DECW$SYSCOMMON:[SYSFONT.DECW.USER_COMMON]"n0         = "DECW$SYSCOMMON:[SYSFONT.DECW.COMMON]"0         = "DECW$SYSCOMMON:[SYSFONT.DECW.SPEEDO]"5         = "DECW$SYSCOMMON:[SYSFONT.DECW.USER_SPEEDO]"a/         = "DECW$SYSCOMMON:[SYSFONT.DECW.TYPE1]" 4         = "DECW$SYSCOMMON:[SYSFONT.DECW.USER_TYPE1]"2         = "DECW$SYSCOMMON:[SYSFONT.DECW.TRUETYPE]"7         = "DECW$SYSCOMMON:[SYSFONT.DECW.USER_TRUETYPE]"e   ------------------------------   Date: 29 Apr 2005 12:40:40 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)4 Subject: Re: What is Different or Special About VMS?, Message-ID: <3deoa8F6s6di1U1@individual.net>  3 In article <d4FcgXqIsRmD@eisner.encompasserve.org>,r> 	koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:Y > In article <3dc4u8F6tagnqU1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:n >>  1 >> I didn't say that.  I said that DEFCON didn't.  > J >    Except that VMS was at DEFCON and was attacked.  RSX and CPM weren't.  H More likely VMS was at DEFCON and a few curious people walked by, tappedG the keyboard, said, "What the heck is this?", shrugged their shoulders, G and walked on.  I have yet to see anything that even hinted at the idearF that people at DEFCON knew enough about VMS to pose the kind of threatG to it that they pose to Windows and Unix.  Security by obscurity.  But, G again, I stress that this does notmean that VMS is in any way unsecure.oF I have no doubt that it is more secure than both Windows and Unix.  MyI only pooint is that DEFCON is not a reasonable measure any more than CERTeK is.  If someone can get NSA or NIST to certify VMS as secure, I would place G value on that evaluation.  But DEFCON, what a joke.  I would be willing G to bet that if you went to DEFCON with a Prime-9950 running the currentoE release of Primos no one there would be able to hack it.  The same isgK probably true of RSX.  And if access was limited to other than the console,eJ even RSTS (hacking passwords on RSTS during the startup is well documentedG on the INTERNET).  However, put Johnny Billquist at the console and allm bets are off.  :-)   bill      --  J 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>   g   ------------------------------    Date: 29 Apr 2005 06:04:15 -0700 From: bob@instantwhip.com 4 Subject: Re: What is Different or Special About VMS?C Message-ID: <1114779855.923045.242210@o13g2000cwo.googlegroups.com>-  1 wrong ... these people had access to hack tcp/ip, * apache, and other common software packages- which the techniques would be the same on anyD5 other platform ... buffer overflows and other attacks 5 use the same logic on all computers ... the hard facts/ is they could not compromise the vms system ... 0 something which is proved over and over again in the real world ...   ------------------------------  # Date: Fri, 29 Apr 2005 13:28:23 GMTt! From: Nigel Barker <nigel@hp.com>t4 Subject: Re: What is Different or Special About VMS?8 Message-ID: <ubd471dop2k28anjvqpdbtk055k7u41tpv@4ax.com>  F On 29 Apr 2005 12:40:40 GMT, bill@cs.uofs.edu (Bill Gunshannon) wrote:  4 >In article <d4FcgXqIsRmD@eisner.encompasserve.org>,? >	koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:cZ >> In article <3dc4u8F6tagnqU1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes: >>> 2 >>> I didn't say that.  I said that DEFCON didn't. >> aK >>    Except that VMS was at DEFCON and was attacked.  RSX and CPM weren't.u >yI >More likely VMS was at DEFCON and a few curious people walked by, tappedxH >the keyboard, said, "What the heck is this?", shrugged their shoulders,H >and walked on.  I have yet to see anything that even hinted at the ideaG >that people at DEFCON knew enough about VMS to pose the kind of threata* >to it that they pose to Windows and Unix.  = Quite the contrary here's a report from one of the DFWUG teamu$ http://vmsone.com/~opcom/index.html   G Also I just realised that it's just over a year since the death of JohnpN Wisniewski. There's another description of DEFCON here as part of a tribute to? John. http://wiz.openvms.org/stories.php?story=05/04/14/43717441   -- Nigel Barker Live from the sunny Cote d'Azuro   ------------------------------    Date: 29 Apr 2005 06:30:19 -07007 From: "Douglas_Graver@idx.com" <Douglas_Graver@idx.com> 4 Subject: Re: What is Different or Special About VMS?C Message-ID: <1114781419.373822.116660@g14g2000cwa.googlegroups.com>m  B I think some of the "coolest" things about OpenVMS Clustering are:F  the ability to move a cluster from one data center to another with no cluster downtime,nA  the ability to upgrade the O/S version with no cluster downtime.a  D I think those are things that, at an introductory level, are easy toB both understand and be impressed with. My peers across the cubicleA aisle in MS-Windows land are certainly impressed by those things.n  E And don't forget the stories about how OpenVMS survived 9-11. I spoke C in person with a VMS Cluster Manager who told me that their trading A system survived 9-11 with no loss of data, and how the unix based. systems were a nightmare.i   ------------------------------    Date: 29 Apr 2005 06:38:03 -0700$ From: "AEF" <spamsink2001@yahoo.com>4 Subject: Re: What is Different or Special About VMS?B Message-ID: <1114781883.081534.60250@z14g2000cwz.googlegroups.com>   Keith Cayemberg wrote: > Craig Dedo wrote:CD > > Recently, I have been asked to make a presentation about the VMSD > > operating system to an introductory computer science class at an areaD > > high school.  The time for the presentation, including Q&A, will last aF > > little under an hour.  Most of the students are at an introductory5 > > level in terms of software development expertise.  > >dG > > The teacher has asked me to concentrate on what makes VMS different F > > from other operating systems, e.g., Windows or the various Unixes. I D > > should answer the question, "Why should a person choose VMS as a( > > development or production platform?" > >A [...]e >s  > - the wonders of logical names >        - search-listsh >        - rooted-logicals >        - concealed-devices5 >        - with modality (kernel,exec,super,user) andt7 >          scope (cluster,system,group,job,process) fore$ >          security and organisationE >        - also useful for for parameterizing/controlling applicationo) >          behavior even during executiono     Also for lnm's:r   Can represent...9 o any subset of file-spec components, including node nameh
 o queue names 
 o queue formsa o queue characteristicsc o mail addresses   Also:  o iterative translationn/ o SYS$INPUT, SYS$OUTPUT, SYS$COMMAND, SYS$ERROR 
 o SYS$SYSROOTh o device independence-0 o as used with OPEN, READ, WRITE, CLOSE commands     [...]7  C And more (my apologies if I'm repeating what others have written ina this thread already):g  B o Using BACKUP/IMAGE and BACKUP/INCREMENTAL restores a disk to itsC final state, without deleted files coming back from the dead, so toe speak.  	 o Editors   E o Memory management (Actually, I don't know much about how other OSes- manage memory)   ------------------------------  % Date: Fri, 29 Apr 2005 13:18:35 -0400D' From: "Main, Kerry" <kerry.main@hp.com>o4 Subject: RE: What is Different or Special About VMS?R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB5ECEC2@tayexc19.americas.cpqcorp.net>   > -----Original Message-----$ > From: bill@triangle.cs.uofs.edu=20A > [mailto:bill@triangle.cs.uofs.edu] On Behalf Of Bill Gunshannong > Sent: April 29, 2005 8:41 AM > To: Info-VAX@Mvb.Saic.Comh6 > Subject: Re: What is Different or Special About VMS? >=205 > In article <d4FcgXqIsRmD@eisner.encompasserve.org>,d@ > 	koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:3 > > In article <3dc4u8F6tagnqU1@individual.net>,=20p, > bill@cs.uofs.edu (Bill Gunshannon) writes: > >>=20 3 > >> I didn't say that.  I said that DEFCON didn't.r > >=20B > >    Except that VMS was at DEFCON and was attacked.  RSX and=20 > CPM weren't. >=20B > More likely VMS was at DEFCON and a few curious people walked=20 > by, tappedA > the keyboard, said, "What the heck is this?", shrugged their=20a > shoulders,@ > and walked on.  I have yet to see anything that even hinted=20
 > at the idearH > that people at DEFCON knew enough about VMS to pose the kind of threat; > to it that they pose to Windows and Unix.  Security by=20l > obscurity.  But,B > again, I stress that this does notmean that VMS is in any way=20 > unsecure.lH > I have no doubt that it is more secure than both Windows and Unix.  My? > only pooint is that DEFCON is not a reasonable measure any=20e > more than CERTB > is.  If someone can get NSA or NIST to certify VMS as secure,=20 > I would placeeA > value on that evaluation.  But DEFCON, what a joke.  I would=20r > be willing@ > to bet that if you went to DEFCON with a Prime-9950 running=20
 > the currentmG > release of Primos no one there would be able to hack it.  The same isD> > probably true of RSX.  And if access was limited to other=20 > than the console,s? > even RSTS (hacking passwords on RSTS during the startup is=20u > well documented < > on the INTERNET).  However, put Johnny Billquist at the=20 > console and alld > bets are off.  :-) >=20 > bill > =20  >=20   Bill,o  E Here is a pointer to a trip report on DEFCON9. As the article states,aC the only VMS specific incident was a social engineering one where aaE hacker using social eng convinced one of the system managers to login E remotely and sniffed his password off the network (the VMS folks were- not using SSH).-  H As to your comment about no one hitting the VMS box or these hackers not1 knowing OpenVMS, as Nigel pointed out, check out:u( http://www.vmsone.com/~opcom/defcon9.htm  C "The Alpha was disconnected from the haxor network, the serial portsG issue (our fault alone) was fixed, and the network was reconnected. TheeF incident did not repeat, nor did any hack whatsoever of the VMS systemG take place during the event. The hackers bombarded the box with telnetstG and ftp attempts to every bizarre port number imaginable, obscure portssH in the 40,000 range and more. The word of the early-morning incident hadC spread, and those seeking glory and a reputation besieged the box."   
 [snip ...]  E "This man turned out to be none other than Cedric Zool, the notoriouss< Belgian hacker. This was a stroke of luck, as we fell into a? conversation, and I discovered that VMS was one of his favorite5G operating systems, and the only one he really trusted for his importantr data."  
 [snip ...]  E "During the closing session at about 4 PM Sunday, the Ghetto Hackers, F which are the most respected and skillful, gave the Green team `props'H because our stuff stayed up and our `root' was the only one they did notH get. We consider this a positive note and a high compliment, coming from this well-accomplished group."   Regards,  
 Kerry Main Senior Consultantd HP Services Canada Voice: 613-592-4660s Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  $ "OpenVMS has always had integrity .. Now, Integrity has OpenVMS .."   ------------------------------  % Date: Fri, 29 Apr 2005 10:10:10 -0700m From: Z <Z@no.spam>j4 Subject: Re: What is Different or Special About VMS?, Message-ID: <PTtce.26622$c42.23548@fe07.lga>   Nigel Barker wrote:w? > Quite the contrary here's a report from one of the DFWUG teame& > http://vmsone.com/~opcom/index.html     What specific link on that page?   ------------------------------    Date: 29 Apr 2005 10:47:01 -0700 From: bob@instantwhip.com 4 Subject: Re: What is Different or Special About VMS?C Message-ID: <1114796820.983274.185330@f14g2000cwb.googlegroups.com>   & http://vmsone.com/~opcom/index=AD.html  G "As events would have it, we had an issue, which we did not understand,:E with the operation of the serial port used as the operators' console.1E At 2:00 a.m. Saturday morning the system manager decided to telnet to:> the box in order to do some routine checks. Using Telnet in anF environment with 5000 hackers on your network is an insecure method ofG administering a computer system. A lot of people were fascinated by theiG VMS system, and had asked many questions about it, shoulder-surfing thel@ console operator, who of course answered their questions in this  friendly game of an environment.  D One of the hackers who had been showing a lot of interest in the VMSD box happened to be sniffing packets from the system manager's PC. HeF discovered the password to the account, a simple procedure any 13 yearE old kid can pull off with ease after a little social engineering. TheeF hacker logged in, and placed a couple text files (his mark for points)E in the manager's user directory, and then notified the system managersB in order to claim the points. There were no points for hacking theD machine because the files were placed in a user directory instead of= the `root' VMS directory. He was awarded 10 points for socialh engineering.  ; Was this an instance of VMS being hacked? No, it was just aoE circumstance where a privileged login session was passed in plaintext0D over a network with 5000 mechanics, social engineers, and hackers onF it. By using a telnet session on an open network, the system managers'F login information was freely made available to any who cared to record; it. Giving away your login info in this way to a hacker whovE subsequently uses it does not constitute being hacked, it constitutes F an error in security procedure. The thought of improved security, suchC as some level of encryption for telnet on VMS, immediately comes tot mind. Be very afraid.   B The Alpha was disconnected from the haxor network, the serial portG issue (our fault alone) was fixed, and the network was reconnected. ThelF incident did not repeat, nor did any hack whatsoever of the VMS systemG take place during the event. The hackers bombarded the box with telnets G and ftp attempts to every bizarre port number imaginable, obscure portsaD in the 40,000 range and more. The word of the early-morning incidentG had spread, and those seeking glory and a reputation besieged the box."s   ------------------------------   End of INFO-VAX 2005.238 ************************