1 INFO-VAX	Mon, 13 Sep 2004	Volume 2004 : Issue 509       Contents:+ Re: BACKUP is ignoring /SINCE=BACKUP.  Why?  DEC3000 boot problems  Re: DEC3000 boot problems  Re: DEC3000 boot problems 2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future.2 Re: I'm giving up computers if this is the future. Re: Impersonate  Re: Impersonate  Re: Impersonate  Re: Impersonate  Re: Impersonate  Re: Impersonate  Re: Impersonate (bit off topic) % Re: Itanium not meeting Intel's goals # Re: Marketing of tech into campuses # Re: Marketing of tech into campuses # Re: Marketing of tech into campuses  PLUG: txt2pdf 7.5 / Posting  ( WAS: why mainframes are still used?) - Re: Posting  - recommendation for News server - Re: Posting  - recommendation for News server $ Reminder: Datamation OpenVMS article Re: Suggestion for file system" Re: why mainframes are still used?" Re: why mainframes are still used?" Re: why mainframes are still used?7 Re: why mainframes are still used? - printer-made music ( Windoze reboots turn into disk rebuilds!, Re: Windoze reboots turn into disk rebuilds!  F ----------------------------------------------------------------------    Date: 13 Sep 2004 07:05:17 -0700. From: spamsink2001@yahoo.com (Alan E. Feldman)4 Subject: Re: BACKUP is ignoring /SINCE=BACKUP.  Why?= Message-ID: <b096a4ee.0409130605.5141942a@posting.google.com>   R John Santos <john@egh.com> wrote in message news:<qPv0d.3252$iS2.1391@trnddc09>... > Tom Adams wrote:O > > I have one machine where the incremental backup is backing up every file on  > > the machine. > > P > > I have 5 Alphas and they are all running the same incremental backup commandM > > procedure.   The others are working normally, but this one is filling the   > > backup tape every other day. > > A > > Does anyone have any idea what might be wrong on this system?  > 7 > What version of VMS and what ECO's have been applied?  > E > The behaviour has changed several times over the last few years.  I E > think it currently once again "does the right thing".  For a while, I > any change to a directory, not including creating or deleting any files G > in it, but including any change in its size (which could be triggered B > randomly by creating or deleting files in it), caused the entire@ > directory and all its subdirectories to be backed up during anE > incremental backup.  If you created a new top-level directory, then % > the entire disk could be backed up.  > D > I think the XQP was changed to no longer touch the "modified" dateD > on a directory except if the directory metadata (protection, ACLs,H > ownership, etc.) were changed or if it was renamed.  (I think renamingE > directories or deleting and creating directories with the same name D > could trick BACKUP into missing changes under the original scheme,C > and the quick fix was to backup EVERYTHING if the "modified" date B > on the parent directory was more recent than the "backup" date.)  E IIRC, what would happen is this: If a .DIR file were to be renamed or F moved in the middle of a sequence of incremental backups; then, duringF a restore operation, files saved under the old directory name would beC restored under that name, even if they were later deleted under the E new directory name. Files saved under the new directory name would be B restored under the new directory name. (The VMS v5.5 Backup manualF acknowledges this and says the system manager must do a "manual merge"D of the new and old directories.) As for directories that are deletedB and recreated, I don't think that would be a problem under the old? scheme as the latest incremental save would contain the correct A directory entires and time stamps for that directory-spec. I also D recall CJL posting an example using SET FILE/ENTER that would not be$ handled correctly by the old scheme.  A I don't know what the latest behavior is, but it seems to me that D BACKUP should unconditionally back up entire subdirectory trees ONLY& if the directory has renamed or moved.   > E > It was around about V7.1-V7.2 that the behaviour changed.  A Google 6 > search should reveal hundreds of posts on the topic. >  > HTH    ------------------------------  # Date: Mon, 13 Sep 2004 16:38:15 GMT ' From: "Mike E." <m_esso@@@@hotmail.com>  Subject: DEC3000 boot problems, Message-ID: <X1k1d.417928$gE.84344@pd7tw3no>  = Hi all, I'm running, ...was running OpenVMS 7.1 on a DEC3000.   M Apparently some software was installed on the weekend (Fortran, I think) and  2 ever since then the machine won't boot at startup.- The error message that I get at boot time is: 
 ?02 KSP INVAL .   PC= 00000000.00000000 PSL= 00000000.00001F00  M I'm not sure what would help determine if this is a HW failure or some other  M problem, but I'm thinking that either the bootblock is corrupt or the hdd is  K failing (though it still sounds normal during operation, and the disks are  K spinning).  I am able to boot to CD,  but I can't mount the hdd to perform  M an ANALYZE/DISK_STRUCTURE or a WRITEBOOT.  When I attempt to mount DKA0, I'm   getting this message:   8 %MOUNT-I-OPRQST, device _DKA0: contains the wrong volume8 %MOUNT-I-NOOPR, no operator available to service request: %MOUNT-I-OPRQST, Please mount volume RAMA in device _DKA0:8 %MOUNT-I-NOOPR, no operator available to service request      9 All ideas welcome, I am a total VMS newb, so be gentle :)   8 --------------------------------------------------------2 At the chiv. prompt this is what things look like:8 -------------------------------------------------------- >>> test
 INIT-S-CPU...  INIT-S-RESET_TC... INIT-S-ASIC...
 INIT-S-MEM... 
 INIT-S-NVR... 
 INIT-S-SCC...  INIT-S-NI... INIT-S-SCSI... INIT-S-ISDN... T-STS-ASIC - OK  T-STS-MEM - OK T-STS-NVR - OK T-STS-SCC - OK
 T-STS-NI - OK  T-STS-SCSI A - OK  T-STS-SCSI B - OK  T-STS-ISDN - OK      >>> boot
 INIT-S-CPU...  INIT-S-RESET_TC... INIT-S-ASIC...
 INIT-S-MEM... 
 INIT-S-NVR... 
 INIT-S-SCC...  INIT-S-NI... INIT-S-SCSI... INIT-S-ISDN... AUDIT_BOOT_STARTS ...  AUDIT_CHECKSUM_GOOD  AUDIT_LOAD_BEGINS  AUDIT_LOAD_DONE   
 ?02 KSP INVAL /   PC= 00000000.00000000 PSL= 00000000.00001F00     ------------------------------  # Date: Mon, 13 Sep 2004 17:17:35 GMT % From: Rob Brown <mylastname@gmcl.com> " Subject: Re: DEC3000 boot problemsE Message-ID: <Pine.LNX.4.61.0409131107300.20414@localhost.localdomain>   $ On Mon, 13 Sep 2004, it was written:  D > I'm not sure what would help determine if this is a HW failure or D > some other problem, but I'm thinking that either the bootblock is F > corrupt or the hdd is failing (though it still sounds normal during D > operation, and the disks are spinning).  I am able to boot to CD, F > but I can't mount the hdd to perform an ANALYZE/DISK_STRUCTURE or a E > WRITEBOOT.  When I attempt to mount DKA0, I'm getting this message:  > : > %MOUNT-I-OPRQST, device _DKA0: contains the wrong volume: > %MOUNT-I-NOOPR, no operator available to service request< > %MOUNT-I-OPRQST, Please mount volume RAMA in device _DKA0:: > %MOUNT-I-NOOPR, no operator available to service request  A To mount the disk without paying attention to the file structure:       $ MOUNT/FOREIGN DKA0    Then you can do things like:  '    $ DUMP/BLOCK=(START=0,COUNT=2) DKA0:   ! To see if you can read the drive.   = To mount the disk when you don't know the name of the volume:       $ MOUNT/OVERRIDE=ID DKA0:  D Then you can use the usual DCL commands to examine the drive.  As a C bonus, the mount command tells you the name of the volume that was   mounted.   hth      --    B Rob Brown                        b r o w n a t g m c l d o t c o mA G. Michaels Consulting Ltd.      (866)438-2101 (voice) toll free! 6 Edmonton                         (780)438-9343 (voice)5                                   (780)437-3367 (FAX) 2                                   http://gmcl.com/   ------------------------------    Date: 13 Sep 2004 12:22:52 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) " Subject: Re: DEC3000 boot problems3 Message-ID: <3EQ+bzYOHWQs@eisner.encompasserve.org>   V In article <X1k1d.417928$gE.84344@pd7tw3no>, "Mike E." <m_esso@@@@hotmail.com> writes: > : > %MOUNT-I-OPRQST, device _DKA0: contains the wrong volume: > %MOUNT-I-NOOPR, no operator available to service request< > %MOUNT-I-OPRQST, Please mount volume RAMA in device _DKA0:: > %MOUNT-I-NOOPR, no operator available to service request      Boot from CD, then:  %    $mount/noassist/over=id dka0: rama   F    That should get you to where you can do analize/disk and writeboot.  D    Meanwhile I think you can turn on boot verboseness either throughE    a console (try >>>HELP before you boot) parameter or SYSBOOT> (via 7    >>> boot -fl=0,1)  so you can get a better idea what ?    program is clearing the PC (usually via trashing the stack). B    Look at sys$system:systartup.com on another system to see what A    options p1, p2, ... provide, they're settable via STARTUP_P1,      STARTUP_P2, ... in SYSBOOT>  E    You can find more on this stuff in the System Manager's manual and     the VMS Installation Guide.  -    It might be possible to do a minimum boot.    ------------------------------  % Date: Mon, 13 Sep 2004 08:02:45 +0200 * From: Paul Sture <nospam@sture.homeip.net>; Subject: Re: I'm giving up computers if this is the future. * Message-ID: <2qkrg9Fvnqh5U1@uni-berlin.de>   Wayne Sewell wrote:  >>From: Z <z@no.spam>  >>X-Newsgroups: comp.os.vms = >>Subject: Re: I'm giving up computers if this is the future. ' >>Date: Sun, 12 Sep 2004 11:55:21 -0700 > >>Organization: Posted via Supernews, http://www.supernews.com2 >>Message-ID: <10k96sddvgnde31@corp.supernews.com> >>" >>VAXman- @SendSpamHere.ORG wrote: >>F >>>I would *never* use a PeeCee for terminal emulation.  Hell, I wouldG >>>never use a PeeCee for anything other than to emulate this customer  G >>>environment.  I have ample real VT terminals, DECWindows on monitors G >>>and a Poerbook which, out of the box, does a fine VT emulation using G >>>a keyboard with has an alternate keypad which does not have the [,]   >>>and [-] keys fused into one.  >>F >>A PC used as a terminal emulator is much better than the real thing H >>when you need to capture output, like from a SHOW CONFIG, for example. >  > 1 > As can a mac emulator, with far fewer screwups.  >   ? Thinking of moderns Macs with no serial port,  do you have any  B recommendations for USB to serial converters which "get it right"?   ------------------------------  % Date: Mon, 13 Sep 2004 06:13:32 +0000 7 From: David B Sneddon - bigpond <dbsneddon@bigpond.com> ; Subject: Re: I'm giving up computers if this is the future. * Message-ID: <41453A8C.8010801@bigpond.com>    Paul Sture mentioned in passing: > A > Thinking of moderns Macs with no serial port,  do you have any  D > recommendations for USB to serial converters which "get it right"? >   B I use a Dolphin USB to Serial Adapter and haven't had any problemsA with it.  Using that and a collection of gender benders and 9 pin F to 25 pin converters and cables I can connect it to any serial device.   Regards, Dave.  --  I David B Sneddon (dbs)    VMS Systems Programmer     dbsneddon@bigpond.com I Sneddo's quick guide ...          http://www.users.bigpond.com/dbsneddon/ I DBS freeware at ...   http://www.users.bigpond.com/dbsneddon/software.htm I "Life is what happens to you while you're busy making other plans" Lennon    ------------------------------  % Date: Mon, 13 Sep 2004 09:55:48 +0100 & From: Elliott Roper <nospam@yrl.co.uk>; Subject: Re: I'm giving up computers if this is the future. 1 Message-ID: <130920040955484579%nospam@yrl.co.uk>   5 In article <2qkrg9Fvnqh5U1@uni-berlin.de>, Paul Sture   <nospam@sture.homeip.net> wrote:   > Wayne Sewell wrote:  <snip>H > >>A PC used as a terminal emulator is much better than the real thing J > >>when you need to capture output, like from a SHOW CONFIG, for example. > >  > > 3 > > As can a mac emulator, with far fewer screwups.  > >  > A > Thinking of moderns Macs with no serial port,  do you have any  D > recommendations for USB to serial converters which "get it right"?  G Keyspan USA18 should work OK. I have not used mine in that application, ' but it should be a good 'un with Zterm. G If it is more than an idle query, ask again and I'll try it for real. I E have not used Zterm for untold years. I note there is an OS X version E with a sly hack for sending breaks - drop it to 50 bit/sec and send a  null!    --  A I thought I would be the last on earth to mung my e-mail address.  fsnospam$elliott$$   ------------------------------  % Date: Mon, 13 Sep 2004 10:15:36 +0000 - From: David B Sneddon <dbsneddon@bigpond.com> ; Subject: Re: I'm giving up computers if this is the future. ( Message-ID: <41457348.70605@bigpond.com>  / David B Sneddon - bigpond was overheard to say: " > Paul Sture mentioned in passing: >  >>B >> Thinking of moderns Macs with no serial port,  do you have any E >> recommendations for USB to serial converters which "get it right"?  >> > D > I use a Dolphin USB to Serial Adapter and haven't had any problemsC > with it.  Using that and a collection of gender benders and 9 pin H > to 25 pin converters and cables I can connect it to any serial device.  F It is a Dolphin USB9103 and I have used it to connect to VAX and Alpha@ serial ports, HSZ70 console ports and DECservers using C-Kermit.   Regards, Dave.  --  I David B Sneddon (dbs)    VMS Systems Programmer     dbsneddon@bigpond.com I Sneddo's quick guide ...          http://www.users.bigpond.com/dbsneddon/ I DBS freeware at ...   http://www.users.bigpond.com/dbsneddon/software.htm I "Life is what happens to you while you're busy making other plans" Lennon    ------------------------------  # Date: Mon, 13 Sep 2004 10:48:13 GMT " From:   VAXman-  @SendSpamHere.ORG; Subject: Re: I'm giving up computers if this is the future. 0 Message-ID: <00A37CE6.21DF14E0@SendSpamHere.ORG>  u In article <4144f83d$0$22747$db0fefd9@news.zen.co.uk>, "Alex Daniels" <AlexNOSPAMTHANKSDaniels@themail.co.uk> writes: . ><VAXman- @SendSpamHere.ORG> wrote in message + >news:00A37C90.9F01A631@SendSpamHere.ORG... 9 >> In article <4144BA8F.D8E87E05@teksavvy.com>, JF Mezei  ) >> <jfmezei.spamnot@teksavvy.com> writes: $ >>>VAXman-, @SendSpamHere.ORG wrote:I >>>> is available, DECterm pisses all over the PeeCee terminal emulator's  >>>> cornflakes. >>> L >>>DECterm misses the ability to display control codes, save images as .jpg  >>>or J >>>other more current formats etc. It is a real shame that VMS management  >>>have 9 >>>decided to stop developping  GUI applications for VMS.  >> >>% >> What VT terminal displayed images?  >> ><SNIP>  > M >That would be the VT1200 (b/w) or the VT1300 (colour). Also the VXT (albeit  & >not quite the VT prefix you specify). >  >VT1300 N >Cpu: KA42-A, CVAX chip @ 90ns, 64KB external cache Memory: 4MB on cpu board. , >4, (12, or 16MB?) boards to max 8MB (32MB?)  2 In which case, DECWindows and I can save an image.   --  < http://www.ProvN.com  for the *best* OpenVMS system security=                       solutions that others only claim to be.  --  , Cyber-Terrorism (si'-ber tayr'-or-iz-em) n.:M   The release of, the sale of, or the use of any Micro$oft software product!   --  K VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM    ------------------------------  # Date: Mon, 13 Sep 2004 10:49:31 GMT " From:   VAXman-  @SendSpamHere.ORG; Subject: Re: I'm giving up computers if this is the future. 0 Message-ID: <00A37CE6.500703DE@SendSpamHere.ORG>  Z In article <60t2ic.lh1.ln@dadsys2.fuller.local>, Stuart Fuller <stufuller@usa.net> writes:$ >  VAXman-  @SendSpamHere.ORG wrote: > I >> In article <10k9n5p9hmr48e3@corp.supernews.com>, Z <z@no.spam> writes: # >>>VAXman- @SendSpamHere.ORG wrote: H >>>>>A PC used as a terminal emulator is much better than the real thingK >>>>>when you need to capture output, like from a SHOW CONFIG, for example.  >>> J >>>> Bullshit!  Use ALPHA_LOGGER (free utility you can find in SYS$EXAMPLEJ >>>> on an Alpha) or Raxco's CarbonCopy on a real terminal.  If DECWindowsI >>>> is available, DECterm pisses all over the PeeCee terminal emulator's  >>>> cornflakes. >>> 4 >>>DECWindows isn't available at the SRM >>> prompt. >>  . >> Huh?  What's the point your trying to make? >>     > G >As a Field Service person, I frequently need to do a SHOW CONFIG at a  J >console prompt, record the information, replace the <whatever stores the L >config>, restore the information.  Using my laptop, with whatever terminal K >emulator, as a console device to save the configuration information saves  C >me a lot of time and effort, vs. writing all the information down.   A ...and this couldn't be done using a system running DECwindows?      --  < http://www.ProvN.com  for the *best* OpenVMS system security=                       solutions that others only claim to be.  --  , Cyber-Terrorism (si'-ber tayr'-or-iz-em) n.:M   The release of, the sale of, or the use of any Micro$oft software product!   --  K VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM    ------------------------------  # Date: Mon, 13 Sep 2004 10:53:33 GMT " From:   VAXman-  @SendSpamHere.ORG; Subject: Re: I'm giving up computers if this is the future. 0 Message-ID: <00A37CE6.E05C0827@SendSpamHere.ORG>  \ In article <41452FD7.48A01B67@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes:' >> > What VT terminal displayed images?  > ! >VT240, 330, 340 and some others.   5 Not JPGs which I believe the original post mentioned.   < Do these terminal emulators support Regis or sixel graphics? --  < http://www.ProvN.com  for the *best* OpenVMS system security=                       solutions that others only claim to be.  --  , Cyber-Terrorism (si'-ber tayr'-or-iz-em) n.:M   The release of, the sale of, or the use of any Micro$oft software product!   --  K VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM    ------------------------------  # Date: Mon, 13 Sep 2004 10:51:00 GMT " From:   VAXman-  @SendSpamHere.ORG; Subject: Re: I'm giving up computers if this is the future. 0 Message-ID: <00A37CE6.856ADDDA@SendSpamHere.ORG>  F In article <10ka0hg8s0c5cd5@corp.supernews.com>, Z <z@no.spam> writes: >Stuart Fuller wrote: I >>>>>>A PC used as a terminal emulator is much better than the real thing L >>>>>>when you need to capture output, like from a SHOW CONFIG, for example. >>>>J >>>>>Bullshit!  Use ALPHA_LOGGER (free utility you can find in SYS$EXAMPLEJ >>>>>on an Alpha) or Raxco's CarbonCopy on a real terminal.  If DECWindowsI >>>>>is available, DECterm pisses all over the PeeCee terminal emulator's  >>>>>cornflakes. >>>>5 >>>>DECWindows isn't available at the SRM >>> prompt.  >>> . >>>Huh?  What's the point your trying to make? >>>    >>   >>  I >> As a Field Service person, I frequently need to do a SHOW CONFIG at a  L >> console prompt, record the information, replace the <whatever stores the N >> config>, restore the information.  Using my laptop, with whatever terminal M >> emulator, as a console device to save the configuration information saves  E >> me a lot of time and effort, vs. writing all the information down.  > 	 >Exactly.   H Exactly what?  You mean to tell me that ONLY a PeeCee running a terminalH emulator is the only way to do this?  You're not worthy to be use VMS if that is the case.    --  < http://www.ProvN.com  for the *best* OpenVMS system security=                       solutions that others only claim to be.  --  , Cyber-Terrorism (si'-ber tayr'-or-iz-em) n.:M   The release of, the sale of, or the use of any Micro$oft software product!   --  K VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM    ------------------------------  % Date: Mon, 13 Sep 2004 08:11:50 -0400 ' From: Stuart Fuller <stufuller@usa.net> ; Subject: Re: I'm giving up computers if this is the future. 0 Message-ID: <6q24ic.6ml.ln@dadsys2.fuller.local>  #   VAXman-  @SendSpamHere.ORG wrote:   @ > In article <60t2ic.lh1.ln@dadsys2.fuller.local>, Stuart Fuller > <stufuller@usa.net> writes: % >>  VAXman-  @SendSpamHere.ORG wrote:  >>J >>> In article <10k9n5p9hmr48e3@corp.supernews.com>, Z <z@no.spam> writes:$ >>>>VAXman- @SendSpamHere.ORG wrote:I >>>>>>A PC used as a terminal emulator is much better than the real thing L >>>>>>when you need to capture output, like from a SHOW CONFIG, for example. >>>>K >>>>> Bullshit!  Use ALPHA_LOGGER (free utility you can find in SYS$EXAMPLE K >>>>> on an Alpha) or Raxco's CarbonCopy on a real terminal.  If DECWindows J >>>>> is available, DECterm pisses all over the PeeCee terminal emulator's >>>>> cornflakes.  >>>>5 >>>>DECWindows isn't available at the SRM >>> prompt.  >>> / >>> Huh?  What's the point your trying to make?  >>>    >>G >>As a Field Service person, I frequently need to do a SHOW CONFIG at a J >>console prompt, record the information, replace the <whatever stores theL >>config>, restore the information.  Using my laptop, with whatever terminalK >>emulator, as a console device to save the configuration information saves D >>me a lot of time and effort, vs. writing all the information down. > A > ...and this couldn't be done using a system running DECwindows?  >   I Not if that system is the only system in the computer room, and the only  + thing it's running is the console firmware.  --             Stu    ------------------------------   Date: 13 Sep 2004 13:06:55 GMT( From: bill@cs.uofs.edu (Bill Gunshannon); Subject: Re: I'm giving up computers if this is the future. + Message-ID: <2qlkbfF10depaU2@uni-berlin.de>   0 In article <6q24ic.6ml.ln@dadsys2.fuller.local>,* 	Stuart Fuller <stufuller@usa.net> writes:% >   VAXman-  @SendSpamHere.ORG wrote:  >>  B >> ...and this couldn't be done using a system running DECwindows? >>   > K > Not if that system is the only system in the computer room, and the only  - > thing it's running is the console firmware.   J You misunderstood.  Instead of carrying a laptop with kermit with you whenG you make service calls you are supposed to carry an Alpha with keyboard  mouse and monitor.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------    Date: 13 Sep 2004 07:54:48 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) ; Subject: Re: I'm giving up computers if this is the future. 3 Message-ID: <uf2FIUqVE69I@eisner.encompasserve.org>   F In article <10k96sddvgnde31@corp.supernews.com>, Z <z@no.spam> writes:  F > A PC used as a terminal emulator is much better than the real thing H > when you need to capture output, like from a SHOW CONFIG, for example.  9   You never used the printer port on you VT to autoprint?    ------------------------------    Date: 13 Sep 2004 07:56:13 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) ; Subject: Re: I'm giving up computers if this is the future. 3 Message-ID: <ZD2uCpXNlTXk@eisner.encompasserve.org>   U In article <00A37C90.9F01A631@SendSpamHere.ORG>,   VAXman-  @SendSpamHere.ORG writes:  > $ > What VT terminal displayed images? >   D    Several with REGIS and/or SIXEL capability, including the VT420 I    have at home.   ------------------------------    Date: 13 Sep 2004 07:57:48 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) ; Subject: Re: I'm giving up computers if this is the future. 3 Message-ID: <F7ayOPz$Czgh@eisner.encompasserve.org>   Z In article <60t2ic.lh1.ln@dadsys2.fuller.local>, Stuart Fuller <stufuller@usa.net> writes: > H > As a Field Service person, I frequently need to do a SHOW CONFIG at a K > console prompt, record the information, replace the <whatever stores the  M > config>, restore the information.  Using my laptop, with whatever terminal  L > emulator, as a console device to save the configuration information saves D > me a lot of time and effort, vs. writing all the information down.  D    Not a problem if you have a proper LA120 console.  Or a proper VT(    console with a LA75 printer attached.   ------------------------------  % Date: Mon, 13 Sep 2004 09:18:38 -0400 # From: "John Smith" <a@nonymous.com> ; Subject: Re: I'm giving up computers if this is the future. , Message-ID: <0fednV_q8f-sA9jcRVn-gg@igs.net>   Bob Koehler wrote:H > In article <10k96sddvgnde31@corp.supernews.com>, Z <z@no.spam> writes: > F >> A PC used as a terminal emulator is much better than the real thing@ >> when you need to capture output, like from a SHOW CONFIG, for >> example.  > ; >   You never used the printer port on you VT to autoprint?     E I have an LA-12C (portable printing terminal with serial and 300 baud ( acoustic coupler) you could use instead.   ------------------------------    Date: 13 Sep 2004 08:14:55 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) ; Subject: Re: I'm giving up computers if this is the future. 3 Message-ID: <rMycGC6+RAEB@eisner.encompasserve.org>   Z In article <6q24ic.6ml.ln@dadsys2.fuller.local>, Stuart Fuller <stufuller@usa.net> writes: > K > Not if that system is the only system in the computer room, and the only  - > thing it's running is the console firmware.   D    Then blame DEC/Compaq/HP for letting you carry a PC into the room"    instead of a proper VMS laptop.   ------------------------------   Date: 13 Sep 2004 13:43:07 GMT( From: bill@cs.uofs.edu (Bill Gunshannon); Subject: Re: I'm giving up computers if this is the future. + Message-ID: <2qlmfbF115feaU1@uni-berlin.de>   , In article <0fednV_q8f-sA9jcRVn-gg@igs.net>,& 	"John Smith" <a@nonymous.com> writes: > Bob Koehler wrote:I >> In article <10k96sddvgnde31@corp.supernews.com>, Z <z@no.spam> writes:  >>G >>> A PC used as a terminal emulator is much better than the real thing A >>> when you need to capture output, like from a SHOW CONFIG, for  >>> example. >>< >>   You never used the printer port on you VT to autoprint? >  > G > I have an LA-12C (portable printing terminal with serial and 300 baud * > acoustic coupler) you could use instead.   + And I thought my Portacom-110 was bad.  :-)    bill      --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Mon, 13 Sep 2004 10:53:56 -0400 ( From: David Froble <davef@tsoft-inc.com>; Subject: Re: I'm giving up computers if this is the future. , Message-ID: <4145B484.6010605@tsoft-inc.com>   Mike Bartman wrote:     E > Yep.  A fault is a fault.  DEC usually did better.  I don't believe D > other VAXstations from that period had similar problems...the 4000C > series for instance...but I may be mistaken about that.  If I am, , > someone here will surely point it out. ;-)    L Ok, since you asked for it.  The machine hasn't been powered up for several M years, but I seem to remember that a 4 GB disk on a VAXStation 4000 model 60  N would not show it's size correctly.  The actual size disk may differ, but the O console program (BIOS) did have limitations.  Once VMS was running, it handled   the disk fine.  M Don't know if using a large disk as the system disk would cause problems.  I  Q believe in keeping the OS seperate and use RZ26 drives for the system disk on my   VAXStations.   Dave     --  4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com 170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  % Date: Mon, 13 Sep 2004 11:16:31 -0400 + From: Lord Isildur <isildur@andrew.cmu.edu> ; Subject: Re: I'm giving up computers if this is the future. H Message-ID: <Pine.LNX.4.60-041.0409131114040.2624@unix44.andrew.cmu.edu>   VAXstation 3100: 1988                      ^ + PeeCee with 1024 cylinders limitation: 1998 +                                           ^ L And that is just a minor, tiny, small, miniscule example of stupidity in theM architecture and implementation. PC's are toy computers, being forced against K their nature to try to do real work because real computers got (relatively) 9 too expensive for shortsighted decision-makers to accept.    Isildur      On Sun, 12 Sep 2004, Z wrote:    > Mike Bartman wrote: M >>> Didn't VAXstations suffer this same problem? I seem to remember a 2GB or  H >>> 1GB disk size limit when I rebuilt a 3100 (?) from spare parts last 	 >>> year.  > H >> Yes, the 3100 machines have a 1.2gig limit for the boot device.  OnceG >> VMS is up, the limit is gone for data disks, but you can't boot from   >> anything bigger than 1.2gigs. > L > Yet you fault PCs from the same time period for the same basic limitation? >    ------------------------------  % Date: Mon, 13 Sep 2004 11:12:17 -0400 + From: Lord Isildur <isildur@andrew.cmu.edu> ; Subject: Re: I'm giving up computers if this is the future. H Message-ID: <Pine.LNX.4.60-041.0409131108540.2624@unix44.andrew.cmu.edu>  I the vaxstations themselves don't have a problem, but the routines in the  E monitor to do disk i/o in VS3100's could only understand 31-bit byte  L offsets, and thus whatever was being booted had to be in the first 2 gigs ofL a disk. With ultrix or BSD or mach or whatnot, that wasn't an issue, but VMSI didn't (doesn't? i'm not too current with vms these days) do partitioning F and to be safe, the boot device should be <2g in size on one of them. # only the VS3100's had this problem.    isildur      On Sat, 11 Sep 2004, Z wrote: O > Didn't VAXstations suffer this same problem? I seem to remember a 2GB or 1GB  G > disk size limit when I rebuilt a 3100 (?) from spare parts last year.    ------------------------------  % Date: Mon, 13 Sep 2004 10:58:58 -0400 ( From: David Froble <davef@tsoft-inc.com>; Subject: Re: I'm giving up computers if this is the future. , Message-ID: <4145B5B2.8040109@tsoft-inc.com>   Bob Koehler wrote:  H > In article <10k96sddvgnde31@corp.supernews.com>, Z <z@no.spam> writes: >  > F >>A PC used as a terminal emulator is much better than the real thing H >>when you need to capture output, like from a SHOW CONFIG, for example. >> > ; >   You never used the printer port on you VT to autoprint?  >  >   O That requires a printer, if you want hardcopy, which may not be desired.  Hard  A to cut-n-paste hardcopy.  Scissors, paste, scanner, not so great.   P If you're going to feed another system from the printer port, why put the VT in  the middle?  Just go direct.   Dave   --  4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com 170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------  % Date: Mon, 13 Sep 2004 17:40:50 +0100 - From: John Laird <nospam@laird-towers.org.uk> ; Subject: Re: I'm giving up computers if this is the future. 8 Message-ID: <3vibk090ahd9mfk6eisjh6meqrlbufsfeq@4ax.com>  I On Mon, 13 Sep 2004 11:12:17 -0400, Lord Isildur <isildur@andrew.cmu.edu>  wrote:  J >the vaxstations themselves don't have a problem, but the routines in the F >monitor to do disk i/o in VS3100's could only understand 31-bit byte M >offsets, and thus whatever was being booted had to be in the first 2 gigs ofIM >a disk. With ultrix or BSD or mach or whatnot, that wasn't an issue, but VMS3J >didn't (doesn't? i'm not too current with vms these days) do partitioningG >and to be safe, the boot device should be <2g in size on one of them. r$ >only the VS3100's had this problem.  I It's a 1Gb limit, caused by the use of some SCSI commands that could onlysK address blocks up to that point.  Console roms were changed to a newer SCSIi command option in later models.t  E Although, as you say, VMS does not understand partitions, it is quiteuD feasible to set up a system disk with anything beyond the 1Gb beyondI assigned to a "container" file which can be served as a disk by a utility'H like LDDRIVER (available as Freeware).  Or you can be brave and trust toK luck - by placing the index file at the beginning of the volume with preseteK space for headers set sufficiently large, an installation or backup restore K of a VMS system will locate all files more than sufficiently close to blockeJ 1 to boot successfully, or at least boot far enough to get proper VMS diskE drivers installed.  As I understand it, the real *risk* is the systemtI dumpfile, which is mapped at a console level - if you let this go outside F the first 1Gb, it really will smoke your data files when/if the system crashes.   --  ; Never itch for anything you aren't willing to scratch for. t   Mail john rather than nospam...t   ------------------------------  # Date: Mon, 13 Sep 2004 17:03:05 GMT " From:   VAXman-  @SendSpamHere.ORG; Subject: Re: I'm giving up computers if this is the future.e0 Message-ID: <00A37D1A.7FF18445@SendSpamHere.ORG>  Z In article <6q24ic.6ml.ln@dadsys2.fuller.local>, Stuart Fuller <stufuller@usa.net> writes:$ >  VAXman-  @SendSpamHere.ORG wrote: >rA >> In article <60t2ic.lh1.ln@dadsys2.fuller.local>, Stuart Fullera >> <stufuller@usa.net> writes:& >>>  VAXman-  @SendSpamHere.ORG wrote: >>>eK >>>> In article <10k9n5p9hmr48e3@corp.supernews.com>, Z <z@no.spam> writes:l% >>>>>VAXman- @SendSpamHere.ORG wrote:eJ >>>>>>>A PC used as a terminal emulator is much better than the real thingM >>>>>>>when you need to capture output, like from a SHOW CONFIG, for example.- >>>>>-L >>>>>> Bullshit!  Use ALPHA_LOGGER (free utility you can find in SYS$EXAMPLEL >>>>>> on an Alpha) or Raxco's CarbonCopy on a real terminal.  If DECWindowsK >>>>>> is available, DECterm pisses all over the PeeCee terminal emulator'sS >>>>>> cornflakes. >>>>>i6 >>>>>DECWindows isn't available at the SRM >>> prompt. >>>> r0 >>>> Huh?  What's the point your trying to make? >>>>   s >>>iH >>>As a Field Service person, I frequently need to do a SHOW CONFIG at aK >>>console prompt, record the information, replace the <whatever stores thehM >>>config>, restore the information.  Using my laptop, with whatever terminaltL >>>emulator, as a console device to save the configuration information savesE >>>me a lot of time and effort, vs. writing all the information down.  >>  B >> ...and this couldn't be done using a system running DECwindows? >>   >cJ >Not if that system is the only system in the computer room, and the only , >thing it's running is the console firmware.  G If that system is the only system in the computer room then he would be # "writing all the information down".    -- g< http://www.ProvN.com  for the *best* OpenVMS system security=                       solutions that others only claim to be.  -- r, Cyber-Terrorism (si'-ber tayr'-or-iz-em) n.:M   The release of, the sale of, or the use of any Micro$oft software product! r -- hK VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COMe   ------------------------------    Date: 13 Sep 2004 12:13:48 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)c; Subject: Re: I'm giving up computers if this is the future. 3 Message-ID: <DKcS+3rsgQd+@eisner.encompasserve.org>i  W In article <4145B5B2.8040109@tsoft-inc.com>, David Froble <davef@tsoft-inc.com> writes:n > R > If you're going to feed another system from the printer port, why put the VT in  > the middle?  Just go direct.  G    If you ran out of ribbons for your LA120 you may find that it's hardf?    to enter information on a serial printer without a keyboard.t   ------------------------------  # Date: Mon, 13 Sep 2004 17:35:47 GMT 3 From: hammond@not@peek.ssr.hp.com (Charlie Hammond)e; Subject: Re: I'm giving up computers if this is the future.u3 Message-ID: <TTk1d.10470$aH2.7580@news.cpqcorp.net>t  I >> In article <10k96sddvgnde31@corp.supernews.com>, Z <z@no.spam> writes:t >> dG >>>A PC used as a terminal emulator is much better than the real thing dI >>>when you need to capture output, like from a SHOW CONFIG, for example.r  % Have you tried SET HOST /LOG=<file> ?o   -- .J       Charlie Hammond -- Hewlett-Packard Company -- Ft Lauderdale  FL  USAF           (hammond@not@peek.ssr.hp.com -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------    Date: 13 Sep 2004 08:01:05 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)c Subject: Re: Impersonate3 Message-ID: <ShMY8gcBu$SG@eisner.encompasserve.org>   W In article <4143EA34.8020006@tsoft-inc.com>, David Froble <davef@tsoft-inc.com> writes:k > K > I cannot see that.  To me, and possibly only me, the DETACH priv was for dQ > allowing a detached process to be created.  What clarity does it gain by being DD > re-named to IMPERSONATE?  Stupidest thing they've done in a while.  H    Exactly!  Anybody can create a detached process, the DETACH privilegeF    allows one to create a detached process under a different username.  #    That's why the name was changed.    ------------------------------  % Date: Mon, 13 Sep 2004 11:05:19 -0400r( From: David Froble <davef@tsoft-inc.com> Subject: Re: Impersonate, Message-ID: <4145B72F.3010004@tsoft-inc.com>   Larry Kilgallen wrote:  Y > In article <41449C9F.7020108@tsoft-inc.com>, David Froble <davef@tsoft-inc.com> writes:o >  > Q >>Maybe spliting the priviledge would have been better, (maybe not!) with DETACH nU >>allowing the usage of CREPRC for the same UIC and IMPERSONATE for using other UICs.N >> > H > Breaking existing programs is not historically popular in VMS circles. >   N And changing the name of a priviledge didn't possibly do that?  At least with P something like a system service they could have multiple names for the routine, J but there's no backward compatibility when the priviledge name is changed.  O Ok, you'll ask how such could happen.  What about an application that uses the t priviledge name?   Dave   -- .4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com 170 Grimplin RoadO Vanderbilt, PA  15486    ------------------------------  % Date: Mon, 13 Sep 2004 11:13:49 -0400R( From: David Froble <davef@tsoft-inc.com> Subject: Re: Impersonate* Message-ID: <4145B92D.40006@tsoft-inc.com>   John Santos wrote:   > David Froble wrote:M >  >> Larry Kilgallen wrote:I >>> >>> In article <4143EA34.8020006@tsoft-inc.com>, David Froble ! >>> <davef@tsoft-inc.com> writes:- >>>- >>>> Larry Kilgallen wrote:4 >>>> >>>MF >>>>> You can defend against it by demanding that VMS Development make! >>>>> privilege names more clear.0 >>>>>0F >>>>> Others have done that in the past, and the response was that theE >>>>> VMS Developers renamed the old DETACH privilege to IMPERSONATE.oH >>>>> It did not gain any capabilities in the renaming, but it certainly >>>>> made things more clear.  >>>>> J >>>> I cannot see that.  To me, and possibly only me, the DETACH priv was J >>>> for allowing a detached process to be created.  What clarity does it J >>>> gain by being re-named to IMPERSONATE?  Stupidest thing they've done  >>>> in a while. >>>> >>>TK >>> The fact that you just skimmedthe documentation is a sample of why theyaK >>> changed the name.  The security consideration of DETACH was always that E >>> the user could create a detached process _running_under_any_UIC_.q >>>  >>E >> Yes, and I've used it both ways.  The problem is that without the $H >> DETACH priv, Ok, it's been a while, a long while since I wrote those $ >> utilities, you cannot use CREPRC. >>C >> Ok, after looking at it from your prespective, there is a valid oJ >> argument for the new name.  I'll still claim that a valid argument for  >> the old name also exists. >>F >> Maybe spliting the priviledge would have been better, (maybe not!) A >> with DETACH allowing the usage of CREPRC for the same UIC and  $ >> IMPERSONATE for using other UICs. >> >> Daveo >> >  > Dave,  > I > You are arguing from a situation that hasn't been true for a very long 3 > time (20 years?)    * I've got utilities at least that old.  :-)    H > In VMS V1.0, DETACH allowed you to create a detached process under anyH > UIC.  Without DETACH privilege, you couldn't create a detached process# > at all (even under your own UIC).c > ? > You could always create subprocesses, if your quotas allowed.a > F > Lots of people complained; they had applications which really neededF > detached processes but no need to create them under anything but theA > creator's UIC and they didn't want to grant DETACH willy-nilly.i    6 I do understand the issues.  It sure is a valid point.    C > About VMS V3.0, $CREPRC (underlying RUN) was changed.  If you had G > DETACH privilege, you could still create detached processes under anyTE > UIC. If not, you could now create detached processes under your own  > UIC (subject to quotas).    P I wondered about that when I posted.  Didn't bother to go looking for the docs. O   But I have to ask, is V3.0 correct?  My poor memory seems to feel that while  N working under V4.* that the priviledge was still required.  Then again, maybe 9 most of that work was pre-V3.0.  The past is so dim.  :-)o    F > So the DETACH priv no longer governed detached processes per se, butE > the name stayed the same for about 15 years, until around V7.0 whenMH > it was changed to IMPERSONATE as being more descriptive.  The functionC > of the priv didn't change at all; it is still the same bit in thee > priv mask and in the UAF.0 >   C As you can tell, I don't always read all of the release notes.  :-)p   Dave   -- h4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com 170 Grimplin Road  Vanderbilt, PA  15486u   ------------------------------  % Date: Mon, 13 Sep 2004 11:24:58 -0400/( From: David Froble <davef@tsoft-inc.com> Subject: Re: Impersonate, Message-ID: <4145BBCA.5050805@tsoft-inc.com>   Z wrote:   > John Santos wrote: > G >> So the DETACH priv no longer governed detached processes per se, butrF >> the name stayed the same for about 15 years, until around V7.0 whenI >> it was changed to IMPERSONATE as being more descriptive.  The function,D >> of the priv didn't change at all; it is still the same bit in the >> priv mask and in the UAF. >  >  > Dumb question: > > > So why does SUBMIT/USER= require CMKRNL and not IMPRESONATE?  Q As I'm sitting here laughing, I'm trying to think what difference there is, from oK a security perspective, in creating a detached process for another UIC and  P submitting  batch job under another user name.  Can't think of any.  Seems that M you've pointed out an inconsistancy.  Then again, someone may point out some - difference.    Dave   -- 44 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com 170 Grimplin Roadu Vanderbilt, PA  15486:   ------------------------------  % Date: Mon, 13 Sep 2004 18:37:36 +0200 0 From: Keith Cayemberg <keith.cayemberg@arcor.de> Subject: Re: ImpersonateB Message-ID: <4145ccd1$0$18562$9b4e6d93@newsread2.arcor-online.net>   David Froble wrote:a > Larry Kilgallen wrote: > = >> In article <41449C9F.7020108@tsoft-inc.com>, David Froble    >> <davef@tsoft-inc.com> writes: >> >>G >>> Maybe spliting the priviledge would have been better, (maybe not!) .B >>> with DETACH allowing the usage of CREPRC for the same UIC and % >>> IMPERSONATE for using other UICs.p >>>w >>I >> Breaking existing programs is not historically popular in VMS circles.y >> > K > And changing the name of a priviledge didn't possibly do that?  At least ,J > with something like a system service they could have multiple names for I > the routine, but there's no backward compatibility when the priviledge 0 > name is changed. > H > Ok, you'll ask how such could happen.  What about an application that  > uses the priviledge name?  >  > Dave >   D To my knowledge, the DETACH privilege is still available in current 2 syntax as a synonym for the IMPERSONATE privilege.   $ show system/noprocJ OpenVMS V7.3-2  on node CH21  13-SEP-2004 18:26:35.50  Uptime  14 05:12:06  % $ set proc/priv=(noall,tmpmbx,netmbx)s $ set proc/priv=detach $ show proc/priv  G 13-SEP-2004 18:28:44.43   User: CYBERHAI         Process ID:   20204D7Es@                            Node: CH21             Process name:  "LKWP___20204D7E"e   Authorized privileges:G   EXQUOTA      GRPNAM       GRPPRV       IMPERSONATE  NETMBX       OPER H   PRMMBX       SETPRV       SYSNAM       SYSPRV       TMPMBX       WORLD   Process privileges:n3   IMPERSONATE          may impersonate another user 0   NETMBX               may create network device3   TMPMBX               may create temporary mailbox-   Process rights:-
   INTERACTIVEo   REMOTE   System rights:   SYS$NODE_CH21    Soft CPU Affinity: off     Cheers!c   Keith Cayembergp   ------------------------------  % Date: Mon, 13 Sep 2004 19:11:46 +0200>0 From: Keith Cayemberg <keith.cayemberg@arcor.de> Subject: Re: ImpersonateB Message-ID: <4145d4d2$0$26115$9b4e6d93@newsread4.arcor-online.net>   Keith Cayemberg wrote:   > David Froble wrote:  >  >> Larry Kilgallen wrote:  >>> >>> In article <41449C9F.7020108@tsoft-inc.com>, David Froble ! >>> <davef@tsoft-inc.com> writes:n >>>t >>>aH >>>> Maybe spliting the priviledge would have been better, (maybe not!) C >>>> with DETACH allowing the usage of CREPRC for the same UIC and n& >>>> IMPERSONATE for using other UICs. >>>> >>>lJ >>> Breaking existing programs is not historically popular in VMS circles. >>>- >>F >> And changing the name of a priviledge didn't possibly do that?  At G >> least with something like a system service they could have multiple dI >> names for the routine, but there's no backward compatibility when the . >> priviledge name is changed. >>I >> Ok, you'll ask how such could happen.  What about an application that - >> uses the priviledge name? >> >> Dave  >> > F > To my knowledge, the DETACH privilege is still available in current 4 > syntax as a synonym for the IMPERSONATE privilege. >  > $ show system/noprocL > OpenVMS V7.3-2  on node CH21  13-SEP-2004 18:26:35.50  Uptime  14 05:12:06 > ' > $ set proc/priv=(noall,tmpmbx,netmbx)S > $ set proc/priv=detach > $ show proc/priv >   D It also appears the VMS Engineers thought about dependencies on the & return string from Lexicals as well...  + $ write sys$output  f$getjpi("","AUTHPRIV")RQ SYSNAM,GRPNAM,DETACH,PRMMBX,SETPRV,TMPMBX,WORLD,OPER,EXQUOTA,NETMBX,SYSPRV,GRPPRV   + $ write sys$output  f$getjpi("","PROCPRIV")  DETACH,TMPMBX,NETMBX   Cheers!    Keith Cayemberg    ------------------------------  % Date: Mon, 13 Sep 2004 09:59:42 -0400n" From: Glenn Everhart <gce@gce.com>( Subject: Re: Impersonate (bit off topic)3 Message-ID: <4145a861$0$2654$61fed72c@news.rcn.com>    John Santos wrote: > David Froble wrote:  >  >> Larry Kilgallen wrote:e >>> >>> In article <4143EA34.8020006@tsoft-inc.com>, David Froble ! >>> <davef@tsoft-inc.com> writes:4 >>>t >>>> Larry Kilgallen wrote:u >>>> >>>4F >>>>> You can defend against it by demanding that VMS Development make! >>>>> privilege names more clear.t >>>>>wF >>>>> Others have done that in the past, and the response was that theE >>>>> VMS Developers renamed the old DETACH privilege to IMPERSONATE.bH >>>>> It did not gain any capabilities in the renaming, but it certainly >>>>> made things more clear.- >>>>>mJ >>>> I cannot see that.  To me, and possibly only me, the DETACH priv was J >>>> for allowing a detached process to be created.  What clarity does it J >>>> gain by being re-named to IMPERSONATE?  Stupidest thing they've done  >>>> in a while. >>>> >>> K >>> The fact that you just skimmedthe documentation is a sample of why theylK >>> changed the name.  The security consideration of DETACH was always thatrE >>> the user could create a detached process _running_under_any_UIC_.  >>>. >>E >> Yes, and I've used it both ways.  The problem is that without the eH >> DETACH priv, Ok, it's been a while, a long while since I wrote those $ >> utilities, you cannot use CREPRC. >>C >> Ok, after looking at it from your prespective, there is a valid uJ >> argument for the new name.  I'll still claim that a valid argument for  >> the old name also exists. >>F >> Maybe spliting the priviledge would have been better, (maybe not!) A >> with DETACH allowing the usage of CREPRC for the same UIC and o$ >> IMPERSONATE for using other UICs. >> >> Daveo >> >  > Dave,  > I > You are arguing from a situation that hasn't been true for a very long n > time (20 years?) > H > In VMS V1.0, DETACH allowed you to create a detached process under anyH > UIC.  Without DETACH privilege, you couldn't create a detached process# > at all (even under your own UIC).f > ? > You could always create subprocesses, if your quotas allowed.n > F > Lots of people complained; they had applications which really neededF > detached processes but no need to create them under anything but theA > creator's UIC and they didn't want to grant DETACH willy-nilly.  > C > About VMS V3.0, $CREPRC (underlying RUN) was changed.  If you hadfG > DETACH privilege, you could still create detached processes under anymE > UIC. If not, you could now create detached processes under your own  > UIC (subject to quotas). > F > So the DETACH priv no longer governed detached processes per se, butE > the name stayed the same for about 15 years, until around V7.0 wheneH > it was changed to IMPERSONATE as being more descriptive.  The functionC > of the priv didn't change at all; it is still the same bit in the  > priv mask and in the UAF.e >   D It's a bit off topic, but if one can arrange for two parties to haveC the ability to have in their hands the same random number remotely,rJ as with a token, then the person as well as the token can be authenticated@ by specifying that the human do some pre-agreed transform on theH digits of the random number - maybe a permutation of a few of them - andH report to the other end, which has to know what the pre-agreed transformJ is. This seems better than just appending another password since anyone inJ the middle only observes random numbers; their derivation remains obscure.  I It keeps getting harder to avoid impersonation with privs only...could bee good to know tricks...   Glenn Everhart   ------------------------------  # Date: Mon, 13 Sep 2004 13:41:59 GMTp& From: jlsue <jefflsxxxz@sbcglobal.net>. Subject: Re: Itanium not meeting Intel's goals8 Message-ID: <bs8bk09ati7qc1fcuhg4rl74v49kk7fb6n@4ax.com>  H On Fri, 10 Sep 2004 09:09:10 -0400, "John Smith" <a@nonymous.com> wrote:   >JF Mezei wrote:  M >As a substitute processor for Itanic, Intel will offer HP a very interestings2 >processor it has up its sleeve, code named "EV8". >    <slaps forehead>   WOW!  I coulda had a (e)V8!n   --- jls'0 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)n   ------------------------------   Date: 13 Sep 2004 12:49:53 GMT( From: bill@cs.uofs.edu (Bill Gunshannon), Subject: Re: Marketing of tech into campuses+ Message-ID: <2qljbhF10depaU1@uni-berlin.de>a  , In article <z-udnVY1OfCo8dncRVn-rw@igs.net>,& 	"John Smith" <a@nonymous.com> writes: > Bob Ceculski wrote:o5 >> "Main, Kerry" <kerry.main@hp.com> wrote in messagey >>N > news:<FD827B33AB0D9C4E92EACEEFEE2BA2FB3E0341@tayexc19.americas.cpqcorp.net>. > .. >>5 >> ok, now when is HP going to do this in the states? ( >> This should have been done years ago! >  > + > You can't force a drunk to stop drinking.  > H > You can take that as a comment about DEC/ChumHPaq and/or university CS > departments as you see fit.n   G What does it have to do with University CS Departments?  CS DepartmentscC aren't shills for any company's products.  They teach concepts, not(C particulars,  And before you jump in here with a complaint about CS E Departments teaching Unix, remember that Unix is the only OS that hasjH been available in source form with an agreement that allowed students toE read (and even play with) it in an academic environment.  VMS doesn't C even have an agreement that makes it practical to use it, much less G disect it in the classroom.  When you add tot hat the fact that it also>H takes special (and expensive) hardware, how can you be surprised that it- has so little remaining presence in academia?>   bill  s   -- iJ 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: Mon, 13 Sep 2004 11:36:05 -0400 ( From: David Froble <davef@tsoft-inc.com>, Subject: Re: Marketing of tech into campuses, Message-ID: <4145BE65.3020308@tsoft-inc.com>   Bill Gunshannon wrote:  . > In article <z-udnVY1OfCo8dncRVn-rw@igs.net>,( > 	"John Smith" <a@nonymous.com> writes: >  >>Bob Ceculski wrote:l >>5 >>>"Main, Kerry" <kerry.main@hp.com> wrote in messager >>>e >>>oN >>news:<FD827B33AB0D9C4E92EACEEFEE2BA2FB3E0341@tayexc19.americas.cpqcorp.net>. >>.. >>5 >>>ok, now when is HP going to do this in the states?a( >>>This should have been done years ago! >>>g >>+ >>You can't force a drunk to stop drinking.l >>H >>You can take that as a comment about DEC/ChumHPaq and/or university CS >>departments as you see fit.  >> >   I > What does it have to do with University CS Departments?  CS DepartmentstE > aren't shills for any company's products.  They teach concepts, not E > particulars,  And before you jump in here with a complaint about CSmG > Departments teaching Unix, remember that Unix is the only OS that haseJ > been available in source form with an agreement that allowed students toG > read (and even play with) it in an academic environment.  VMS doesn't.E > even have an agreement that makes it practical to use it, much less I > disect it in the classroom.  When you add tot hat the fact that it alsoyJ > takes special (and expensive) hardware, how can you be surprised that it/ > has so little remaining presence in academia?h >  > bill >  i >  >    So let me range afield a bit.h  L Students are studying subjects with an interest in working in the aerospace P industry.  They need to learn about current state of the art of rocket engines. N   Using your rationalization, they'd use bottle rockets since they are cheap, L and you're teaching concepts (reaction engines).  This sure leaves out many Q things that you could teach, using liquid fuel for cooling the rocket nozzle and e such.   N I'm guessing that you don't feel that there is value in learning about things M like a distributed lock manager.  I may be stretching things a bit, but that @T sounds like being a shill for Microsoft's concepts.  Then again, I'm a biased bigot.   Dave   -- t4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com 170 Grimplin Roade Vanderbilt, PA  15486-   ------------------------------   Date: 13 Sep 2004 17:33:24 GMT( From: bill@cs.uofs.edu (Bill Gunshannon), Subject: Re: Marketing of tech into campuses+ Message-ID: <2qm3v4F118nmiU1@uni-berlin.de>3  , In article <4145BE65.3020308@tsoft-inc.com>,+ 	David Froble <davef@tsoft-inc.com> writes:l > Bill Gunshannon wrote: > / >> In article <z-udnVY1OfCo8dncRVn-rw@igs.net>,i) >> 	"John Smith" <a@nonymous.com> writes:- >> - >>>Bob Ceculski wrote: >>>S6 >>>>"Main, Kerry" <kerry.main@hp.com> wrote in message >>>> >>>>O >>>news:<FD827B33AB0D9C4E92EACEEFEE2BA2FB3E0341@tayexc19.americas.cpqcorp.net>.e >>>..D >>>a6 >>>>ok, now when is HP going to do this in the states?) >>>>This should have been done years ago!l >>>> >>>e, >>>You can't force a drunk to stop drinking. >>>,I >>>You can take that as a comment about DEC/ChumHPaq and/or university CSi >>>departments as you see fit. >>>r >>  J >> What does it have to do with University CS Departments?  CS DepartmentsF >> aren't shills for any company's products.  They teach concepts, notF >> particulars,  And before you jump in here with a complaint about CSH >> Departments teaching Unix, remember that Unix is the only OS that hasK >> been available in source form with an agreement that allowed students to-H >> read (and even play with) it in an academic environment.  VMS doesn'tF >> even have an agreement that makes it practical to use it, much lessJ >> disect it in the classroom.  When you add tot hat the fact that it alsoK >> takes special (and expensive) hardware, how can you be surprised that itA0 >> has so little remaining presence in academia? >>   >  > So let me range afield a bit.r > N > Students are studying subjects with an interest in working in the aerospace R > industry.  They need to learn about current state of the art of rocket engines. P >   Using your rationalization, they'd use bottle rockets since they are cheap, N > and you're teaching concepts (reaction engines).  This sure leaves out many S > things that you could teach, using liquid fuel for cooling the rocket nozzle and i > such.u  G Assuming they actually attended a school that taught "rocketry" (Yes, IqI know of at least one that goes into that much detail) they would not find:F themselves assembling Shuttle engines.  They would, at best, build andD launch simple solid fuel rockets.  But most of what they would learn- would be from books.  The same is true of CS.i   > P > I'm guessing that you don't feel that there is value in learning about things $ > like a distributed lock manager.    L As a concept, certainly.  One companies specific implementation, not really.  O >                                   I may be stretching things a bit, but that qV > sounds like being a shill for Microsoft's concepts.  Then again, I'm a biased bigot.  G None of the CS classes specifically address Microsoft either.  Granted, E the literacy classes that everyone takes as part of General EducationhE is MS based, but that's not CS. (And I don't agree with that either!)   ? If you want to be an MCSE go to a Trade School that teaches MS.t6 If you want to be a VMS Sys Admin go to the Boot Camp.2 If you want to be a Solaris Sys Admin talk to Sun.C If you want to learn Computer Science that can be applied to any of 7 these specific OS implementations, go to a real school.|  @ Why do you think so many of us continue to argue for some way toB get VMS systems made available for student use?  They aren't goingB to get it in the classroom, but they can be exposed to it.  Sadly,@ HP doesn't seem to share the same zeal for this that a few of us here do.   bill   -- mJ 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>   v   ------------------------------    Date: 13 Sep 2004 05:46:05 -0700) From: mail@sanface.com (SANFACE Software)s Subject: PLUG: txt2pdf 7.5= Message-ID: <8c682947.0409130446.394ce298@posting.google.com>o  / We would like to announce txt2pdf 7.5 version.  # http://www.sanface.com/txt2pdf.htmlmE txt2pdf is shareware; it is a very flexible and powerful Perl5 scriptiB that converts text files to PDF format files, so you can use it in> every operating systems supported by Perl5, including OpenVMS.( Remember to read "txt2pdf on OpenVMS" at# http://www.sanface.com/openvms.htmld; It's simple to design background like invoices, orders etc.e) Here nice examples made using txt2pdf PROk- http://www.sanface.com/pdf/Purchase_Order.pdf ) http://www.sanface.com/pdf/oldinvoice.pdf4$ http://www.sanface.com/pdf/hfmus.pdf) http://www.sanface.com/pdf/heraldbill.pdf @ If you prefer we also distribute executables for Windows, Linux,@ Solaris, AIX, HP-UX, and Mac OS X. Inside the Windows version is Visual txt2pdf, a VB GUI.    What's new in this version  D To put one or more lines before the beginning of the PDF now you can use inside txt2pdf.cfg prepdf.F This feature can be very useful. On Xerox printers + DocuSP controllerF you can use Xerox Job ticket commands. The "options" must be placed atC the beginning of the pdf file, before the %PDF. To print duplex your can usemB prepdf: %XRXbegin: 001.0300\n%XRXPDLformat: PDF\n%XRXrequirements: duplex\n%XRXend  (\n means go to a new line)g= If you need more information about Xerox Job ticket or how toe* configure it don't hesitate to contact us.E Pay attention: accordingly with PDF Specification 1.5 (specifically -0@ Implementation Notes 13 and 14) PDF Header in form "%PDF-M.m" orF "%!PS-Adobe-N.n PDF-M.m" should appear somewhere within the first 1024C bytes of the file. We suggest you to use this feature only to print  the pdf!  B If you use email txt2pdf feature with priority now you can set the> email priority. The default 3 means X-Priority: 3 (Normal) andE X-MSMail-Priority: Normal. You can use also 1 X-Priority: 1 (Highest)e; and X-MSMail-Priority: High or 5 X-Priority: 5 (Lowest) and- X-MSMail-Priority: Low   Test txt2pdf 7.5! 6 You can find it at http://www.sanface.com/txt2pdf.html   ------------------------------  % Date: Mon, 13 Sep 2004 08:58:12 -0700e# From: "Tom Linden" <tom@kednos.com>08 Subject: Posting  ( WAS: why mainframes are still used?)( Message-ID: <opsd9vzawuzgicya@hyrrokkin>  K On Mon, 13 Sep 2004 11:34:37 -0400, David M Smith <dsmit115@csc.com> wrote:e  C > On 11 Sep 2004 10:23:15 -0700, tom@kednos.com (Tom Linden) wrote:2 >g1 >> Nigel Barker <nigel@hp.com> wrote in message  ,7 >> news:<7ep3k09qiqgod3oevem3ee6og1c0ceos4h@4ax.com>...tJ >>> On Fri, 10 Sep 2004 13:02:59 -0400, David M Smith <dsmit115@csc.com>  
 >>> wrote: >>>2I >>> >Tom -- if it's all the same to you, I would love for you to change    >>> your "quote"K >>> >character in your newsreader from < to >. It would make reading your  r >>> posts so, >>> >much easier (for me, and maybe others). >>>?L >>> I'll vote for that too. If you could trim to context too that would be   >>> even >>> better.0 >> >> Is that better? >aD > It's wonderful! Thanks, your posts are much easier to read in my   > newsreader > program (Forte Agent).  H Mainly accomplished (or made easier by switching from Outlook to Opera   Mail.e  J This brings up another question, whether to use Newsreader or Info-VAX.  IH have tried several different (free) newservers and they all seem to haveI about a 24 hour latency.  Not sure why they need to take so long.  Anyoner have any recommendation   K > -------------------------------------------------------------------------iK > David M. Smith 302.391.8533                       dsmit115 at csc dot comtK > Computer Sciences Corporation     (Opinions are those of the writer only)hK > -------------------------------------------------------------------------  >        -- lC Using Opera's revolutionary e-mail client: http://www.opera.com/m2/    ------------------------------  % Date: Mon, 13 Sep 2004 17:12:07 +0100n- From: Roy Omond <Roy.Omond@BlueBubble.UK.Com>-6 Subject: Re: Posting  - recommendation for News server* Message-ID: <2qlvbmFvos4fU1@uni-berlin.de>   Tom Linden wrote:I  M > On Mon, 13 Sep 2004 11:34:37 -0400, David M Smith <dsmit115@csc.com> wrote:h >  >> [... snip ...] >>  E >> It's wonderful! Thanks, your posts are much easier to read in my   $ >> newsreader program (Forte Agent). >oJ > Mainly accomplished (or made easier by switching from Outlook to Opera   > Mail.l > L > This brings up another question, whether to use Newsreader or Info-VAX.  IJ > have tried several different (free) newservers and they all seem to haveK > about a 24 hour latency.  Not sure why they need to take so long.  Anyonev > have any recommendationi  A I'm a very happy user of news://news.individual.net which is verytA kindly supplied to anyone "kostenlos" by the most esteemed peoplea) at the Freie Universitt Berlin, Germany.c  7 I don't even notice *any* latency, never mind 25 hours.w  6 You can register for free at http://www.individual.net  	 Roy Omondu+ Blue Bubble Ltd. (aka Blaue Seifenblase :-)t   ------------------------------  % Date: Mon, 13 Sep 2004 09:40:52 -0700h# From: "Tom Linden" <tom@kednos.com>r6 Subject: Re: Posting  - recommendation for News server( Message-ID: <opsd9xyeb0zgicya@hyrrokkin>  / On Mon, 13 Sep 2004 17:12:07 +0100, Roy Omond  n$ <Roy.Omond@BlueBubble.UK.Com> wrote:   > Tom Linden wrote:o >eI >> On Mon, 13 Sep 2004 11:34:37 -0400, David M Smith <dsmit115@csc.com>  s	 >> wrote:l >> >>> [... snip ...] >>u >cG >>> It's wonderful! Thanks, your posts are much easier to read in my   g% >>> newsreader program (Forte Agent).l >>L >> Mainly accomplished (or made easier by switching from Outlook to Opera    >> Mail.C >>  This brings up another question, whether to use Newsreader or  l >> Info-VAX.  I K >> have tried several different (free) newservers and they all seem to havetG >> about a 24 hour latency.  Not sure why they need to take so long.   -	 >> Anyone- >> have any recommendation > C > I'm a very happy user of news://news.individual.net which is veryIC > kindly supplied to anyone "kostenlos" by the most esteemed people.+ > at the Freie Universitt Berlin, Germany.4 >69 > I don't even notice *any* latency, never mind 25 hours.e >w8 > You can register for free at http://www.individual.net  = I did about an hour ago, but haven't got the passwd back yet,t@ also tried news.f.de.plusline.net and I got your response before) info-VAX, but it didn't allow me to post.. >e > Roy Omond - > Blue Bubble Ltd. (aka Blaue Seifenblase :-)e >G       -- >C Using Opera's revolutionary e-mail client: http://www.opera.com/m2/i   ------------------------------  # Date: Mon, 13 Sep 2004 17:33:38 GMTe6 From: "Kenneth Farmer" <kfarmer@NOSPAM.spyderbyte.com>- Subject: Reminder: Datamation OpenVMS articleo; Message-ID: <SRk1d.355$zA3.144410@twister.southeast.rr.com>W   ----- Original Message ----- e# From: "John Smith" <a@nonymous.com>l Newsgroups: comp.os.vmsf( Sent: Friday, September 10, 2004 4:27 PM Subject: VMS in Datmation.     > Remaining Vehemently OpenVMS >oD > Tossing aside thoughts of the latest operating systems, the Albert EinsteinJ > Healthcare Network in Philadelphia just bought a new OpenVMS system. ForA > them, it's definitely not out with the old and in with the new.. >s: > http://itmanagement.earthweb.com/erp/article.php/3406641    E Just a reminder to everyone about this article since it was posted onmJ Friday.  The more of us that visit the better.  They decide content by howG many visits they get to these articles...obviously.  The last couple oft6 Datamation VMS articles they ran had a good hit count.   Ken   # OpenVMS.org, http://www.OpenVMS.orgw ________________________" Kenneth R. Farmer <>< 336-736-7376$ SpyderByte: http://www.SpydeByte.com   ------------------------------    Date: 13 Sep 2004 08:13:17 -0500 From: briggs@encompasserve.org' Subject: Re: Suggestion for file systemm3 Message-ID: <8EPNMikclHM7@eisner.encompasserve.org>e  s In article <4141f588$0$436$9b4e6d93@newsread2.arcor-online.net>, Keith Cayemberg <keith.cayemberg@arcor.de> writes:  > JF Mezei wrote:k > N >> When you consider the various servers on VMS systems which generate, out of> >> security necessity, large log files, here is my suggestion: >> KI >> File system has both a start of file offset and an end of file offset.s ...DE > Interesting idea. Sort of a mini Spiralog capability for log files.1 > K > With a fixed-size log file the file beginning could be lost in step with oF > the new  blocks appended at the end of the file. The implementation B > could resemble a "circular queue" implemented as a file. With a H > variable-sized file a type of  check pointing could be implemented to I > expire blocks from the beginning of the file according to a designated ,
 > delta-time.:  F The implementation scheme that sprang to mind for me was a rudimentaryE "sparsely populated file".  At the file system level, you have one ordB more null mapping pointers in the file header for the front of the	 log file.n  E At the RMS level you then add beginning-of-file information analogousoH to the existing EOF, EBK fields.  Call them BOF, FUB (beginning of file, first used byte), perhaps.  D And the critical operation is then the "truncate-at-the-front" where> you arbitrate file locks (establishing exclusive access to theB truncation zone), update BOF and FUB, release initial space to theC free pool and allocate null mapping pointer(s) (care being taken ifyE this requires mapping pointers to be bubbled on up into the extension0A header chain or if it allows mapping pointers to be shuffled downlA from the extension header chain) and update BOF, FUB information.c  A And now you need to flush the initial entries in any active WCB'sd? so that subsequent access to the initial segment can error out.   G > Sounds like this capability would require at least its own file type eF > designation if it were to be integrated into the existing RMS level K > services. That would also require all relevant applications to know what AF > to do when encountering such a file type. Or, should/must this be a J > capability integrated at the I/O System Services level? Then I guess we H > really would be talking about a new ODS-level file system. Either way 1 > were talking about a potentially major project.@  . Major change for minor functionality, I agree.  H If one were going to do it at all, it'd probably be worth going for full blown sparse files.t   	John Briggs   ------------------------------  % Date: Mon, 13 Sep 2004 11:09:17 +0100i< From: Andrew Harrison <andrew_remove__s._harrison@s_u_n.com>+ Subject: Re: why mainframes are still used? 0 Message-ID: <ci3rkg$kfa$1@new-usenet.uk.sun.com>   Alex Daniels wrote:aN > "Andrew Harrison SUNUK Consultancy" <Andrew_No.Harrison_No@nospamn.sun.com> = > wrote in message news:chsc6f$h0i$1@new-usenet.uk.sun.com...  >  >>Alex Daniels wrote:c >>M >>>"Andrew Harrison" <andrew_remove__s._harrison@s_u_n.com> wrote in message h- >>>news:chp31m$f1u$1@new-usenet.uk.sun.com...l >>>hD >>>>The Sun E10K took the I/O crown away from IBM and it hasn't been
 >>>>returned.  >>> 	 >>><SNIP>i >>> > Andrew Harrison  >>> G >>>Are there any Sun boxes with greater maximum I/O bandwidth than the  C >>>AlphaServer GS1280 at 222 GB/s ? If so can you please provide a r
 >>>reference.a >>>i >>A >>You apear to be labouring under the impression that I suggested04 >>that the E10K still holds the I/O bandwidth crown. >>
 >>It doesn't.: >>> >>Nor was I suggesting that Sun holds the I/O bandwidth crown. >>< >>My point was that IBM lost the I/O throughput crown a long' >>time ago and have never recovered it.t >>@ >>I would also be very carefull about suggesting that the GS1280B >>can do 222 GB/s in I/O bandwidth as you appear to be attempting,> >>that is the maximum bandwidth, measured bandwidth using realA >>applications is much lower than that at 15 GB/s, lower than the,  >>measured bandwidth on an F25K. >>	 >>regardse >>Andrew HarrisonD >  > M > I asked a question, as I'm looking at getting some new boxes for one of my gA > customers, I didn't mean for you to read anything else into it.c > O > As you mention the F25k however, I had a look at the STREAMS results for it,   > compared to the GS1280 eK > (http://www.cs.virginia.edu/stream/top20/Bandwidth.html). The GS1280 (64 uG > CPUs) is 7 places above the F25k (144 CPUs, 72 Dual Processor chips).a > C > Is this not an accurate/representative or appropriate comparison?  >   ; Streams measures the speed of the system interconnect (ish)>: it does not measure the I/O throughput of a system, nor is9 it really an application. If for example you had touted ae9 streams number at an IBM mainframe protagonist they woulde7 have countered with my IBM can sustain X GB/s through as1 DB2 DBMS to disk storage with a real application.a   Regardso Andrew Harrisonh   ------------------------------  % Date: Mon, 13 Sep 2004 11:10:50 +0100l< From: Andrew Harrison <andrew_remove__s._harrison@s_u_n.com>+ Subject: Re: why mainframes are still used?,0 Message-ID: <ci3rnc$kfa$2@new-usenet.uk.sun.com>   Fabio Cardoso wrote:  v > Andrew Harrison <andrew_remove__s._harrison@s_u_n.com> wrote in message news:<chp31m$f1u$1@new-usenet.uk.sun.com>... >  >>JF Mezei wrote:a >>$ >>>Mainframes have 2 selling points: >>>uM >>>On a technical basis, I think that where IBM mainframes historically had atT >>>large edge was in IO throughput. (I think this could be debated nowadays though). >>>a >>B >>The Sun E10K took the I/O crown away from IBM and it hasn't been >>returned.p >>F >>SGI fans might claim that SGI snatched it first but SGI Origins haveI >>never really ventured into the large commercial batch processing space.a >  > B > A few years ago the company saved money not buying a new Origin.K > Instead they mounted a Beowulf with 32 Compaq DL servers (2 x CPUs) each.oJ > And this system outperformed the Origin in price (> 50%) and processing. >   D Sure but Beowulf is an MPP system and because of that it is unlikelyC to figure highly in anyones lst of suitable platforms to host largew& commercial batch processing workloads.   Regardst Andrew Harrisone	 > Regardsb >  > Fc   >  >  o > 	 >>Regards  >>Andrew Harrisoni >>O >>>From a management point of view, it isn't so much the mainframe per say, buthP >>>the software that runs on it. If you are a bank and your banking software andQ >>>databases are already on the mainframe since they have always been there, thendL >>>you are more likely to continue to run that solid software and either addP >>>software to your mainframe, or connect some lesser machines to your mainframe* >>>to make use of the mainframe databases. >>>iP >>>Consider banking web site or telephone applications. The web or telephone appO >>>may run on UNXI, but they connect to the IBM mainframe to perform the actualeL >>>transactions. Same for ATM and interbank networks which may run on tandemP >>>machines for total uptime, but the tandem ends up sending the transactions to >>>the IBM mainframe.d >>>tQ >>>Once you've built an ecosystem around your mainframe, it is quite difficult to0- >>>remove/change the heart of that ecosystem.    ------------------------------  % Date: Mon, 13 Sep 2004 11:34:37 -0400 & From: David M Smith <dsmit115@csc.com>+ Subject: Re: why mainframes are still used? 8 Message-ID: <3ffbk0pqrbe6us8nu5vf0mn32a06ksiqg8@4ax.com>  A On 11 Sep 2004 10:23:15 -0700, tom@kednos.com (Tom Linden) wrote:E  b >Nigel Barker <nigel@hp.com> wrote in message news:<7ep3k09qiqgod3oevem3ee6og1c0ceos4h@4ax.com>...N >> On Fri, 10 Sep 2004 13:02:59 -0400, David M Smith <dsmit115@csc.com> wrote: >>  S >> >Tom -- if it's all the same to you, I would love for you to change your "quote" Q >> >character in your newsreader from < to >. It would make reading your posts son+ >> >much easier (for me, and maybe others).u >> iN >> I'll vote for that too. If you could trim to context too that would be even
 >> better. >  >Is that better?  K It's wonderful! Thanks, your posts are much easier to read in my newsreaderb program (Forte Agent).I -------------------------------------------------------------------------FI David M. Smith 302.391.8533                       dsmit115 at csc dot comRI Computer Sciences Corporation     (Opinions are those of the writer only),I -------------------------------------------------------------------------d   ------------------------------  % Date: Mon, 13 Sep 2004 10:22:17 +0200a! From: Soterro <soterroatyahoocom>e@ Subject: Re: why mainframes are still used? - printer-made music: Message-ID: <4145583c$0$5456$4d4ef98e@read.news.ch.uu.net>   Mike Bartman wrote: E > Ever hear anyone play a tune on a line printer?  People had so much>5 > more spare time back before we all had computers...   I I loved it, and there might be more out there to like this 'symphony for i dot matrix pinters':  5 http://www.sat.qc.ca/the_user/dotmatrix/en/intro.htmlg   Sn   ------------------------------    Date: 13 Sep 2004 06:52:43 -0700( From: bob@instantwhip.com (Bob Ceculski)1 Subject: Windoze reboots turn into disk rebuilds! = Message-ID: <d7791aa1.0409130552.1d39ba9d@posting.google.com>.  7 this is hilarious ... users now instead of rebooting ton5 fix problems for xp now just rebuild with the factory 4 os rebuild disk ... instead of progression xp is now regressing ... :)u  ) http://www.theinquirer.net/?article=18401T   ------------------------------  % Date: Mon, 13 Sep 2004 16:33:15 +01002+ From: Rodrigo Ventura <yoda@isr.ist.utl.pt> 5 Subject: Re: Windoze reboots turn into disk rebuilds!h) Message-ID: <m3wtyychfo.fsf@pixie.isrnet>4  9 >>>>> "Bob" == Bob Ceculski <bob@instantwhip.com> writes:   D     Bob> this is hilarious ... users now instead of rebooting to fixE     Bob> problems for xp now just rebuild with the factory os rebuildID     Bob> disk ... instead of progression xp is now regressing ... :)  F That's really cool! And the hardware everyday faster, so that re-buildC process is faster. Next thing they invest a scheme to automaticallyeE rebuild the system partition each time the system boot, which is even-, cooler, since viruses do not survive a boot!   Cheers,r   RodrigoY   -- C  : *** Rodrigo Martins de Matos Ventura <yoda@isr.ist.utl.pt>. ***  Web page: http://www.isr.ist.utl.pt/~yoda0 ***   Teaching Assistant and PhD Student at ISR:7 ***    Instituto de Sistemas e Robotica, Polo de Lisboa 4 ***     Instituto Superior Tecnico, Lisboa, PORTUGALH *** PGP fingerprint = 0119 AD13 9EEE 264A 3F10  31D3 89B3 C6C4 60C6 4585   ------------------------------   End of INFO-VAX 2004.509 ************************! >>> <davef@tsoft-inc.com> writes:n >>>t >>>aH >>>>W6XhPeD3+ĤȩETZ%?AmDL
L>cDQ~agrBziĂhl]bP~x)a1),}^:zYX*WVr=IaⴴZUMC~f{u<݌D[0ꨶh
FTFc0A8uuۃar+*ȳΏ⩫(U9N- "'hl6~Xs)M~5tFL4ZZE
Ɂ8m5K,'JP6R-bbX*)MWUd:c'))nv]P]<MxbEw>N7g7,|%@=1'qU㇆dZ1l
PN8G3ڃtCň09D*%j}RrJ]^fvO%sX݄C!tcdT]DQR>pHqpoBb/nK]\*v[UZ3Nl_@$7-B,1z-|U\
"s*%[KY=`xPTr*`ǭălL-!LϦpԅ.)I!ʦJ~{B9f~
ONkG5`bDzk-ZJdOXŹ:XL+_I-Wxp"2ƹ宸bиM#ECl7`FRP!cBc;<&ZUD'1SA
ҥ9WdPd莅a1ѢҘRvm %#yPA<B=#*l^H[gǸ>QbY@͑
CFdJ>IEzaCmZ-Dx}rCI`L's&/$12,}.\ -Tq6D&6ON`wѸh)2ӢXbGfv-
TبWn^T\-vz,U]DfWV_Ԁ͊kJV[Fx嗴"ŲDuNi\9@*VzFnb_rKɱW&r9=XS[*&h([zLv~V[zLXZ
[dKAgdT#*ġG~qFV]M*2\+!TLAjQ,-
[(ne
T
/\kD~պh7Kb^$f 169-9; =~ƂUW·J#;vy
УTza;b-l\a#
]9XRLM`Vi̍/b9n \;PL]<F <O+qD4$
S4&n}(b*sDU8KXR0/9[$\G- G+i?ᰫXC,/TMaRWS鱺Qs.XLEތћzF)lXzi˂uw&[!! {
	=PzؤOU1ֱ'%D4tM{89O(ȱG@aVl`ǍᚍLF9wd^ˏIvAP*4yMPBʄYI.6R@j@6P4hA8
i&N6!	
h=@oGFQr$H;A1d6' 2APEvdePI۔
ߴIz)[	eiePvaF2^WlxK:%Poإ
Am=<уe`Y	a|XL"0GC=T*UA){C{A3t C8eaQCBdΧD1Y}h!E|Z~"g!x@ĆF