1 INFO-VAX	Sun, 01 May 2005	Volume 2005 : Issue 241       Contents: Re: DEC/MMJ connectors8 Re: Maybe HP should get out of the hardware business....8 Re: Maybe HP should get out of the hardware business....8 Re: Maybe HP should get out of the hardware business....8 Re: Maybe HP should get out of the hardware business....8 Re: Maybe HP should get out of the hardware business....8 Re: Maybe HP should get out of the hardware business....8 Re: Maybe HP should get out of the hardware business....8 Re: Maybe HP should get out of the hardware business....8 Re: Maybe HP should get out of the hardware business....8 Re: Maybe HP should get out of the hardware business....0 Re: more questions on Cyrillic fonts and Mozilla) Re: MySQL and PHP Authentication Protocol  OpenVMS and Mac OS X 10.4 $ Re: Sandia says alpha the best chip! Re: Setting up an NTP server Re: Setting up an NTP server Re: Setting up an NTP server Re: Slow Filesystem I/O  Re: Slow Filesystem I/O  Re: Slow Filesystem I/O  Re: Slow Filesystem I/O  Re: Slow Filesystem I/O  Re: Slow Filesystem I/O  RE: Slow Filesystem I/O  Re: Slow Filesystem I/O ) Re: SYS$SYSROOT and similar logical names ) Re: SYS$SYSROOT and similar logical names + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + RE: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + RE: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS? + Re: What is Different or Special About VMS?   F ----------------------------------------------------------------------  % Date: Sat, 30 Apr 2005 17:36:11 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>  Subject: Re: DEC/MMJ connectors B Message-ID: <1114896985.a0a53cb9bc52a4212b12cacf0d0f73f8@teranews>   David J Dachtera wrote: A > Some variants also have a DB25M. Likewise, some VT320 and VT510  > terminals.  H Those are the terminals that had been sold outside the USA. Not only did? they have DC25 AND MMJ, but also more complete setup menus with D language/keyboard selections, and more selections in the port (modemH signals). I believe they were called the "international" models and were# more expensive than the USA models.    ------------------------------  % Date: Sat, 30 Apr 2005 20:54:15 +0200 + From: Karsten Nyblad <nospam@nospam.nospam> A Subject: Re: Maybe HP should get out of the hardware business.... = Message-ID: <4273d464$0$67257$157c6196@dreader2.cybercity.dk>    Alan Greig wrote: D > And right now it would probably be a better bet for HP to buy AMD.  H Hmmmm.  One of the costs of developing computer chips is developing new H semiconductor processes suitable for high speed microprocessors.  Intel G can let their x86 business pay for that, but AMD barely have the money  H for it, and they have had to get help from IBM.  Do you think IBM would I help HP?  Further, do you think Sun and IBM would continue their Opteron  " based product lines if HP buy AMD?   ------------------------------  % Date: Sat, 30 Apr 2005 21:00:50 +0200 + From: Karsten Nyblad <nospam@nospam.nospam> A Subject: Re: Maybe HP should get out of the hardware business.... = Message-ID: <4273d5ef$0$78288$157c6196@dreader1.cybercity.dk>    icerq4a@spray.se wrote:  > The latest rumour I G > heard was that no system builder have a plan to buy the Opteron Horus D > chipset, which is sad, but that does not mean that we will not see > larger x86 systems.   I Newisys normally lives from designing complete systems that others sells  I as OEM.  My guess is that the marked for Opteron based systems with more  I than 8 processors is too small for firms to design their own systems.  I  D would wait and see if Newisys can make other firms sell Horus based  systems on OEM basis.    ------------------------------   Date: 30 Apr 2005 20:24:06 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)A Subject: Re: Maybe HP should get out of the hardware business.... , Message-ID: <3di7r6F6p61toU1@individual.net>  = In article <4273d464$0$67257$157c6196@dreader2.cybercity.dk>, . 	Karsten Nyblad <nospam@nospam.nospam> writes: > Alan Greig wrote: E >> And right now it would probably be a better bet for HP to buy AMD.  > J > Hmmmm.  One of the costs of developing computer chips is developing new J > semiconductor processes suitable for high speed microprocessors.  Intel I > can let their x86 business pay for that, but AMD barely have the money  J > for it, and they have had to get help from IBM.  Do you think IBM would K > help HP?  Further, do you think Sun and IBM would continue their Opteron  $ > based product lines if HP buy AMD?  F If the performance proceeded like it has up to now and their customers6 continued to express interest, certainly they would.     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: 30 Apr 2005 13:26:23 -0700* From: "Alan Greig" <greigaln@netscape.net>A Subject: Re: Maybe HP should get out of the hardware business.... C Message-ID: <1114892783.062179.208890@f14g2000cwb.googlegroups.com>    Karsten Nyblad wrote:   D > Newisys normally lives from designing complete systems that others sells E > as OEM.  My guess is that the marked for Opteron based systems with  moreG > than 8 processors is too small for firms to design their own systems.   IE > would wait and see if Newisys can make other firms sell Horus based  > systems on OEM basis.   8 AMD liked Horus so much they just bought the designer...  O http://www.amd.com/us-en/Corporate/VirtualPressRoom/0,,51_104_543~96057,00.html   B "Rich Oehler, Renowned Microprocessor and Multi-Processing Systems( Architect, joins AMD as Corporate Fellow@ -Oehler, former IBM Fellow and chief technology officer (CTO) at8 Newisys, will focus on increasing the alignment of AMD'sC enterprise-class computing strategies with the needs of the Fortune  1000-   F SUNNYVALE, CA -- March 28, 2005 --AMD (NYSE: AMD) announced today that: Rich Oehler, a 40-year industry veteran well known for hisE groundbreaking work in microprocessor, symmetric multi-processing and D mission-critical systems design, has joined AMD's Office of Strategy+ and Technology reporting to CTO Fred Weber.   @ "It is a real honor to have Rich join the AMD team," said Weber.C "His unique depth of experience in enterprise computing enhances an = already immensely talented group focused on strengthening the F performance, reliability, availability and scalability features of our 64-bit solutions."  @ Oehler will have broad responsibility for closely aligning AMD's> multi-core, enterprise-class computing strategies with the topE priorities of Fortune 1000 Chief Information Officers and Information  Technology management."  ---   * And in a recent post in comp.arch he says:   Rich Oehler 	  Mar 31, 7:28 am Newsgroups: comp.arch ) From: "Rich Oehler" <rich.oeh...@amd.com>   Date: 31 Mar 2005 07:28:25 -0800  Local: Thurs,Mar 31 2005 7:28 am" Subject: Re: Rich Oehler joins AMD    = Let me assure you that the Horus effort at Newisys continues.   B At Newisys, I accomplished what I wanted to do: define a technicalC vision for Opteron based servers, create an environment that valued A initiative and technical excellence, and put together a team that / execute that vision in the Newisys environment.   F During our first five years, the Newisys team has produced world classF Opteron servers (the Newisys 2100 and 4300 servers), has developed theB next generation of out of band system management function, and has6 taken Opteron beyond its scaling limits (4-8 sockets).  0 The Newisys team continues to execute very well.  @ It was simply time for me to apply my talents to new challenges.   Rich   ---   C I think there's a company and a man with confidence in their future 	 products.  --  
 Alan Greig   ------------------------------  % Date: Sat, 30 Apr 2005 17:48:08 -0400 ' From: Dave Froble <davef@tsoft-inc.com> A Subject: Re: Maybe HP should get out of the hardware business.... 0 Message-ID: <1177v8kb5qtsn04@corp.supernews.com>   Alan Greig wrote:  > Dave Froble wrote: >  >>Alan Greig wrote:  >>B >>HP controlling AMD would be the kiss of death.  Several reasons. >  > F > Didn't say they should do it just that it would be a better bet thanD > buying all the Itanium rights and engineers back from Intel. WhichE > would be a terrible idea and likely result in even more pain for HP  > product lines. >  > E >>AMD is just becoming acceptable to many, and having a rival control  >  > AMD  >  >>would quickly end that.  >  > 2 > Not if HP got out of the desktop PC business :-)  I Dell will probably take care of that.  Unless they continue to be stupid  
 about AMD.   --  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: Sat, 30 Apr 2005 17:55:57 -0400 ' From: Dave Froble <davef@tsoft-inc.com> A Subject: Re: Maybe HP should get out of the hardware business.... 0 Message-ID: <1177vncn2uapvd5@corp.supernews.com>   Karsten Nyblad wrote:  > Alan Greig wrote:  > E >> And right now it would probably be a better bet for HP to buy AMD.  >  > J > Hmmmm.  One of the costs of developing computer chips is developing new B > semiconductor processes suitable for high speed microprocessors.  G Yes, I usually don't seperate the two, but definitely, the tech needed  J to etch silicon and such has not a damn thing to do with processor design.  D There is some middle ground where processor designers must know the H capabilities of the various processes.  A good area for finger pointing.  F Still, one would think that a foundry that appriciates customers, and H customers that are willing to work closely with a foundry is a workable 
 situation.  F The computer industry didn't grow up that way, and maybe it couldn't, D but with hindsight, there are probably a few processor designs that ' would be viable today that are history.    --  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: Sat, 30 Apr 2005 18:00:06 -0400 ' From: Dave Froble <davef@tsoft-inc.com> A Subject: Re: Maybe HP should get out of the hardware business.... 0 Message-ID: <1177vv5ql6cmed0@corp.supernews.com>   Alan Greig wrote:  > Karsten Nyblad wrote:  >  > D >>Newisys normally lives from designing complete systems that others >  > sells  > E >>as OEM.  My guess is that the marked for Opteron based systems with  >  > more > G >>than 8 processors is too small for firms to design their own systems.  >  >  I > E >>would wait and see if Newisys can make other firms sell Horus based  >>systems on OEM basis.  >  > : > AMD liked Horus so much they just bought the designer... > Q > http://www.amd.com/us-en/Corporate/VirtualPressRoom/0,,51_104_543~96057,00.html  > D > "Rich Oehler, Renowned Microprocessor and Multi-Processing Systems* > Architect, joins AMD as Corporate FellowB > -Oehler, former IBM Fellow and chief technology officer (CTO) at: > Newisys, will focus on increasing the alignment of AMD'sE > enterprise-class computing strategies with the needs of the Fortune  > 1000-  > H > SUNNYVALE, CA -- March 28, 2005 --AMD (NYSE: AMD) announced today that< > Rich Oehler, a 40-year industry veteran well known for hisG > groundbreaking work in microprocessor, symmetric multi-processing and F > mission-critical systems design, has joined AMD's Office of Strategy- > and Technology reporting to CTO Fred Weber.  > B > "It is a real honor to have Rich join the AMD team," said Weber.E > "His unique depth of experience in enterprise computing enhances an ? > already immensely talented group focused on strengthening the H > performance, reliability, availability and scalability features of our > 64-bit solutions." > B > Oehler will have broad responsibility for closely aligning AMD's@ > multi-core, enterprise-class computing strategies with the topG > priorities of Fortune 1000 Chief Information Officers and Information  > Technology management."  > ---  > , > And in a recent post in comp.arch he says: >   > Rich Oehler 	  Mar 31, 7:28 am > Newsgroups: comp.arch + > From: "Rich Oehler" <rich.oeh...@amd.com> " > Date: 31 Mar 2005 07:28:25 -0800" > Local: Thurs,Mar 31 2005 7:28 am$ > Subject: Re: Rich Oehler joins AMD >  > ? > Let me assure you that the Horus effort at Newisys continues.  > D > At Newisys, I accomplished what I wanted to do: define a technicalE > vision for Opteron based servers, create an environment that valued C > initiative and technical excellence, and put together a team that 1 > execute that vision in the Newisys environment.  > H > During our first five years, the Newisys team has produced world classH > Opteron servers (the Newisys 2100 and 4300 servers), has developed theD > next generation of out of band system management function, and has8 > taken Opteron beyond its scaling limits (4-8 sockets). > 2 > The Newisys team continues to execute very well. > B > It was simply time for me to apply my talents to new challenges. >  > Rich >  > ---  > E > I think there's a company and a man with confidence in their future  > products.   C I think that anyone that thinks that AMD isn't aiming right at the  = target Intel wanted to hit had better re-think their opinion.   * Possibly the spirit of Alpha does live on.   --  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: Sat, 30 Apr 2005 18:03:16 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> A Subject: Re: Maybe HP should get out of the hardware business.... B Message-ID: <1114898597.a224847d656ccad7084f8f32bc9ce380@teranews>   Karsten Nyblad wrote: I > Hmmmm.  One of the costs of developing computer chips is developing new I > semiconductor processes suitable for high speed microprocessors.  Intel H > can let their x86 business pay for that, but AMD barely have the money	 > for it,     : This is one big misundestanding. Chips is more than CPUs.   F Alpha's last years was a very good example. Compaq/Digital didn't haveD to worry about process. It would find a supplied of FABbing servicesE such as Intel, Samsung, IBM etc who had good process, and then supply F them with the blueprints for the alpha achip and ask them to produce a! few thousand (or whatever number)   E AMD has 2 businesses: the FABbing and the designing. Just because you D only hear about Opteron doesn't mean that their plants don't produceE other chips for other customers. Just like IBM's plants don't produce  only Power chips.   F It s true however that CPUs do require the latest and greatest in chipH manufacturing process. But if you have a 65nm plant that is only 50% useD for chips, it doesn't mean you can't use it to produce lower qualityG stuff such as flash memory etc in the spare capacity your plant offers.    ------------------------------  % Date: Sat, 30 Apr 2005 18:02:56 -0400 ' From: Dave Froble <davef@tsoft-inc.com> A Subject: Re: Maybe HP should get out of the hardware business.... 0 Message-ID: <117804edgqdl579@corp.supernews.com>   Alan Greig wrote:   : > AMD liked Horus so much they just bought the designer...  F Bad form some say to follow one's own post, but, after I hit the send C button, it occured to me, (yeah, I'm slow), that AMD wouldn't hire  G someone like Rich Oehler without having some purpose in mind.  They've  I already got Opteron doing well without him.  Wonder what's on their mind?    --  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: Sun, 01 May 2005 06:17:44 +0200 + From: Karsten Nyblad <nospam@nospam.nospam> A Subject: Re: Maybe HP should get out of the hardware business.... = Message-ID: <42745875$0$78288$157c6196@dreader1.cybercity.dk>    Dave Froble wrote: > Karsten Nyblad wrote:  >  >> Alan Greig wrote: >>F >>> And right now it would probably be a better bet for HP to buy AMD. >> >> >>G >> Hmmmm.  One of the costs of developing computer chips is developing  G >> new semiconductor processes suitable for high speed microprocessors.  >  > I > Yes, I usually don't seperate the two, but definitely, the tech needed  L > to etch silicon and such has not a damn thing to do with processor design. > F > There is some middle ground where processor designers must know the J > capabilities of the various processes.  A good area for finger pointing. > H > Still, one would think that a foundry that appriciates customers, and J > customers that are willing to work closely with a foundry is a workable  > situation.  C The problems is that you needed a foundry that is willing to share  H process information with their close costumers years before the process H is ready to be used in production.  Further, you need a foundry that is ? willing to optimize their processes for microprocessor designs.   H > The computer industry didn't grow up that way, and maybe it couldn't, F > but with hindsight, there are probably a few processor designs that ) > would be viable today that are history.  > I The semiconductor processes were developed for, e.g., DRAM until the mid  I 90's.  It is only after those days the latest and greatest semiconductor  6 processes have had their first use on microprocessors.   ------------------------------  % Date: Sat, 30 Apr 2005 17:31:18 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 9 Subject: Re: more questions on Cyrillic fonts and Mozilla B Message-ID: <1114896698.6a7c7fb5574f1cc390bba1f59efeaa50@teranews>  / Phillip Helbig---remove CLOTHES to reply wrote:   So Mozilla is8 > definitely finding the fonts, but it is using only one > size/strength/shape. >  > What am I missing?    G You'll need to go through Mozilla's documentation and/or resource files B to get a hint of what you need to do next to provide for a greater? variety of font support.  Do you know which font Mozilla uses ? E (exactly) ? This may give you some hint on how Mozilla decides to use 
 that font.  D Based on your table, if, for instance, you said you wanted a 10 pontF font by default, then Mozilla might simply use the one font that is at 10 points in your list.   L Seems to me that this is now at an application level, not the os or X level.   ------------------------------    Date: 30 Apr 2005 17:57:54 -0700 From: Kevin.R.Morris@gmail.com2 Subject: Re: MySQL and PHP Authentication ProtocolC Message-ID: <1114909074.898743.169960@g14g2000cwa.googlegroups.com>   G Yes, actually the VAMP board is where I found the tidbit of information ? that pointed me to the answer that got me over the finish line.    Thanks,  Kevin    ------------------------------   Date: 1 May 2005 03:50:02 GMT  From: healyzh@aracnet.com " Subject: OpenVMS and Mac OS X 10.4+ Message-ID: <d51jla0rbp@enews3.newsguy.com>   J I ended up getting 10.4 a lot sooner than I intended to, and realistically! sooner than I feel I should have.   I One big problem for me is that it would appear that it no longer supports K classic Appletalk shares, so I can no longer access the Appletalk shares on H my OpenVMS 7.3-2 system.  Realistically I should have taken care of thisK issue on my own and stopped using it a few years ago, and it was one of the H issues that kept me on Mac OS 9 for as long as I was (initially Mac OS XK didn't support them).  Still I find it odd that they appear to have removed > the support, since they went to the trouble to add it back in.  K At this point the largest issue for me is some fairly bad instability in my L apps.  I've had Eudora, Safari, and quite a few xterms blow up on me today. E I find this *very* disturbing, and have thought about rolling back to   10.3.9, though I probably won't.  B On a possitive note, they look to have updated font handling underG X-Windows, and while xterm continues to lack support for the -xrm flag, K while using my modified .Xmodmap file, I have very good VT emulation now in H my xterms.  I couldn't believe it when I saw that doublewidth fonts wereK working!!!  Fonts also seem to be handled better on the one X-Windows app I  use regularly besides xterm.  H On the Mac specific side, Mail.app is still to underpowered for me.  I'mL quite honestly not sure what I think of Spotlight (their new document searchK feature).  I've a feeling it's going to be one of thos "ask me in 2-3 years L sorts of things).  On a surprising note, I think I may actually come to like7 Dashboard (up until today I thought it sounded stupid).    		Zane   ------------------------------  % Date: Sat, 30 Apr 2005 18:16:12 -0400 ' From: Dave Froble <davef@tsoft-inc.com> - Subject: Re: Sandia says alpha the best chip! 0 Message-ID: <11780tln2hhf20f@corp.supernews.com>   Alan Greig wrote:  > Dave Froble wrote: >  >>icerq4a@spray.se wrote:  >>7 >>>Most Alpha people are working Itanium, some on Xeon, D >>>and they are especially on the future common system architecture. >>>  >>? >>Recent news is that the ex-Alpha people were thrown off their  > 
 > project. > < >>Because the ex-HP people didn't like what they were doing. >  > G > Likely Intel just needed an excuse to move its best people (the Alpha G > folks) into its top priority X86-64 projects. Let the ex-HP engineers * > fiddle with the death rattle of Itanium. >   H That isn't how it sounded.  The HP people didn't like the processor the  Alpha people were developing.   ) http://www.theinquirer.net/?article=20286   I As we reported earlier today from a different source, HP didn't like the  C Alpha team's Tukwila design, so while it will still use the system  I interface unit that team designed, the new Tukwila duck will be designed   in Fort Collins.  F The HP Fort Collins team, it has emerged, had tried to kill the Alpha G Tukwila as long ago as two years, and visited Massachusetts armed with  F numbers showing the design ran too hot and wouldn't get anywhere near  the performance it wanted.  F Still, it appears to have taken two years for the HP designers to get H their way. Unfortunately, the HP designers might not have realised soon D enough that soon enough they'd be working for the Intel Corporation.   --  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: Sat, 30 Apr 2005 13:58:48 -0400 ' From: Dave Froble <davef@tsoft-inc.com> % Subject: Re: Setting up an NTP server 0 Message-ID: <1177hqmscl2hfef@corp.supernews.com>   David J Dachtera wrote:    > $ > Try (text retrieved from W/98-SE): >  > C:\> NET/? > 6 > For more information about a specific Microsoft NET 0 > command, type the command name followed by /?  > (for example, NET VIEW /?).  > 8 > NET CONFIG   Displays your current workgroup settings.@ > NET DIAG     Runs the Microsoft Network Diagnostics program toA >              display diagnostic information about your network. 6 > NET HELP     Provides information about commands and >              error messages.A > NET INIT     Loads protocol and network-adapter drivers without 2 >              binding them to Protocol Manager.  > > NET LOGOFF   Breaks the connection between your computer and= >              the shared resources to which it is connected. 9 > NET LOGON    Identifies you as a member of a workgroup. + > NET PASSWORD Changes your logon password. 6 > NET PRINT    Displays information about print queues' >              and controls print jobs.  > NET START    Starts services.  > NET STOP     Stops services.D > NET TIME     Displays the time on or synchronizes your computer's ? >              clock with the clock on a Microsoft Windows for  J >              Workgroups, Windows NT, Windows 95, or NetWare time server.7 > NET USE      Connects to or disconnects from a shared 5 >              resource or displays information about  >              connections. ; > NET VER      Displays the type and version number of the  3 >              workgroup redirector you are using.  6 > NET VIEW     Displays a list of computers that share6 >              resources or a list of shared resources& >              on a specific computer. >  > C:\> NET TIME /? > 7 > Displays the time on or synchronizes your computer's  9 > clock with the shared clock on a Microsoft Windows for  0 > Workgroups, Windows NT, Windows 95, or NetWare > time server. > 9 > NET TIME [\\computer | /WORKGROUP:wgname] [/SET] [/YES]  > 8 >   computer    Specifies the name of the computer (time7 >               server) whose time you want to check or 7 >               synchronize your computer's clock with. > >   /WORKGROUP  Specifies that you want to use the clock on a < >               computer (time server) in another workgroup.> >   wgname      Specifies the name of the workgroup containing; >               a computer whose clock you want to check or : >               synchronize your computer's clock with. If7 >               there are multiple time servers in that : >               workgroup, NET TIME uses the first one it  >               finds.; >   /SET        Synchronizes your computer's clock with the 6 >               clock on the computer or workgroup you >               specify.8 >   /YES        Carries out the NET TIME command without= >               first prompting you to provide information or   >               confirm actions. > A > The text from NET on W2K is rather more terse and less helpful.  > J > As to what constitutes "decent documentation" these days, that's another > topic entirely.  >   I Well, you're a glutton for punishment if you still have anything running  	 W98.  :-)   H I've been through the above before.  Nowhere does it say that it's more < than a one time thing.  Systems seem to not be continuously  synchronizing the time.   D Stan Quayle posted some good (I hope) info.  I do hate the registry.   Thanks   --  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: Sat, 30 Apr 2005 14:01:46 -0400 ' From: Dave Froble <davef@tsoft-inc.com> % Subject: Re: Setting up an NTP server 0 Message-ID: <1177i06c5ot842d@corp.supernews.com>   Dirk Munk wrote: > Dave Froble wrote: >  > <snip> > I >> Now if I could only figure out how to get the Windows 2000 systems to  ! >> read the time from the server.  >> >> Dave  >>G > doubleclick on the clock of your Windows system, and you will find a  C > Internet time provider tab where you can specify the sntp server.   G Just tried that.  Windows 2000 Pro, service pack 4.  Nothing like what  ? you describe.  Are you possibly talking about an XP capability?    --  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: Sat, 30 Apr 2005 22:40:53 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net>% Subject: Re: Setting up an NTP server + Message-ID: <42744FC5.72911DFE@comcast.net>    Dave Froble wrote: > [snip]J > Well, you're a glutton for punishment if you still have anything running > W98.  :-)   H Believe it or else, I just "recently" "up"graded from W95-OSR2. Frankly,B the prior was *VASTLY* more stable and less troublesome, even when running Interhose Exploder.   I > I've been through the above before.  Nowhere does it say that it's more = > than a one time thing.  Systems seem to not be continuously  > synchronizing the time.   G I recall something about a scheduler ("Agent?"). Could be set up to run  periodically from a .BAT file.  > Also, found a similar reference on W2K, but allows for (S)NTP:   The syntax of this command is:   C:\> net time /?  1 NET TIME [\\computername | /DOMAIN[:domainname] |  /RTSDOMAIN[:domainname]] [/SET] $          [\\computername] /QUERYSNTP4          [\\computername] /SETSNTP[:ntp server list]  $ C:\> net time /setsntp:132.163.4.103  F ...worked for me. I'm looking for the scheduling agent now to see if IA can set it up to run at least once a day. Then, I'll see if I can E convince W2K to share that info with my W/9x machines if they issue a B NET TIME /SET command pointing to the W2K machine (so far, no luck there).    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  % Date: Sat, 30 Apr 2005 14:06:54 -0400 ' From: Dave Froble <davef@tsoft-inc.com>   Subject: Re: Slow Filesystem I/O0 Message-ID: <1177i9ufeusmka7@corp.supernews.com>   Thomas OToole wrote: >  > Bill Todd wrote: >  >>Rob Young wrote: >> >>...  >>; >>   Newer hardware has masked the problem you describe for  >>. >>>      the most part (smallness of buffers). >>I >>If so, it would seem strange that people are still complaining about it I >>then.  Rather, it seems more likely that you're just typically clueless = >>- a question which, fortunately, we can examine right here.  >  > B > Yet it's UNfortunate that you don't spend even a fraction of theF > considerable bandwidth you bring to bear bashing Rob doing somethingG > much more useful, like pissing on bathgatesOS. Considering the thread C > started with a complaint about a backported gartner(TM) compliant I > billyapp, the least you should do is provide comment about the numerous  > other causes of G > inefficiency within such an app. YOU are most unfair, considering the J > opportunies for tuning that VMS provides at both the application and theH > system level, which, no doubt, have not even been considered, and thatJ > goes double considering the billions of dollars of thrust that have beenF > applied getting the pig (apologies to pigs everywhere) known as billO > gates airborne. So what's the deal? Are you pwned by the bathgates syndicate?  >  > -Tom O'Toole  F I gotta sort of agree.  While Rob sometimes does warrant ridicule, in C this case I think he mentioned some good points.  I don't think he   deserved it this time.  I Are you letting it get personal Bill, or perhaps acting on reflex?  "Oh,  > it's Rob, gotta dump on him."  Care to re-think your response?   --  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: Sat, 30 Apr 2005 14:36:26 -0400 ( From: Bill Todd <billtodd@metrocast.net>  Subject: Re: Slow Filesystem I/O= Message-ID: <h_GdnUyuS5m2Te7fRVn-2g@metrocastcablevision.com>    Dave Froble wrote: > Thomas OToole wrote: >  >> >> Bill Todd wrote:  >> >>> Rob Young wrote: >>>  >>> ...  >>> < >>>   Newer hardware has masked the problem you describe for >>> / >>>>      the most part (smallness of buffers).  >>>  >>> K >>> If so, it would seem strange that people are still complaining about it K >>> then.  Rather, it seems more likely that you're just typically clueless ? >>> - a question which, fortunately, we can examine right here.  >> >> >>C >> Yet it's UNfortunate that you don't spend even a fraction of the G >> considerable bandwidth you bring to bear bashing Rob doing somethingeH >> much more useful, like pissing on bathgatesOS. Considering the threadD >> started with a complaint about a backported gartner(TM) compliantJ >> billyapp, the least you should do is provide comment about the numerous >> other causes ofH >> inefficiency within such an app. YOU are most unfair, considering theK >> opportunies for tuning that VMS provides at both the application and theuI >> system level, which, no doubt, have not even been considered, and that K >> goes double considering the billions of dollars of thrust that have beenhG >> applied getting the pig (apologies to pigs everywhere) known as billvF >> gates airborne. So what's the deal? Are you pwned by the bathgates 
 >> syndicate?e >> >> -Tom O'Toole  >  > H > I gotta sort of agree.  While Rob sometimes does warrant ridicule, in E > this case I think he mentioned some good points.  I don't think he i > deserved it this time. > K > Are you letting it get personal Bill, or perhaps acting on reflex?  "Oh, S@ > it's Rob, gotta dump on him."  Care to re-think your response?  H No - but it sounds as if you might benefit from re-reading it until you  understand it better.r   - bill   ------------------------------    Date: 30 Apr 2005 15:18:17 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)i  Subject: Re: Slow Filesystem I/O3 Message-ID: <M4YCghb+FLtE@eisner.encompasserve.org>i  ` In article <4273A95E.4B2ECDD5@comcast.net>, David J Dachtera <djesys.nospam@comcast.net> writes: > Thomas OToole wrote: >> e >> Larry Kilgallen wrote:i >> >v >> > In article <24xGylcV4xfC@eisner.encompasserve.org>, koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:e >> > > In article <slrnd6nk9v.78k.usenet@gaia.roc2.gblx.net>, Dan Foster <usenet@evilphb.org> writes:- >> > >>-K >> > >> That's a shame (Spiralog). Wonder why then-DEC dropped it if it wasr >> > >> pretty good? >> > >K >> > >    They dropped it because it didn't work.  You could get impressive5K >> > >    speed in laboratory conditions.  In the real world you didn't.  ArI >> > >    nasty problem showed up and there was no real justification forh6 >> > >    fixing it with no real benefit to customers. >> >: >> > Spiralog was an ideal concept for write-mostly files. >> >H >> > When they started talking about it at DECUS some of us wondered who; >> > had write-mostly applications.  Apparently nobody did." >> l4 >> Like disk to disk backup or database journaling?? > I > What would the advantage of Spiralog (versus the existing) have been ina > those cases?  = Well for the disk-to-disk backup it would be a way to prevent ; using the backup disk as a replacement, since it would have6& different performance characteristics.   ------------------------------  % Date: Sat, 30 Apr 2005 18:23:09 -0400.' From: Dave Froble <davef@tsoft-inc.com>g  Subject: Re: Slow Filesystem I/O0 Message-ID: <11781acet3nuaf7@corp.supernews.com>   Bill Todd wrote: > Dave Froble wrote: >  >> Thomas OToole wrote:A >> >>>  >>> Bill Todd wrote: >>>n >>>> Rob Young wrote:h >>>> >>>> ... >>>>= >>>>   Newer hardware has masked the problem you describe foro >>>>0 >>>>>      the most part (smallness of buffers). >>>> >>>> >>>>L >>>> If so, it would seem strange that people are still complaining about itL >>>> then.  Rather, it seems more likely that you're just typically clueless@ >>>> - a question which, fortunately, we can examine right here. >>>r >>>e >>>. >>> D >>> Yet it's UNfortunate that you don't spend even a fraction of theH >>> considerable bandwidth you bring to bear bashing Rob doing somethingI >>> much more useful, like pissing on bathgatesOS. Considering the thread E >>> started with a complaint about a backported gartner(TM) compliantlK >>> billyapp, the least you should do is provide comment about the numerouse >>> other causes of I >>> inefficiency within such an app. YOU are most unfair, considering theeL >>> opportunies for tuning that VMS provides at both the application and theJ >>> system level, which, no doubt, have not even been considered, and thatL >>> goes double considering the billions of dollars of thrust that have beenH >>> applied getting the pig (apologies to pigs everywhere) known as billG >>> gates airborne. So what's the deal? Are you pwned by the bathgates a >>> syndicate? >>>a >>> -Tom O'Toole >> >> >>I >> I gotta sort of agree.  While Rob sometimes does warrant ridicule, in  F >> this case I think he mentioned some good points.  I don't think he  >> deserved it this time.c >>G >> Are you letting it get personal Bill, or perhaps acting on reflex?  uF >> "Oh, it's Rob, gotta dump on him."  Care to re-think your response? >  > J > No - but it sounds as if you might benefit from re-reading it until you  > understand it better.  >  > - bill >   D Well, Ok, the way I read it is that some of the hardware on top end B stuff is able to mask problems with writing large amounts of data C quickly.  Get a bunch of cache and battery backup and you can mask  D slower write performance, whether based on the filesystem design or H other.  Only problem would be if you rammed the data at the storage for ( a long period of time, say several days.   What did I miss?   -- e4 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: Sat, 30 Apr 2005 18:26:18 -0400 ' From: Dave Froble <davef@tsoft-inc.com>p  Subject: Re: Slow Filesystem I/O0 Message-ID: <11781g93qcd27ce@corp.supernews.com>   Larry Kilgallen wrote:b > In article <4273A95E.4B2ECDD5@comcast.net>, David J Dachtera <djesys.nospam@comcast.net> writes: >  >>Thomas OToole wrote: >> >>>Larry Kilgallen wrote:m >>>Ou >>>>In article <24xGylcV4xfC@eisner.encompasserve.org>, koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:s >>>>c >>>>>In article <slrnd6nk9v.78k.usenet@gaia.roc2.gblx.net>, Dan Foster <usenet@evilphb.org> writes:r >>>>>,I >>>>>>That's a shame (Spiralog). Wonder why then-DEC dropped it if it wasl >>>>>>pretty good? >>>>>yI >>>>>   They dropped it because it didn't work.  You could get impressiveiI >>>>>   speed in laboratory conditions.  In the real world you didn't.  ArG >>>>>   nasty problem showed up and there was no real justification forr4 >>>>>   fixing it with no real benefit to customers. >>>>9 >>>>Spiralog was an ideal concept for write-mostly files.r >>>>G >>>>When they started talking about it at DECUS some of us wondered whor: >>>>had write-mostly applications.  Apparently nobody did. >>> 4 >>>Like disk to disk backup or database journaling?? >>I >>What would the advantage of Spiralog (versus the existing) have been inc >>those cases? >  > ? > Well for the disk-to-disk backup it would be a way to prevente= > using the backup disk as a replacement, since it would haveb( > different performance characteristics.  B The concepts behind Spiralog work well for a system used for data  acquisition.  F Most of us use the system more as a source of data, not a destination.   --  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 Roadr Vanderbilt, PA  15486c   ------------------------------  % Date: Sat, 30 Apr 2005 21:27:06 -0400t( From: Bill Todd <billtodd@metrocast.net>  Subject: Re: Slow Filesystem I/O= Message-ID: <OumdnfxbLvP2renfRVn-hQ@metrocastcablevision.com>i   Dave Froble wrote:   ...i  F > Well, Ok, the way I read it is that some of the hardware on top end D > stuff is able to mask problems with writing large amounts of data E > quickly.  Get a bunch of cache and battery backup and you can mask 0F > slower write performance, whether based on the filesystem design or J > other.  Only problem would be if you rammed the data at the storage for * > a long period of time, say several days. >  > What did I miss?  D 1.  Stable write-back controller cache can *help* write performance G somewhat, but (as Rob himself noted) it still has latency on the order rI of a millisecond, so *at best* it can improve upon the latency of a fast  H disk by a factor of less than 10.  Write-back system cache, as I noted, I can improve upon the latency of a fast disk by a factor of 100 or more -  F plus whatever additional gains you get from not having to write out a I significant amount of data at all (the data that gets deleted while it's uF still in the cache), plus whatever the far greater write bandwidth to 7 system cache gets you in terms of avoiding bottlenecks.i  E Rob's contention was not the entirely reasonable one that write-back e5 controller cache can *help* mask *some* of the major rF default-write-performance handicap that VMS has compared with systems B that use larger and better-streamed write-back buffering:  it was C (first) that it masked the problem "for the most part" and (later)  F flatly that "the 'problem' of a slow filesystem is a non-problem" and H "Won't matter how slow the filesystem is, the hardware will be there to ? mask it".  It's pretty incompetent to claim that spotting your rF competition a 10:1 write-latency advantage (and an even greater write F bandwidth advantage) is a 'non-problem' simply because you've reduced D that handicap from more like 100:1 if you didn't use the controller B caching at all - but then I'm guessing that Rob doesn't have much F experience with modern Unix performance (and possibly that you don't,  either).  F 2.  Your 'long period of time' is nothing like 'several days'.  Let's B take RMS's pathetic 8 KB default buffer as an example.  If you're A stuffing sequential-file data at the disk as fast as RMS can (by tH default) destage it synchronously, you're placing that 8 KB on the disk D   each disk revolution (i.e., every 4 - 6 ms. with a 15K or 10K rpm ? FC/SCSI drive).  That's a data rate of a measly 1.3 - 2 MB/sec.d  F Well, that measly 1.3 - 2 MB/sec amounts to a not-so measly 4.7 - 7.2 G GB/hour, and that's just the data rate that you want to *improve upon* iF by using the write-back controller cache.  So if, for example, you're G shooting just for the 10x improvement that Rob was boasting about (not oI the 100x improvement you can get by using a system cache) you're talking hA about gobbling up 40 - 60 GB of cache, per hour, per application eA (because since you're accepting data about 5x faster than you'll  ? eventually be able to get rid of it to disk after taking queue xE optimizations into account, most of the data that you stuff into the i= cache during the hour will still be there at the hour's end).-  B No, not all applications will stress the controller cache to this G degree, so by all means decrease that amount by as much as an order of oG magnitude (as long as you just want to cater to *average* environments .I rather than to the most demanding ones, anyway):  then on average you'll mI fill up only 4 - 6 GB of cache per hour, per application.  Just how much >H (mirrored and battery-backed, of course) cache did you say you had, and E just how much more expensive was it than unmirrored, volatile system oI memory which also, not so incidentally, could be put to better uses when f they had priority?  I 3.  Then comes the question of what happens when your long (or, as noted gD above, perhaps not-so-long) period of time is up, and your cache is B full:  what does your data rate drop back to then?  Well, in this A example, right back to 2.6 - 4 MB/sec (I'm assuming that queuing aF optimizations will be able to double the IOPS for the disk, though if C the disk is being shared with other applications both this and the AF original 1.3 - 2 MB/sec figures should be reduced somewhat to account C for additional seek overhead) - and that's where Spiralog came in, DF because (even as implemented, and I believe that improvements on that H were possible) it should have been able to sustain 20 - 40 MB/sec under I that kind of load (the disks we're talking about here are today's disks,  H not those from Spiralog's era), which is where my comment about happily H buying 5x - 20x as much disk just to compensate for the deficiencies of $ VMS's default data-handling came in.  @ The bottom line is that hardware cannot *wholly* compensate for H poorly-designed data-handling, and just compensating *partially* is not H cheap (large amounts of mirrored, battery-backed controller cache, many H times as many disks - these are not down-in-the-noise expenses, even in  a high-end system).t  I Boasting that VMS doesn't perform as pitifully as it used to compared to  F Unix if you throw enough expensive hardware at it (which Unix doesn't C require) is not exactly the best way to win converts, I'll suggest.p  F You're right:  I don't like Rob.  While I don't consider him to be as I much of an active sleaze as I've found Kerry and often Terry to be, they lD all vigorously defended the Alphacide while studiously ignoring (or E actively attempting to spin away) mounting evidence that it was just  ? what we suspected it was, and continue to do so to this day by  F uncritically touting Itanic (though I've heard that Terry may finally F have seen something of the light in this area).  Kerry has an obvious F job-related motive for pimping for cHumPaq and my impression was that I Terry hoped to acquire one, whereas Rob's motivation is less clear:  for pH some reason he seems to develop organizational allegiances which cannot ; be weakened by any factual evidence - which I guess may be  E characteristic of faith-based rather than analytical minds but which mH still doesn't explain the strength of the allegiance in the first place.  G If Rob were simply an uncritical, enthusiastic supporter of the things tE that he (for whatever reason) decides are worthy of his support, I'd mF find him a lot more tolerable.  It's his continual attempt to portray > them as *relatively* far better than they are compared to the I competition which causes me to slap him down hard, though without any of iH the distortions (intentional or sloppy - it's often hard to tell) which H he himself so readily indulges in to support his (current) pet products E (as we saw with Alpha, while his allegiance to organizations remains rE steadfast, his allegiance to particular products can turn on a dime).u  F I really don't give a damn whether people like the way I treat Rob or F not:  as with cHumPaq, I'll stop when, and only when, I see some real F attempt to make amends for what are (going on 4 years now) continuing E transgressions (not to suggest that Rob's are anything comparable in  G significance to cHumPaq's, but they both exude a similar stench).  But  F I'm as careful with the facts when responding to Rob as I am in other E situations - not that I give a damn about people who can't recognize a
 that, either.o   - bill   ------------------------------  % Date: Sat, 30 Apr 2005 23:03:28 -0400 ' From: "Main, Kerry" <kerry.main@hp.com>   Subject: RE: Slow Filesystem I/OR Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB5ECEFC@tayexc19.americas.cpqcorp.net>   > -----Original Message-----4 > From: Bill Todd [mailto:billtodd@metrocast.net]=20 > Sent: April 30, 2005 9:27 PM > To: Info-VAX@Mvb.Saic.Com " > Subject: Re: Slow Filesystem I/O >=20 > Dave Froble wrote: >=20 > ...  >=20J > > Well, Ok, the way I read it is that some of the hardware on top end=20H > > stuff is able to mask problems with writing large amounts of data=20I > > quickly.  Get a bunch of cache and battery backup and you can mask=20aJ > > slower write performance, whether based on the filesystem design or=20B > > other.  Only problem would be if you rammed the data at the=20 > storage for=20, > > a long period of time, say several days. > >=20 > > What did I miss? >=20H > 1.  Stable write-back controller cache can *help* write performance=20A > somewhat, but (as Rob himself noted) it still has latency on=20a > the order=20; > of a millisecond, so *at best* it can improve upon the=20  > latency of a fast=20@ > disk by a factor of less than 10.  Write-back system cache,=20 > as I noted,=20? > can improve upon the latency of a fast disk by a factor of=20t > 100 or more -=20J > plus whatever additional gains you get from not having to write out a=20B > significant amount of data at all (the data that gets deleted=20 > while it's=20 J > still in the cache), plus whatever the far greater write bandwidth to=209 > system cache gets you in terms of avoiding bottlenecks.0 >=20I > Rob's contention was not the entirely reasonable one that write-back=20y9 > controller cache can *help* mask *some* of the major=20-J > default-write-performance handicap that VMS has compared with systems=20F > that use larger and better-streamed write-back buffering:  it was=20G > (first) that it masked the problem "for the most part" and (later)=20HJ > flatly that "the 'problem' of a slow filesystem is a non-problem" and=20@ > "Won't matter how slow the filesystem is, the hardware will=20 > be there to=20C > mask it".  It's pretty incompetent to claim that spotting your=20 J > competition a 10:1 write-latency advantage (and an even greater write=20J > bandwidth advantage) is a 'non-problem' simply because you've reduced=20H > that handicap from more like 100:1 if you didn't use the controller=20F > caching at all - but then I'm guessing that Rob doesn't have much=20J > experience with modern Unix performance (and possibly that you don't,=20
 > either). >=20  E Mmmm... You do realize VMS also supports write back caching now via a   dynamic sysgen parameter, right?H http://h71000.www7.hp.com/doc/731FINAL/6657/6657pro_007.html (5.11.1 and 5.11.3)e   Or - on VMS V7.3-1 or later:! SYSGEN>  HELP SYS RMS_SEQFILE_WBH   C The doc's simply state that you need to be careful using it becausetF should a disk path error occur after the application has been told theG write has already been completed, then you may have some data integrityy issues to deal with.   [snip..]   >=20J > You're right:  I don't like Rob.  While I don't consider him to be as=20A > much of an active sleaze as I've found Kerry and often Terry=20f > to be, they=20H > all vigorously defended the Alphacide while studiously ignoring (or=20I > actively attempting to spin away) mounting evidence that it was just=20mC > what we suspected it was, and continue to do so to this day by=20.J > uncritically touting Itanic (though I've heard that Terry may finally=20J > have seen something of the light in this area).  Kerry has an obvious=20J > job-related motive for pimping for cHumPaq and my impression was that=20A > Terry hoped to acquire one, whereas Rob's motivation is less=20  > clear:  for=20? > some reason he seems to develop organizational allegiances=20n > which cannot=20k? > be weakened by any factual evidence - which I guess may be=20eI > characteristic of faith-based rather than analytical minds but which=20,@ > still doesn't explain the strength of the allegiance in the=20 > first place. >=20   Bill .. Bad hair day, eh?    :-)t   [Snip..]  
 Kerry Main Senior Consultantn HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477- kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  $ "OpenVMS has always had integrity .. Now, Integrity has OpenVMS .."   ------------------------------  % Date: Sat, 30 Apr 2005 23:28:47 -0400r( From: Bill Todd <billtodd@metrocast.net>  Subject: Re: Slow Filesystem I/O= Message-ID: <BaidnVh9Nett0enfRVn-uw@metrocastcablevision.com>f   Main, Kerry wrote: >>-----Original Message-----2 >>From: Bill Todd [mailto:billtodd@metrocast.net]  >>Sent: April 30, 2005 9:27 PM >>To: Info-VAX@Mvb.Saic.Com " >>Subject: Re: Slow Filesystem I/O >> >>Dave Froble wrote: >> >>...  >> >>G >>>Well, Ok, the way I read it is that some of the hardware on top end eE >>>stuff is able to mask problems with writing large amounts of data sF >>>quickly.  Get a bunch of cache and battery backup and you can mask G >>>slower write performance, whether based on the filesystem design or e? >>>other.  Only problem would be if you rammed the data at the a >> >>storage for  >>+ >>>a long period of time, say several days.n >>>I >>>What did I miss?S >>F >>1.  Stable write-back controller cache can *help* write performance ? >>somewhat, but (as Rob himself noted) it still has latency on i >>the order 9 >>of a millisecond, so *at best* it can improve upon the t >>latency of a fast > >>disk by a factor of less than 10.  Write-back system cache,  >>as I noted, = >>can improve upon the latency of a fast disk by a factor of 9 >>100 or more - H >>plus whatever additional gains you get from not having to write out a @ >>significant amount of data at all (the data that gets deleted 
 >>while it's hH >>still in the cache), plus whatever the far greater write bandwidth to 9 >>system cache gets you in terms of avoiding bottlenecks.i >>G >>Rob's contention was not the entirely reasonable one that write-back -7 >>controller cache can *help* mask *some* of the major cH >>default-write-performance handicap that VMS has compared with systems D >>that use larger and better-streamed write-back buffering:  it was E >>(first) that it masked the problem "for the most part" and (later) rH >>flatly that "the 'problem' of a slow filesystem is a non-problem" and > >>"Won't matter how slow the filesystem is, the hardware will  >>be there to A >>mask it".  It's pretty incompetent to claim that spotting your 0H >>competition a 10:1 write-latency advantage (and an even greater write H >>bandwidth advantage) is a 'non-problem' simply because you've reduced F >>that handicap from more like 100:1 if you didn't use the controller D >>caching at all - but then I'm guessing that Rob doesn't have much H >>experience with modern Unix performance (and possibly that you don't, 
 >>either). >> >  > G > Mmmm... You do realize VMS also supports write back caching now via a " > dynamic sysgen parameter, right?J > http://h71000.www7.hp.com/doc/731FINAL/6657/6657pro_007.html (5.11.1 and	 > 5.11.3)l  B Er, what part of the definition of *default* behavior do you find  difficult to understand?  H VMS provides all the tools that you need to do almost anything you want F to - include recreate a complete file system just the way you want it  within your own application.  H Clearly, however, rolling your own file system because VMS's doesn't do D what you'd like it to do involves a non-trivial amount of work.  By H contrast, the default behavior which Unix provides is pretty damn close I to what most people want (as evidenced by the fact that even RMS buffers aE up multiple small writes to sequential files before batching them to eA disk - it just does so so anachronistically as to be effectively t incompetent at it).w   >  > Or - on VMS V7.3-1 or later:# > SYSGEN>  HELP SYS RMS_SEQFILE_WBHw > E > The doc's simply state that you need to be careful using it becausegH > should a disk path error occur after the application has been told theI > write has already been completed, then you may have some data integrityT > issues to deal with.  F That's the problem with trying to fix performance at too low a level: G either it's expensive (as with stable, mirrored, write-back controller  I cache) or you quietly break the assumptions upon which higher levels are  E basing the guarantees which they provide, without making that at all uG obvious (unless people read the note which the system manager taped up i" inside the door of his office...).   > 
 > [snip..] >  > H >>You're right:  I don't like Rob.  While I don't consider him to be as ? >>much of an active sleaze as I've found Kerry and often Terry e >>to be, they F >>all vigorously defended the Alphacide while studiously ignoring (or G >>actively attempting to spin away) mounting evidence that it was just  A >>what we suspected it was, and continue to do so to this day by PH >>uncritically touting Itanic (though I've heard that Terry may finally H >>have seen something of the light in this area).  Kerry has an obvious H >>job-related motive for pimping for cHumPaq and my impression was that ? >>Terry hoped to acquire one, whereas Rob's motivation is less   >>clear:  for = >>some reason he seems to develop organizational allegiances   >>which cannot  = >>be weakened by any factual evidence - which I guess may be sG >>characteristic of faith-based rather than analytical minds but which o> >>still doesn't explain the strength of the allegiance in the  >>first place. >> >  >  > Bill .. Bad hair day, eh?e  E While I don't pay too much attention to my hair, I'm thankful that I nF still have most of it should I ever decide to take that up as a hobby.  C And it keeps my brain at a sufficiently comfortable temperature to  H function just fine, so that when I decide to call a spade a spade I can % do so both coherently and accurately.h   - bill   ------------------------------  % Date: Sat, 30 Apr 2005 14:24:37 -0400.' From: Dave Froble <davef@tsoft-inc.com>p2 Subject: Re: SYS$SYSROOT and similar logical names0 Message-ID: <1177jb3kvghsq8c@corp.supernews.com>  / Phillip Helbig---remove CLOTHES to reply wrote:t > This is the default setup: > % > $ dir sys$manager:systartup_vms.comc >   > Directory SYS$SYSROOT:[SYSMGR] >  > SYSTARTUP_VMS.COM;2  >  > Total of 1 file. >  > Directory SYS$COMMON:[SYSMGR]/ >  > SYSTARTUP_VMS.COM;202S >  > Total of 1 file. > ( > Grand total of 2 directories, 2 files. > $ sh log sys$sysrootM >    "SYS$SYSROOT" = "DSA133:[SYS0.]" [concealed,terminal] (LNM$SYSTEM_TABLE)  >         = "SYS$COMMON:"cV > 1  "SYS$COMMON" = "DSA133:[SYS0.SYSCOMMON.]" [concealed,terminal] (LNM$SYSTEM_TABLE) > & > This raises a number of questions.   > E > First, why is SYS$SYSROOT used in two contexts: as the logical namek; > itself, and as the alias for the concealed directory nameuH > DSA133:[SYS0.]?  In other words, as a logical name, SYS$SYSROOT pointsJ > to two directories, namely DSA133:[SYS0.] and DSA133:[SYS0.SYSCOMMON.]. J > However, with the DIRECTORY command, SYS$SYSROOT shows up as the name ofJ > only the first directory, the second showing up as SYS$COMMON (which, of; > course, is in turn defined as DSA133:[SYS0.SYSCOMMON.]). t > A > It seems it would make much more sense to define SYS$SYSROOT as  > $ >    "SYS$SYSROOT" = "SYS$SPECIFIC:"L > 1  SYS$SPECIFIC = "DSA133:[SYS0.]" [concealed,terminal] (LNM$SYSTEM_TABLE) >         = "SYS$COMMON:"iQ > 1  "SYS$COMMON" = "DSA133:[SYS0.SYSCOMMON.]" [concealed,terminal] (LNM$SYSTEM_Tr > H > This would have several advantages.  First, it would put SYS$SPECIFIC K > and SYS$COMMON on the same footing.  (Note that the SYS$SPECIFIC logical hA > is defined by default, but it is not used in the definition Of  H > SYS$SYSROOT.)  Second, it would make clear that SYS$SYSROOT refers to D > both SYS$SPECIFIC and SYS$COMMON, as now in the definition of the K > logical SYS$SYSROOT (except that the translation of SYS$SPECIFIC, rather  D > than SYS$SPECIFIC, is used in that definition) and not create any G > confusion with the DIRECTORY command by suggesting that it refers to 5 > only the former. > I > Second question: Since SYS$SYSROOT is defined as a search list, why not ? > use VMS$COMMON:[000000.] rather than [SYS0.SYSCOMMON.] in thetK > definition?  This would make it unnecessary to have [SYS0.SYSCOMMON.] be  B > an alias for VMS$COMMON:[000000.] in the first place.  The only I > advantage I can see in the current scheme is that one can do something e8 > like $ DIR [SYS0...] and have it pick up the stuff in ' > VMS$COMMON:[000000...] via the alias.u > H > (At first I thought that the search-list functionality might be neededH > during system startup before the software (which might be somewhere in9 > the search list) is running, thus one would have to use K > [SYS0.SYSCOMMON.] explicitly rather than using SYS$SYSROOT.  However, in E@ > that case one could just use VMS$COMMON:[000000.] explicitly.) > G > Third, it is probably not supported, but it would make sense to me to0I > add a third translation for SYS$SYSROOT, namely pointing to a directorygF > on a non-system disk shared by all members of a cluster, for exampleF > containing common SYSUAF.DAT etc.  This would allow one redefinitionJ > (which could be done, for example, only if this directory is available) D > rather than defining SYSUAF and all the other logicals explicitly. > G > Actually, one might want to have two such additional definitions, one'B > preceeding the standard two and one following them.  Things likeI > SYSUAF.DAT could be in the first directory of the search list, so that rJ > all machines (satellites or boot nodes) would use them no matter what.  J > In the fourth directory in the search list, one could have things which K > are used only if there is no entry in a previous directory in the search  I > list, which would allow one to put stuff in SYS$COMMON on a particular SG > system disk so that it is used by all nodes booting from that system  K > disk, but not by those booting from other system disks (which would pick 16 > it up from the fourth directory in the search list). >   G One possible explanation is that not all the above has always existed. lF I believe VMS$COMMON is a, relatively, new item.  Ok, VMS V3 or V4 or H thereabouts.  If VMS was starting from scratch right now, then probably F many things could be different.  But VMS is very good at not breaking ! things when issuing new versions.S  D Leaving the original constructs in place doesn't hurt anything, and E changing them just might hurt something somewhere for someone.  Damn n good reason in my opinion.   -- 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 Roado Vanderbilt, PA  15486R   ------------------------------  % Date: Sat, 30 Apr 2005 17:56:26 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 2 Subject: Re: SYS$SYSROOT and similar logical namesB Message-ID: <1114898190.c4054c093a2d4c81c5c749426d2ace58@teranews>   Dave Froble wrote:H > One possible explanation is that not all the above has always existed.G > I believe VMS$COMMON is a, relatively, new item.  Ok, VMS V3 or V4 orp > thereabouts.    F Actually, for me, it was V5.0. I started on MicroVMS which didn't haveH the specific/common hiearchy and I still remember being so confused whenC I blindly installed 5.0 the day I received the tapes (I hadn't readhG about the VAXorcist yet) and saw these strange directory listing givingr 2 directories :-)t   ------------------------------   Date: 30 Apr 2005 18:20:04 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)4 Subject: Re: What is Different or Special About VMS?, Message-ID: <3di0ikF6tk33dU2@individual.net>  3 In article <W7XGj34KL$Yw@eisner.encompasserve.org>,t0 	Kilgallen@SpamCop.net (Larry Kilgallen) writes:Y > In article <3dg968F6rrre3U1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:o > D >> point.  Surely if VMS is still in the favor of the government andG >> DOD as many here keep insisting you can come up with a more credibleo >> source to show it.s > : > Many government agencies have rules against publicity.    < I wasn't talking about Public Affairs driven dog&pony shows.< In order to be used in many places in the goverment, systems> have to meet certain criteria.  This is done before the agency@ involved even gets permission to buy it.  Publishing the resultsA of these certifications would go a lot further than the ramblingsT3 of some teenage computer geek without a girlfriend.F  D And one more comment on the "they pounded on the VMS machine" story.A Did any one ever check how much of this was specifically targeted-B VMS attacks and how much was just the usual script-kiddie nonsense@ that anyone can download from the INTERNET?  Everyone knows thatF 99% of todays hacking consists of attacking 5 year old vulnerabilitiesB using programs the hacker doesn't even understandi all of which is@ specifically targeted at MS and sometimes Unix so it would be noC real surprise that these attacks failed to compromise VMS. (By theroF way, the question was rhetorical.  Nobody, not even the people running= the "conference" have a clue what anybody is actually doing.)R   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: Sat, 30 Apr 2005 14:15:08 -0400s' From: Dave Froble <davef@tsoft-inc.com>t4 Subject: Re: What is Different or Special About VMS?0 Message-ID: <1177ipan30tan91@corp.supernews.com>   Larry Kilgallen wrote:Y > In article <3dg968F6rrre3U1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:t >  > C >>point.  Surely if VMS is still in the favor of the government andhF >>DOD as many here keep insisting you can come up with a more credible >>source to show it. >  > D > Many government agencies have rules against publicity.  I did someD > consulting work for a distinctly non-DoD government agency and the@ > purchase order had specific boilerplate terms forbidding me toB > publicize that relationship.  I can safely say the motive was toC > prohibit any sort of implicit US government endorsement -- not ase > a security measure.i  D I'm pretty sure that such practices aren't good.  For one, it's not % their money, it's the people's money.n  I I'd think that such practices cause there to be a small group of vendors rE who specialize in government work, and such keeps the prices we, the i people, pay for such services.  I If anyone choses a product or service, that in itself IS an endorsement. 7D   Not necessarily a proclimation that the product or service is the : greatest, but at least that they chose it for some reason.  C Not that the government will change because of any opinions I have.h   -- j4 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: 30 Apr 2005 18:40:24 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)4 Subject: Re: What is Different or Special About VMS?, Message-ID: <3di1onF6t99vjU1@individual.net>  R In article <FD827B33AB0D9C4E92EACEEFEE2BA2FB5ECEEE@tayexc19.americas.cpqcorp.net>,* 	"Main, Kerry" <kerry.main@hp.com> writes: >  >> -----Original Message-----  >> From: bill@cs.uofs.edu=20  >> Sent: April 29, 2005 10:35 PM >> To: Info-VAX@Mvb.Saic.Com7 >> Subject: Re: What is Different or Special About VMS?a >>=20  >  > [snip ..]f >  >>=20sH >> I have never said that VMS was not secure.  But the fact remains thatG >> without the attention that Windows and Unix get we'll probably neverdG >> know if there isn't some hole waiting to be exploited.  How many manhI >> hours have been spent trying to hack Windows?  How many man hours havehB >> been spent trying to hack Unix?  Howmany man hours have been=20 >> spent tryinghH >> to hack VMS?  What was they said about an infinite number of monkeys? >>=20c > < > Given the bigger targets i.e. stock exchanges, banks, chipI > manufacturing, health, lotteries, would not the rewards be much bigger?: > H > Another way to look at this - "how many man hours have spent trying toD > hack OpenVMS and the hackers simply got frustrated and moved on toD > easier pickings where the rewards i.e. success in actually hackingB > something to brag about, stand a much better chance of success?" >  > B >> My biggest problem is with the insistence on using something asC >> irrelevant as DEFCON (or number of CERT advisories) to prove thenD >> point.  Surely if VMS is still in the favor of the government andG >> DOD as many here keep insisting you can come up with a more credibler >> source to show it.  >>=20> > G > OpenVMS is still very much a huge part of secure Govt work. As others H > have stated, these are Cust's who obviously do not want others to know > what they are running.=20g  ? But in order to get into those positions in the first place thei= systems had to meet specific criteria.  You can advertise the > criteria ans to what level you meet it without ever mentioning> any the organization.  Remember the Orange Book?  Most serious= profesionals would place a lot more value on a C3 rating thanM the rantings of DEFCON.p   >  >> >=20eB >> > My opinion (and others have expressed similar thoughts) is=20 >> that even ifs@ >> > OpenVMS were more popular from a mainstream perspective,=20 >> there would not@ >> > be anywhere near the number of security issues (notice I=20 >> did not say no' >> > issues)=20o >>=20 F >> And I agree. But surely you realize that VMS has profited from muchD >> of this because I wold like to think that when a hole is found inF >> one of the others someone in VMS Engineering takes the time to lookB >> at the equivalent code in VMS to see that there is, in fact, noE >> vulnerability there.  Of course, if one is found and fixed, no onep6 >> would ever know because VMS was never attacked. =20 > B > When holes are id'ed in any of the OpenSource programs in use byF > OpenVMS, they are reviewed to see if it is applicable to OpenVMS. InH > most cases, even if the vulnerability might affect the process runningI > that OpenSource program, but it does not usually result in the attackertG > getting any elevated priv's as what often happens on other platforms.n >  >> But prudent practicet  >> would be to check anyway. =20 > ! > As stated they do get reviewed.  > + >> Because of differences in the underlyingdF >> OS model I am sure many of these exploits do not exist, such as theE >> dreaded "buffer overflow".  But, guess what, that problem also didIH >> not affect RSX, RSTS or Primos (my other favorite examples.)  Many ofI >> the more well known exploits have been closed in most versions of Unix1 >> as well.=20 >>=20aI >> >         seen today on other platforms e.g. the 100+ security patches-C >> > released publically by Red Hat for its Linux platforms (you=20  >> said NSA was91 >> > using Linux? Now that makes me feel safe...)n >>=20oG >> First of all, I hardly expect any of NSA'a machines are available tooE >> the hackers.  Second, NSA hardened Linux several years ago.  ThosehF >> modifications are only now being put into core distributions.  Why?I >> Because (again, as I have frequently said here) Linux is not a seriousi9 >> operating system it is a toy written by a child and=20. >> maintained by self- >> proclained computer experts.g > I > Well, keep in mind that something like 60+% of computer security issuespI > are directed from internal employees. That is typically what is of mostdF > concern to those who are into the high end security requirements.=20  E Oh, I'm well aware of that.  It was one of the answers on my IA final> last month. :-)    >  >> =20@ >> >                                             since Jan of=20 >> this year ore4 >> > what ever number has been released for Windows. >>=20aH >> Windows has no excuse other than incompetence and greed. (Unlike manyH >> in the industry today I do not now nor have I ever thought Bill Gates$ >> was any kind of computer genius.) >>=20r >> >=20-C >> > Every OS has had some security issues in the past, but as I=20  >> stated in an = >> > earlier thread, there are at least 2 major parts to a=20  >> secure solution -> >> > the capabilities of the Admin's and the base OS design=20 >> itself. The bestsB >> > Admin's will not be able to stop holes in the OS that they=20 >> are not aware@ >> > of, hence the base OS design is absolutely critical. The=20 >> less OS issuesl* >> > the more secure the overall solution. >> >=20cB >> > The point is that, regardless if you want to believe it or=20 >> not, one ofH >> > OpenVMS's claims i.e. very high security, was tested at this DEFCONJ >> > event and beat up in an environment full of knowledgeable hackers.=20 >>=20cI >> See, there you go again.  Knowing how few real VMS professionals there B >> are left in the world and knowing that schools stopped using=20 >> VMS several> >> years ago and knowing that running VMS yourself requires=20 >> special hardwareuH >> as well as mandatory licenses, just where did all those pre-pubescentB >> computer geeks get all that knowledge?  They have Windows at=20 >> home.  They@ >> have Linux.  Some probably have some flavor of BSD.  It is=20 >> highly unlikelyB >> they are all running VMS in their basements in order to look=20
 >> for holes.  >>=20s > 
 > [snip..] >  >>=20eC >> >                                                            =20n >>        ManyB >> > of these were hackers were looking for the bragging rights=20 >> to say they@ >> > hacked into the "VAX" system. [actually Alpha OpenVMS as=20 >> noted in thea
 >> > article]l >>=20yH >> There you prove my point again.  They don't even know what the systemI >> is.  They don;t even know the name of the OS and yet you expect peoplerA >> to believe they are knowledable enough about it to launch a=20s >> real attackH >> and thus be properly repelled by the security of the OS.  The fact isI >> they didn't know what it was and in most cases had probably never seen-; >> a VMS system before in their life.  I wonder how many=20  >> attempts to login; >> as "root" were logged?  More importantly, it would be=20i >> interesting to seeoI >> how many attemts were made to login to accounts such as MAINT (which Is> >> believe was one of the default maintenance accounts back=20 >> inthe DEC days.)lA >> That at least would have shown some little understanding of=20o >> VMS even if >> somewhat dated. >>=20n > I > Since you were not there, have no idea of who was there, and no idea ofmF > what level of skills these attendee's had, how can you make any suchG > "out of the air" assumptions that these were rookie hackers with onlya! > Windows and *nux experience?=20e > H > Even when the article stated that one of the more well know hackers atI > that event stated he only trusted OpenVMS to store his files (which youSG > would think implies a very good understanding of OpenVMS), you simplyeE > poo-poo this as not being important as *you* had not heard of him??i  E Actually, I don't think anything any of them has to say is important.uF Any more than I look at Star Trek Conventioneers for scientific break-: throughs.  I prefer to put my trust in real professionals.      >> >=20BA >> > In the end, they were not successful. The one incident of=20t >> social eng was B >> > not a reflection on the OS itself, but rather the Admin's.=20 >> i.e. social> >> > eng is not "hacking" a system, but rather using social=20 >> trickery to getB >> > valid passwords etc which ths system views as a valid access. >> >=20iH >> > So, instead of a hacker having the bragging rights, OpenVMS now has) >> > kudo's from the event instead. =3D205 >>=20rF >> Faint praise.  It's just too bad that so many serious professionals, >> fail to see it as nothing more than that. >>=20aE >> There are published criteria for secure systems.  VMS can probablytH >> pass them at a much higher level than Windows or any version of Unix.I >> That would be something to brag about.  And it would make great press.aI >> Unfortunately, HP doesn't seem to care and so we are left with relyingiH >> on DEFCON for bragging rights.  That's like going to a Star Trek Con-@ >> vention to get up to date on the latest finding in physics=20 >> and science.u >>=20m > F > You know that Windows and Linux have passed many of these "published( > secure systems criteria" requirements   H Actually, they haven't.  But they know how to word a good press release.E Most of the Windows systems I am aware of that have officially passedtF security criteria only passed when kept disconnected from the network.F Of course, you won't see that mentioned in MS's press release.  In anyG case, surely you will admit that VMS is likely to pass at a much higher-K level of trust than any Windows or Linux box?  If VMS is still as prevalenttD in government as we are led to believe then the record of VMS havingH passed these criteria should be available.  Unless, for reasons of theirF own, HP chooses to not allow the publication of this information (kindJ of like the government agencies that choose to not announce they are using VMS.)o  D >                                       - despite all of the monthlyG > security patches released each month for these two platforms. [Over a J > 100 publically available security patches for RH Linux since Jan of this > year alone]. > J > So, perhaps I am missing something, but saying OpenVMS also passed theseH > same "published secure systems crtiteria" gives what type of marketing > material?c  E Something of more value than the droolings of DEFCON.  You don't say,tF "VMS passed the same criteria as Linux".  You say, "VMS got XXX RatingG and in order to get this rating it had to to X and Y and Z and not do Mx
 and N and O."   C Tell you what Kerry.  The next time you make a call on the CIO of ayE Fortune 50 company before you tell him anything ask him about DEFCON.rD Find out what he knows about it and how much value he places in what+ the people there have to say.  Let us know.    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>   h   ------------------------------  % Date: Sat, 30 Apr 2005 15:57:39 -0400a' From: "Main, Kerry" <kerry.main@hp.com>g4 Subject: RE: What is Different or Special About VMS?R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB5ECEF1@tayexc19.americas.cpqcorp.net>   > -----Original Message-----$ > From: bill@triangle.cs.uofs.edu=20A > [mailto:bill@triangle.cs.uofs.edu] On Behalf Of Bill Gunshannone > Sent: April 30, 2005 2:40 PM > To: Info-VAX@Mvb.Saic.Coma6 > Subject: Re: What is Different or Special About VMS? >=20  	 [Snip...]a     > >=20B > > So, perhaps I am missing something, but saying OpenVMS also=20 > passed these@ > > same "published secure systems crtiteria" gives what type=20 > of marketing
 > > material?h >=20G > Something of more value than the droolings of DEFCON.  You don't say, H > "VMS passed the same criteria as Linux".  You say, "VMS got XXX Rating? > and in order to get this rating it had to to X and Y and Z=20p > and not do M > and N and O."i >=20  F Personal opinion as I am not sure what latest cert's are for OpenVMS -E Problem is that these ratings take a very long time to get certified,pC hence are very expensive. Now if Customers were demanding this, youlF would likely see a larger number of vendors going after these ratings,H but that is not the case much anymore (amazing as this may sound in view of all the issues today).R  G Unfortunately, as you indicated, Customer RFP's will only ask "XYZ CertaF required". So, even if they did a bare minimum to pass, both Linux andE Windows will have passed the XYZ Cert,  and so are fully compliant ton the RFP.  E > Tell you what Kerry.  The next time you make a call on the CIO of asG > Fortune 50 company before you tell him anything ask him about DEFCON.oF > Find out what he knows about it and how much value he places in what- > the people there have to say.  Let us know.-  H You are phrasing the question in the wrong manner. I would ask that CIO:  # - Are you concerned about security?mH - Are you concerned about having to QA monthly security patches that areE recommended by the vendors of Windows/Linux platforms before they aret released to production?=F - Are you concerned about your Customer information getting leaked out via the Internet?y  H When he says yes to each of the above, explain a bit more about having aG rock solid secure platform that has zero known viruses, conforms to thelF latest industry standard application applications and interfaces (e.g.F J2EE, Java, XML, Apache, PHP, PERL, SOAP etc), can scale to over 3,000C cpu's in a active-active cluster spread across multiple sites up to D 800km apart and has proven itself in disaster scenario's such as the= tragic 9/11 event in the US (after losing an entire site, the0D application continued to run fine with zero lost data or applicationH uptime) and also has Windows and Java based GUI management products that) his Operations folks can continue to use.   H And then - oh, by the way, the judges at a hackers conference also voted OpenVMS "cool and unhackable".   :-)a  
 Kerry Main Senior Consultant0 HP Services Canada Voice: 613-592-4660e Fax: 613-591-4477d kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  $ "OpenVMS has always had integrity .. Now, Integrity has OpenVMS .."   ------------------------------  % Date: Sat, 30 Apr 2005 21:47:56 +02002+ From: Karsten Nyblad <nospam@nospam.nospam>f4 Subject: Re: What is Different or Special About VMS?= Message-ID: <4273e0f9$0$67261$157c6196@dreader2.cybercity.dk>n   Main, Kerry wrote:< > Given the bigger targets i.e. stock exchanges, banks, chipI > manufacturing, health, lotteries, would not the rewards be much bigger?  > H I would assume that such systems are placed with at least two firewalls F between the Internet and the really critical systems with web servers > placed between the two firewalls.  Then the web servers would A communicate with the critical servers through protocals that are tI difficult to hack, e.g., no ODBC or HTML or any other protocol, that may  H be used to send program code through the second firewall.  Thus it will E be very difficult to get to a position where you can send abbretatry tG messages to the VMS systems, and if you do, then the chances are small  I that the VMS systems will interprete the messages as code to be executed.-  D If I was a hacker, I would consentrate on systems more likely to be K connected directly to the Internet or with only one firewall to protect it.    ------------------------------    Date: 30 Apr 2005 14:52:06 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)04 Subject: Re: What is Different or Special About VMS?3 Message-ID: <fI2o66fSgimL@eisner.encompasserve.org>G  ` In article <4273AAB1.1406F825@comcast.net>, David J Dachtera <djesys.nospam@comcast.net> writes: > Larry Kilgallen wrote: >> cZ >> In article <3dg968F6rrre3U1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes: >> oF >> > point.  Surely if VMS is still in the favor of the government andI >> > DOD as many here keep insisting you can come up with a more credible  >> > source to show it.f >> KE >> Many government agencies have rules against publicity.  I did somehE >> consulting work for a distinctly non-DoD government agency and thesA >> purchase order had specific boilerplate terms forbidding me toe  >> publicize that relationship.  > B > So, don't. Publicize the product/service, *NOT* who buys it - no > prohibition against that, eh?    See the References section at:  , 	http://www.ljk.com/ljk/marketing_fluff.html   ------------------------------    Date: 30 Apr 2005 15:16:05 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)e4 Subject: Re: What is Different or Special About VMS?3 Message-ID: <vUPAFuIDugUo@eisner.encompasserve.org>2  W In article <3di1onF6t99vjU1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:aT > In article <FD827B33AB0D9C4E92EACEEFEE2BA2FB5ECEEE@tayexc19.americas.cpqcorp.net>,, > 	"Main, Kerry" <kerry.main@hp.com> writes:  H >> OpenVMS is still very much a huge part of secure Govt work. As othersI >> have stated, these are Cust's who obviously do not want others to know7 >> what they are running.=20 > A > But in order to get into those positions in the first place the,? > systems had to meet specific criteria.  You can advertise thea@ > criteria ans to what level you meet it without ever mentioning@ > any the organization.  Remember the Orange Book?  Most serious? > profesionals would place a lot more value on a C3 rating than: > the rantings of DEFCON.-  8 There was no C3 rating.  It went C1, C2, B1, B2, B3, A1.  D But the Orange Book has been replaced by the Common Criteria and VMSB has not been evaluated against the Common Criteria.  At successiveB DECUS and successor meetings VMS product managers have asked thoseA attending how important Common Criteria compliance is to them andoC been met with a deafening roar of silence.  Presumably the HP folksuD who deal with the US Federal Government have asked the same questionD "What criteria are important to you" and the answer came back "COE".E The fact that COE is a Solaris Emulation certificate is not the faultaC of VMS sales folk, and the federal customers were not interested intE the fact that Tru64 Unix is COE compliant -- the customers wanted COEs on VMS.)  G >> You know that Windows and Linux have passed many of these "publishedi) >> secure systems criteria" requirements s > J > Actually, they haven't.  But they know how to word a good press release.G > Most of the Windows systems I am aware of that have officially passed H > security criteria only passed when kept disconnected from the network.@ > Of course, you won't see that mentioned in MS's press release.  J Windows NT was evaluated at the C2 level disconnected from any networking.  F VMS was evaluated at the C2 level disconnected from any network except for clustering.o  = The state of network security evaluation is still "emerging".k   ------------------------------  % Date: Sat, 30 Apr 2005 17:50:39 -0400t- From: JF Mezei <jfmezei.spamnot@teksavvy.com>t4 Subject: Re: What is Different or Special About VMS?B Message-ID: <1114897850.bc5419b39a7d81f345ca730fd36077f0@teranews>   re: VMS as a hacker's target  G Many often argue that hackers don't target VMS because there are so few>F sites around, and a virus targetted for Microsoft has far mroe chancesC of propagating and finding a "home" since there are so many around.r  7 This is not untrue, but it is also not the whole story.e  G VMS has an *image* of being hard to hack. Windows has an image of beingyG easy to hack with far more of a payback probability since it is easy toe find infectable machines.a  D VMS has an image of being managed by seasoned people, Windows has anG image of being managed by windows weenies who concentrate more on bells F and whistles and don't really understand system/network security. ThatS makes windows system far mroe likely to be poorly setup and thus easy to penetrate.L  D HOWEVER, for a serious hacker, a VMS system would offer a far biggerG challenge and more glory when he gets caught after having hacked into a F military system for instance. Far more glory in than that hacking intoF some grand mother's wintel box. But most hackers are not sophisticated! enough to hack into a VMS system.l  H Windows hackers are akin to pick pocket thieves. VMS hackers are akin toG Goldfinger planning a heist on Fort Knox. (or Ocean's 11 or Tomas Crownt? Affair). You need far more sophistication, planning and usually 3 connections within the organisation to get into it.a  H Note that in terms of industrial espoonage, Westjet (a canadian airline)G is king, and it hacked into web pages available to ex eemployees of AirfE Canada and Jetsgo which provided Wesjet with confidential information F about load factors for each individual flights and average yield. (eg:E if 5 persons paid $20 but 95 paid $100, then the yield is $96 on thatcH flight, and that is information that is extremely confidential since theD public only ever sees the ads for the $20 fare and competitors neverB knwo how many seats at $20 are available on a competitor's flight.    F This had nothing to do with system security or hacking per say. It hadF to do with stupid application managers who didn't realise that many exA empooyees were now employees at a competitor and that they reallyaG shouldn't have access to this information, and also never thought about @ putting a daily transaction limit on each ex employee usernames.- (westjet would do thousands of transactions).l   ------------------------------  % Date: Sat, 30 Apr 2005 17:53:02 -0400K- From: JF Mezei <jfmezei.spamnot@teksavvy.com>i4 Subject: Re: What is Different or Special About VMS?B Message-ID: <1114897991.9f16a4330c29f8137ccc56ade97b95b5@teranews>  0 RE: Military not revealing details of contracts.  C Unless of course, it is a big deal with Microsoft to have its large'D shipos wun on Windows, at which point, the military even brags aboutG saving money by going with commodity windows systems and has no problemr revealing which OS they chose.   ------------------------------  % Date: Sat, 30 Apr 2005 18:30:32 -0400l' From: Dave Froble <davef@tsoft-inc.com>e4 Subject: Re: What is Different or Special About VMS?0 Message-ID: <11781o6nsd0vhe6@corp.supernews.com>   Bill Gunshannon wrote:5 > In article <W7XGj34KL$Yw@eisner.encompasserve.org>,h2 > 	Kilgallen@SpamCop.net (Larry Kilgallen) writes: > Y >>In article <3dg968F6rrre3U1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:o >> >>D >>>point.  Surely if VMS is still in the favor of the government andG >>>DOD as many here keep insisting you can come up with a more crediblea >>>source to show it.l >>: >>Many government agencies have rules against publicity.   >  > > > I wasn't talking about Public Affairs driven dog&pony shows.> > In order to be used in many places in the goverment, systems@ > have to meet certain criteria.  This is done before the agencyB > involved even gets permission to buy it.  Publishing the resultsC > of these certifications would go a lot further than the ramblingsp5 > of some teenage computer geek without a girlfriend.y > F > And one more comment on the "they pounded on the VMS machine" story.C > Did any one ever check how much of this was specifically targetedND > VMS attacks and how much was just the usual script-kiddie nonsenseB > that anyone can download from the INTERNET?  Everyone knows thatH > 99% of todays hacking consists of attacking 5 year old vulnerabilitiesD > using programs the hacker doesn't even understandi all of which isB > specifically targeted at MS and sometimes Unix so it would be noE > real surprise that these attacks failed to compromise VMS. (By theroH > way, the question was rhetorical.  Nobody, not even the people running? > the "conference" have a clue what anybody is actually doing.)  >  > bill >   G Well, one could think that something like DEFCON would draw the people O? who devised the techniques that the me-to hackers use.  Or not.s   -- 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 Roadn Vanderbilt, PA  15486T   ------------------------------  % Date: Sat, 30 Apr 2005 19:39:21 -0400:- From: JF Mezei <jfmezei.spamnot@teksavvy.com>)4 Subject: Re: What is Different or Special About VMS?B Message-ID: <1114904359.bb6304eac286efbc95097fd048536a3a@teranews>   JF Mezei wrote:a > 2 > RE: Military not revealing details of contracts.     Here is another example:  _ http://news.com.com/Defense+Department+signs+Red+Hat+deal/2100-1012_3-5690643.html?tag=nefd.hedt  9 Defebnse dept signs a big deal with Red Hat (aka: Linux).Y    D The argument that that defense dept can't publisize platform choices2 seems to apply only to VMS and to no other system.   ------------------------------  % Date: Sat, 30 Apr 2005 20:07:12 -0400 ' From: "Main, Kerry" <kerry.main@hp.com>o4 Subject: RE: What is Different or Special About VMS?R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB5ECEF6@tayexc19.americas.cpqcorp.net>   > -----Original Message-----9 > From: JF Mezei [mailto:jfmezei.spamnot@teksavvy.com]=20  > Sent: April 30, 2005 5:51 PM > To: Info-VAX@Mvb.Saic.ComS6 > Subject: Re: What is Different or Special About VMS? >=20	 [snip ..]a   >=20A > Windows hackers are akin to pick pocket thieves. VMS hackers=20 
 > are akin to @ > Goldfinger planning a heist on Fort Knox. (or Ocean's 11 or=20
 > Tomas Crown]A > Affair). You need far more sophistication, planning and usuallyn5 > connections within the organisation to get into it.o >=20  : Hey, great analogy - can I steal (oops, borrow) that line?   :-),   [snip..]   >=20H > This had nothing to do with system security or hacking per say. It hadH > to do with stupid application managers who didn't realise that many exC > empooyees were now employees at a competitor and that they reallyn> > shouldn't have access to this information, and also never=20 > thought abouteB > putting a daily transaction limit on each ex employee usernames./ > (westjet would do thousands of transactions).T >=20  B This also proves the point about the damage that internal (even ifD former) employees can have i.e. that most of the security issues are caused by internal issues.   Regards   
 Kerry Main Senior Consultantt HP Services Canada Voice: 613-592-4660r Fax: 613-591-4477r kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  $ "OpenVMS has always had integrity .. Now, Integrity has OpenVMS .."   ------------------------------   Date: 1 May 2005 00:59:17 GMT ( From: bill@cs.uofs.edu (Bill Gunshannon)4 Subject: Re: What is Different or Special About VMS?, Message-ID: <3dinv5F6n6sabU1@individual.net>  0 In article <11781o6nsd0vhe6@corp.supernews.com>,* 	Dave Froble <davef@tsoft-inc.com> writes: > Bill Gunshannon wrote:6 >> In article <W7XGj34KL$Yw@eisner.encompasserve.org>,3 >> 	Kilgallen@SpamCop.net (Larry Kilgallen) writes:l >> kZ >>>In article <3dg968F6rrre3U1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes: >>>t >>>aE >>>>point.  Surely if VMS is still in the favor of the government andlH >>>>DOD as many here keep insisting you can come up with a more credible >>>>source to show it. >>>a; >>>Many government agencies have rules against publicity.  t >> b >> n? >> I wasn't talking about Public Affairs driven dog&pony shows.n? >> In order to be used in many places in the goverment, systemsmA >> have to meet certain criteria.  This is done before the agencymC >> involved even gets permission to buy it.  Publishing the results4D >> of these certifications would go a lot further than the ramblings6 >> of some teenage computer geek without a girlfriend. >> lG >> And one more comment on the "they pounded on the VMS machine" story.eD >> Did any one ever check how much of this was specifically targetedE >> VMS attacks and how much was just the usual script-kiddie nonsensegC >> that anyone can download from the INTERNET?  Everyone knows thatFI >> 99% of todays hacking consists of attacking 5 year old vulnerabilitiesyE >> using programs the hacker doesn't even understandi all of which is C >> specifically targeted at MS and sometimes Unix so it would be noaF >> real surprise that these attacks failed to compromise VMS. (By therI >> way, the question was rhetorical.  Nobody, not even the people runninge@ >> the "conference" have a clue what anybody is actually doing.) >> s >> billr >> y > I > Well, one could think that something like DEFCON would draw the people  A > who devised the techniques that the me-to hackers use.  Or not.   jB Actually, giving it a little bit of thought, it is not the hackersC that go to DEFCON that I would be worried about.  It's the ones whosD don't.  Successful thieves don;t advertise their trade and criminalsF who brag about their accomplishments usually end out taking a vacation at government expense.   bill   -- VJ 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: 1 May 2005 01:08:43 GMTi( From: bill@cs.uofs.edu (Bill Gunshannon)4 Subject: Re: What is Different or Special About VMS?, Message-ID: <3diogqF6n6sabU2@individual.net>  3 In article <vUPAFuIDugUo@eisner.encompasserve.org>,r0 	Kilgallen@SpamCop.net (Larry Kilgallen) writes:Y > In article <3di1onF6t99vjU1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:eU >> In article <FD827B33AB0D9C4E92EACEEFEE2BA2FB5ECEEE@tayexc19.americas.cpqcorp.net>,k- >> 	"Main, Kerry" <kerry.main@hp.com> writes:  > I >>> OpenVMS is still very much a huge part of secure Govt work. As others J >>> have stated, these are Cust's who obviously do not want others to know >>> what they are running.=20u >> tB >> But in order to get into those positions in the first place the@ >> systems had to meet specific criteria.  You can advertise theA >> criteria ans to what level you meet it without ever mentioningnA >> any the organization.  Remember the Orange Book?  Most seriousd@ >> profesionals would place a lot more value on a C3 rating than >> the rantings of DEFCON. > : > There was no C3 rating.  It went C1, C2, B1, B2, B3, A1.  ? I just pulled a code out of my hat.  I think everyone knows theeC colored books went away, but it was the point I was trying to make.- > F > But the Orange Book has been replaced by the Common Criteria and VMS7 > has not been evaluated against the Common Criteria.  t  D If VMS is still being sold under government procurements (as we keep( being told) I find this hard to believe.  C >                                                     At successivetD > DECUS and successor meetings VMS product managers have asked thoseC > attending how important Common Criteria compliance is to them anda. > been met with a deafening roar of silence.    ? So maybe they don't really rate security as high as some people @ seem to think.  If Common Criteria is not important, I doubt the2 rantings from DEFCON are considered more valuable.  D >                                            Presumably the HP folksF > who deal with the US Federal Government have asked the same questionF > "What criteria are important to you" and the answer came back "COE".G > The fact that COE is a Solaris Emulation certificate is not the faultnE > of VMS sales folk, and the federal customers were not interested innG > the fact that Tru64 Unix is COE compliant -- the customers wanted COEt	 > on VMS.e  E Well, according to the class I just got back from, Common Criteria isnH not optional.  As for COE and "a Solaris Emulation certificate" it mightG interest people here to know that Solaris is no longer taught at all tooF Information Systems Technicians at the Warrant Officer Basic Course at Ft. Gordon.    > H >>> You know that Windows and Linux have passed many of these "published* >>> secure systems criteria" requirements  >> ,K >> Actually, they haven't.  But they know how to word a good press release.tH >> Most of the Windows systems I am aware of that have officially passedI >> security criteria only passed when kept disconnected from the network.eA >> Of course, you won't see that mentioned in MS's press release.. > L > Windows NT was evaluated at the C2 level disconnected from any networking.  L Why would this surprise you?  If hackers can't get at the machine they can't break into it.   > H > VMS was evaluated at the C2 level disconnected from any network except > for clustering.  > ? > The state of network security evaluation is still "emerging".f  E And which one do you think will be able to maintain that rating once   networking is included?t   bill   -- aJ 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: 1 May 2005 01:13:53 GMTh( From: bill@cs.uofs.edu (Bill Gunshannon)4 Subject: Re: What is Different or Special About VMS?, Message-ID: <3dioqhF6n6sabU3@individual.net>  R In article <FD827B33AB0D9C4E92EACEEFEE2BA2FB5ECEF1@tayexc19.americas.cpqcorp.net>,* 	"Main, Kerry" <kerry.main@hp.com> writes: >  > F >> Tell you what Kerry.  The next time you make a call on the CIO of aH >> Fortune 50 company before you tell him anything ask him about DEFCON.G >> Find out what he knows about it and how much value he places in whatg. >> the people there have to say.  Let us know. > J > You are phrasing the question in the wrong manner. I would ask that CIO: > % > - Are you concerned about security?vJ > - Are you concerned about having to QA monthly security patches that areG > recommended by the vendors of Windows/Linux platforms before they ares > released to production?oH > - Are you concerned about your Customer information getting leaked out > via the Internet?s > J > When he says yes to each of the above, explain a bit more about having aI > rock solid secure platform that has zero known viruses, conforms to the H > latest industry standard application applications and interfaces (e.g.H > J2EE, Java, XML, Apache, PHP, PERL, SOAP etc), can scale to over 3,000E > cpu's in a active-active cluster spread across multiple sites up tosF > 800km apart and has proven itself in disaster scenario's such as the? > tragic 9/11 event in the US (after losing an entire site, theeF > application continued to run fine with zero lost data or applicationJ > uptime) and also has Windows and Java based GUI management products that+ > his Operations folks can continue to use.u  G I agree that these are the questions to ask when trying to make a sale.w@ But the question at hand was as to the value of DEFCON rantings.   > J > And then - oh, by the way, the judges at a hackers conference also voted  > OpenVMS "cool and unhackable".  = Which would likely mean nothing to the CIO who has other moree/ professional criteria for making his decisions.v   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>   a   ------------------------------    Date: 30 Apr 2005 20:06:36 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)f4 Subject: Re: What is Different or Special About VMS?3 Message-ID: <AOHuLn0FKuQL@eisner.encompasserve.org>d  k In article <4273e0f9$0$67261$157c6196@dreader2.cybercity.dk>, Karsten Nyblad <nospam@nospam.nospam> writes:, > Main, Kerry wrote:= >> Given the bigger targets i.e. stock exchanges, banks, chip J >> manufacturing, health, lotteries, would not the rewards be much bigger? >> mJ > I would assume that such systems are placed with at least two firewalls H > between the Internet and the really critical systems with web servers @ > placed between the two firewalls.  Then the web servers would C > communicate with the critical servers through protocals that are kK > difficult to hack, e.g., no ODBC or HTML or any other protocol, that may aJ > be used to send program code through the second firewall.  Thus it will G > be very difficult to get to a position where you can send abbretatry iI > messages to the VMS systems, and if you do, then the chances are small xK > that the VMS systems will interprete the messages as code to be executed.. > F > If I was a hacker, I would consentrate on systems more likely to be M > connected directly to the Internet or with only one firewall to protect it.f   Not if you had inside access.y   ------------------------------  % Date: Sun, 01 May 2005 07:03:33 +0200r+ From: Karsten Nyblad <nospam@nospam.nospam> 4 Subject: Re: What is Different or Special About VMS?= Message-ID: <42746332$0$78286$157c6196@dreader1.cybercity.dk>    Larry Kilgallen wrote:m > In article <4273e0f9$0$67261$157c6196@dreader2.cybercity.dk>, Karsten Nyblad <nospam@nospam.nospam> writes:h >  >>Main, Kerry wrote: >>= >>>Given the bigger targets i.e. stock exchanges, banks, chipIJ >>>manufacturing, health, lotteries, would not the rewards be much bigger? >>>  >>J >>I would assume that such systems are placed with at least two firewalls H >>between the Internet and the really critical systems with web servers @ >>placed between the two firewalls.  Then the web servers would C >>communicate with the critical servers through protocals that are  K >>difficult to hack, e.g., no ODBC or HTML or any other protocol, that may sJ >>be used to send program code through the second firewall.  Thus it will G >>be very difficult to get to a position where you can send abbretatry eI >>messages to the VMS systems, and if you do, then the chances are small .K >>that the VMS systems will interprete the messages as code to be executed.n >>F >>If I was a hacker, I would consentrate on systems more likely to be M >>connected directly to the Internet or with only one firewall to protect it.d >  >  > Not if you had inside access.0  F True.  However, normally there will be a firewall not allowing, e.g., G ODBC and HTML between the computer center and the rest of the company. eC I think most of the insider hacking is based on social engineering.e   ------------------------------   End of INFO-VAX 2005.241 ************************