0 INFO-VAX	Thu, 19 Feb 2004	Volume 2004 : Issue 99      Contents: Re: Competitive Update Re: Competitive Update4 Re: COMputable image linker map program counter help4 Re: COMputable image linker map program counter help DCL Numerical symbols  Re: DCL Numerical symbols  RE: DCL Numerical symbols  RE: DCL Numerical symbols  Re: DCL Numerical symbols  RE: DCL Numerical symbols  Re: DCL Numerical symbols  RE: DCL Numerical symbols  Re: DCL Numerical symbols  Re: DECnet/OSI over TCP/IP Re: DVD writer on ES40 fopen binary and fdopen  Re: fopen binary and fdopen   Re: GD graphics lib on CSWS 1.3?) Re: HP leaks news of Intel x86-64 product : Re: It is almost certain now, INTEL will have 64bit x86 !!: Re: It is almost certain now, INTEL will have 64bit x86 !!: Re: It is almost certain now, INTEL will have 64bit x86 !!: Re: It is almost certain now, INTEL will have 64bit x86 !!: Re: It is almost certain now, INTEL will have 64bit x86 !!: Re: It is almost certain now, INTEL will have 64bit x86 !! Re: MySQL 4.0.18 for OpenVMS! Re: New ITRC Patch Site Interface  Re: new TCPIP patch for 5.3 > Re: Open Source developers are not to be thrusted says article Re: OpenVMS Feelings RE: OpenVMS Feelings@ Re: OpenVMS thrashes slowaris and aix in cluster tco/virus study@ RE: OpenVMS thrashes slowaris and aix in cluster tco/virus studyP Re: OpenVMS thrashes slowaris and aix in cluster tco/virus study ... again! ... P Re: OpenVMS thrashes slowaris and aix in cluster tco/virus study ... again! ... $ Re: Restricting access to directory.$ RE: Restricting access to directory.$ Re: Restricting access to directory. Re: SAN or NAS ? Re: SAN or NAS ?6 SCO/Microsoft supporters may have created Netsky Virus4 Re: Solved: "Known problems with V7.3 security MUP?" Re: system directory confusion. Re: VMS Fortran RTL restriction on file names?. Re: VMS Fortran RTL restriction on file names?. Re: VMS Fortran RTL restriction on file names? Re: Why was VAX abandoned ?  Re: Why was VAX abandonned ? Re: Why was VAX abandonned ? Re: Why was VAX abandonned ?7 Re: will these third-party CD-ROM drives work with VMS? 1 Re: [OT] Solaris crashes itself when /tmp is full 1 Re: [OT] Solaris crashes itself when /tmp is full 1 Re: [OT] Solaris crashes itself when /tmp is full 1 Re: [OT] Solaris crashes itself when /tmp is full 1 Re: [OT] Solaris crashes itself when /tmp is full M Re: [VMS V7.3-2/AMDS/AVAILMAN] AVAILMAN installation deletes AMDS$CONSOLE.EXE   F ----------------------------------------------------------------------  % Date: Wed, 18 Feb 2004 12:28:13 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>  Subject: Re: Competitive Update 0 Message-ID: <c0vlou$6bg$1@new-usenet.uk.sun.com>   Bill Todd wrote:N > Since Keith isn't likely to be as eager to let you know what the competition@ > is doing as he is to tout HP, I'll fill in the gaps this time: > J > AMD announced yesterday 2 low-power Opterons, presumably mostly aimed atI > blade installations:  a 2 GHz model at 55W, and a 1.4 GHz model at 30W.  > N > Intel announced yesterday at IDF its own 64-bit x86 architecture, which willN > start shipping in limited market segments later this year and is effectively > compatible with AMD's. >   < And IBM who announced that they would be doing a 64 way Xeon< based system despite the disadvantage of it only being 32bit; have announced that the chipset they will be using allready 9 supports the 64bit Intel processor, spooky how could they  have known ?    C > On another front, Itanic no longer holds the TPC-C record:  a new N > 32-processor p690 (using 1.9 GHz POWER4+ CPUs that I hadn't heard of before)G > just blew past both the 1,000,000 tpmC mark and the best 64-processor F > Superdome score - and at significantly lower $/tpmC.  Looks as if myL > 2,000,000 tpmC estimate for the up-coming POWER5 box may have been way too > low. >   D It was inevitable that IBM would overhaul HP when they had access to the faster Power4+ CPU's.   B HP still have the fastest TPC-C TPM number with a cluster of 4 wayA rx5600's but clusters and TPC-C have a long and horrid history so > we can scratch that from the list, though I have no doubt thatG Keith and the rest of the Choir will be clinging to it like shipwrecked 5 sailors now that it is their only source of buoyancy.    Regards  Andrew Harrison    ------------------------------  % Date: Thu, 19 Feb 2004 16:05:10 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>  Subject: Re: Competitive Update 0 Message-ID: <c12mrp$99t$2@new-usenet.uk.sun.com>   Norman Lastovica wrote: B > Does SUN have TPC benchmarks of worth to brag about these days?      Is TPC-C worth bragging about ?   A If you can explain why it is then you can challenge Sun to do one # but I seriously doubt that you can.    Regards  Andrew Harrison  > * > Andrew Harrison SUNUK Consultancy wrote: >  >>Bill Todd wrote: >>O >>>Since Keith isn't likely to be as eager to let you know what the competition A >>>is doing as he is to tout HP, I'll fill in the gaps this time:  >>> K >>>AMD announced yesterday 2 low-power Opterons, presumably mostly aimed at J >>>blade installations:  a 2 GHz model at 55W, and a 1.4 GHz model at 30W. >>> O >>>Intel announced yesterday at IDF its own 64-bit x86 architecture, which will O >>>start shipping in limited market segments later this year and is effectively  >>>compatible with AMD's.  >>>  >>> >>And IBM who announced that they would be doing a 64 way Xeon> >>based system despite the disadvantage of it only being 32bit= >>have announced that the chipset they will be using allready ; >>supports the 64bit Intel processor, spooky how could they  >>have known ? >> >>D >>>On another front, Itanic no longer holds the TPC-C record:  a newO >>>32-processor p690 (using 1.9 GHz POWER4+ CPUs that I hadn't heard of before) H >>>just blew past both the 1,000,000 tpmC mark and the best 64-processorG >>>Superdome score - and at significantly lower $/tpmC.  Looks as if my M >>>2,000,000 tpmC estimate for the up-coming POWER5 box may have been way too  >>>low.  >>>  >>F >>It was inevitable that IBM would overhaul HP when they had access to >>the faster Power4+ CPU's.  >>D >>HP still have the fastest TPC-C TPM number with a cluster of 4 wayC >>rx5600's but clusters and TPC-C have a long and horrid history so @ >>we can scratch that from the list, though I have no doubt thatI >>Keith and the rest of the Choir will be clinging to it like shipwrecked 7 >>sailors now that it is their only source of buoyancy.  >>	 >>Regards  >>Andrew Harrison  >  >    ------------------------------    Date: 19 Feb 2004 07:59:21 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) = Subject: Re: COMputable image linker map program counter help 3 Message-ID: <g8na4TjaWQKd@eisner.encompasserve.org>   l In article <4b6ec350.0402181231.7b526be4@posting.google.com>, JimStrehlow@data911.com (Jim Strehlow) writes:A > We have a production program on an OpenVMS 7.3 AlphaServer that + > occasionally hangs in a COMputable state. 7 > We have a full linker map of the troublesome program. G > The program links in Oracle Callable Interface modules (C), our BASIC - > and C++ code, etc. which include AST logic.  > H > Various routines attempt some exception handling; but one module mightB > erroneously generate an exception that loops to itself and never > exits. > E > We did a $SHOW PROCESS/CONTINUOUS to get several PC address values. . > The PC ranges from 006F7510 through 00ADA638 > > > Is there a formula (math) to relate the PC (program counter)@ > information to the program modules as shown in the linker map? > G    If the address is in your code, then it will be in the range covered :    by one of the $CODE PSECTs described in the linker map.  F    After finding the range in the PSECT, subtract the start address ofG    the module.  The gives you an offset.  The offset corresponds to the 0    offsets given in a listing with machine code.  H    If the address is not listed in you $CODE PSECTs, the its either in aF    shareable library or in dynamically linked code.  Adding the /full C    option when making the linker map will tell you about shareable  G    libraries.  C++ does some dynamic linking, but generally data moves  G    around, code tends to stay put; most of the dynamic linking for code N    is to implement delayed resolution of code addresses, not to change them.  J    Anything using LIB$FIND_IMAGE_SYMBOL will get results the linker can't     readily predict.      Example:  
 a.c contains:  #include <stdio.h> main() {  	printf("Hello World\n");  }   E Address 20040 is in the $CODE PSECT for module A (from the map file):   ? Psect Name      Module Name       Base     End           Length ? ----------      -----------       ----     ---           ------ I $CODE$                          00020000 000200DF 000000E0 (        224.) I                 A               00020000 000200DF 000000E0 (        224.)   E Since module A starts at 20000 then address 20040 is offset 40 in A.  - From the listing file (with /machine option):     22FD0024     003C		LDA	R23, argv! B6FE0008     0040		STQ	R23, 8(SP)   22FD0020     0044		LDA	R23, envp  8 So address 20040 is where the STQ instruction is stored.   ------------------------------  # Date: Thu, 19 Feb 2004 14:44:41 GMT & From: Jilly <jilly@clarityconnect.com>= Subject: Re: COMputable image linker map program counter help @ Message-ID: <b587d2304e6cb319e5293897ca762638@news.teranews.com>   Jim Strehlow wrote:   A > We have a production program on an OpenVMS 7.3 AlphaServer that + > occasionally hangs in a COMputable state. 7 > We have a full linker map of the troublesome program. G > The program links in Oracle Callable Interface modules (C), our BASIC - > and C++ code, etc. which include AST logic.  > H > Various routines attempt some exception handling; but one module mightB > erroneously generate an exception that loops to itself and never > exits. > E > We did a $SHOW PROCESS/CONTINUOUS to get several PC address values. . > The PC ranges from 006F7510 through 00ADA638 > > > Is there a formula (math) to relate the PC (program counter)@ > information to the program modules as shown in the linker map? > 1 > What hex math is needed to what "base" address?  > 1 > What do we need from which sections in the map?  > Image Section Synopsis ? > Program Section Synopsis ? > % > Do we need from the Image Synopsis: < > User transfer address:                            000101A0< > Debugger transfer address:                        00000340< > OTS transfer address - LIB$INITIALIZE:            0006DA70 >  > Thank you for the help. 3 > JimStrehlow, OpenVMS Systems Manager, Data911.com   A Get into SDA and look to see where sections of code are mapped to   	 $ ANA/SYS  READ/EXEC/NOLOG  READ SYS$LOADABLE_IMAGES:SYSDEF  SHOW PROCESS/IMAGE/INDEX={pid}  K If the PC addresses show up in your main line code then yes you can use the A link map and a machine code listing to find the sections of code.    --  C Jilly - Working from Home in the Chemung River Valley - Waverly, NY H       - jilly@clarityconnect.com                      - Brett Bodine fanH       - Mark.Jilson@hp.com                            - since 1975 or soH       - http://www.jilly.baka.com           - http://www.brettbodine.com   ------------------------------    Date: 19 Feb 2004 07:25:31 -0800" From: daveparboo@hotmail.com (DXP) Subject: DCL Numerical symbols< Message-ID: <819f1cc4.0402190725.eb778e6@posting.google.com>   Greetings...  0 Can anyone help/explain the following behaviour?  " $ a=9999999999       ! 10 nines...
 $ sh sym a6   A = 1410065407   Hex = 540BE3FF  Octal = 12402761777    " $ a=99999999999      ! 11 nines...
 $ sh sym a6   A = 1215752191   Hex = 4876E7FF  Octal = 11035563777     B I need to be able to do some basic arithmetic with numbers of thisE magnitude in a dcl procedure, but if the symbols are not doing what I  ask it is somewhat difficult.    (OpenVMS V7.3-1, ES40)   Many Thanks      Dave   ------------------------------  % Date: Thu, 19 Feb 2004 08:35:47 -0700 + From: "Barry Treahy, Jr." <Treahy@MMaz.com> " Subject: Re: DCL Numerical symbols% Message-ID: <4034D7D3.60600@MMaz.com>   
 DXP wrote:  
 >Greetings...  > 1 >Can anyone help/explain the following behaviour?  > # >$ a=9999999999       ! 10 nines...  >$ sh sym a 7 >  A = 1410065407   Hex = 540BE3FF  Octal = 12402761777  >  > # >$ a=99999999999      ! 11 nines...  >$ sh sym a 7 >  A = 1215752191   Hex = 4876E7FF  Octal = 11035563777  >  > C >I need to be able to do some basic arithmetic with numbers of this F >magnitude in a dcl procedure, but if the symbols are not doing what I >ask it is somewhat difficult. >  >    > ; You are exceeding the size of what can be represented in a  E signed-longword value.  If you need to manipulate values this large,   you'll need to write code...   Barry    --    > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                            ------------------------------  % Date: Thu, 19 Feb 2004 08:36:38 -0700 B From: "Tillman, Brian (AGRE)" <Brian.Tillman@smiths-aerospace.com>" Subject: RE: DCL Numerical symbolsO Message-ID: <11721EF39C7D7F47A55447158274CAF7A59889@cossmgmbx01.email.corp.tld>   
 DXP wrote:  2 > Can anyone help/explain the following behaviour? >=0D& > $ a=3D9999999999       ! 10 nines... > $ sh sym a> >   A =3D 1410065407   Hex =3D 540BE3FF  Octal =3D 12402761777 >=0D >=0D& > $ a=3D99999999999      ! 11 nines... > $ sh sym a> >   A =3D 1215752191   Hex =3D 4876E7FF  Octal =3D 11035563777  E 10 nines exceeds the limit of a 32 bit integer.  10 nines is actually F 2540BE3FF while 11 nines is actually 174876E7FF.  The upper end of the values are simply dropped.  C The highest (unsigned) integer representable in a 32 bit integer is H 4294967295.  In practice, the highest integer is 2147483647, the highest6 signed integer in a two's-complement notational system --=0D  Brian Tillman        =0D Smiths Aerospace 3290 Patterson Ave. SE, MS 1B3 Grand Rapids, MI 49512-1991 > Brian.Tillman is the name, smiths-aerospace.com is the domain.	       =0D : I don't speak for Smiths, and Smiths doesn't speak for me.  * ******************************************G The information contained in, or attached to, this e-mail, may contain= D  confidential information and is intended solely for the use of the=G  individual or entity to whom they are addressed and may be subject to= H  legal privilege.  If you have received this e-mail in error you should=H  notify the sender immediately by reply e-mail, delete the message from=L  your system and notify your system manager.  Please do not copy it for any=F  purpose, or disclose its contents to any other person.  The views or=I  opinions presented in this e-mail are solely those of the author and do= G  not necessarily represent those of the company.  The recipient should= I  check this e-mail and any attachments for the presence of viruses.  The= A  company accepts no liability for any damage caused, directly or= 4  indirectly, by any virus transmitted in this email.* ******************************************   ------------------------------  % Date: Thu, 19 Feb 2004 10:40:08 -0500 * From: Kenneth.Robinson@VerizonWireless.com" Subject: RE: DCL Numerical symbolsR Message-ID: <61CE0AD83F4ED2119DE600805FBBCA6DE46257@hqbedbh2.uswin.ad.vzwcorp.com>  = >From: daveparboo@hotmail.com [mailto:daveparboo@hotmail.com] 
 >Greetings...  > 1 >Can anyone help/explain the following behaviour?  > # >$ a=9999999999       ! 10 nines...  >$ sh sym a 7 >  A = 1410065407   Hex = 540BE3FF  Octal = 12402761777  >  > # >$ a=99999999999      ! 11 nines...  >$ sh sym a 7 >  A = 1215752191   Hex = 4876E7FF  Octal = 11035563777    F It's called 'number too big'. The largest positive number that DCL can
 handle is: 	$ i = %x7fffffff  	$ sho sym i5 	I = 2147483647   Hex = 7FFFFFFF  Octal = 17777777777   C >I need to be able to do some basic arithmetic with numbers of this F >magnitude in a dcl procedure, but if the symbols are not doing what I >ask it is somewhat difficult.  < You need to write code in your favorite language to do this.   Ken    ------------------------------  % Date: Thu, 19 Feb 2004 15:44:35 +0000 - From: Roy Omond <Roy.Omond@BlueBubble.UK.Com> " Subject: Re: DCL Numerical symbols: Message-ID: <c12lnl$1btnag$1@ID-225674.news.uni-berlin.de>  
 DXP wrote: > Greetings... > 2 > Can anyone help/explain the following behaviour? > $ > $ a=9999999999       ! 10 nines... > $ sh sym a8 >   A = 1410065407   Hex = 540BE3FF  Octal = 12402761777 >  > $ > $ a=99999999999      ! 11 nines... > $ sh sym a8 >   A = 1215752191   Hex = 4876E7FF  Octal = 11035563777 >    > D > I need to be able to do some basic arithmetic with numbers of thisG > magnitude in a dcl procedure, but if the symbols are not doing what I  > ask it is somewhat difficult.  >  > (OpenVMS V7.3-1, ES40)  ) Sorry to be the bearer of bad tidings ...   H This appears to be the problem - you're running an ES40 (note the figureG 40).  It means you only have that number of bits to play with, and some 8 of these are used for other purposes in the SCSI driver.  C You'll need to wait for VMS to come out on the IA64 (commonly known D as Itanium, which has 64 bits, most of which are available to you asE a user - the SCSI driver has been rewritten to use less of these than F on the ES40, but remember the number of registers on the IA64 platform6 will also reduce the number of bits available to you).  E If you're not able to wait until you get an IA64 platform possibly as E a replacement for that ancient ES40 thingy, you'll need to migrate to G a Macintosh platform, where bits are not sacrificed by the SCSI driver.   G Don't be fooled by anyone who might suggest moving to Solaris - Solaris F has a similar problem, and can even crash if your temporary space getsB all used up by such large numbers, especially if your program getsE swapped to the same place, which is highly likely given the magnitude $ of the numbers you're trying to use.                     P.s. *sigh* ...    ------------------------------  % Date: Thu, 19 Feb 2004 08:51:54 -0700 B From: "Tillman, Brian (AGRE)" <Brian.Tillman@smiths-aerospace.com>" Subject: RE: DCL Numerical symbolsO Message-ID: <11721EF39C7D7F47A55447158274CAF7A5988F@cossmgmbx01.email.corp.tld>    Roy Omond wrote:  ? > This appears to be the problem - you're running an ES40 (note  > the figure: > 40).  It means you only have that number of bits to play > with, and some: > of these are used for other purposes in the SCSI driver. >=0DE > You'll need to wait for VMS to come out on the IA64 (commonly known F > as Itanium, which has 64 bits, most of which are available to you asG > a user - the SCSI driver has been rewritten to use less of these than H > on the ES40, but remember the number of registers on the IA64 platform8 > will also reduce the number of bits available to you).  H What does either of these points have to do with the OP's problem?  DCL,F whether on a 40 bit machine or on a 1,000,000 bit machine performs itsF symbol arithmetic with 32 bit integers.  If the OP HAD 40 bits to playE with, he'd be having no problem, since 40 bits can easily represent a G ten or 11 digit integer.  The highest signed value for a 40 bit integer  is 549755813887 (12 digits).  & The OP may find the following amusing:   $ a =3D 6442450944 $ sho sym a =   A =3D -2147483648   Hex =3D 80000000  Octal =3D 20000000000  --=0D  Brian Tillman        =0D Smiths Aerospace 3290 Patterson Ave. SE, MS 1B3 Grand Rapids, MI 49512-1991 > Brian.Tillman is the name, smiths-aerospace.com is the domain.	       =0D : I don't speak for Smiths, and Smiths doesn't speak for me.      * ******************************************G The information contained in, or attached to, this e-mail, may contain= D  confidential information and is intended solely for the use of the=G  individual or entity to whom they are addressed and may be subject to= H  legal privilege.  If you have received this e-mail in error you should=H  notify the sender immediately by reply e-mail, delete the message from=L  your system and notify your system manager.  Please do not copy it for any=F  purpose, or disclose its contents to any other person.  The views or=I  opinions presented in this e-mail are solely those of the author and do= G  not necessarily represent those of the company.  The recipient should= I  check this e-mail and any attachments for the presence of viruses.  The= A  company accepts no liability for any damage caused, directly or= 4  indirectly, by any virus transmitted in this email.* ******************************************   ------------------------------  % Date: Thu, 19 Feb 2004 15:57:57 +0000 - From: Roy Omond <Roy.Omond@BlueBubble.UK.Com> " Subject: Re: DCL Numerical symbols0 Message-ID: <4034DD05.8030206@BlueBubble.UK.Com>   Tillman, Brian (AGRE) wrote:    J > What does either of these points have to do with the OP's problem?  DCL,H > whether on a 40 bit machine or on a 1,000,000 bit machine performs itsH > symbol arithmetic with 32 bit integers.  If the OP HAD 40 bits to playG > with, he'd be having no problem, since 40 bits can easily represent a I > ten or 11 digit integer.  The highest signed value for a 40 bit integer  > is 549755813887 (12 digits). > ( > The OP may find the following amusing:  A For goodness sake Brian ... did I *really* have to add a smiley ?   4 I think the *sigh* at the end pretty much summed up.  	 Roy Omond  Blue Bubble Ltd.   ------------------------------  % Date: Thu, 19 Feb 2004 09:02:19 -0700 B From: "Tillman, Brian (AGRE)" <Brian.Tillman@smiths-aerospace.com>" Subject: RE: DCL Numerical symbolsO Message-ID: <11721EF39C7D7F47A55447158274CAF7A59895@cossmgmbx01.email.corp.tld>    Roy Omond wrote:  C > For goodness sake Brian ... did I *really* have to add a smiley ?   H I guess so.  Sorry.  Perhaps it's due to all the time I've been spendingE in the Microsoft newsgroups, reading posts from eternally clueless PC   users.  Yeah, that's the ticket! --=0D  Brian Tillman        =0D Smiths Aerospace 3290 Patterson Ave. SE, MS 1B3 Grand Rapids, MI 49512-1991 > Brian.Tillman is the name, smiths-aerospace.com is the domain.	       =0D : I don't speak for Smiths, and Smiths doesn't speak for me.      * ******************************************G The information contained in, or attached to, this e-mail, may contain= D  confidential information and is intended solely for the use of the=G  individual or entity to whom they are addressed and may be subject to= H  legal privilege.  If you have received this e-mail in error you should=H  notify the sender immediately by reply e-mail, delete the message from=L  your system and notify your system manager.  Please do not copy it for any=F  purpose, or disclose its contents to any other person.  The views or=I  opinions presented in this e-mail are solely those of the author and do= G  not necessarily represent those of the company.  The recipient should= I  check this e-mail and any attachments for the presence of viruses.  The= A  company accepts no liability for any damage caused, directly or= 4  indirectly, by any virus transmitted in this email.* ******************************************   ------------------------------  % Date: Thu, 19 Feb 2004 17:40:43 +0100 * From: Paul Sture <nospam@sture.homeip.net>" Subject: Re: DCL Numerical symbols0 Message-ID: <4034F51B.3E85B76B@sture.homeip.net>   Roy Omond wrote: >  > DXP wrote: > > Greetings... > > 4 > > Can anyone help/explain the following behaviour? > > & > > $ a=9999999999       ! 10 nines... > > $ sh sym a: > >   A = 1410065407   Hex = 540BE3FF  Octal = 12402761777 > >  > > & > > $ a=99999999999      ! 11 nines... > > $ sh sym a: > >   A = 1215752191   Hex = 4876E7FF  Octal = 11035563777 > >  > > F > > I need to be able to do some basic arithmetic with numbers of thisI > > magnitude in a dcl procedure, but if the symbols are not doing what I ! > > ask it is somewhat difficult.  > >  > > (OpenVMS V7.3-1, ES40) > + > Sorry to be the bearer of bad tidings ...  > J > This appears to be the problem - you're running an ES40 (note the figureI > 40).  It means you only have that number of bits to play with, and some : > of these are used for other purposes in the SCSI driver. > E > You'll need to wait for VMS to come out on the IA64 (commonly known F > as Itanium, which has 64 bits, most of which are available to you asG > a user - the SCSI driver has been rewritten to use less of these than H > on the ES40, but remember the number of registers on the IA64 platform8 > will also reduce the number of bits available to you). > G > If you're not able to wait until you get an IA64 platform possibly as G > a replacement for that ancient ES40 thingy, you'll need to migrate to I > a Macintosh platform, where bits are not sacrificed by the SCSI driver.  > I > Don't be fooled by anyone who might suggest moving to Solaris - Solaris H > has a similar problem, and can even crash if your temporary space getsD > all used up by such large numbers, especially if your program getsG > swapped to the same place, which is highly likely given the magnitude & > of the numbers you're trying to use. >  > P.s. *sigh* ...    LOL!   --  
 Paul Sture   ------------------------------  % Date: Thu, 19 Feb 2004 17:35:42 +0000 ) From: Antonio Carlini <arcarlini@iee.org> # Subject: Re: DECnet/OSI over TCP/IP = Message-ID: <Nt6Zb.1508$wJ5.38@newsfep4-winn.server.ntli.net>    Rob Brooks wrote: O > The name change from DECnet/OSI to DECnet-plus was primarily to indicate that O > a node running DECnet/OSI could still communicate with a DECnet Phase IV node K > running the NSP transport -- the DECnet/OSI product was not simply an OSI ' > stack -- it was DECnet (NSP) and OSI.  > G > I'm sure that Antonio could fill in here with a bit more information.   / No, I think you covered pretty much everything.   E DECnet-VAX Extensions was a transitional product: WANDD and PSI were  E ready to ship but Routing was not, so Extensions was a way of getting E something out of the door. (Or so I'm told, that was a bit before my  F time). Actually, some of the WANDD stuff (DSW42 and DSF32 drivers for F example) only ever existed as a Phase V driver: there was some wicked F hack that faked enough of EMAA and NCL on a Phase IV box to allow the E driver to run anyway. So some of you may have been (still be?) using  $ Phase V stuff without knowing it :-)  B The first version of DECnet/OSI did support V5.5-2 so you *could* F install that (V5.5 or V5.6 I think). But you would be much better off : upgrading the OS to V7.3 and using the latest DECnet-Plus.  H The ability to run DECnet over IP came sometime in the V6 timeframe. So I if that is a requirement, you do have to upgrade. It depended on the UCX  H   Pathworks API (usually called PWIP) and pretty much all the 3rd party > stacks supported that, so DECnet should work over any of them.  I The alternative would be to pick one of the 3rd party stacks which would  F   tunnel Phase IV DECnet over IP. Actually being able to find the kit  now might be a problem though.  I > When the product was still known as DECnet/OSI (if that's what you mean J > when you say "In the days of MR/X.400"), one could tunnel DECnet over IP > using UCX and DECnet/OSI.   I MRX lasted well into the DECnet-Plus days. I'm not sure exactly when MRX  G ceased and the replacement (Mailbus??) took over. I do remember having  G to get deeper into MRX than was good for me as a result of some change  H in the way DECnet handled NSAPs; luckily the MRX guys lived on the same G floor of the building! (ISTR that most of the difficulty was having to  D run INSPECT [a security lockdown tool] on everything, even the test H machines; most of the mail gateway products had issues with INSPECT and 4 had to be "tweaked" in non-standard ways to comply).   Antonio    --   --------------- - Antonio Carlini             arcarlini@iee.org    ------------------------------    Date: 18 Feb 2004 15:19:14 +0100C From: vaxinf@chclu.chemie.uni-konstanz.de (Eberhard Heuser-Hofmann)  Subject: Re: DVD writer on ES40 - Message-ID: <40337462$1@news.uni-konstanz.de>   G In article <1036q8l87538l18@news.supernews.com>, "Island Computers USA" & <dbturner@islandco.com.nospam> writes: >Tom > 5 >This sounds like a job for "Eberhard Heuser-Hofmann"  >  >  >  >  > D >"Tom Simpson" <thomas.simpson1@nospam.comcast.net> wrote in message+ >news:tPwYb.340087$xy6.1680859@attbi_s02... H >> I'd like to install a HP 400i DVD writer on our ES40 system using theD >> built-in IDE controller and open drive bay.  Has anyone done this >before?A >> Will it work on the IDE bus?  I'm assuming the ES40 has an IDE 
 >interface...  >>E >> I want to use it daily for archiving data in RW mode, if possible.  >>" >> I'm running VMS 7.3-1, update2. >> >> Regards,  >> Tom >> >> >  >  >   : Machine and OS version are OK. The burner is not tested wi DVDwrite but the  predecessor.  D RW media are not that meaningful. You should use DVD-RAM and a drive that supports this (ie. LG).   eberhard   ------------------------------    Date: 19 Feb 2004 02:50:32 -0800- From: bhushann@hotmail.com (Bhushan Narkhede)   Subject: fopen binary and fdopen< Message-ID: <8a3b834.0402190250.4c5f8736@posting.google.com>   Hi Guys,  <    I have a piece of code that I am not able to get working.D I have a code that generates a binary file say test.lib with fopen() and : the other code that tries to use open() and then fdopen().> Open() succeeds and fdopen() with binary mode "rb" on the file descriptor fails with error 22.   " Example : code generating test.lib #include <stdio.h>  
 int main() {          FILE* fp = NULL;$         fp = fopen("test.lib","wb");  &         fwrite("Hello World",13,1,fp);           fclose(fp);                    return 0;  }    Example: code opening test.lib   #include <stdio.h> #include <fcntl.h> #include <errno.h>2 #include <sys/stat.h>           /* umask, mkdir */0 #include <unistd.h>           /* umask, mkdir */0 #include <limits.h>           /* umask, mkdir */   /*#define O_BINARY 0  */  
 int main() {          int fd;          FILE* fp = NULL;         char* str = ""; %         char* filename = "test.lib";  :         /*printf("\ncreate binary file , fp = %d\n",fp);*/             0                 fd = open(filename,O_RDWR,NULL);4                           printf("\n fd = %d\n",fd);                  fp = fdopen(fd, "rb");  )         printf("\nerrorno = %d\n",errno);          printf("fp = %x\n",fp);          fread(str,13,1,fp);          printf("\n%s",str);       close(fd);          return 0;  }   ' Can anyone please guide me on the same.    Rgds,  Bhushan    ------------------------------  % Date: Thu, 19 Feb 2004 11:16:25 -0000 * From: "Richard Brodie" <R.Brodie@rl.ac.uk>$ Subject: Re: fopen binary and fdopen+ Message-ID: <c125ub$lvo@newton.cc.rl.ac.uk>   : "Bhushan Narkhede" <bhushann@hotmail.com> wrote in message6 news:8a3b834.0402190250.4c5f8736@posting.google.com...  < >  I have a piece of code that I am not able to get working.F > I have a code that generates a binary file say test.lib with fopen()A > and  the other code that tries to use open() and then fdopen(). @ > Open() succeeds and fdopen() with binary mode "rb" on the file! > descriptor fails with error 22.   @ If you have "b" on the fdopen(), you need "ctx=bin" on the open.? Or you could just use plain file IO and just use fopen() again.    ------------------------------  % Date: Thu, 19 Feb 2004 14:38:44 +0000 ) From: Witchy <news@sruasonidyranib.co.uk> ) Subject: Re: GD graphics lib on CSWS 1.3? 8 Message-ID: <7hi9301lpitnfsofsaegnmnv7803pam9gb@4ax.com>  / On Thu, 19 Feb 2004 13:57:35 GMT, Mike Rechtman ' <michael.rechtman.nospam@hp.com> wrote:    >Witchy wrote: >>   >> Hi folks, >>  F >> Thanks to Rick Barry I've got an almost fully functioning webserver >> running on hobbyist VMS 7.3   > 
 ><snip...> >  > 
 >> Thanks! >>   >> -- 
 >> cheers, >>   >> witchy/binarydinosaurs  > $ >Alpha, or by a happy chance - VAX??  E Alpha 3000-300L:X. I was planning on using my VAX3100 with NetBSD but ( the Alpha came along just in time.......   -- cheers,    witchy/binarydinosaurs   ------------------------------  % Date: Wed, 18 Feb 2004 09:54:02 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> 2 Subject: Re: HP leaks news of Intel x86-64 product0 Message-ID: <c0vcnr$32n$1@new-usenet.uk.sun.com>   Dean Woodward wrote:+ > http://www.theinquirer.net/?article=14187C    - The details have now been announced by Intel.E  - The points that the choir were disputing havel all been confirmed to be true.  , They are providing 64bit support for Xeon it- will be available in 2 CPU systems code namede- Nocona in the first half of 2004. And it will ) also be available in Prescott and Potomac  later this year.  . This was pretty obvious to anyone who has seen. pictures of the Prescott die, the Integer Unit was way to big.n  5 They demoed this in a Dell box and IBM announced thata2 they have a new chipset for their high end Xseries- SMP servers which allready supports the 64bite CPU's.  3 The IBM announcments are interesting, they recentlyt4 announced a 64 way Xeon based system to be delivered2 in 2005, at the time lack of 64bit support was one1 issue that IBM mentioned counterbalanced by XeonsP2 price performance advantage over Itanium the other5 CPU IBM looked at. It would appear from IBM's chipseta0 announcements that they were just being cute and. waiting for the Intel x86-64 launch which they knew all about.y  7 Intel also announced that it will be x86-64 compatible,s, another point that the choir were disputing.  3 If you want a bit of a laugh here is the HP spin on  this.s  G "From a customer perspective, this is all about choice and our support nH for industry-standard architectures. Building upon HP's rich history of F innovation with Intel across the IA-32 and Itanium platforms, we will F offer customers solutions utilizing the Intel Itanium 2 processors as B well as Intel Xeon processors with 64-bit extension technology to B provide the best performance, availability and scalability in the H market," said Shane Robison, chief strategy and technology officer, HP. D "No other company has more experience, or a clearer vision, to help E businesses implement complete solutions - built on 32-bit and 64-bit C technologies - than HP."   Regards] Andrew Harrisoni   ------------------------------  % Date: Thu, 19 Feb 2004 11:13:56 -0500 * From: "Bill Todd" <billtodd@metrocast.net>C Subject: Re: It is almost certain now, INTEL will have 64bit x86 !!]2 Message-ID: <XKmdnYy_OP06fandRVn-hw@metrocast.net>  @ "Robert Klute" <robert_klute_removethis@hp.com> wrote in message2 news:9mt730hp1tj18i51hhj09ifmaeolgm1n18@4ax.com...   ...-  G > Even though it was announced in July (if I remember correctly), thererC > are still no 8ways based on the Opteron 800.  Intel appears to bejJ > focusing in on the desktop to 2way space with it's offering.  Cray stillD > talks about Red Storm with lots of caveats and warnings in its SECB > disclosures.  So, right now, I see little evidence of the x86-64J > architectures being generally used in anything other than the desktop to
 > 4way space.b  H Well, Sun stated nearly a year ago that it was working on larger Opteron3 systems.  Newisys announced such work even earlier.t  H But the real, and immediate, threat comes from IBM.  IBM *already* sellsL Xeon platforms up to 16 processors in size, and has announced a 64-processorK Xeon platform for the future - including support for the x86-64 extensions.y   - bill   ------------------------------  # Date: Thu, 19 Feb 2004 16:40:53 GMTe9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com>oC Subject: Re: It is almost certain now, INTEL will have 64bit x86 !! 0 Message-ID: <pG5Zb.688$fS7.409@news.cpqcorp.net>  7 "JF Mezei" <jfmezei.spamnot@istop.com> wrote in message'# news:4033D3D6.5F45354D@istop.com...b > Fred Kleinsorge wrote:J > > Eh?  Like what?  VMS continues to execute to it's roadmap, and exceeds it's > > forecasts. >  >,  > Which version of the roadmap ?  H The one done in 1979 that said we'd still be on VAX.  I think that's the only one you really want.    ------------------------------  % Date: Thu, 19 Feb 2004 16:59:32 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>lC Subject: Re: It is almost certain now, INTEL will have 64bit x86 !! 0 Message-ID: <c12q1n$aa2$1@new-usenet.uk.sun.com>   Robert Klute wrote: J > On Wed, 18 Feb 2004 16:06:34 -0500, JF Mezei <jfmezei.spamnot@istop.com> > wrote: >  >  > M >>HP need to tackle the issue. They cannot pretend nothing happened otherwise0N >>they will lose credibility. IA64 has just been relegated to low volume nicheL >>market and has absolutely 0 of the originally promised benefits (low cost,M >>industry standard, commodity). Goeing forward with IA64 today means that HPtN >>will need to support 4 additional platforms (VMS, NSK, HPUX, Linux) on theirF >>IA64 boxes, and then support them on the 8086, on to of the original  >>VAX/Alpha/MIPS/PaRisc support. >  > J > Why all the platform support?  HP could support VMS, NSK, HPUX, Windows,J > and Linux on IA-64.  Then support 32-bit and x86-64 Windows and Linux onF > on its Industry Standard Server (ISS) boxes (whether they use AMD orB > Intel chips).  The legacy support would slowly dissappear as the > architectures are retired. >   < Its difficult to see what future the Windows on IPF port has- now that the x86 landscape has been reshaped.D  ; I am sure that Microsoft would like to sell Windows 2003 onn8 large 8+ way servers but in reality that market is tiny.  > Windows platforms with more than 8+ CPU's are not new Sequent,= Unisys and IBM all have or had platforms in this space. Salesi0 ot these platforms have been very dissapointing.  = If you take non x86 platforms then NT on Alpha also supportedr/ 8+ CPU systems and again it failed to take off.e  C Windows on Itanium was always more of a marketing exercise allowing ? Microsoft access to the 64bit club than a product that had muchE$ practical use on very large systems.  @ But the introduction of Windows on AMD64 and Intels new platform= means that Windows on Itanium is now one of 2 64bit platformsC> supported by Microsoft and a platform that has huge advantages< for MS. They can for example port SQL-Server and a few other< apps that may benefit from 64bit support but not bother with< the rest of their products that won't. They also do not have: to participate in the same level of ISV capture to the new< platform because only a minority of apps benefit from 64 bit support.  ) Windows on Itanium will struggle against:h  G Customers unwillingness to deploy very large SMP systems using Windows.eH The availability of large x86 SMP based systems running essentially the 6 same OS but with better price performance than Itanium? A very small software portfolio compared with Windows on AMD64.r  C This has all the characteristics of the situation that hastened the  demise of NT on Alpha.  A The same also applies to Linux on Itanium, which basically leavesa@ IPF as the migration platform for Alpha, MIPS (Tandem) and HP-PA? customers, which would be fine except that the economics of IPFo5 don't suggest that this would be a sustainable model.y   Regards  Andrew Harrison G > Even though it was announced in July (if I remember correctly), there C > are still no 8ways based on the Opteron 800.  Intel appears to be J > focusing in on the desktop to 2way space with it's offering.  Cray stillD > talks about Red Storm with lots of caveats and warnings in its SECB > disclosures.  So, right now, I see little evidence of the x86-64J > architectures being generally used in anything other than the desktop to
 > 4way space.g > H > That leaves IA-64 as Intel's offering in the large SMP space.  It is aF > place Microsoft wants to be.  Its gives HP a single architecture forC > enterprise systems and a separate one for ISS.  Coming from a BigeE > Iron/Tin background, I just don't see what being able to run a 8086t0 > derivative ISA brings to the enterprise table. >  >  > Robert Klute,d: > Who speaks for himself and not the company he works for. >  >  >    ------------------------------   Date: 19 Feb 2004 17:47:21 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)C Subject: Re: It is almost certain now, INTEL will have 64bit x86 !!h: Message-ID: <c12sr9$10t4k0$1@ID-135708.news.uni-berlin.de>  8 In article <9mt730hp1tj18i51hhj09ifmaeolgm1n18@4ax.com>,6 	Robert Klute <robert_klute_removethis@hp.com> writes: > H > That leaves IA-64 as Intel's offering in the large SMP space.  It is aF > place Microsoft wants to be.  Its gives HP a single architecture for2 > enterprise systems and a separate one for ISS.    ? But wasn't the desire to only support one architecture given aseC the reason for killing off the Alpha?  What you describe is exactly4! what they had prior to Alphacide.A   bill   -- fJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Thu, 19 Feb 2004 18:58:14 +0100  From: Dirk Munk <munk@home.nl>C Subject: Re: It is almost certain now, INTEL will have 64bit x86 !!22 Message-ID: <c12udg$qc6$1@news3.tilbu1.nb.home.nl>   John Vottero wrote:0- > "Dirk Munk" <munk@home.nl> wrote in message . > news:c11o91$hgi$1@news2.tilbu1.nb.home.nl... > ...- > L >>I have seen video statements by Carly that "OpenVMS is a strategic product >  > for> > L >>HP". Why not send brochures with this statement to HP customers? Why don't >  > they > > >>explicitely revoke Bob Palmer's strategy in such a brochure? >>H >>We all know that nothing is so difficult as reversing a negative image > 	 > about ap > J >>product. And what public efforts have Compaq and HP done for VMS in thisJ >>respect? Nothing, absolutely nothing. HP's VMS non-marketing division isJ >>effectively working as SUN's marketing division, and very well too I can >  > assure >  >>you. >> >  > L > Nothing?  That's simply not true.  Have you seen the new OpenVMS brochure?
 > It's at: > 8 > http://h71000.www7.hp.com/openvms/brochures/index.html > N > A lot of the material available on that page is new, you should take a look. > # > Direct links to the brochure are:n > B > http://h71000.www7.hp.com/openvms/brochures/openvms_brochure.pdf > B > http://h71000.www7.hp.com/openvms/brochures/openvms_brochure.htm >  > H I haven't read them yet, but that isn't the point. I know those kind of O brochures, and they make great reading for us believers. But ask your managers  O if they got one of those brochures in the mail. I can give you the answer *NO*.   P That is the problem. We can get all the upbeat material we need from the HP web L site. But what does HP marketing want us to do with it? Print it out, put a I staple through the pages, and send it by internal mail to the management?Y  L WE know VMS is great. Hell, even our SUN, HP UX and Tru64 support staff are N convinced that VMS is a superior operating system. Even our management knows! O But as long as HP fails to give us any public marketing support, we CAN'T SELL  I IT. It is as simple as that, and that is why I used the term 'betraying'.    ------------------------------  % Date: Thu, 19 Feb 2004 11:42:40 -0700 + From: "Barry Treahy, Jr." <Treahy@MMaz.com> C Subject: Re: It is almost certain now, INTEL will have 64bit x86 !!t' Message-ID: <403503A0.9030008@MMaz.com>-   --=_IS_MIME_Boundary$ Content-Type: multipart/alternative;0  boundary="------------080208070409060201060901"  , This is a multi-part message in MIME format.& --------------0802080704090602010609019 Content-Type: text/plain; charset=us-ascii; format=flowed9 Content-Transfer-Encoding: 7bith   Dirk Munk wrote:   >>G > I haven't read them yet, but that isn't the point. I know those kind .F > of brochures, and they make great reading for us believers. But ask F > your managers if they got one of those brochures in the mail. I can  > give you the answer *NO*.e >uG > That is the problem. We can get all the upbeat material we need from  E > the HP web site. But what does HP marketing want us to do with it? -H > Print it out, put a staple through the pages, and send it by internal  > mail to the management?u >cD > WE know VMS is great. Hell, even our SUN, HP UX and Tru64 support H > staff are convinced that VMS is a superior operating system. Even our B > management knows! But as long as HP fails to give us any public I > marketing support, we CAN'T SELL IT. It is as simple as that, and that 9% > is why I used the term 'betraying'.  > I The old /DEC Professional/ and /VAX Professional/ mags were often called oE the mouth pieces of Digital, but they served a valued purposes, that iE being the conveying of technical and marketing information about the sE Digital line of products...  Other pubs of that era also existed but nE nothing like that exists today for the specific VMS markets; You can tI find them for PC's, Windows, Internet, Gaming, Security, even Linux, but a
 VMS, nope!     Barry    --    > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                               & --------------080208070409060201060901) Content-Type: text/html; charset=us-asciih Content-Transfer-Encoding: 7bitL  ? <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">9 <html> <head>I   <meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">e   <title></title>  </head>9 <body> Dirk Munk wrote:<br>G <blockquote cite="midc12udg$qc6$1@news3.tilbu1.nb.home.nl" type="cite">    <blockquote type="cite"><br>   </blockquote>.G I haven't read them yet, but that isn't the point. I know those kind of6E brochures, and they make great reading for us believers. But ask yourtG managers if they got one of those brochures in the mail. I can give youf the answer *NO*.   <br>   <br>D That is the problem. We can get all the upbeat material we need fromB the HP web site. But what does HP marketing want us to do with it?E Print it out, put a staple through the pages, and send it by internal  mail to the management?    <br>   <br>G WE know VMS is great. Hell, even our SUN, HP UX and Tru64 support staffl? are convinced that VMS is a superior operating system. Even our ? management knows! But as long as HP fails to give us any public*F marketing support, we CAN'T SELL IT. It is as simple as that, and that# is why I used the term 'betraying'.n   <br>   <br>
 </blockquote>nE The old <i>DEC Professional</i> and <i>VAX Professional</i> mags were B often called the mouth pieces of Digital, but they served a valued= purposes, that being the conveying of technical and marketingeJ information about the Digital line of products...&nbsp; Other pubs of thatD era also existed but nothing like that exists today for the specificC VMS markets; You can find them for PC's, Windows, Internet, Gaming,i( Security, even Linux, but VMS, nope!<br> <br> <br>	 Barry<br>o( <pre class="moz-signature" cols="72">--    Barry Treahy, Jr                       E-mail: <a class="moz-txt-link-abbreviated" href="mailto:Treahy@MMaz.com">Treahy@MMaz.com</a>> Midwest Microwave                          Phone: 480/314-1320B Vice President &amp; CIO                         FAX: 480/661-7028                        -   </pre> </body>  </html>,  ( --------------080208070409060201060901-- --=_IS_MIME_Boundary* Content-Type: text/plain;charset=us-ascii;!      name="VirusWall_Message.txt"8 Content-Transfer-Encoding: 7bit.  Content-Disposition: attachment;%      filename="VirusWall_Message.txt"c  3 ----------------------------------------- (on MML2)f  . [AZ] email-body was scanned and no virus found9 ---------------------------------------------------------3   --=_IS_MIME_Boundary--   ------------------------------    Date: 19 Feb 2004 09:51:03 -0800& From: jordan@ccs4vms.com (Rich Jordan)% Subject: Re: MySQL 4.0.18 for OpenVMS'= Message-ID: <cc5619f2.0402190951.1f69607b@posting.google.com>%  p Jean-Franois Pironne <jf.pieronne@laposte.net> wrote in message news:<c10jv7$211$1@news-reader5.wanadoo.fr>...2 > MySQL 4.0.18 is the latest MySQL stable version. > S > The kit also include the client shareable library (compatible with previous one).  > N > Download from http://www.pi-net.dyndns.org/anonymous/kits/ or any of the two
 > mirrors. >  .- .  .e >  >  > Jean-Franois3   Thanks very much!m   Rich JordanB CCSb   ------------------------------  % Date: Thu, 19 Feb 2004 08:33:46 -0500g/ From: Geoge Pagliarulo <georgepag@adelphia.net> * Subject: Re: New ITRC Patch Site Interface1 Message-ID: <PtmdndeJ4aQhJ6ndRVn-gQ@adelphia.com>   E The functionality of V7.1-1H1 and V7.1-1H2 was all rolled into V7.1. =I So, any patches that applied to V7.1-1H1 or V7.1-1H2 were issued as V7.1  I patches. On the old site, the patches in the V7.-1H1(2) directories were tG the same patches that were in the V7.1 directory. On the new site, you nE need to look in the V7.1 directories for them.  There were never any  & V7.1-1H1 or V7.1-1H2 specific patches.  G 	I'll talk to the ITRC support folks about the directory structure and  I see if we can get V7.1-1H1 and V7.2-1H2 directories listed.  Since it is uD the archive site I don't think there will be a problem.  If we were G talking about the active side, having multiple entries of the same kit  - would wreak have with the indexing tools etc.a 	m George     Bob Ceculski wrote:hk > "George Pagliarulo" <georgepag@adelphia.net> wrote in message news:<NNLYb.578$JJ6.65@news.cpqcorp.net>..., >  >>HiF >>The new version of the ITRC patch site will finally be rolled out onC >>February 23rd. Improvements to the site consist of the following:y >>" >>** New installation Rating icons >>** Easier to see IconsK >>** A "Browse Patch List" search will order results by Installation Ratinga) >>then Issue Date then Alphabetical Issueo* >>** Dates will be shown on search resultsN >>** "Recommended" and Most Recent" fields replaced by one "Most Recent" field& >>to better reflect the VMS   process.K >>** Patch quality ratings now reflect the VMS test process rather than thei >>pre-merger HP process. >>M >>I believe these changes reflect most of what customers were looking for andhM >>should go a long way towards making the site more usable. Once the new site0H >>comes up, if you have ideas for further enhancements or see additional1 >>problems that need to be corrected let me know.1 >> >>George PagliaruloF >>ECO Release Processi  >>OpenVMS Sustaining Engineering >>george.pagliarulo@hp.com >  > ? > have you ever fixed the problem to be able to get patches foru> > older versions of vms?  We have to go to a site in austraila= > to get patches for 7.1-1h2 ... the old site worked fine ...o? > the new site has lost all of the prior version patches!  I've  > put in several complaints ...,   ------------------------------    Date: 19 Feb 2004 10:04:06 -0800& From: jordan@ccs4vms.com (Rich Jordan)$ Subject: Re: new TCPIP patch for 5.3= Message-ID: <cc5619f2.0402191004.456bca1b@posting.google.com>n  o "Shael Richmond" <ksrich@bellsouth.net> wrote in message news:<fMzWb.37916$8a5.15504@bignews1.bellsouth.net>...e; > "Jerry Alan Braga" <jabraga@flanagan.ca> wrote in messagei( > news:Mp4Wb.21064$Qa3.16705@edtnps89...9 > > Had the same problem as well as yes I did log on too.r > >,% > > I was told it was a bind problem.  > >r6 > > if you type tcpip show mx validdomain.com it works: > > if you type tcpip show mx invaliddomain.com it crashes > K > It has something to do with using an alternate gateway and the mx record. N > On mine if I do the show mx validdomain.com it crashes.  If I issue the same. > command on a pre-patch system it works fine. >  > ShaelI  E Anyone have any updated information on this problem?  Heard back fromiB HP?  We have a 'patch day' scheduled next weekend which would have@ included this update if not for the reported problems.  Any info appreciated.   Rich CCSt   ------------------------------  # Date: Wed, 18 Feb 2004 14:39:48 GMTt& From: jlsue <jefflsxxxz@sbcglobal.net>G Subject: Re: Open Source developers are not to be thrusted says articlel8 Message-ID: <stt6305lhgamb88kmvj8t4fknnnckqpqdi@4ax.com>  H On Sun, 15 Feb 2004 01:09:38 -0500, JF Mezei <jfmezei.spamnot@istop.com> wrote:  O >From an auditing point of view, isn't open-source better since auditors at theoC >user site can in fact look at the source code whereas when you buyiI >pre-compiled software you have no way to verify that there are no hiddent8 >treasures that could be nefarous to your organisation ?  9 That's a tougher question to answer than one might think.eG And I'd bet it depends on how you weigh the balance between the companyuH (non-open source) that wishes to survive (and thus provide good code) vsI the capabilities of the open source community to truly understand all the- code they put into the library.5  J Very, very, very few places would even take the time to 'verify that thereJ are no hidden treasures that could be nefarious to your organization' evenI if they had the code.  And it actually goes deeper than that, because nothD only don't they take the time to review the entire Linux source codeJ distribution, they don't build the business processes (including training,I experience, and job promotion/career advancement) that's really necessarysA to do this regularly enough to address the implied argument 'for'iH open-source.  And if the IT department doesn't have the expertise, don't+ expect the auditing departments to have it.o  H It's similar to the way programming is done in most organizations today.H It used to be that you'd study algorithms to get the most efficient workE out of your code.  Today that is pretty rare for most IT shops - it's0I cheaper to buy more hardware, and they just get the features out the doorjI to their customer departments.  Likewise, in most IT shops they just wantFK to install the distribution as-is... perhaps with a recompilation, but whatk does that really mean?   --- jlsa0 The preceding message was personal opinion only.6 I do not speak in any authorized capacity for anyone,  and certainly not my employer.- (get rid of the xxxz in my address to e-mail)t   ------------------------------   Date: 18 Feb 2004 15:43:32 GMT( From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: OpenVMS Feelingss: Message-ID: <c10173$1a43q9$1@ID-135708.news.uni-berlin.de>  O In article <11721EF39C7D7F47A55447158274CAF7A59640@cossmgmbx01.email.corp.tld>,*E 	"Tillman, Brian (AGRE)" <Brian.Tillman@smiths-aerospace.com> writes:  >  > David J. Dachtera wrote: > : >> and The Three Stooges are the senior surgical staff.=0D > 1 > Paging Doctor Fine, Doctor Howard, Doctor Fine!m >   C Close, but no cigar.  2 Howards, Moe and Curly (3 actually as Shempe' was also a Howard) and one Fine, Larry.d   Nyuck, Nyuck, Nyuck......e   bill   -- oJ 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>   a   ------------------------------  % Date: Thu, 19 Feb 2004 07:24:07 -0700-B From: "Tillman, Brian (AGRE)" <Brian.Tillman@smiths-aerospace.com> Subject: RE: OpenVMS Feelings O Message-ID: <11721EF39C7D7F47A55447158274CAF7A59840@cossmgmbx01.email.corp.tld>.   bill@gw5.cs.uofs.edu wrote:*  E > Close, but no cigar.  2 Howards, Moe and Curly (3 actually as Shemp1) > was also a Howard) and one Fine, Larry.u  G You should have seen that moments after I posted my original, I correctE it.B --=0Dh Brian Tillman        =0D Smiths Aerospace 3290 Patterson Ave. SE, MS 1B3 Grand Rapids, MI 49512-1991 > Brian.Tillman is the name, smiths-aerospace.com is the domain.	       =0Dh: I don't speak for Smiths, and Smiths doesn't speak for me.      * ******************************************G The information contained in, or attached to, this e-mail, may contain=1D  confidential information and is intended solely for the use of the=G  individual or entity to whom they are addressed and may be subject to=mH  legal privilege.  If you have received this e-mail in error you should=H  notify the sender immediately by reply e-mail, delete the message from=L  your system and notify your system manager.  Please do not copy it for any=F  purpose, or disclose its contents to any other person.  The views or=I  opinions presented in this e-mail are solely those of the author and do='G  not necessarily represent those of the company.  The recipient should=aI  check this e-mail and any attachments for the presence of viruses.  The=tA  company accepts no liability for any damage caused, directly or=t4  indirectly, by any virus transmitted in this email.* ******************************************   ------------------------------  % Date: Thu, 19 Feb 2004 16:15:14 +0000aO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>lI Subject: Re: OpenVMS thrashes slowaris and aix in cluster tco/virus study 0 Message-ID: <c12nel$9h4$1@new-usenet.uk.sun.com>   Main, Kerry wrote:  	 > Andrew,a > I > Please - give it a rest .. The old "well, the reason we could not breakpJ > in is because no one tried", or "its not really of interest.." is really > getting old. >   C No Kerry you give it a rest, the TechFoolish report is exactly thate? foolish and has holes big enough to drive a london bus through.   A The fact that it keeps being reissued in warmed over form every 94? months or so without any change to its basic flaws is testimentj; to its authors persistence not the quality of their report..  < Any report that compares the cost of downtime of a mid range: system running OS A vs a range of high end systems runningA OS's BCD etc is always going to show that OS A's cost of downtime 9 is lower given the same level of outages on all the OS's.-  @ This is a classic paid for TCO trick used over and over again to? get the "right" result for the sponsor of the analysis. Is juste= a shame that you seem intent on defending what is basically at' marketing excercise not done very well.i  I > If you'd like the real reason they found it difficult to hack, refer tonD > Keith Cayemberg's recent response on the newsgroup about OpenVMS's > security design features.s >   > If you can explain why this has any impact on the TCO analysis= then I would be interested in hearing what your reasoning is.r  	 Can you ?a   Regardse Andrew Harrisone
 > Regards, >  > Kerry Main > Senior Consultant  > HP Services Canada > Voice: 613-592-4660y > Fax: 613-591-4477) > Email: kerryDOTmainAThpDOTcomn. > (remove the DOT's and AT for email address)  >  >    ------------------------------  % Date: Thu, 19 Feb 2004 13:23:55 -0500u' From: "Main, Kerry" <kerry.main@hp.com>tI Subject: RE: OpenVMS thrashes slowaris and aix in cluster tco/virus studyIR Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB278E3B@tayexc19.americas.cpqcorp.net>   > -----Original Message-----, > From: Andrew Harrison SUNUK Consultancy=203 > [mailto:Andrew_No.Harrison_No@nospamn.sun.com]=20s" > Sent: February 19, 2004 11:15 AM > To: Info-VAX@Mvb.Saic.Come> > Subject: Re: OpenVMS thrashes slowaris and aix in cluster=20 > tco/virus studyP >=20 > Main, Kerry wrote: >=20 > > Andrew,h > >=20H > > Please - give it a rest .. The old "well, the reason we could not=20? > > break in is because no one tried", or "its not really of=20e > interest.."=20 > > is really getting old. > >=20 >=20; > No Kerry you give it a rest, the TechFoolish report is=20c= > exactly that foolish and has holes big enough to drive a=20r > london bus through.  >=20> > The fact that it keeps being reissued in warmed over form=20B > every 9 months or so without any change to its basic flaws is=20G > testiment to its authors persistence not the quality of their report.h >=20A > Any report that compares the cost of downtime of a mid range=20e? > system running OS A vs a range of high end systems running=20t= > OS's BCD etc is always going to show that OS A's cost of=20pD > downtime is lower given the same level of outages on all the OS's. >=20B > This is a classic paid for TCO trick used over and over again=20B > to get the "right" result for the sponsor of the analysis. Is=20; > just a shame that you seem intent on defending what is=20-5 > basically a marketing excercise not done very well.0 >=20? > > If you'd like the real reason they found it difficult to=20s > hack, refer=20B > > to Keith Cayemberg's recent response on the newsgroup about=20 > OpenVMS's=20 > > security design features.m > >=20 >=20: > If you can explain why this has any impact on the TCO=20H > analysis then I would be interested in hearing what your reasoning is. >=20 > Can you ?e >=20	 > Regards  > Andrew Harrison  > > Regards, > >=20 > > Kerry Main > > Senior Consultantn > > HP Services Canada > > Voice: 613-592-4660  > > Fax: 613-591-4477I! > > Email: kerryDOTmainAThpDOTcom / > > (remove the DOT's and AT for email address)g > >=20 > >=20   Andrew,>  G As I said, give it a rest .. my reply (as indicated by the quotes in myeB reply that you snipped) was in direct response to your response toF OpenVMS security as "well, the reason we could not break in is becauseH no one really tried", or "its not really of interest.." or "no one knows0 or cares about it in the hacking community.."=20  H You know as well as everyone else here the huge majority of hackers willC always migrate to what is easiest to hack and will leave the really  tough system hacking to others.e  G If you contiunue to make statements like that then you can expect the ai similar reply i.e.=20t  D "If you'd like the real reason why hackers find OpenVMS difficult toG hack, refer to Keith Cayemberg's recent response on the newsgroup about ? comparing UNIX security to OpenVMS's security design features."   7 Move on until you can respond properly to Keith's note.n   Regards,  
 Kerry Main Senior Consultanto HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477u Email: kerryDOTmainAThpDOTcomn. (remove the DOT's and AT for email address)=20   ------------------------------  % Date: Wed, 18 Feb 2004 14:00:46 +0000 O From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> Y Subject: Re: OpenVMS thrashes slowaris and aix in cluster tco/virus study ... again! ...  0 Message-ID: <c0vr6e$8et$1@new-usenet.uk.sun.com>   david20@alpha1.mdx.ac.uk wrote: c > In article <zA9Xb.27339$yE5.102621@attbi_s54>, glen herrmannsfeldt <gah@ugcs.caltech.edu> writes:i > % >>>Andrew Harrison SUNUK Consultancy   >>; >><Andrew_No.Harrison_No@nospamn.sun.com> wrote in message  . >>news:<c0i9ip$mmv$1@new-usenet.uk.sun.com>... >> >>(snip) >> >>< >>Hacking rates are not so much related to how good a system= >>is, but how much of a target it is.   I still remember that < >>we were still running SunOS machines when Solaris became a= >>hacking target.  SunOS had more known bugs, but Solaris was < >>popular for running web servers.  Since that time, Windows> >>NT and 2000 became popular web server OS, and joined Solaris >>as a hacking target. >>? >>If VMS can approach Solaris and Win2K in terms of running web5A >>servers, then I am sure it will also increase as a hack target.I >> >  > I > This is the same garbage Microsoft tries to sell to show that it isn't  J > really insecure it's just noone else has as many systems to be targeted. > N > The fact that IIS has historically been much less secure than Apache despiteL > Apache having a much bigger market share shows that this is total rubbish. >  > David Webb > VMS and Unix team leader > CCSS > Middlesex University >  >   D However Microsoft is not the exception that proves the rule. HackersD do have to be familiar with an OS to be able to hack it sucessfully.  A You could argue that Microsoft is in the worst possible position,o> sloppy codeing, poor architecture, a high degree of visibility? and a hacker community well versed in its products. I don't sayoA this in any attempt to defend Microsoft, having broadband at homeo? has brought home to me how huge MS's problem appears to be from : the volume of patches I get asked to apply to my XP box on a weekly basis.   B The SuSE desktop I also run at home provides similar functionalityA to my XP box actually slightly more because it has Apache and JESg@ components running on it as well, the volume of security related@ patches for SuSE however is much lower than for XP. I expect the@ volume to increase as linux gets more popular on the desktop but? I don't expect it to approach the level that Windows users see.4  B OpenVMS's relative obscurity as an OS is one reason why it is lessB likely to be hacked over and above any technical merit it may have from a security standpoint.e   RegardsE Andrew Harrisona   ------------------------------  % Date: Thu, 19 Feb 2004 15:56:26 +0000eO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>eY Subject: Re: OpenVMS thrashes slowaris and aix in cluster tco/virus study ... again! ... a0 Message-ID: <c12mbd$924$1@new-usenet.uk.sun.com>   Bob Ceculski wrote:i > Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> wrote in message news:<c0vr6e$8et$1@new-usenet.uk.sun.com>...h > ! >>david20@alpha1.mdx.ac.uk wrote:e >>d >>>In article <zA9Xb.27339$yE5.102621@attbi_s54>, glen herrmannsfeldt <gah@ugcs.caltech.edu> writes: >>>n >>> ' >>>>>Andrew Harrison SUNUK Consultancy   >>>>= >>>><Andrew_No.Harrison_No@nospamn.sun.com> wrote in message n0 >>>>news:<c0i9ip$mmv$1@new-usenet.uk.sun.com>... >>>>
 >>>>(snip) >>>> >>>>> >>>>Hacking rates are not so much related to how good a system? >>>>is, but how much of a target it is.   I still remember thatn> >>>>we were still running SunOS machines when Solaris became a? >>>>hacking target.  SunOS had more known bugs, but Solaris wase> >>>>popular for running web servers.  Since that time, Windows@ >>>>NT and 2000 became popular web server OS, and joined Solaris >>>>as a hacking target. >>>>A >>>>If VMS can approach Solaris and Win2K in terms of running weboC >>>>servers, then I am sure it will also increase as a hack target.  >>>> >>>f >>>uJ >>>This is the same garbage Microsoft tries to sell to show that it isn't K >>>really insecure it's just noone else has as many systems to be targeted.o >>>eO >>>The fact that IIS has historically been much less secure than Apache despitegM >>>Apache having a much bigger market share shows that this is total rubbish.a >>>i
 >>>David Webb  >>>VMS and Unix team leader  >>>CCSS  >>>Middlesex Universityb >>>  >>>o >>F >>However Microsoft is not the exception that proves the rule. HackersF >>do have to be familiar with an OS to be able to hack it sucessfully. >>C >>You could argue that Microsoft is in the worst possible position,a@ >>sloppy codeing, poor architecture, a high degree of visibilityA >>and a hacker community well versed in its products. I don't sayaC >>this in any attempt to defend Microsoft, having broadband at homevA >>has brought home to me how huge MS's problem appears to be fromr< >>the volume of patches I get asked to apply to my XP box on >>a weekly basis.g >>D >>The SuSE desktop I also run at home provides similar functionalityC >>to my XP box actually slightly more because it has Apache and JESoB >>components running on it as well, the volume of security relatedB >>patches for SuSE however is much lower than for XP. I expect theB >>volume to increase as linux gets more popular on the desktop butA >>I don't expect it to approach the level that Windows users see.e >>D >>OpenVMS's relative obscurity as an OS is one reason why it is lessD >>likely to be hacked over and above any technical merit it may have >>from a security standpoint.r >>	 >>Regardse >>Andrew Harrisont >  > < > so you can't answer to any of the above or the IBM guy, so: > now you are back to the security thru obscurity mantra, = > because that is all you have to argue with ... and I run ana9 > openvms box from my home, and don't need to worry aboutD; > patches or hacks, because it is unhackable as declared by.: > defcon9 ... and I would say these guys know a few things > about computers!      < I did Bob, discussions about how well architected OpenVMS is; from a security standpoint are irrelevant in the context of 8 your CERT related Trolling. The quicker you realise that+ the less pain you will inflict on yourself.t   Regardse Andrew Harrison    ------------------------------    Date: 19 Feb 2004 08:07:03 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)r- Subject: Re: Restricting access to directory.a3 Message-ID: <PLCvX1stTSll@eisner.encompasserve.org>s  ] In article <KhYYb.654$Wz7.242@news.cpqcorp.net>, "Rahul" <rahul.kulkarni@digital.com> writes:: > Hello,H >    I want to restrict the creation of a new file or modification of anL > existing file in a particular directory. This rule should be applicable toJ > all the accounts (accounts which have BYPASS privilege) including system
 > account.9 >   or, is there any way to make files read_only in VMS ?>  E    You cannot restrict the file access of an account that has BYPASS.HG    That's what BYPASS means.  If this is a problem, then those accounts C    should not have BYPASS.  If you have a problem with the owner ofh$    SYSTEM, then see his/her manager.  F    To really make files read-only, you burn them on a CD.  This is not    unique to VMS.    ------------------------------  % Date: Thu, 19 Feb 2004 07:20:21 -0700 B From: "Tillman, Brian (AGRE)" <Brian.Tillman@smiths-aerospace.com>- Subject: RE: Restricting access to directory.wO Message-ID: <11721EF39C7D7F47A55447158274CAF7A5983A@cossmgmbx01.email.corp.tld>a   Rahul wrote:  H >    I want to restrict the creation of a new file or modification of an> > existing file in a particular directory. This rule should beG > applicable to all the accounts (accounts which have BYPASS privilege)i > including system
 > account.9 >   or, is there any way to make files read_only in VMS ?s  @ There is no way to protect a file from an account holding BYPASSE privilege.  BYPASS means that all file protection is ignored.  That'ss$ the entire purpose of the privilege. --=0Di Brian Tillman        =0D Smiths Aerospace 3290 Patterson Ave. SE, MS 1B3 Grand Rapids, MI 49512-1991a> Brian.Tillman is the name, smiths-aerospace.com is the domain.	       =0D : I don't speak for Smiths, and Smiths doesn't speak for me.      * ******************************************G The information contained in, or attached to, this e-mail, may contain=sD  confidential information and is intended solely for the use of the=G  individual or entity to whom they are addressed and may be subject to= H  legal privilege.  If you have received this e-mail in error you should=H  notify the sender immediately by reply e-mail, delete the message from=L  your system and notify your system manager.  Please do not copy it for any=F  purpose, or disclose its contents to any other person.  The views or=I  opinions presented in this e-mail are solely those of the author and do=aG  not necessarily represent those of the company.  The recipient should=rI  check this e-mail and any attachments for the presence of viruses.  The=sA  company accepts no liability for any damage caused, directly or= 4  indirectly, by any virus transmitted in this email.* ******************************************   ------------------------------  % Date: Thu, 19 Feb 2004 17:37:38 +0100h* From: Paul Sture <nospam@sture.homeip.net>- Subject: Re: Restricting access to directory. 0 Message-ID: <4034F462.6CEB3CBA@sture.homeip.net>   Tillman, Brian (AGRE) wrote: >  > Rahul wrote: > J > >    I want to restrict the creation of a new file or modification of an@ > > existing file in a particular directory. This rule should beI > > applicable to all the accounts (accounts which have BYPASS privilege). > > including system > > account.; > >   or, is there any way to make files read_only in VMS ?t > B > There is no way to protect a file from an account holding BYPASSG > privilege.  BYPASS means that all file protection is ignored.  That'sa& > the entire purpose of the privilege.  F Nitpick. The security system will only use BYPASS as a last resort, soG if you have access via "less privileged" means it will use those before82 it gets to checking to see if BYPASS is available.  H A useful reference here is the flowchart in the Guide to System SecurityG - Section 4.3 (in my somewhat dated copy of the manual) "How the Systemb4 Determines If a User Can Access a Protected Object".   --  
 Paul Sture   ------------------------------  % Date: Wed, 18 Feb 2004 14:47:27 +0000rO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>e Subject: Re: SAN or NAS ? 0 Message-ID: <c0vtu0$9vv$1@new-usenet.uk.sun.com>   John Brandon wrote:s= > Just wanted to see what the opinions of SAN or NAS were ...r > - > I have a SAN environment on my VMS servers.u >  > The UNIX group has their NAS.p > O > They (UNIX) run ORACLE  and other database related applications on their NAS.a > 7 > They (UNIX) think it a good idea - saves money - etc.i > N > Even one (UNIX) states that there is no difference between SAN and NAS.  LOL >  >  > Or is the joke on me?s >    No.e  D 1.	The latency on NAS based storage should be rather worse than SAN     	based storage.  @ 2.	Unless the Servers that are attached to the NAS based storage= 	are using Network adaptors with TOE support then driving the1> 	NAS network will consume more CPU and memory than driving the> 	same throughput on a SAN. This is because of the NAS protocol- 	overhead, multiple copies of the blocks etc.e  @ 3.	If you are using something like a DBMS you will find that the@ 	DBMS has to be specifically qualified for NAS storage and oftenA 	the specific NAS storage, this isn't generally the case with SANt 	based storage.    So you were right to laugh.   E What NAS does do in theory is give you more flexibility and allow you G to access shared storage using lower cost components, almost everythingaH you can buy nowdays has at least a 100baseT network connection most haveF Gigabit, so you don't need to add additional adaptors to get access to5 your storage. That is if you don't mind the 3 issues.g  @ The client I work for has both, a SAN for the systems where diskB throughput and latency is an issue (not necssarely the same thing)A and a NAS strategy for smaller systems like Web servers etc whicht? don't require high performance disk I/O but which might benefit.+ from a centralised managed pool of storage.   D They would also laugh with you rather than against you because theirA view is that neither solution NAS or SAN is the right solution inh every circumstance.e   regardsr Andrew Harrison  >  > J*o*h*n B*r*a*n*d*o*nn > VMS Systems Administratore, > firstname.lastname.spam.me.not@dalsemi.com   ------------------------------    Date: 19 Feb 2004 08:20:26 -0600+ From: young_r@encompasserve.org (Rob Young)a Subject: Re: SAN or NAS ?t3 Message-ID: <DjhumVOyx5Ga@eisner.encompasserve.org>   a In article <RQWYb.17844$Dg1.17047@twister.rdc-kc.rr.com>, "Mike Naime" <mnaime@kc.rr.com> writes:  >    > I > Unless your network backbone is also 1GIG, they cannot compete with theeN > throughput of a SAN.  Even if they are at 1GIG, you could be at 2 GIG today.3 > 10 GIG in a year or two.  (I forget the timeline)r >   = 	4 Gig this year.  10 Gig?  Ouch!  Will it happen in the nexte	 	5 years?   4 http://news.earthweb.com/storage/article.php/3295601    
 Storage News   January 6, 2004 , Hitachi Demos First 4-Gig FC Implementation  By Paul Shread      M Four-gigabit Fibre Channel received yet another big boost on Monday, with the & first 4-Gig FC product implementation.   [snip]  N 4-Gig FC has caught on quickly as a bridge between current 1-2 Gig systems andJ the 10-Gig systems of the future. One of 4-Gig's biggest selling points isL that, unlike 10-Gig, it is backward-compatible with 1-2 Gig FC. 4G will also' cost about the same as 1-2 Gig systems.8   [snip]  N The 4Gb/s FCAL specification is designed to allow for a data transfer of up to> 400MB/s per port and up to 800MB/s in a dual port environment.     ---   > 	Since 4-Gig is backwards compatible, you don't have to switch= 	out your infrastructure.  Since going to 10-gig will requirerF 	a switch out of infrastructure (switches, HBAs, GBICs, etc.), 10-gig C 	may be for dreamers.  That and the fact by the time we (collective @ 	we) have demand for 10-gig, will be quite a ways out (as we all@ 	drill down and look at bandwidth usage and what would be ideal,@ 	for instance - a single tape stream will be 100 MByte/sec write> 	speed in 2 years - a guess.  The 4-gig provides 400 MByte/sec$ 	bandwidth per switched connection).     				Rob    ------------------------------    Date: 19 Feb 2004 01:04:58 -0800' From: doug_mentohl@yahoo.co.uk (Daeron)a? Subject: SCO/Microsoft supporters may have created Netsky Virush= Message-ID: <da46811d.0402190104.37fda469@posting.google.com>P  % Software feud may have provoked virusw0 Rachel Ross feb 19 2004 http://tinyurl.com/2x8nr  F A new computer virus spread over the Internet yesterday in an apparent+ attempt to destroy one of its predecessors.a  D The Netsky.b virus e-mails itself to addresses stored on an infectedC computer. It also copies itself to folders that appear to be sharedd with other computers.,   [..]  D Once a computer has been infected, the virus tries to deactivate twoC versions of the Mydoom virus that hit the Internet late last month.w   [..]  C Sebbag said it's possible someone who supports Microsoft or SCO mayO0 have written Netsky.b in retaliation for Mydoom.   http://www.thestar.com/NASApp/cs/ContentServer?pagename=thestar/Layout/Article_Type1&c=Article&cid=1077145814587&call_pageid=968350072197&col=969048863851   ------------------------------  % Date: Wed, 18 Feb 2004 14:04:28 +0000tO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>t= Subject: Re: Solved: "Known problems with V7.3 security MUP?"e0 Message-ID: <c0vrdc$8et$2@new-usenet.uk.sun.com>   Bob Koehler wrote: > In article <c0vbj0$2n8$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: > C >>Really so how about providing some examples to back up that claimo >  > B >    I'm sure google is full of them.  Meanwhile it's time to stop >    feeding the troll.. >   ; Precisely, since you havn't provided answers to a perfectlyB: sensible challenge I guess you now know who has been doing= the trolling. Thanks BTW looks like you will be going hungry.a   Regardsa Andrew Harrison>   ------------------------------    Date: 19 Feb 2004 05:02:32 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)i' Subject: Re: system directory confusiont3 Message-ID: <mtkqE5kVpdWF@eisner.encompasserve.org>*  W In article <9wXYb.29222$D_5.26478@edtnps84>, Alder <PGDEHMKOKIMD@spammotel.com> writes: F > As a newcomer to VMS I was eager to get my hands on "OpenvMS System H > Management Guide, 2nd Edition" by Baldwin, Hoffman, and Miller.  It's C > stuffed full of great information that's helped me get my system e5 > organized and generally explains things quite well.u > J > Chapter 5 deals with system startup and suggests (page 64), for reasons D > of easing into clusters, moving the default site-specific startup ? > procedures from their default location in SYS$MANAGER to the L/ > cluster-common startup directories (page 64).   ? I haven't read the book, and trusting the authors does not give > enough context to understand where they are coming from in the section to which you refer.h  K > Right above Table 5.5 the authors state that these files should be moved  J > to the SYS$COMMON:[SYS$STARTUP] directory, then in Table 5.5 they state I > the directory should be SYS$COMMON:[SYSMGR]. Are these two directories 9F > effectively the same at system startup?  Will SYSTARTUP_VMS.COM and : > SYLOGICALS.COM be executed if I choose either directory?  C Most access is via SYS$STARTUP:, which looks first in [SYS$STARTUP]r@ and then in [SYSMGR].  The two files you name are better left in? [SYSMGR] because that is where those who follow you will expect-
 to find them.-   $ SHOW LOGICAL SYS$STARTUPA    "SYS$STARTUP" = "SYS$SYSROOT:[SYS$STARTUP]" (LNM$SYSTEM_TABLE)-         = "SYS$MANAGER"5< 1  "SYS$MANAGER" = "SYS$SYSROOT:[SYSMGR]" (LNM$SYSTEM_TABLE) $ SHOW LOGICAL SYS$SYSROOT;    "SYS$SYSROOT" = "EISNER$DRA0:[SYS0.]" (LNM$SYSTEM_TABLE)m         = "SYS$COMMON:"nD 1  "SYS$COMMON" = "EISNER$DRA0:[SYS0.SYSCOMMON.]" (LNM$SYSTEM_TABLE) $   H Of course none of this has to do with cluster-common startup directoriesG unless your entire cluster uses the same system disk.  If you depend oniD the entire cluster using the same system disk, you will be in a heapF of trouble when that changes (such as running multiple VMS versions inG the same cluster, perhaps because new hardware requires a new version).-  D Cluster-common startup directories should only be located on storageA media which all systems can reach, such as via CI, Fibre Channel,t? multiport SCSI bus, etc.  Except for CI, that leaves out VAXes. 9 Except for Fibre Channel, that leaves out multiport SCSI.V   ------------------------------    Date: 19 Feb 2004 07:32:54 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler).7 Subject: Re: VMS Fortran RTL restriction on file names? 3 Message-ID: <c8co5mNQAsyB@eisner.encompasserve.org>o  f In article <0uf730l52132d6a5j809i74hjflmfkdsf3@4ax.com>, Steve Lionel <Steve.Lionel@intel.com> writes: > P > The RTL that ships with the compiler is always newer than what comes with VMS.O > It may well be, though, that the RTL internally uses NAM$C_MAXRSS as a lengthdG > limit. It CAN handle ODS-5 filenames, but perhaps not very long ones.e > 1 > I think a USEROPEN should work here, with care.  > F    OK.  That's what I was planning to try anyhow.  Any plans to updateG    the RTL for long file names?  Is there any limit in the ANSI Fortran<    standard?   ------------------------------    Date: 19 Feb 2004 07:33:31 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)e7 Subject: Re: VMS Fortran RTL restriction on file names?w3 Message-ID: <JawL+aNGURpn@eisner.encompasserve.org>   Y In article <MRPYb.617$L_6.588@news.cpqcorp.net>, John Reagan <john.reagan@hp.com> writes:t > J > With a quick scan through the listings, I didn't see any support in the D > FORRTL for long names.  I see uses of FAB, RAB64, and NAM, but no  > extended NAMs.  (    Exactly what I was guessing.  Thanks.   ------------------------------  % Date: Thu, 19 Feb 2004 10:07:07 -0500r+ From: Steve Lionel <Steve.Lionel@intel.com> 7 Subject: Re: VMS Fortran RTL restriction on file names? 8 Message-ID: <16k9301tuhhgk57pgc8pt3cop8tll19asr@4ax.com>  K On 19 Feb 2004 07:32:54 -0600, koehler@eisner.nospam.encompasserve.org (Bobn Koehler) wrote:a  G >   OK.  That's what I was planning to try anyhow.  Any plans to updatesH >   the RTL for long file names?  Is there any limit in the ANSI Fortran
 >   standard?y  M You'll have to ask HP about updates - I am no longer involved in that side of?L things. The ANSI Fortran standard says nothing at all about filename lengthsM (nor about the minimum supported length of a character string, though one canh3 make a good argument for 11 being the lower bound.)-       Steve Lionel Software Products Division Intel Corporation 
 Nashua, NH  8 User communities for Intel Software Development Products"   http://softwareforums.intel.com/ Intel Fortran Supportm7   http://developer.intel.com/software/products/support/s   ------------------------------    Date: 19 Feb 2004 10:37:37 -08007 From: jones.computer.srv@worldnet.att.net (Daryl Jones)e$ Subject: Re: Why was VAX abandoned ?= Message-ID: <8a646952.0402191037.2c04927c@posting.google.com>,   Dear JF Mezei:  E You should read the Foreward section Digital Technical Journal, AlphaoF AXP Architecture and Systems, Volume 4 Number 4, page 17. You can find this at:  J http://www.research.compaq.com/wrl/DECarchives/DTJ/DTJ800/axp-foreword.txt  # Here is an portion of that article.   > "Nonetheless, senior managers and engineers saw trouble ahead.F Workstations had displaced VAX VMS from its original technical market.F Networks of personal computers were replacing timesharing. Application9 investment was moving to standard, high-volume computers. F Microprocessors had surpassed the performance of traditional mid-rangeA computers and were closing in on mainframes. And advances in RISCmD technology threatened to aggravate all of these trends. Accordingly,@ the Executive Committee asked Engineering to develop a long-term? strategy for keeping Digital's systems competitive. Engineering5+ convened a task force to study the problem.t  F The task force looked at a wide range of potential solutions, from theC application of advanced pipelining techniques in VAX systems to thewA deployment of a new architecture. A basic constraint was that the B proposed solution had to provide strong compatibility with currentE products. After several months of study, the team concluded that onlyeD a new RISC architecture could meet the stated objective of long-termE competitiveness, and that only the existing VMS and UNIX environmentsoC could meet the stated constraint of strong compatibility. Thus, thetD challenge posed by the task force was to design the most competitive> RISC systems that would run the current software environments.  8 Key groups in Engineering responded to this challenge. AB cross-functional team from hardware and software defined the basicA architecture. Advanced development teams began work on the knottyLC engineering problems: in the semiconductor group, the specificationeE and design of a fast microprocessor, and the automatic translation oftA executable binary images; in the operating systems groups, on thetB porting of ULTRIX and of VMS (which was not portable!); and in theD compiler group, on superscalar code generation. In the fall of 1989,F Alpha became an officially sanctioned advanced development program.[2]? In the summer of 1990, it transitioned to product development."e     Regards, Daryl Jonesf    [ JF Mezei <jfmezei.spamnot@istop.com> wrote in message news:<40263256.4AD301E6@istop.com>...n > Duncan Macdonald wrote:sO > > RISC core and a RISC to CISC converter with a large cache). Leaving out theuO > > converter (and making the compilers do the work) made it possible to make aw > > fast chip - the ALPHA. > O > Yes, it is quite possible that Alpha was an architecture wich was less costly 
 > to improve.c > P > However, had they stuck with VAX, they wouldn't have lost so many customers inH > the transition from Vax to Alpha (Digital lost all messaging business,L > remember it used to be world leader is messaging with a large portfolio ofO > gateways to many disparate systems, all that went down the drain when MessageE" > Router was not ported to Alpha). > P > And you wouldn't have had customers "stuck" on vax because ISvs didn't port toI > Alpha, and then abandonned VAX development because there weren't enougha > customers left.c > E > One has to look at it from a business point of view as opposed to a  > performance point of view. > M > Consider also that DEC had to develop totally new compilers, and maintain 2 3 > separate versions of VMS (no single source tree).h > C > What was missing from Alpha-VMS was an image activator that would.; > automatically invoke Vest when it encountered VAX images.  > P > Remember that at the time the decision was made to go to Alpha, the 8086 wouldK > probably still have been at the 286 stage. Didn't the 386 arrive in early O > 1990s ? And Windows was still not a credible application and most were eitherc! > on MS-DOS or MACs before 1990s.D   ------------------------------    Date: 18 Feb 2004 23:17:51 -08007 From: jones.computer.srv@worldnet.att.net (Daryl Jones).% Subject: Re: Why was VAX abandonned ?a= Message-ID: <8a646952.0402182317.6acab987@posting.google.com>u   Dear Richard B. Gilbert:  A "If you use DEC Fortran, local variables are naturally aligned by E default. To control alignment of record structures and common blocks,  use /ALIGN qualifier."  A This is from Migrating an Application from OpenVMS VAX to OpenVMS E Alpha, VMS Version 7.0, page 2-10, under the title of Eliminating the* Problem.  C Of course, this doesn't prevent a programmer from self destruction.    Regards, Daryl Jones1  k "Richard B. Gilbert" <rgilbert88@comcast.net> wrote in message news:<Paqdnc7lqOn_Y67dRVn-gQ@comcast.com>...aE > The Alpha compilers will align the data for you unless you make it eK > impossible for them to do so.  It has been a few years since I wrote any -H > Fortran and I haven't any Alpha specific manuals to refer to.  Having F > said that, there are basically two situations in which the compiler K > can't align data for you.  The first is in a COMMON block which requires lI > that the data be stored and aligned as declared; the only way to align oK > it is to do it manually.  The easiest way is to order the data by length eH > so that quadwords come first, then longwords, words, and bytes.  Even I > better is to not use COMMON at all; in the dark ages it was the method .H > you used to declare data that would be stored in the root module of a K > program so that it would be available to overlays (remember those in the oC > PDP11?).  Now, COMMON is used by lazy programmers to destroy the VK > modularity of programs.  It was not uncommon, a few years ago, to find a $C > large program with dozens of named COMMON blocks and hundreds or  K > thousands of variables in COMMON.  Each function or subroutine would use  H > one or two variables from each COMMON block so that each function and K > subroutine had to declare most or all of them.   This meant that none of eD > this code could be reused anywhere else because it was hopelessly G > dependent on it's COMMON environment.  Of course you could, and some rD > did, move the common environment right along with the function or F > subroutine. . . .  See "Composite/Structured Design" by Glenford J. C > Myers for chapter and verse on the evils of COMMON.  The man was gE > amazing; years before objected oriented programming came along, he i? > described something very like the "object" as we now know it.  > I > The second situation in which the compiler can't align data for you is  H > when you declare a structure that is not naturally aligned and forbid K > the compiler to use "filler" to align it.  You would normally forbid the tH > compiler to use filler only when your structure mapped a record which G > you were reading from a binary file.  If you have this situation and t; > can't change the file/record design, you are truly stuck!  >  > Daryl Jones wrote: > & > >Dear Richard B. Gilbert and others: > >lH > >I am having a hard time trying to keep up on your discussion. Anyway,E > >somewhere in this discussion, the topic is about the alignment and I > >difference between the VAX and Alpha. Listed below is discussion about3D > >data alignment from "Migrating an Application from OpenVMS VAX to@ > >OpenVMS Alpha", version 7.0, page 2-9, paragraph 2.5.1.1 Data
 > >Alignment.  > >0G > >"Accessing data not naturally aligned in memory incurs a significant:I > >performance penalty both on the VAX and Alpha systems. On VAX systems, C > >most languages align data on the next available byte boundary by>G > >default, because the VAX architecture provides hardware support thatcF > >minimizes the performance penalty in referencing unaligned data. OnF > >Alpha systems, the default is to align each data item naturally, soB > >Alpha, like other RISC architectures, does not provide hardwareG > >support to minimize the performance degradation from using unalignedt	 > >data."T > >dF > >I recall that most VAX compilers did the data alignment for you butB > >not in the Alpha. The big key is that the VAX provided hardwareI > >support for alignment problems and the Alpha and other RISC processors I > >do not. However, it is stated in the above source that Alpha compliersoI > >do correct the most potential alignment problems and you can catch the G > >rest by adding the "/WARNING=ALIGNMENT" qualifier. It is possible tos" > >correct the alignment problems. > >tG > >In summer of 1992, the NVAX chip claimed to be dramatically improvedt? > >performance over previous VAX microprocessors and results inaH > >performance that approaches and may exceed the performance of popularI > >industry RISC microprocessors. This was accomplished by using pipelineuC > >techniques traditionally associated with reduced instruction setuH > >computer (RISC) CPU. There you have it, a VAX microprocessor that may3 > >have had the raw performance of some RISC chips.t > >nF > >The VAX processor was design with the idea of supporting programmedF > >languages, which includes branching, where as the Alpha wasn't. The> > >Alpha processor is a Load/Store RISC architecture. The RISCH > >architecture uses the compiler to optimize the program languages thusB > >branching. The raw power of the Alpha vs other chips was not asF > >stellar as one would think. The later versions of the Alpha systemsG > >performances were due to compiler optimization of the Alpha hardware H > >configuration. Why was the VAX replaced by the Alpha? It is easier toG > >design RISC architecture and write a compiler to optimize the use ofyH > >the hardware configuration than to design a CISC chip to do the same. > > % > >Here is my two cents that subject.a > >  > >Regards,a > >Daryl Jones > >o > >  mn > >"Richard B. Gilbert" <rgilbert88@comcast.net> wrote in message news:<uLednbnOh_U3GqzdRVn-jw@comcast.com>... > >  n > >eL > >>Yes, the VAX is byte addressable.  So, I believe, is the Alpha.  If you K > >>address your data as bytes there is no alignment problem!  If you want   > >>larger units, align them.h > >>J > >>The VAX architecture had a known penalty for misalignment.  The Alpha K > >>just has a larger penalty.  I think most RISC processors have a larger eG > >>penalty for misalignment than their CISC predecessors.  Should the  M > >>engineers have slowed down the processor and made it more expensive (for eL > >>everybody) just to reduce the penalties for misalignment?  They elected G > >>to have it lean, fast, and cheap (relatively).  For many years the  6 > >>Alphas were the fastest iron that money could buy! > >>F > >>The VMS Fortran compiler aligns the start of COMMON on a quadword K > >>boundary.    Perhaps you meant to say that the programmer neglected to aH > >>put the REAL*16 first, followed by the REAL*8, REAL|INTEGER*4, .....L > >>I followed that rule (descending order by length) for twenty-four years K > >>of programming in Fortran.   I haven't written any lately.  COMMON and eL > >>Structures (in later versions of Fortran) are about the only places you N > >>need to worry about alignment; the compiler takes care of everything else. > >> > >> > >>Tom Linden wrote:a > >> > >>     > >>  > >>> -----Original Message-----> > >>> From: Richard B. Gilbert [mailto:rgilbert88@comcast.net]- > >>> Sent: Sunday, February 15, 2004 4:05 PMe > >>> To: Info-VAX@Mvb.Saic.Com + > >>> Subject: Re: Why was VAX abandonned ?e > >>>w > >>>  > >>> Right on,  Antonio!o > >>>gK > >>> I first met the issue of boundary alignment ca. 1970 with the IBM 360eK > >>> Model 91.  It is just part of the way processors work.  The designers M > >>> can make varying amounts of effort to "fix up" misaligned variables and0L > >>> can charge greater or lesser penalties for them.  I would imagine thatM > >>> there are substantial costs involved both in engineering and in silicont0 > >>> to get maximum fixup with minimum penalty. > >>>>I > >>> Why should the rest of us pay extra so that incompetent programmerstJ > >>> don't have to think about alignment?  Especially since the compilersI > >>> will do most of the work unless you insist on a structure of {byte,aI > >>> word, longword, quadword} with no filler.  There is no way that thesK > >>> compiler can align that!  For those not following this, the structureiI > >>> aligns beautifully if you make it {quadword, longword, word, byte};rM > >>> start it on a quadword boundary and everything falls neatly into place.  > >>>uE > >>>Yes, but you don't have the freedom to align data except for newn > >>>applications.O > >>>I would venture to say that the bulk of FORTRAN employs COMMON that is notdP > >>>aligned.  So what does the bright cpu engineer do to overcome the alignment
 > >>>penalty,oO > >>>he increases the memory bandwidth and inserts a barrel shifter.  What does  > >>>theH > >>>engineer without proper oversight do, he ignores the problem as not > >>>important.  > >>>sN > >>>In PL/I you can give a structure the aligned attribute, but in FORTAN you > >>>can't;oA > >>>moreover a lot of I/O to structures presumes data is packed.c > >>> @ > >>>Last I looked, the VAX was a byte addressable architecture. > >>>a > >>> Antonio Carlini wrote: > >>>t > >>> >  > >>> >eN > >>> >> Note that the Power PC has only a one tick penalty for unaligned data > >>> >> accessoN > >>> >> (which the Alpha engineers arrogantly refer to as misaligned access!) > >>> >a > >>> >aM > >>> > Misaligned is misaligned. Most VAX processors also have penalties for-I > >>> > misaligned accesses. The compilers hide these, where possible andUL > >>> > allowed, by padding structures. MACRO-32 programmers are expected toG > >>> > be able to count! Plenty of other processors do the same thing.s > >>> >c > >>> > Antonios > >>> >u
 > >>> > -- > >>> >i > >>> > ---------------o5 > >>> > Antonio Carlini             arcarlini@iee.orgc > >>> >  > >>>a	 > >>> ---v, > >>> Incoming mail is certified Virus Free.@ > >>> Checked by AVG anti-virus system (http://www.grisoft.com).F > >>> Version: 6.0.566 / Virus Database: 357 - Release Date: 1/22/2004 > >>>> > >>>---+ > >>>Outgoing mail is certified Virus Free.>? > >>>Checked by AVG anti-virus system (http://www.grisoft.com).eE > >>>Version: 6.0.566 / Virus Database: 357 - Release Date: 1/22/2004> > >>>a > >>>  > >>>  > >>>      e > >>>o >  > --   ------------------------------    Date: 18 Feb 2004 07:46:03 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)e% Subject: Re: Why was VAX abandonned ?m3 Message-ID: <hi8Pe02UW0lD@eisner.encompasserve.org>n  y In article <c0u3dh$1cdhr4$1@ID-141708.news.uni-berlin.de>, "Peter Weaver" <WeaverConsultingServices@sympatico.ca> writes:g > < > You can do that now with the latest version of CHARON-VAX.  H    Nice.  But is there a hobbyist version again?  My Pico-VAX expired in
    September.>   ------------------------------    Date: 19 Feb 2004 07:37:34 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)a% Subject: Re: Why was VAX abandonned ?e3 Message-ID: <RWmxtfx7V95S@eisner.encompasserve.org>s  j In article <4mWYb.17330$wD5.1878@nwrddc03.gnilink.net>, "Stephen Eickhoff" <operagost@example.com> writes: > J > "Bob Koehler" <koehler@eisner.nospam.encompasserve.org> wrote in message/ > news:0JgXIW0mcmbs@eisner.encompasserve.org...t > C >>    On my DEC 3000 Model 600S there is no halt button.  ^P works.t > 9 > Check again. It's on the back, below the AUI connector.u  G    Oh, jeez.  It's been hidden under the 10BaseT transciever since day rI    one and I forgot it was there.  I don't think I've seen it since 1994.e   ------------------------------  % Date: Thu, 19 Feb 2004 12:58:46 +0100f- From: "Winfried Bergmann" <dummy@empuron.com>e@ Subject: Re: will these third-party CD-ROM drives work with VMS?: Message-ID: <c128dn$1dj2p7$1@ID-170759.news.uni-berlin.de>  L "Phillip Helbig---remove CLOTHES to reply" <helbig@astro.multiCLOTHESvax.de>5 schrieb im Newsbeitrag news:c110fk$3ai$1@online.de...a> > In article <c10tqq$ur$1@titan.btinternet.com>, "Jeff Humber"$ > <matrix01@globalnet.co.uk> writes: >vG > > I believe you need to configure the CD-ROM for 512 bytes to achieve  OpenVMS  > > compatibility. >b > Right. > G > > Check out the link below, it mentions the Plextor & Toshiba models.  > >i/ > > http://narpes.com/aw/vax/dec-cdrom-list.txts >>: > Thanks.  The link appears to be not working now, though. >aJ > I've gotten the Plextor and Toshiba to work.  Interestingly, they wantedA > to have the termination jumper set.  OK, makes sense.  With the B > cyberdrive, it showed up as 8*7 devices at the console prompt ifJ > terminated.  When not terminated, however, it showed up as only 8.  I'veI > seen this before, for example with an old EXABYTE tape drive.  It showsaJ > up as 7 devices in VMS as well, but only on ALPHA, not on VAX.  However,G > even on ALPHA, MKA500:, say, works (while MKA501--MKA507 appear to berJ > bogus devices).  Within VMS, the cyberdrive just shows up as one device,G > SHOW DEVICE shows its name etc, but when trying to MOUNT something, IsH > get a message saying that I should mount the volume etc and everything > hangs. >-? > OK, 2 out of 3 ain't bad (actually 2 out of 4, since I have 2i > Cyberdrives).p >eI > Looks like I'm out of look with the Cyberdrives, since there appears tosC > be no way to set them to 512.  (Interestingly, the other two work.B > WITHOUT the "block" jumper set.  Looks like 512 is the default.) >t  K Could be, a SCSI command to set the block size (MODE SENSE?) to 512 byte isi supported by those drives.   ------------------------------  % Date: Wed, 18 Feb 2004 11:42:09 +0000nO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com>s: Subject: Re: [OT] Solaris crashes itself when /tmp is full0 Message-ID: <c0vj2i$5eb$1@new-usenet.uk.sun.com>   Bob Koehler wrote: > In article <c0isv9$5h$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes:h >  > % >>Read the rest of the thread thread.  >  >  >    I did read the thread.  >  >    I ask again:  who?  >  > = >>So what you have is a problem that might of might not crash : >>a Solaris box with a similar problem that might or might >>not crash an OpenVMS box.  >  > F >    No.  You don't have a problem on VMS because an unprivileged userF >    can't do the DOS in the default configuration.  A system manager D >    has to set up a system specifically to get VMS into any sort ofH >    problem, that problem has to specifically be selected to be crash, C >    and then the user would have to work hard to get it to happen.lG >    It's always possible to misconfgure a system to open up an attack.y$ >    passwords in the computer case. >   > You seem to be at odds with some of your fellow OpenVMS admins; with respect to printing and C2 auditing. I will leave thate argument to you and them.     I >    There's nothing an unpriviledged user can do which will set this up.n >   C Except apparently print things and generate auditable events as fars= as the C2 audit module is concerned. Apart from that nothing.d  6 Its the old glass house stone scenario all over again.   Regards* Andrew Harrison    ------------------------------  + Date: Wed, 18 Feb 2004 12:30:41 +0000 (UTC)o From: david20@alpha2.mdx.ac.uk: Subject: Re: [OT] Solaris crashes itself when /tmp is full) Message-ID: <c0vlth$bg9$1@news.mdx.ac.uk>e   In article <c0vj2i$5eb$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: >Bob Koehler wrote:  >> In article <c0isv9$5h$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: >> s >>  & >>>Read the rest of the thread thread. >>   >> o >>    I did read the thread. >> e >>    I ask again:  who? >>   >> d> >>>So what you have is a problem that might of might not crash; >>>a Solaris box with a similar problem that might or mightc >>>not crash an OpenVMS box. >> b >>  G >>    No.  You don't have a problem on VMS because an unprivileged user*G >>    can't do the DOS in the default configuration.  A system manager 0E >>    has to set up a system specifically to get VMS into any sort ofiI >>    problem, that problem has to specifically be selected to be crash,  D >>    and then the user would have to work hard to get it to happen.H >>    It's always possible to misconfgure a system to open up an attack.% >>    passwords in the computer case.u >> o > ? >You seem to be at odds with some of your fellow OpenVMS adminsS< >with respect to printing and C2 auditing. I will leave that >argument to you and them. >a >nJ >>    There's nothing an unpriviledged user can do which will set this up. >> A > D >Except apparently print things and generate auditable events as far> >as the C2 audit module is concerned. Apart from that nothing. > 7 >Its the old glass house stone scenario all over again.h > " No its you getting it wrong again.O With the default settings the system will not crash if the system disk is full. I The one person who posted saying that the system would crash if the auditeG server ran out of space was wrong. The default setting is not to crash.e  J As for printing even if print files filled up the system disk it would notF crash because filling the system disk won't cause the system to crash.  K You accuse others of trolling. In general those you accuse are users and/oru7 supporters of VMS. You are neither a user or supporter.sL There are enough users and supporters of VMS who are able to provide balanceG on issues like IA64 without needing the support of someone whose agendaeC with respect to a bright future for VMS must seem a little suspect.C    + So please please stop trolling comp.os.vms.eB Last time I looked you seemed only to ever post to this newsgroup.H I know in this case you will argue that you were only defending Solaris.O But why do you care ? This is a VMS newsgroup. In other newsgroups dedicated toAC other systems you will find critism of Windows, Linux, Solaris etc  K It's the nature of such newsgroups. Your presence almost certainly leads togM more posting of critisms of Solaris in this newsgroup than would otherwise be 	 the case.d  mI Occasional posts from people outside the VMS community giving a different K viewpoint are generally welcomed. But this appears to be your fulltime job.n  
 David Webb VMS and Unix team leader CCSS Middlesex University     >Regards >Andrew Harrison >a   ------------------------------    Date: 18 Feb 2004 07:48:59 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)h: Subject: Re: [OT] Solaris crashes itself when /tmp is full3 Message-ID: <KEM+s93dH$f7@eisner.encompasserve.org>o   In article <c0vj2i$5eb$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: >> d > @ > You seem to be at odds with some of your fellow OpenVMS admins= > with respect to printing and C2 auditing. I will leave thato > argument to you and them.o  G    The C2 requirement is met by the option to stop processing, no crasht    required.  *    And nobody said a thing about printing.   ------------------------------  # Date: Wed, 18 Feb 2004 14:27:02 GMTr& From: jlsue <jefflsxxxz@sbcglobal.net>: Subject: Re: [OT] Solaris crashes itself when /tmp is full8 Message-ID: <62t6301que2eaasauch1lptqcan8in4mq8@4ax.com>  E On Wed, 18 Feb 2004 11:42:09 +0000, Andrew Harrison SUNUK Consultancy . <Andrew_No.Harrison_No@nospamn.sun.com> wrote:   >Bob Koehler wrote:a   >> .G >>    No.  You don't have a problem on VMS because an unprivileged user G >>    can't do the DOS in the default configuration.  A system manager :E >>    has to set up a system specifically to get VMS into any sort of I >>    problem, that problem has to specifically be selected to be crash, iD >>    and then the user would have to work hard to get it to happen.H >>    It's always possible to misconfgure a system to open up an attack.% >>    passwords in the computer case.o >>   >0? >You seem to be at odds with some of your fellow OpenVMS adminsi< >with respect to printing and C2 auditing. I will leave that >argument to you and them.  K You do realize the difference between a default system weakness or setting,eF and one that the business/sysadmin take for specific security reasons, right?  J The only way that filling the system disk will cause the crash issue is onH a system where the business had deemed security to be critical enough toJ prevent someone (including fully-privileged system managers) from covering
 their tracks.e  E But it takes a decisive, overt action to enable such security levels.    >i >CJ >>    There's nothing an unpriviledged user can do which will set this up. >> C >sD >Except apparently print things and generate auditable events as far> >as the C2 audit module is concerned. Apart from that nothing.  G Hopfully, you also realize that by doing the research to implement suchyI security measures that would cause the system to take more drastic actioni@ in certain situations, they would also learn how to relocate allG user-writable areas (e.g., print queues, TCP log files, etc) off of thedE system disk to prevent a typical DOS attack, and hiding their tracks.a  I And note, also, that the printing and audit issues you mention are mostlyeJ available to only AUTHORIZED system users - i.e., an inside job.  It's notK going to be available to someone who's just trying to crack into the system- or cause havoc on it.-  H And besides, anyone worried about C2 level of security would most likelyJ also be monitoring their systems closely, and be made aware of significant  audit messages when they happen.     >t7 >Its the old glass house stone scenario all over again.w  ; Only if you ignore the facts of the implementation details.p  K Do you really want to go head-to-head on the technical aspects of this party8 of VMS?  It's definitely not as you want to describe it.   --- jlsd0 The preceding message was personal opinion only.6 I do not speak in any authorized capacity for anyone,  and certainly not my employer.- (get rid of the xxxz in my address to e-mail)t   ------------------------------  % Date: Thu, 19 Feb 2004 16:02:41 +0000aO From: Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> : Subject: Re: [OT] Solaris crashes itself when /tmp is full0 Message-ID: <c12mn4$99t$1@new-usenet.uk.sun.com>   david20@alpha1.mdx.ac.uk wrote:1 > In article <c1056q$cg1$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes: > ! >>david20@alpha2.mdx.ac.uk wrote:U >> >>>In article <c0vqgd$803$1@new-usenet.uk.sun.com>, Andrew Harrison SUNUK Consultancy <Andrew_No.Harrison_No@nospamn.sun.com> writes:$ >>>@ >>>s# >>>>david20@alpha2.mdx.ac.uk wrote:: >>>> >>>>P >>>>>You accuse others of trolling. In general those you accuse are users and/or< >>>>>supporters of VMS. You are neither a user or supporter.Q >>>>>There are enough users and supporters of VMS who are able to provide balanceiL >>>>>on issues like IA64 without needing the support of someone whose agendaH >>>>>with respect to a bright future for VMS must seem a little suspect. >>>>>I >>>>@ >>>>What is the title of this thread ??????????????????????????? >>>>. >>>>Solaris crashes itself when /tmp is full ! >>>>A >>>>So now you explain to me why I should not respond, or perhaps6B >>>>explain why was this posted to this thread in the first place. >>>>D >>>>Last time I looked comp.os.vms wasn't a Solaris admin newsgroup. >>>>2 >>>>Sorry but that really wasn't one of your best. >>>>1 >>>>I have snipped the rest of your post for yourV. >>>>benefit, you can thank me at your leisure. >>>> >>>e >>>GL >>>As usual you snip to support your own position without regard to what was >>>actually posted.t >>>l >>>As I posted t >>>  >>>a >>>"K >>>I know in this case you will argue that you were only defending Solaris. R >>>But why do you care ? This is a VMS newsgroup. In other newsgroups dedicated toE >>>other systems you will find critism of Windows, Linux, Solaris etcnN >>>It's the nature of such newsgroups. Your presence almost certainly leads toP >>>more posting of critisms of Solaris in this newsgroup than would otherwise be >>>the case. >>>" >>C >>Chicken, egg. I am here because one Rob Young started an anti Suna> >>tirade to deflect criticism from the then owners of OpenVMS. >> >  > 4 > On 24/11/2003 in an article "RE: IA64 TPC results" > you said   > L > "Rob as you know perfectly well I started posting to this group because of: > your attempts to FUD Sun over ebays reliability issues."  P > That first post in 1997 seems to have been on the settlement between Intel andJ > DEC and you were responding to Terry Shannon. Rob doesn't appear to have$ > contributed to that thread at all. >   C And if you read the post you would find that it was crossposted andh= it doesn't refer to Alpha/Digital etc in any detrimental way.R  B Having a crossposted article turn up on this newsgoup which barely( refers to Alpha is hardly a crime is it.   RegardsL Andrew HarrisonL   ------------------------------  # Date: Thu, 19 Feb 2004 11:38:26 GMTm0 From: "Barry Kierstein" <Barry.Kierstein@HP.Com>V Subject: Re: [VMS V7.3-2/AMDS/AVAILMAN] AVAILMAN installation deletes AMDS$CONSOLE.EXE0 Message-ID: <Se1Zb.660$yF7.573@news.cpqcorp.net>   Bart,c  H     This is on the plate to do.  I definitely agree, and we see it every time we install also.,   Barry    ------------------------------   End of INFO-VAX 2004.099 ************************