0 INFO-VAX	Wed, 22 Jan 2003	Volume 2003 : Issue 44      Contents:5 Re: (OT) lots of spam in vmsnet.sdk.openvms.fieldtest 5 Re: (OT) lots of spam in vmsnet.sdk.openvms.fieldtest 5 RE: (OT) lots of spam in vmsnet.sdk.openvms.fieldtest 3 Re: (UNIX is OVER) was  . . . Re: Alpha for HP UX ? 3 Re: (UNIX is OVER) was  . . . Re: Alpha for HP UX ? 3 Re: (UNIX is OVER) was  . . . Re: Alpha for HP UX ? 4 Re: Access linux and/or windows directories with vms1 Re: AlphaServer users given lifeline till Itanium 1 Re: AlphaServer users given lifeline till Itanium 1 Re: AlphaServer users given lifeline till Itanium 1 Re: AlphaServer users given lifeline till Itanium 1 Re: AlphaServer users given lifeline till Itanium 1 Re: AlphaServer users given lifeline till Itanium 1 Re: AlphaServer users given lifeline till Itanium 1 Re: AlphaServer users given lifeline till Itanium 1 Re: AlphaServer users given lifeline till Itanium 1 RE: AlphaServer users given lifeline till Itanium 1 Re: AlphaServer users given lifeline till Itanium 1 Re: AlphaServer users given lifeline till Itanium  Re: automatic screensaver  Re: automatic screensaver  Re: automatic screensaver  Re: automatic screensaver  Changed DVDWRITE homepage path. Command/API to list locked files in a cluster?2 RE: Command/API to list locked files in a cluster?C copyright (was: RE: EV7 hits, blows away Sun, IBM, 2-3x VMS gains!)  Determining outbound identity ! Re: Determining outbound identity > Re: Function keys no longer work after TCPware upgrade to 5.6.$ Re: Galaxy support Profile for ES80? Re: How to Backup OSX  Re: How to Backup OSX  Re: How to Backup OSX  Re: How to Backup OSX  Re: How to Backup OSX  Re: How to Backup OSX  Re: How to Backup OSX  Re: How to Backup OSX  Re: How to Backup OSX  Re: How to Backup OSX  Re: How to Backup OSX  RE: How to Backup OSX , Re: How to flash VAXstation 4000/90 FEPROMS?0 Re: Interrupt stack with EMC fibre channel disks Re: Is that possible- Re: Itanium with alpha goodies on the way ... - RE: Itanium with alpha goodies on the way ... 
 keeping it up  RE: Marvel Performance Re: Multinet FTP Question & Re: MX 5.3-2 ignoring vmsmail forwards Re: NIST Servers and ACTS  Re: NIST Servers and ACTS P Re: OpenVMS.org: Marvel article and HP's press release for Marveland Alpha RetaiP Re: OpenVMS.org: Marvel article and HP's press release for Marveland Alpha Retai/ Re: Recovering from deleted .PCSI$DATABASE file 7 Re: Security threats seen big for 2003 ... not for VMS! > Re: shadow-set consistency on simultaneous loss of all members> Re: shadow-set consistency on simultaneous loss of all members> Re: shadow-set consistency on simultaneous loss of all members- Re: Standards for sending of VMS binary files @ Re: Storage network with VMS cluster and other OSses, ESS, EMC2?@ Re: Storage network with VMS cluster and other OSses, ESS, EMC2?< TCP/IP Services v5.3 - how to increase the number of "mbufs"@ Re: TCP/IP Services v5.3 - how to increase the number of "mbufs"@ Re: TCP/IP Services v5.3 - how to increase the number of "mbufs"@ Re: TCP/IP Services v5.3 - how to increase the number of "mbufs") Re: The HP marketing juggernaut grinds on ) Re: The HP marketing juggernaut grinds on  Versastor surfaces Re: Versastor surfaces
 Re: VMS Jokes  YA OpenVMS Newbie  Re: YA OpenVMS Newbie  Re: YA OpenVMS Newbie  Re: YA OpenVMS Newbie   F ----------------------------------------------------------------------  % Date: Wed, 22 Jan 2003 11:02:40 -0000 * From: "Richard Brodie" <R.Brodie@rl.ac.uk>> Subject: Re: (OT) lots of spam in vmsnet.sdk.openvms.fieldtest, Message-ID: <b0ltoh$14k6@newton.cc.rl.ac.uk>  b "Tom Linden" <tom@kednos.com> wrote in message news:CIEJLCMNHNNDLLOOGNJIIECGGHAA.tom@kednos.com...E > Speaking of which, how does one correlate an IP with a geographical D > region.  I am aware of arin and the like but how is the allocationD > made, and by whom?  So you could block, for example, Korea but not > ther Aussies?   K For Korea in particular, you can use http://korea.services.net/ . If you're @ sufficiently bloody minded you can blackhole them at the router.   ------------------------------    Date: 22 Jan 2003 09:18:50 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) > Subject: Re: (OT) lots of spam in vmsnet.sdk.openvms.fieldtest3 Message-ID: <Ul66TtTVoIkX@eisner.encompasserve.org>   B Perhaps due to Shane's efforts, the original problem seems to have abated, at least for now.    ------------------------------  % Date: Wed, 22 Jan 2003 09:35:49 -0800 $ From: Shane Smith <ssmith@icius.com>> Subject: RE: (OT) lots of spam in vmsnet.sdk.openvms.fieldtest0 Message-ID: <01C2C1F9.B6BB9780@sulfer.icius.com>  H I hope I helped at least. If it happens again, let me know and I'll take another swing.   Shane    -----Original Message-----: From: Kilgallen@SpamCop.net [mailto:Kilgallen@SpamCop.net]) Sent: Wednesday, January 22, 2003 7:19 AM  To: Info-VAX@Mvb.Saic.Com > Subject: Re: (OT) lots of spam in vmsnet.sdk.openvms.fieldtest    B Perhaps due to Shane's efforts, the original problem seems to have abated, at least for now.    ------------------------------   Date: 22 Jan 2003 13:19:05 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)< Subject: Re: (UNIX is OVER) was  . . . Re: Alpha for HP UX ?5 Message-ID: <b0m5o8$r17c8$1@ID-135708.news.dfncis.de>   @ In article <20030121233651.69077.qmail@web20207.mail.yahoo.com>,1 	Fabio Cardoso <fabiopenvms@yahoo.com.br> writes:  > 2 > There is no escape ! The standard UNIX will die  >   2 Here we go (yet) again.  Unix is dead, news at 11.  @ This has been a constant theme ever since 1980 when Byte printed their headline, "Is UNIX Dead?"   C It wasn't then, it isn't now, and it shows no sign of dying anytime A in the foreseeable future.  Unlike the many proprietary OSes that " have come and gone in the interim.   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: Wed, 22 Jan 2003 14:55:20 GMT 0 From: rw@vorpalbunnyeircom.net (Russell Wallace)< Subject: Re: (UNIX is OVER) was  . . . Re: Alpha for HP UX ?0 Message-ID: <3e2eafdf.774597927@news.eircom.net>  7 On Tue, 21 Jan 2003 15:36:51 -0800 (PST), Fabio Cardoso ! <fabiopenvms@yahoo.com.br> wrote:   A >There is no escape ! The standard UNIX will die as the companies 5 >will consolidate processors (ex. Itanium or Opteron)  > F >May be SUN will be the last on with an owned processor. IBM will haveB >Itanium or Opteron servers. SGI will have Itanium servers. HP !!!, >May be the Cobalt apliances can save SUN !   F Consolidation of processors does not imply the death of either Unix orE its vendors. Sun, to take your example, don't make money on the Sparc F chip per se, but on their expertise in building big scalable SMP boxesE with an operating system that can make use of them. There's no reason C they couldn't start putting Opterons instead of Sparcs inside those E boxes, eliminate the Sparc R&D costs, and still have happy customers.   = (Though it seems that Sun are wise enough to not announce the ; termination of their own chip at least until after a proven B alternative is actually in place, unlike certain other companies.)   --  3 "Mercy to the guilty is treachery to the innocent." + Remove killer rodent from address to reply. ! http://www.esatclear.ie/~rwallace    ------------------------------  % Date: Wed, 22 Jan 2003 12:45:44 -0500 A From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> < Subject: Re: (UNIX is OVER) was  . . . Re: Alpha for HP UX ?. Message-ID: <3e2ed8c9$1_1@hpb10302.boi.hp.com>  5 "Bill Gunshannon" <bill@cs.uofs.edu> wrote in message / news:b0m5o8$r17c8$1@ID-135708.news.dfncis.de... B > In article <20030121233651.69077.qmail@web20207.mail.yahoo.com>,2 > Fabio Cardoso <fabiopenvms@yahoo.com.br> writes: > > 3 > > There is no escape ! The standard UNIX will die  > >  > 4 > Here we go (yet) again.  Unix is dead, news at 11. > B > This has been a constant theme ever since 1980 when Byte printed! > their headline, "Is UNIX Dead?"  >   L While the title is nonsense, the underlying argument (I think) is that LinuxJ sales volume is reaching critical mass, and that Linux will be THE UNIX ofF the future.  Proprietary UNIX implementations will become "legacy" andH wither as A) the base Linux OS becomes robust and scaleable enough to beH used for everything from the UNIX desktop user to mission critical largeJ scale servers.  B) Vendors find ways to sell Linux added-value software toH differentiate themselves.  C) A single CPU architecture becomes dominantB enough to spawn a shrink-wrap business for application and library developers.   J After all, Sun taught us in the beginning that UNIX was "Open" and UNIX isJ UNIX at least at the standards level.  So will Sun, HP, IBM, or anybody beK able to keep up with the new evolving UNIX standard called LINUX with their J proprietary UNIX implementations.  And why *should* they spend millions ofJ dollars to continue to develop and maintain their own UNIX implementation,J when they can piggyback on the worldwide LINUX development community - andA then they can concentrate on other value added SW and HW support.   L It's an interesting question.  I'm sure that my friends in HP-UX, and at SUNI don't see it shaking out that way.  Just as the dozen or so LINUX heads I $ know can't imagine it not happening.   ------------------------------  + Date: Wed, 22 Jan 2003 11:26:25 +0000 (UTC) + From: david20@alpha2.mdx.ac.uk (David Webb) = Subject: Re: Access linux and/or windows directories with vms + Message-ID: <b0lv51$8vp$1@aquila.mdx.ac.uk>   W In article <3E2DEB28.8050905@spammotel.com>, Alder <PGDEHMKOKIMD@spammotel.com> writes:  >Franz-Jrgen Tollmann wrote: > >> "Alder" <TFTAJLLYMXZP@spammotel.com> schrieb im Newsbeitrag9 >> news:a6840bf2.0301191835.1460bfa@posting.google.com...  >>  4 >>>fdc@columbia.edu (Frank da Cruz) wrote in message >>>  >>> L >>>>Remember we are talking about PCs with Windows and a VMS server.  Kermit >>>  >> 95  >>  F >>>>for Windows supports the following secure TCP/IP connection types: >>>>  >>>> . SSL/TLS terminal sessions4 >>>> . Telnet, FTP, and HTTP secured with SSL or TLSE >>>> . Telnet, Rlogin, and FTP secured with Kerberos IV or Kerberos V & >>>> . Telnet and FTP secured with SRP >>>> . SSH v1 terminal sessions  >>>> . SSH v2 terminal sessions  >>>>J >>>>Thus you can use practically any kind of security that is installed on >>>  >> VMS.  >>  I >>>I for one have no aversion to using a Windows telnet client capable of C >>>using Kerberos authentication and TLS encryption to connect from E >>>Windows to VMS.  Trouble is, what server options are available for D >>>VMS?  This is a hobbyist talking, so freeware is what I'd prefer.
 >>>Anyone? >>   >>   >> Right!!!  >>   >>   >  >I thought so :-)  >    Well lets see :-  K Kerberos    -      http://www.openvms.compaq.com/openvms/products/kerberos/   M The latest version of DEC TCPIP Services supports a kerberized telnet server. C Multinet and TCPWARE have supported kerberos for a number of years.   F SSH         -      Public domain version 1.5 ssh server available fromA                    http://www.er6.eng.ohio-state.edu/~jonesd/ssh/   <                    Public domain ssh client  available from +                    http://www.free.lp.fish/ N                    Richard Levitte is working on an improved public domain sshI                    client (Bamse) but I haven't heard how long it will be *                    before that is usable.   N Dec TCPIP services doesn't currently support SSH - though version 2 client and( server are promised in a future release.  K Multinet and TCPWARE have their own (I believe both now SSH v2) clients and  servers.  N SSL  - All the main VMS web servers (OSU, WASD, Apache - and probably purveyorN but I haven't checked) support SSL. For OSU and WASD you will need to downloadO the OpenSSL library from http://www.openssl.org/. If you have VMS 7.3-1 then an & older version of openssl is included.     
 David Webb VMS and Unix team leader CCSS Middlesex University     ------------------------------  % Date: Wed, 22 Jan 2003 07:01:14 -0500 2 From: rdeininger@mindspring.com (Robert Deininger): Subject: Re: AlphaServer users given lifeline till ItaniumL Message-ID: <rdeininger-2201030701140001@user-2ive3rg.dialup.mindspring.com>  < In article <c5cf6e8.0301210728.2cc621a4@posting.google.com>,) baby_p_nut@yahoo.com (Baby Peanut) wrote:   7 >http://news.zdnet.co.uk/story/0,,t269-s2129026,00.html  > . >AlphaServer users given lifeline till Itanium  >09:45 Tuesday 21st January 2003
 >Peter Judge   > A >ENSA@Work 2003: Users of HP's AlphaServer range have seven years & >before they have to move to Itanium.   I If they choose to start that article with an inaccuracy, I guess the rest F might not be worth reading.  HP has never said anything about "have to move" off of alpha.    ------------------------------  % Date: Wed, 22 Jan 2003 07:03:25 -0500 2 From: rdeininger@mindspring.com (Robert Deininger): Subject: Re: AlphaServer users given lifeline till ItaniumL Message-ID: <rdeininger-2201030703250001@user-2ive3rg.dialup.mindspring.com>  F In article <CIEJLCMNHNNDLLOOGNJIMEDMGHAA.tom@kednos.com>, "Tom Linden" <tom@kednos.com> wrote:   H >>"Oracle has also taken clustering technology from us, and they will goD >>into HP UX not just on Itanium," said HP's Ken Surplice. "ClustersC >>will work with PA RISC, as well as older system including Digital A >>Vaxes. Users will be able to mix all three architectures in one  >>cluster."  > 7 >He must have misspoken here, since this excludes Alpha     I Yes, a mis-speak or a mis-quote.  VAX, PA-RISC, and HP-UX together?  Huh?    ------------------------------   Date: 22 Jan 2003 12:17:21 GMT( From: nmm1@cus.cam.ac.uk (Nick Maclaren): Subject: Re: AlphaServer users given lifeline till Itanium0 Message-ID: <b0m24h$h9n$1@pegasus.csx.cam.ac.uk>   In article <rdeininger-2201030701140001@user-2ive3rg.dialup.mindspring.com>, rdeininger@mindspring.com (Robert Deininger) writes: ? |> In article <c5cf6e8.0301210728.2cc621a4@posting.google.com>, , |> baby_p_nut@yahoo.com (Baby Peanut) wrote: |>  : |> >http://news.zdnet.co.uk/story/0,,t269-s2129026,00.html |> >1 |> >AlphaServer users given lifeline till Itanium # |> >09:45 Tuesday 21st January 2003  |> >Peter Judge  |> >D |> >ENSA@Work 2003: Users of HP's AlphaServer range have seven years) |> >before they have to move to Itanium.   |>  L |> If they choose to start that article with an inaccuracy, I guess the restI |> might not be worth reading.  HP has never said anything about "have to  |> move" off of alpha.  A They have, however, said that they will stop maintaining them and   even stopping the sale of parts.     Regards, Nick Maclaren,* University of Cambridge Computing Service,> New Museums Site, Pembroke Street, Cambridge CB2 3QH, England. Email:  nmm1@cam.ac.uk/ Tel.:  +44 1223 334761    Fax:  +44 1223 334679    ------------------------------  % Date: Wed, 22 Jan 2003 12:24:33 +0000 , From: Chris Quayle <lightwork@aerosys.co.uk>: Subject: Re: AlphaServer users given lifeline till Itanium- Message-ID: <3E2E8D81.169FE476@aerosys.co.uk>    Robert Deininger wrote:  > > > In article <c5cf6e8.0301210728.2cc621a4@posting.google.com>,+ > baby_p_nut@yahoo.com (Baby Peanut) wrote:  > 9 > >http://news.zdnet.co.uk/story/0,,t269-s2129026,00.html  > > 0 > >AlphaServer users given lifeline till Itanium" > >09:45 Tuesday 21st January 2003 > >Peter Judge > > C > >ENSA@Work 2003: Users of HP's AlphaServer range have seven years ' > >before they have to move to Itanium.  > K > If they choose to start that article with an inaccuracy, I guess the rest H > might not be worth reading.  HP has never said anything about "have to > move" off of alpha.   F What interested me was the fact that the new Alphaserver announcementsC come from the very top, which would suggest that HP are starting to G hedge their bets over IA-64. Eventually, even she must notice that they H already have an industry leading cpu which is being downplayed (for waht@ reason ?) in favour of an architecture which still can't cut theD mustard, despite billions of dollars of investment. So why bother to continue with it ?.   B Even the most blinkered management drones must become aware of the obvious at some stage...   Chris    ------------------------------  # Date: Wed, 22 Jan 2003 13:53:36 GMT # From: "John Smith" <a@nonymous.com> : Subject: Re: AlphaServer users given lifeline till ItaniumH Message-ID: <AnxX9.74329$ej1.42559@news02.bloor.is.net.cable.rogers.com>  9 "Chris Quayle" <lightwork@aerosys.co.uk> wrote in message ' news:3E2E8D81.169FE476@aerosys.co.uk...  > Robert Deininger wrote:  > > @ > > In article <c5cf6e8.0301210728.2cc621a4@posting.google.com>,- > > baby_p_nut@yahoo.com (Baby Peanut) wrote:  > > ; > > >http://news.zdnet.co.uk/story/0,,t269-s2129026,00.html  > > > 2 > > >AlphaServer users given lifeline till Itanium$ > > >09:45 Tuesday 21st January 2003 > > >Peter Judge > > > E > > >ENSA@Work 2003: Users of HP's AlphaServer range have seven years ) > > >before they have to move to Itanium.  > > D > > If they choose to start that article with an inaccuracy, I guess the restA > > might not be worth reading.  HP has never said anything about  "have to > > move" off of alpha.  > : > What interested me was the fact that the new Alphaserver
 announcements E > come from the very top, which would suggest that HP are starting to D > hedge their bets over IA-64. Eventually, even she must notice that theyE > already have an industry leading cpu which is being downplayed (for  wahtB > reason ?) in favour of an architecture which still can't cut theF > mustard, despite billions of dollars of investment. So why bother to > continue with it ?.  > D > Even the most blinkered management drones must become aware of the > obvious at some stage...    D But a real hedge-your-bets would include parallel development of the. next generation EV8 and that has ceased, IIRC.   ------------------------------  + Date: Wed, 22 Jan 2003 14:31:42 +0000 (UTC) + From: david20@alpha2.mdx.ac.uk (David Webb) : Subject: Re: AlphaServer users given lifeline till Itanium+ Message-ID: <b0ma0e$c51$1@aquila.mdx.ac.uk>   n In article <AnxX9.74329$ej1.42559@news02.bloor.is.net.cable.rogers.com>, "John Smith" <a@nonymous.com> writes: > : >"Chris Quayle" <lightwork@aerosys.co.uk> wrote in message( >news:3E2E8D81.169FE476@aerosys.co.uk... >> Robert Deininger wrote: >> >A >> > In article <c5cf6e8.0301210728.2cc621a4@posting.google.com>, . >> > baby_p_nut@yahoo.com (Baby Peanut) wrote: >> >; >> What interested me was the fact that the new Alphaserver  >announcementsF >> come from the very top, which would suggest that HP are starting toE >> hedge their bets over IA-64. Eventually, even she must notice that  >they F >> already have an industry leading cpu which is being downplayed (for >waht C >> reason ?) in favour of an architecture which still can't cut the G >> mustard, despite billions of dollars of investment. So why bother to  >> continue with it ?. >>E >> Even the most blinkered management drones must become aware of the  >> obvious at some stage...  >  > E >But a real hedge-your-bets would include parallel development of the / >next generation EV8 and that has ceased, IIRC.  >  >   M Which was probably the biggest complaint at the time. If Compaq had announced N porting of VMS and Tru64 to IA64 but not simultaneously announced the droppingK of work on EV8 (and hence in peoples mind the death of the Alpha chip) then I they would have garnered almost universal support from the people in this L newsgroup. It would have been seen as a prudent move just in case IA64 wouldM deliver on it's promises rather than the gigantic leap of faith and imprudent ) burning of boats which actually occurred.   
 David webb VMS and Unix team leader CCSS Middlesex University   ------------------------------    Date: 22 Jan 2003 09:24:21 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) : Subject: Re: AlphaServer users given lifeline till Itanium3 Message-ID: <VC53uGXnvzWg@eisner.encompasserve.org>   Y In article <b0ma0e$c51$1@aquila.mdx.ac.uk>, david20@alpha2.mdx.ac.uk (David Webb) writes:   O > Which was probably the biggest complaint at the time. If Compaq had announced P > porting of VMS and Tru64 to IA64 but not simultaneously announced the droppingM > of work on EV8 (and hence in peoples mind the death of the Alpha chip) then K > they would have garnered almost universal support from the people in this  > newsgroup.  K And they would have lost the opportunity for the payments offered by Intel.   I Which of those opportunities best matches the letter and spirit of the US E securities laws requiring them to do what is financially best for the 
 stockholders?   E If Alpha were suddenly a tremendous hit, HP still has the legal right 
 to pursue it.    ------------------------------  + Date: Wed, 22 Jan 2003 15:48:16 +0000 (UTC) + From: david20@alpha1.mdx.ac.uk (David Webb) : Subject: Re: AlphaServer users given lifeline till Itanium+ Message-ID: <b0mefv$dqj$1@aquila.mdx.ac.uk>   c In article <VC53uGXnvzWg@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes: Z >In article <b0ma0e$c51$1@aquila.mdx.ac.uk>, david20@alpha2.mdx.ac.uk (David Webb) writes: > P >> Which was probably the biggest complaint at the time. If Compaq had announcedQ >> porting of VMS and Tru64 to IA64 but not simultaneously announced the dropping N >> of work on EV8 (and hence in peoples mind the death of the Alpha chip) thenL >> they would have garnered almost universal support from the people in this
 >> newsgroup.  > L >And they would have lost the opportunity for the payments offered by Intel. >   ; Were the payments not just for porting their OS' to IA64 ?  < Are you really saying that Intel paid Compaq to kill Alpha ?J Paying Compaq to port their OS's to IA64 is one thing but if there are anyK official documents proving that Intel linked this payment to Compaq killing < off their chip then I'd think Intel would be in big trouble.    
 David Webb VMS and Unix team leader CCSS Middlesex University      J >Which of those opportunities best matches the letter and spirit of the USF >securities laws requiring them to do what is financially best for the >stockholders? > F >If Alpha were suddenly a tremendous hit, HP still has the legal right >to pursue it.   ------------------------------    Date: 22 Jan 2003 10:25:28 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) : Subject: Re: AlphaServer users given lifeline till Itanium3 Message-ID: <s+k1KODmlcZE@eisner.encompasserve.org>   Y In article <b0mefv$dqj$1@aquila.mdx.ac.uk>, david20@alpha1.mdx.ac.uk (David Webb) writes: e > In article <VC53uGXnvzWg@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes: [ >>In article <b0ma0e$c51$1@aquila.mdx.ac.uk>, david20@alpha2.mdx.ac.uk (David Webb) writes:  >>Q >>> Which was probably the biggest complaint at the time. If Compaq had announced R >>> porting of VMS and Tru64 to IA64 but not simultaneously announced the droppingO >>> of work on EV8 (and hence in peoples mind the death of the Alpha chip) then M >>> they would have garnered almost universal support from the people in this  >>> newsgroup. >>M >>And they would have lost the opportunity for the payments offered by Intel.  >> > = > Were the payments not just for porting their OS' to IA64 ?  > > Are you really saying that Intel paid Compaq to kill Alpha ?  I Compaq has not killed Alpha.  The payments were made in return for making H IA64 the chosen environment for servers in the future.  I presume CompaqK (now HP) is free to enhance Alpha as much as possible for operating systems  made by others (e.g., Apple).   L > Paying Compaq to port their OS's to IA64 is one thing but if there are anyM > official documents proving that Intel linked this payment to Compaq killing > > off their chip then I'd think Intel would be in big trouble.  G So you are right, in a nit-picking sort of way.  Compaq can honor their H agreement with Intel (we presume) by continuing to develop Alpha but not support the new Alphas on VMS.   ------------------------------  % Date: Wed, 22 Jan 2003 09:46:09 -0800 $ From: Shane Smith <ssmith@icius.com>: Subject: RE: AlphaServer users given lifeline till Itanium0 Message-ID: <01C2C1FB.1DBD1C00@sulfer.icius.com>  D It would be perfectly in character for Intel to officially offer theF money for the port, and unofficially mention that support for the ideaH might dry up within Intel management if the Alpha were to continue to be
 a threat.   A When the Athlon first came out, rumours were rife that Intel were G hinting to motherboard manufacturers that there might be a support chip H shortage if they were to produce an Athlon motherboard. I'm not saying IG have direct knowledge of this (if you're listening, Intel lawyers), but 4 I heard of it from several usually reliable sources.  G And only recently, they were taken to court for popping AMD balloons on F motherboard manufacturers stands at a trade show. Can you believe that? one? They actually won the case because they apparently had the C permission of the stand owners. You can just imagine it, can't you: E Excuse me Sir, you appear to have an AMD balloon here. You know, it's A strange but chip shortages have this strange correlation with AMD ; balloon usage. Not quite sure how that happens, but.... ;-)    Shane    -----Original Message-----@ From: david20@alpha1.mdx.ac.uk [mailto:david20@alpha1.mdx.ac.uk]) Sent: Wednesday, January 22, 2003 7:48 AM  To: Info-VAX@Mvb.Saic.Com : Subject: Re: AlphaServer users given lifeline till Itanium    3 In article <VC53uGXnvzWg@eisner.encompasserve.org>, / Kilgallen@SpamCop.net (Larry Kilgallen) writes: L >In article <b0ma0e$c51$1@aquila.mdx.ac.uk>, david20@alpha2.mdx.ac.uk (David
 Webb) writes:  > F >> Which was probably the biggest complaint at the time. If Compaq had	 announced H >> porting of VMS and Tru64 to IA64 but not simultaneously announced the droppingN >> of work on EV8 (and hence in peoples mind the death of the Alpha chip) thenL >> they would have garnered almost universal support from the people in this
 >> newsgroup.  > L >And they would have lost the opportunity for the payments offered by Intel. >   ; Were the payments not just for porting their OS' to IA64 ?  < Are you really saying that Intel paid Compaq to kill Alpha ?F Paying Compaq to port their OS's to IA64 is one thing but if there are any C official documents proving that Intel linked this payment to Compaq  killing < off their chip then I'd think Intel would be in big trouble.    
 David Webb VMS and Unix team leader CCSS Middlesex University      J >Which of those opportunities best matches the letter and spirit of the USF >securities laws requiring them to do what is financially best for the >stockholders? > F >If Alpha were suddenly a tremendous hit, HP still has the legal right >to pursue it.   ------------------------------  # Date: Wed, 22 Jan 2003 18:03:42 GMT # From: "John Smith" <a@nonymous.com> : Subject: Re: AlphaServer users given lifeline till ItaniumH Message-ID: <22BX9.75775$ej1.69246@news02.bloor.is.net.cable.rogers.com>  : "Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in message- news:VC53uGXnvzWg@eisner.encompasserve.org... F > In article <b0ma0e$c51$1@aquila.mdx.ac.uk>, david20@alpha2.mdx.ac.uk (David Webb) writes: > C > > Which was probably the biggest complaint at the time. If Compaq 
 had announced E > > porting of VMS and Tru64 to IA64 but not simultaneously announced  the droppingD > > of work on EV8 (and hence in peoples mind the death of the Alpha
 chip) thenE > > they would have garnered almost universal support from the people  in this  > > newsgroup. > F > And they would have lost the opportunity for the payments offered by Intel. > D > Which of those opportunities best matches the letter and spirit of the USC > securities laws requiring them to do what is financially best for  the  > stockholders?  > A > If Alpha were suddenly a tremendous hit, HP still has the legal  right  > to pursue it.   C Pursue what? By the time HP wakes up and smells the coffee, several A other architecture families will have bypassed IA-64 and make the ? catch-up effort with EV8 too little too late. Which team of CPU 3 architects will HP pull out of a hat to do the job?   D De jure and de facto are two vastly different propositions - you may5 legally have the right but in the market you're dead.    ------------------------------  + Date: Wed, 22 Jan 2003 18:10:25 +0000 (UTC) + From: david20@alpha1.mdx.ac.uk (David Webb) : Subject: Re: AlphaServer users given lifeline till Itanium+ Message-ID: <b0mmqg$gg2$1@aquila.mdx.ac.uk>   c In article <s+k1KODmlcZE@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes: Z >In article <b0mefv$dqj$1@aquila.mdx.ac.uk>, david20@alpha1.mdx.ac.uk (David Webb) writes:f >> In article <VC53uGXnvzWg@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes:\ >>>In article <b0ma0e$c51$1@aquila.mdx.ac.uk>, david20@alpha2.mdx.ac.uk (David Webb) writes: >>> R >>>> Which was probably the biggest complaint at the time. If Compaq had announcedS >>>> porting of VMS and Tru64 to IA64 but not simultaneously announced the dropping P >>>> of work on EV8 (and hence in peoples mind the death of the Alpha chip) thenN >>>> they would have garnered almost universal support from the people in this >>>> newsgroup.  >>> N >>>And they would have lost the opportunity for the payments offered by Intel. >>>  >>  > >> Were the payments not just for porting their OS' to IA64 ? ? >> Are you really saying that Intel paid Compaq to kill Alpha ?  > J >Compaq has not killed Alpha.  The payments were made in return for makingI >IA64 the chosen environment for servers in the future.  I presume Compaq L >(now HP) is free to enhance Alpha as much as possible for operating systems >made by others (e.g., Apple). >   M >> Paying Compaq to port their OS's to IA64 is one thing but if there are any N >> official documents proving that Intel linked this payment to Compaq killing? >> off their chip then I'd think Intel would be in big trouble.  > H >So you are right, in a nit-picking sort of way.  Compaq can honor theirI >agreement with Intel (we presume) by continuing to develop Alpha but not  >support the new Alphas on VMS.   N No not supporting VMS and Tru64 on any new alphas produced is exactly the sameK as "killing" the chip since apart from a relatively small linux usage those " were the OSs which used that chip.  M Any papers proving that Intel linked it's payments to Compaq not just porting M it's Operating systems to IA64 but to either stopping development of Alpha or H to committing to only run those OSs on IA64 rather than on both would be equally as damaging for Intel.N If Compaq made that decision on its own then thats fine since the chip and OSsD are theirs to do with as they wish. However Intel offering financial, inducements for Compaq to do that is not OK.  
 David Webb VMS and Unix team leader CCSS Middlesex University   ------------------------------  % Date: Wed, 22 Jan 2003 08:33:44 +0100 ' From: JOUKJ <joukj@hrem.stm.tudelft.nl> " Subject: Re: automatic screensaver2 Message-ID: <3E2E4958.6050205@hrem.stm.tudelft.nl>   Tom Flick wrote:F > I have an Alpha running OpenVMS 6.2.  How can I get a screensaver toJ > automatically appear if the mouse or keyboard has been idle for a while? > G Another resonder already pointeed out you coud use xscreensaver if you   have an X-server running. A An alternative is a combination of Xautolock and Xlockmore (also  B applicable only if you have it on a X11-screen). Pointer to these % programs can be found on my web-page: (    http://nchrem.tnw.tudelft.nl/openvms/                 Jouk   ------------------------------  # Date: Wed, 22 Jan 2003 12:09:36 GMT % From: "Tom Flick" <tomarg@cfl.rr.com> " Subject: Re: automatic screensaver> Message-ID: <4SvX9.115017$Sa3.2847859@twister.tampabay.rr.com>  % Thanks very much.  I'll check it out.    Tom    ------------------------------  % Date: Wed, 22 Jan 2003 12:50:19 -0500 A From: "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> " Subject: Re: automatic screensaver. Message-ID: <3e2ed9dc$1_2@hpb10302.boi.hp.com>  K Uh.  How about installing V7.3-1 and Motif V1.2-6?  It's there.  After all,4K V6.2 is like what - 7 years old?  Lots of improvments between then and now.     0 "Tom Flick" <tomarg@cfl.rr.com> wrote in message5 news:ZWlX9.27822$o8.584498@twister.tampabay.rr.com...eF > I have an Alpha running OpenVMS 6.2.  How can I get a screensaver toJ > automatically appear if the mouse or keyboard has been idle for a while? >P > Tom F. >c >s   ------------------------------  # Date: Wed, 22 Jan 2003 18:40:15 GMTi" From:   VAXman-  @SendSpamHere.ORG" Subject: Re: automatic screensaver0 Message-ID: <00A1A5A4.4421D46D@SendSpamHere.ORG>  r In article <3e2ed9dc$1_2@hpb10302.boi.hp.com>, "Fred Kleinsorge" <kleinsorge@star-dot-zko-dot-dec-dot-com> writes: >eL >Uh.  How about installing V7.3-1 and Motif V1.2-6?  It's there.  After all,L >V6.2 is like what - 7 years old?  Lots of improvments between then and now.  2 Except for something to replace DisplayPostscript!   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM             e5   "Well my son, life is like a beanstalk, isn't it?" e   ------------------------------  % Date: Wed, 22 Jan 2003 08:50:48 +0100aC From: Eberhard Heuser-Hofmann <vaxinf@chclu.chemie.uni-konstanz.de> ' Subject: Changed DVDWRITE homepage pathP> Message-ID: <00A1A57B.D2CDCD88.7@CHCLU.CHEMIE.UNI-KONSTANZ.DE>   Hi,w  D I have switched my ADSL-contract from volume to flat with the result4 that my DVDwrite-homepage+email-address has changed.   So use  # http://home.tiscali.de/dvd4openvms a  < if you want to learn about a DVD-burning program that works 5 for DVD-R(W) (i.e. Pioneer DVR-A03/04/05) as well as  ' for DVD+R(W) (i.e. HP200i/Ricoh5125A). B  u regardse	 Eberhard cO ===============================================================================    Dr. Eberhard Heuser-Hofmann  Univ. Konstanz Fakultaet fuer Chemiet Universitaets-Strasse 10 D-78464 Konstanz Germanyb. Phone: +49-7531-88-2026, FAX: +49-7531-88-3139* email: vaxinf@chclu.chemie.uni-konstanz.de  O ===============================================================================    ------------------------------    Date: 22 Jan 2003 08:52:54 -0800* From: cgilley@bravewc.com (Charles Gilley)7 Subject: Command/API to list locked files in a cluster?r< Message-ID: <139d5a58.0301220852.959da01@posting.google.com>  C Subject pretty much says it all - I have a cluster of Alphas, and IoD need to be able to list who has what files locked, preferably across the cluster.  C Haven't found anything yet, but reading about file locking makes mee! want to believe something exists.n   Thanks   ------------------------------  % Date: Wed, 22 Jan 2003 09:49:45 -0800e$ From: Shane Smith <ssmith@icius.com>; Subject: RE: Command/API to list locked files in a cluster?O0 Message-ID: <01C2C1FB.A0DE7200@sulfer.icius.com>  F It isn't cross cluster, but "SHOW DEV/FILES diskname" shows open filesD within a single node. It's always been enough for me. Or do you mean6 locked in a specific way? ie ALLOW NONE or ALLOW READ?   Shanen   -----Original Message-----6 From: cgilley@bravewc.com [mailto:cgilley@bravewc.com]) Sent: Wednesday, January 22, 2003 8:53 AMe To: Info-VAX@Mvb.Saic.Comu7 Subject: Command/API to list locked files in a cluster?     C Subject pretty much says it all - I have a cluster of Alphas, and ISD need to be able to list who has what files locked, preferably across the cluster.  C Haven't found anything yet, but reading about file locking makes me ! want to believe something exists.>   Thanks   ------------------------------  + Date: Wed, 22 Jan 2003 17:22:58 +0100 (MET)n9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>nL Subject: copyright (was: RE: EV7 hits, blows away Sun, IBM, 2-3x VMS gains!); Message-ID: <01KRJKO8C22Q8WW5NX@sysdev.deutsche-boerse.com>i  J > I think that there is a question of fair use here. Posting an article inI > a newsgroup does not invalidate the copyright of an article. The writer E > still maintains rights over the article and ayone wanting to use itu7 > commercially would still need to contact the author. t > B > But if an article is posted in a public forum, doesn't it become' > implicitely publicly distributable ? L  	 Check outn  0    http://www.templetons.com/brad/copymyths.html  F At http://www.templetons.com/brad/ (which is also where you end up if F you go to http://www.templetons.com/ ), there is a lot of interesting I stuff by, shall we say, an internet pioneer.  Brew up a cup of tea, take  I your shoes off, and surf for a couple of hours.  There are links to lots TE of interesting things, including stuff that puts Bill Gates's wealth 6E into perspective (sort of like Carl Sagan did by compressing Earth's iD history into a year---nothing more complex than algae until August, > dinosaurs around for a month or so, humans for a few minutes).   ------------------------------  % Date: Wed, 22 Jan 2003 03:51:52 -0400i0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>& Subject: Determining outbound identity/ Message-ID: <3E2E4D8D.8AFF6FCA@vl.videotron.ca>a  6 Ok, foreign mail protocol generates an outbound email.  L Is there a standard way, across all TCPIP stacks, to find out what the From: identity would be ?X  
 Can I assume:.  / username@ tcpip$inet_host . tcpip$inet_domain ?$  = Or are those logical names non-standard across TCPIP stacks ?q  E Or should I just ask the system manager to add the domain name in the 5 software's config and not rely on any logical names ?w   ------------------------------    Date: 22 Jan 2003 11:13:22 +0100' From: huber@mppmu.mpg.de (Joseph Huber) * Subject: Re: Determining outbound identity+ Message-ID: <31fm53vdgga8@vms.mppmu.mpg.de>   b In article <3E2E4D8D.8AFF6FCA@vl.videotron.ca>, JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes:8 > Ok, foreign mail protocol generates an outbound email.N > Is there a standard way, across all TCPIP stacks, to find out what the From: > identity would be ?  > Can I assume:g1 > username@ tcpip$inet_host . tcpip$inet_domain ? ? > Or are those logical names non-standard across TCPIP stacks ? G > Or should I just ask the system manager to add the domain name in the 7 > software's config and not rely on any logical names ?c  < Probably no, there is no standard amongst the TCP/IP stacks.  D For TCPIP/IP some items (detailed description in the TCPIP doc set):  . Standard is "From: username@tcpip$inet_host" .M At user level, logical TCPIP$SMTP_FROM defines the "From:" field individuallyn- (but this can be forbidden by system logical: 9       "DEFINE/SYSTEM TCPIP$SMTP_PROHIBIT_USER_HEADERS 1" w ).  7 TCPIP config can  define the @domain field system-wide:wG       SET CONFIGURATION SMTP /SUBSTITUTE_DOMAIN=(HIDDEN,NAME=mydomain) U   --  N Joseph "Sepp" Huber   mailto:joseph.huber@web.de   http://www.huber-joseph.de/   ------------------------------    Date: 22 Jan 2003 07:57:56 -0800+ From: steve.spires@torex.com (Steve Spires) G Subject: Re: Function keys no longer work after TCPware upgrade to 5.6.h= Message-ID: <57136ccc.0301220757.7c875b1e@posting.google.com>o  h peter@langstoeger.at (Peter LANGSTOEGER) wrote in message news:<ieHW9.83763$TY.811020@news.chello.at>...l > In article <57136ccc.0301170404.34def9e@posting.google.com>, steve.spires@torex.com (Steve Spires) writes:( > >Has anybody seen this problem before? >  > It depends ;-) > C > >We carried out an upgrade of both VMS and TCPware, so we are noweD > >sitting at VMS 7.3 and TCPware 5.6 [on Alpha in case this makes aG > >difference] but since doing the upgrade function keys [F1 to F14] no5H > >longer work correctly - it looks like the escape sequences might have! > >changed, I'm not sure exactly.a > L > What function keys ? Do you mean the TCPware TELNET.EXE which does a VT200H > or a 3270 communication to a remote system or do you mean a (?) client$ > connecting to the TCPware system ? > F > I've seen the former case with the upgrade from TCPware V5.1 to V5.2I > while connecting (with TN3270) to an AS400. The problem was never fixedaH > and so I used the TELNET.EXE of V5.1 up to the V5.6 installation (when > connecting to the AS400).e > ' > I can't comment on the latter case...p    B Thanks. It looks like [I'm getting this problem repotred second orA third hand...!] the problem is when doing rsh commands, where theeC ESCAPE key is sending a '$' rather than the expected sequence. Thisu8 seems to be cuasing problems with function keys as well.  B I REALLY must try and get a 'proper' idea of the exact problem but' this is what I have to go on currently.U  B I will try and get the site to allow me to carry out the operation; myself and see if I get a clearer idea of what's happening.t  E Any ideas from anyone in the meantime would be once more appreciated.p   Cheers   Steve Spires Technical Consultant Torex Health   ------------------------------  % Date: Wed, 22 Jan 2003 07:24:21 -0500d2 From: rdeininger@mindspring.com (Robert Deininger)- Subject: Re: Galaxy support Profile for ES80?uL Message-ID: <rdeininger-2201030724210001@user-2ive3rg.dialup.mindspring.com>  
 In articleI <BADE2849CBA500D2.2B9DEA68AE01B2BA.8483892FFE446D18@lp.airnews.net>, "Hall Kuff" <kuff@comcast.net> wrote:p  L >    Does anyone have the Galaxy support profile for ES80 systems...? I hearM >that software partitions only are supported... that doesn't seem right given E >the isolation possible with the PCI bus architecture on the ES80...?t  4 Read the release notes.  No support for a while yet:  < http://www.openvms.compaq.com/doc/731final/6665/6665pro.html   (Section 1.11)G Galaxy support on ES47 and GS1280 AlphaServer series systems is not yetiH available but is expected to be qualified soon after the initial releaseF of these systems. Galaxy support, which includes support of the sharedJ memory cluster interconnect, will be announced on the OpenVMS web page at:    ...  J I expect ES80 will support partitions as small as 2P each, if you have allA the need outlying hardware bits.  I don't know if 1P partions aren	 possible.m   ------------------------------  # Date: Wed, 22 Jan 2003 06:48:30 GMT>0 From: Ron Shepard <ron-shepard@NOSPAM.attbi.com> Subject: Re: How to Backup OSXA Message-ID: <ron-shepard-B21C3A.00482922012003@netnews.attbi.com>C  E Throughout the history of computing (since the 50's at least), there lD has always been some kind of media that is cheaper than the primary A disk/drum/etc storage.  Back in the 70's for example, disk space  C cost about $100/MB and tape storage cost about $1/MB.  In the past mD few years, the cost of disk space has decreased tremendously, about E a factor of 100,000 since the 70's.  HD storage is now about $1/GB.  aC However, tape costs have not changed as much, a GB of tape storage oC now costs maybe 4 or 5 times less than it did 10 years ago.  DVD-R MC is still about $2/GB (not counting the drive, just the media), and  ( CD-R is comparable.  DLT is about $1/GB.  C We now have a situation where hard disk storage is either cheaper, tD or comprable, to the cost of the media that is used to back it up.  E It makes you wonder what is the point.  Why not simply buy more hard oD drives, back up the data, power them down and store them in a vault?  A How will this situation change in the future?  Disk storage will  > almost certainly get cheaper in the future.  Is there any new @ storage media that has the ability to keep up with disk storage A costs?  What happens in 10 years if disk storage is cheaper by a  E factor of 1000 than current prices, but tape/DVD costs are only 4 or L 5 times cheaper?   $.02 -Ron Shepard    ------------------------------  # Date: Wed, 22 Jan 2003 09:53:53 GMTg% From: Tony Lawrence <tony@pcunix.com>  Subject: Re: How to Backup OSX? Message-ID: <RStX9.659308$GR5.492561@rwcrnsc51.ops.asp.att.net>d   Rob Young wrote:c > In article <lZlGvXhDQJ1e@eisner.encompasserve.org>, young_r@encompasserve.org (Rob Young) writes:  > Y >>In article <JCiX9.775606$P31.573375@rwcrnsc53>, Tony Lawrence <tony@pcunix.com> writes:u >> >>>Rob Young wrote:o >>J >>>>	And yes, if you narrow down to the small world you are talking about,G >>>>	do full backups of everything all the time - because you can, and e7 >>>>	because they are easier to understand and support!d >>>CL >>>Which is what I have been saying.  But there's more reason than that, as L >>>I detailed at http://aplawrence.com/Unixart/diffbacks.html : damage to a G >>>tape in the middle of an incremental set can be a very serious loss rK >>>because the next valid tape that has certain data may be much older and <J >>>in some cases (related files) may cause the entire set to be abandoned. >>>u* >>>Been there, seen that.  More than once. >>>2 >>@ >>	Yes, in your small world.  The TSM backend is fault tolerant,> >>	keep that in mind.  I'm sure there is a document explaining; >>	how it works somewhere but you really aren't interested.  >> >  > A > 	Clarifying.  A tape may fail while writing to it.  That countsw? > 	as a miss.  You have to run the backup again from the client D > 	side.  The tape that failed twin comes back from offsite to synch > 	back up.-  H A tape may fail ANYTIME.  Tapes fail while being read.  Tapes get eaten H by rats (well, yeah, that wasn't one of my higher class clients), lost, E mislabeled, stepped on etc.  That's why the more redundancy you have rE with tapes, the safer you are.  Again, I've been there and seen this mI kind of thing first hand.  Apparently in your oh so larger than my world t   the unexpected never happens.s       -- w
 Tony LawrenceH, Free SCO, Mac OS X and  Linux Skills Tests: % http://aplawrence.com/skillstest.htmlC   ------------------------------    Date: 22 Jan 2003 09:15:21 +0100C From: vaxinf@chclu.chemie.uni-konstanz.de (Eberhard Heuser-Hofmann)t Subject: Re: How to Backup OSX- Message-ID: <3e2e5319$1@news.uni-konstanz.de>-  E In article <ron-shepard-B21C3A.00482922012003@netnews.attbi.com>, Ron-. Shepard <ron-shepard@NOSPAM.attbi.com> writes:G |>Throughout the history of computing (since the 50's at least), there nF |>has always been some kind of media that is cheaper than the primary C |>disk/drum/etc storage.  Back in the 70's for example, disk space pE |>cost about $100/MB and tape storage cost about $1/MB.  In the past  F |>few years, the cost of disk space has decreased tremendously, about G |>a factor of 100,000 since the 70's.  HD storage is now about $1/GB.  /E |>However, tape costs have not changed as much, a GB of tape storage lE |>now costs maybe 4 or 5 times less than it did 10 years ago.  DVD-R >E |>is still about $2/GB (not counting the drive, just the media), and  * |>CD-R is comparable.  DLT is about $1/GB. |>E |>We now have a situation where hard disk storage is either cheaper, hF |>or comprable, to the cost of the media that is used to back it up.  G |>It makes you wonder what is the point.  Why not simply buy more hard wF |>drives, back up the data, power them down and store them in a vault? |>  @ I've read an article in a german computer magazine, that states,H that it is nowadays cheaper to throw off the DLT-Tape library and to buy a IDE-based raidsystem.   C |>How will this situation change in the future?  Disk storage will a@ |>almost certainly get cheaper in the future.  Is there any new B |>storage media that has the ability to keep up with disk storage C |>costs?  What happens in 10 years if disk storage is cheaper by a iG |>factor of 1000 than current prices, but tape/DVD costs are only 4 or g |>5 times cheaper?    ? There is difference between backup and archiving. For archivingo9 purpose use CD/DVD's for backup tape or disks. DVD's withv= much higher capacity than today will be out this or next year = (keyword "blue laser"). But the drives will be very expensiver in the beginning as usual.   |> |>$.02 -Ron Shepardo |>   eberhard   ------------------------------  # Date: Wed, 22 Jan 2003 09:50:51 GMTs% From: Tony Lawrence <tony@pcunix.com>a Subject: Re: How to Backup OSX/ Message-ID: <%PtX9.781687$P31.593619@rwcrnsc53>e   Rob Young wrote:Y > In article <JCiX9.775606$P31.573375@rwcrnsc53>, Tony Lawrence <tony@pcunix.com> writes:t >  >>Rob Young wrote: >  > I >>>	And yes, if you narrow down to the small world you are talking about,tF >>>	do full backups of everything all the time - because you can, and 6 >>>	because they are easier to understand and support! >>K >>Which is what I have been saying.  But there's more reason than that, as  K >>I detailed at http://aplawrence.com/Unixart/diffbacks.html : damage to a  F >>tape in the middle of an incremental set can be a very serious loss J >>because the next valid tape that has certain data may be much older and I >>in some cases (related files) may cause the entire set to be abandoned.9 >>) >>Been there, seen that.  More than once.> >> >  > @ > 	Yes, in your small world.  The TSM backend is fault tolerant,> > 	keep that in mind.  I'm sure there is a document explaining; > 	how it works somewhere but you really aren't interested., >   H As I've tried to explain, TSM is completely irrelevant to this.  It's a C wonderful thing, yes, but it would be HOW YOU BACK THAT UP that is g important and relevant.m    I And my world is hardly small :-)  Kind of a funny thing for a VMS guy to k say, you know?           -- r
 Tony Lawrencee, Free SCO, Mac OS X and  Linux Skills Tests: % http://aplawrence.com/skillstest.htmlr   ------------------------------  # Date: Wed, 22 Jan 2003 10:09:23 GMTr% From: Tony Lawrence <tony@pcunix.com>  Subject: Re: How to Backup OSX? Message-ID: <n5uX9.659454$GR5.493142@rwcrnsc51.ops.asp.att.net>p   Ron Shepard wrote:G > Throughout the history of computing (since the 50's at least), there  F > has always been some kind of media that is cheaper than the primary C > disk/drum/etc storage.  Back in the 70's for example, disk space dE > cost about $100/MB and tape storage cost about $1/MB.  In the past tF > few years, the cost of disk space has decreased tremendously, about G > a factor of 100,000 since the 70's.  HD storage is now about $1/GB.  tE > However, tape costs have not changed as much, a GB of tape storage  E > now costs maybe 4 or 5 times less than it did 10 years ago.  DVD-R nE > is still about $2/GB (not counting the drive, just the media), and -* > CD-R is comparable.  DLT is about $1/GB. > E > We now have a situation where hard disk storage is either cheaper, rF > or comprable, to the cost of the media that is used to back it up.  G > It makes you wonder what is the point.  Why not simply buy more hard oF > drives, back up the data, power them down and store them in a vault? > C > How will this situation change in the future?  Disk storage will .@ > almost certainly get cheaper in the future.  Is there any new B > storage media that has the ability to keep up with disk storage C > costs?  What happens in 10 years if disk storage is cheaper by a aG > factor of 1000 than current prices, but tape/DVD costs are only 4 or w > 5 times cheaper?  I If the drives are cheap enough that you can maintain the depth of backup oI that you need, why not?  Indeed, many of us could afford to do that now: dI while I don't particularly want to tie up the cost of a dozen or so disk lF drives, that really isn't all that much money nowadays.  As the disks C are much more rugged than they used to be, it wouldn't be scary to k% transport them home offsite casually.e  A But if we free up the radio spectrum and have high speed "pipes" rG everywhere, maybe we don't need to physically move those drives around d at all?      -- I
 Tony Lawrence , Free SCO, Mac OS X and  Linux Skills Tests: % http://aplawrence.com/skillstest.htmln   ------------------------------  # Date: Wed, 22 Jan 2003 10:13:32 GMT % From: Tony Lawrence <tony@pcunix.com>n Subject: Re: How to Backup OSX? Message-ID: <g9uX9.659504$GR5.493139@rwcrnsc51.ops.asp.att.net>p   Eberhard Heuser-Hofmann wrote:G > In article <ron-shepard-B21C3A.00482922012003@netnews.attbi.com>, Ron'0 > Shepard <ron-shepard@NOSPAM.attbi.com> writes:I > |>Throughout the history of computing (since the 50's at least), there MH > |>has always been some kind of media that is cheaper than the primary E > |>disk/drum/etc storage.  Back in the 70's for example, disk space .G > |>cost about $100/MB and tape storage cost about $1/MB.  In the past lH > |>few years, the cost of disk space has decreased tremendously, about I > |>a factor of 100,000 since the 70's.  HD storage is now about $1/GB.  rG > |>However, tape costs have not changed as much, a GB of tape storage aG > |>now costs maybe 4 or 5 times less than it did 10 years ago.  DVD-R lG > |>is still about $2/GB (not counting the drive, just the media), and  , > |>CD-R is comparable.  DLT is about $1/GB. > |>G > |>We now have a situation where hard disk storage is either cheaper,  H > |>or comprable, to the cost of the media that is used to back it up.  I > |>It makes you wonder what is the point.  Why not simply buy more hard eH > |>drives, back up the data, power them down and store them in a vault? > |> > B > I've read an article in a german computer magazine, that states,J > that it is nowadays cheaper to throw off the DLT-Tape library and to buy > a IDE-based raidsystem.  >   I But it's not suitable for backup.  Not by itself.  That's just a cheaper uB and less functional TSM system: you still need to back that up to  removable media.  As you say:G     > A > There is difference between backup and archiving. For archivingr; > purpose use CD/DVD's for backup tape or disks. DVD's witha? > much higher capacity than today will be out this or next yearh? > (keyword "blue laser"). But the drives will be very expensivei > in the beginning as usual.  B Right.  I like DVDRAM for small systems and if they can boost the A capacity up it has advantages over tapes in some ways (though it l- probably will never match tape write speeds).h       --  
 Tony Lawrenceo, Free SCO, Mac OS X and  Linux Skills Tests: % http://aplawrence.com/skillstest.html.   ------------------------------  # Date: Wed, 22 Jan 2003 10:02:41 GMT % From: Tony Lawrence <tony@pcunix.com>q Subject: Re: How to Backup OSX= Message-ID: <5%tX9.41879$4u5.38250@rwcrnsc52.ops.asp.att.net>f   David J. Dachtera wrote: > Tony Lawrence wrote: >  >>Howard S Shubs wrote:h >>0 >>>In article <Lj%W9.71429$1q3.10159@sccrnsc01>,* >>> Tony Lawrence <tony@pcunix.com> wrote: >>>  >>>e >>>nJ >>>>I have no idea what you are referring to.  Many businesses have a 8 toH >>>>16 hour window in which to do backup.  If their entire system can beA >>>>backed up to removable media in that time frame, they are noto! >>>>constrained by space or time.t >>>  >>>eG >>>In my experience, this is vanishingly rare.  Most people like to get0H >>>home rather than watch tapes spin, even if they can take down time in# >>>order to get a nice cold backup.e >>>- >>J >>I must be an idiot.  I've been arguing with you about the relative valueI >>of backup methods and you obviously are so wet behind the ears that youC# >>think people do backups manually.  >  > 0 > If you know how to automate backups including: >  > o Shutdown the running system  > o Boot up the stand-alone CD  @ Not necessary under any conditions.  I suspect you haven't much  experience at this.a  ! > o Kick-off a system-disk backupT# > o Shutdown the stand-alone systemn3 > o Reboot the live, production system, sans app.'sn+ > o Backup the production application diskse > o Restart the applications.d > I > ...you'll not only become a very wealthy man, you'll fulfill the dreamsh! > of many denizens of this group.f  E Sigh.  All of those things are done every day.  Your first two steps fF aren't necessary, but even that could be easily automated and in fact H I've done things like that for other reasons.  Not necessary for backup  though.      > G > Actually, I could see doing that with a very sophisticated Reflections( > script (RCL or RBS, take your choice). > ; > Oh, yeah, there's that zero-downtime thing to consider....  A Few of my clients get that, although we can come close with disk nI snapshots.  If you REALLY need zero down time, that requires cooperation  I from the application software: it has to be able to bring all data files iB into sync and buffer all transactions while the snapshot is being I prepared.  As that literally takes seconds,  that should be no hardship, d, but I don't have anything with that feature.  E There's nothing technically difficult about it though and everything 0? else is simple and BEING DONE EVERY DAY.  Nothing particularly a sophisticated about it.n       -- i
 Tony Lawrence , Free SCO, Mac OS X and  Linux Skills Tests: % http://aplawrence.com/skillstest.htmlt   ------------------------------  # Date: Wed, 22 Jan 2003 10:31:22 GMT - From: bdc@world.std.com (Brian 'Jarai' Chase)  Subject: Re: How to Backup OSX& Message-ID: <H942KA.L3C@world.std.com>  A In article <ron-shepard-B21C3A.00482922012003@netnews.attbi.com>,m2 Ron Shepard  <ron-shepard@NOSPAM.attbi.com> wrote:  G > Throughout the history of computing (since the 50's at least), there aF > has always been some kind of media that is cheaper than the primary C > disk/drum/etc storage.  Back in the 70's for example, disk space  E > cost about $100/MB and tape storage cost about $1/MB.  In the past oF > few years, the cost of disk space has decreased tremendously, about G > a factor of 100,000 since the 70's.  HD storage is now about $1/GB.  nE > However, tape costs have not changed as much, a GB of tape storage iE > now costs maybe 4 or 5 times less than it did 10 years ago.  DVD-R  E > is still about $2/GB (not counting the drive, just the media), and N* > CD-R is comparable.  DLT is about $1/GB.  J Well, LTO-1 (100GB uncompressed), runs me about $0.60/GB in bulk. I expectJ it drop even further now that LTO-2 technology is readying for prime time.  E > We now have a situation where hard disk storage is either cheaper, iF > or comprable, to the cost of the media that is used to back it up.  G > It makes you wonder what is the point.  Why not simply buy more hard @F > drives, back up the data, power them down and store them in a vault? > C > How will this situation change in the future?  Disk storage will n@ > almost certainly get cheaper in the future.  Is there any new B > storage media that has the ability to keep up with disk storage C > costs?  What happens in 10 years if disk storage is cheaper by a lG > factor of 1000 than current prices, but tape/DVD costs are only 4 or t > 5 times cheaper?  I I'd agree that disk is gaining on tape in terms of $/GB, but I think it'soJ gaining a little more slowly than you might think.  Other useful things toJ consider are things like cm^3/GB and g/GB--because you've got to store and& may need to move around those backups.  F Another important thing to consider is the amount of data required forJ your backups.  Today, it's quite feasible to cost effectively backup a fewH tens of TB using disk only solutions.  But disk solutions don't scale asD well as tape solutions do in terms of overall costs.  Once you startF talking about /many/ tens of TB of backups, or hundreds of TB, tape is still the best way to go.d   -brian.m -- sF --- Brian Chase | bdc@world.std.com | http://world.std.com/~bdc/ ----->       IBM 1130 Club Motto:  "Computer People Are Happy People"   ------------------------------    Date: 22 Jan 2003 22:27:48 +0800, From: Paul Repacholi <prep@prep.synonet.com> Subject: Re: How to Backup OSX- Message-ID: <87vg0hb6or.fsf@prep.synonet.com>a  E vaxinf@chclu.chemie.uni-konstanz.de (Eberhard Heuser-Hofmann) writes:   B > I've read an article in a german computer magazine, that states,F > that it is nowadays cheaper to throw off the DLT-Tape library and to > buy a IDE-based raidsystem.    RIGHT...  B Now burn down the building. Tell me when you get the data off your	 IDE raid.i  4 BTW, you do know why it is  called IDE, don't you :)   -- t< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------    Date: 22 Jan 2003 16:31:10 +0100C From: vaxinf@chclu.chemie.uni-konstanz.de (Eberhard Heuser-Hofmann)a Subject: Re: How to Backup OSX- Message-ID: <3e2eb93e$1@news.uni-konstanz.de>f  < In article <87vg0hb6or.fsf@prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes:cG |>vaxinf@chclu.chemie.uni-konstanz.de (Eberhard Heuser-Hofmann) writes:e |>D |>> I've read an article in a german computer magazine, that states,H |>> that it is nowadays cheaper to throw off the DLT-Tape library and to |>> buy a IDE-based raidsystem.3 |>
 |>RIGHT... |>D |>Now burn down the building. Tell me when you get the data off your |>IDE raid.v  9 And tell me when you get the data off your DLT-Library ifx there is fire in that room.O   |>6 |>BTW, you do know why it is  called IDE, don't you :)  7 Use raid10 (Raid5+mirroring). Even if a IDE-disk fails,a you will have no data loss.u   eberhard   |> |>-- >> |>Paul Repacholi                               1 Crescent Rd.,9 |>+61 (08) 9257-1001                           Kalamunda.$B |>                                             West Australia 6076, |>comp.os.vms,- The Older, Grumpier Slashdot0 |>Raw, Cooked or Well-done, it's all half baked.H |>EPIC, The Architecture of the future, always has been, always will be. |>   ------------------------------  # Date: Wed, 22 Jan 2003 16:24:55 GMTh% From: Tony Lawrence <tony@pcunix.com>e Subject: Re: How to Backup OSX= Message-ID: <BzzX9.45777$4u5.39217@rwcrnsc52.ops.asp.att.net>    Eberhard Heuser-Hofmann wrote:> > In article <87vg0hb6or.fsf@prep.synonet.com>, Paul Repacholi! > <prep@prep.synonet.com> writes: I > |>vaxinf@chclu.chemie.uni-konstanz.de (Eberhard Heuser-Hofmann) writes:C > |>F > |>> I've read an article in a german computer magazine, that states,J > |>> that it is nowadays cheaper to throw off the DLT-Tape library and to! > |>> buy a IDE-based raidsystem.u > |> > |>RIGHT... > |>F > |>Now burn down the building. Tell me when you get the data off your
 > |>IDE raid.o > ; > And tell me when you get the data off your DLT-Library ifr > there is fire in that room.y  : You get it from the tapes you rotated off site, of course.     > |>8 > |>BTW, you do know why it is  called IDE, don't you :) > 9 > Use raid10 (Raid5+mirroring). Even if a IDE-disk fails,  > you will have no data loss.o  < Well, no data loss unless the two drives of the raid5 fail..  H Backing up to a disk drive is quick, convenient, and insufficient.  You I ALWAYS need removable media for archival storage.  Before some damn fool eH says "just buy a bigger drive and store multiple sets there", THAT puts $ too much stuff at risk in one place.       -- a
 Tony Lawrence , Free SCO, Mac OS X and  Linux Skills Tests: % http://aplawrence.com/skillstest.html    ------------------------------  % Date: Wed, 22 Jan 2003 09:51:54 -0800f$ From: Shane Smith <ssmith@icius.com> Subject: RE: How to Backup OSX0 Message-ID: <01C2C1FC.0C284400@sulfer.icius.com>  # Integrated Drive Electronics, IIRC.    -----Original Message-----3 From: Paul Repacholi [mailto:prep@prep.synonet.com]d) Sent: Wednesday, January 22, 2003 6:28 AM- To: Info-VAX@Mvb.Saic.Com  Subject: Re: How to Backup OSX    E vaxinf@chclu.chemie.uni-konstanz.de (Eberhard Heuser-Hofmann) writes:-  B > I've read an article in a german computer magazine, that states,F > that it is nowadays cheaper to throw off the DLT-Tape library and to > buy a IDE-based raidsystem.r   RIGHT...  B Now burn down the building. Tell me when you get the data off your	 IDE raid.x  4 BTW, you do know why it is  called IDE, don't you :)   -- e< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------    Date: 22 Jan 2003 08:12:34 -0800( From: robert_kersey@bat.com (Rob Kersey)5 Subject: Re: How to flash VAXstation 4000/90 FEPROMS?-= Message-ID: <f936a854.0301220812.63a69392@posting.google.com>i  ? Can this firmware flashing be done for the 4000/100 vax family?t  A If so, where can i get the firmware from? Any FTP sites anywhere?    Cheers   Robs   ------------------------------  % Date: Wed, 22 Jan 2003 12:03:32 +0000a( From: Nic Clews <sendspamhere@127.0.0.1>9 Subject: Re: Interrupt stack with EMC fibre channel diskst) Message-ID: <3E2E8894.BF1F1342@127.0.0.1>i   Kenneth wrote: > L > I am using VMS 7.2-1 with the EMC fibre channel disks, the interrupt stackJ > for CPU-0 may go up as high as 50% of the CPU. Is there any relationship> > between them and is there any way I can minimise it. Thanks.  @ Two responses have advised 7.3, 7.3-1 has additional support for FibreChannel over 7.3.  D I have some "very happy bunnies" using a 7.3-1 and FC connected SAN. -- e? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences- nclews at csc dot com>   ------------------------------  + Date: Wed, 22 Jan 2003 11:38:27 +0000 (UTC)f+ From: david20@alpha2.mdx.ac.uk (David Webb)L Subject: Re: Is that possible + Message-ID: <b0lvrj$8vp$2@aquila.mdx.ac.uk>i  g In article <f948cf20.0301212135.2e2e23ef@posting.google.com>, taupin974@hotmail.com (taupin974) writes: 8 >i've tried to built all the files, but it doesn't work.3 >i've already read the simh installation procedure.e >2  L Unfortunately as I said at the beginning of this thread I haven't personallyO used any of these emulators. Hopefully someone else on the list will be able to.N give you more guidance. It might help if you explained how the installation is failing in more detail.w  
 David Webb VMS and Unix team leader CCSS Middlesex University  _ >david20@alpha2.mdx.ac.uk (David Webb) wrote in message news:<b0kkdb$ppj$1@aquila.mdx.ac.uk>...rj >> In article <f948cf20.0301211200.6da4070c@posting.google.com>, taupin974@hotmail.com (taupin974) writes:* >> >So there is only simh which is free ?!5 >> >Where can i download a version of this software ?= >> > >>  $ >> Well the source is available from7 >> http://simh.trailing-edge.com/sources/simhv210-2.zip  >> -J >> Whether there are prebuilt binaries (so you wouldn't need visual c++ or >> Mingw gcc) I don't know.k >> n >>  
 >> David Webb= >> VMS and Unix team leader= >> CCSS= >> Middlesex University  >> - >>   ------------------------------    Date: 22 Jan 2003 23:16:11 +0800, From: Paul Repacholi <prep@prep.synonet.com>6 Subject: Re: Itanium with alpha goodies on the way ...- Message-ID: <87n0ltb4g4.fsf@prep.synonet.com>.  & Shane Smith <ssmith@icius.com> writes:  B > Hmmm, I just had a thought. Given that it'll take a long time toC > shoehorn Alpha features into Itanic, and given Intel's history oftD > (ahem) appropriating technologies, I wonder if they started adding3 > Alpha features /before/ the agreement was signed?- ...-   The Pentium Pro is due when? 4  3 Or do you mean fitting a life jacket to the itanic?:   -- m< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.r@                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Wed, 22 Jan 2003 09:25:47 -0800 $ From: Shane Smith <ssmith@icius.com>6 Subject: RE: Itanium with alpha goodies on the way ...0 Message-ID: <01C2C1F8.409A0100@sulfer.icius.com>  F The stuff the designers came up with or got working after Palmer left.   -----Original Message-----7 From: JF Mezei [mailto:jfmezei.spamnot@vl.videotron.ca]e' Sent: Tuesday, January 21, 2003 9:22 PMh To: Info-VAX@Mvb.Saic.Com-6 Subject: Re: Itanium with alpha goodies on the way ...     Shane Smith wrote:I > put it past them. If that's the case, it's possible the 2005 chip couldt > have some Alpha inheritance.  D More likely that the 8086 will have some Alpha inheritance since the 8086 ise< philosophically more compatible to Alpha than the epic IA64.  F Also, between the time Palmer gave away some of Alpha to Intel and the timeB Curly donated the remains to Intel, was there much done to Alpha ?  C I.E. what could Curly have given to Intel that Intel hadn't alreadya gotten
 from Palmer ?a   ------------------------------  % Date: Wed, 22 Jan 2003 15:13:33 +0000l( From: Nic Clews <sendspamhere@127.0.0.1> Subject: keeping it up) Message-ID: <3E2EB51D.500C7614@127.0.0.1>p   WildStaringEyes wrote: > L > If you are an embedded programmer using VxWorks, then how do you ever have > time for sex?   ) OpenVMS:- No v*agra needed to keep it up!    --  ? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciencesu nclews at csc dot com    ------------------------------  % Date: Wed, 22 Jan 2003 09:22:02 -0800h$ From: Shane Smith <ssmith@icius.com> Subject: RE: Marvel Performances0 Message-ID: <01C2C1F7.BDCE9330@sulfer.icius.com>   <Snip> David Dachtera wrote:  >John Smith wrote: >> Ad 1a >> -----? >> Pictures of Florida hurricane evacuation - highways clogged.a& >> Pictures of earthquake devastation.: >> Voiceover: "Your business can't avoid the unavoidable." >> d >> Pictures of explosions.! >> Pictures of buildings on fire. = >> Voiceover: "Your business can't predict the unpredictable"a >> d >> Pictures of Marvel.8 > Voiceover: " But your business can choose hardware andE >>                    software that make events like these immaterial 6 >>                    to the survival of your company. >>  > >>                   Servers and OpenVMS software from HP have= >>                   been keeping companies in business aftere >> catastrophic-) >>                   events for 25 years.1 >:! >Picture: Toaster, bread in slots( >e2 >>                   All those 'other' companies,  > # >Picture: Toaster, toast popping upp >e, >>                   ...they're just toast." >c5 >Picture: OpenVMS logo dominant, HP logo sub-dominant  >k, >		OpenVMS - when downtime is not an option. >i3 >> Copyright 2003, John Smith. All Rights Reserved.s >o9 >Additions (C)2003 David J Dachtera, All Rights Reserved.s  8 Picture: Winged pig flying over Battersea power station.   (Sorry.)   Shanel   ------------------------------  % Date: Wed, 22 Jan 2003 11:33:19 -0330e, From: "Jeff Barnes" <barnesjw@dfo-mpo.gc.ca>" Subject: Re: Multinet FTP Question* Message-ID: <CsyX9.313$84.94621@localhost>  J Thanks for your responses.  As you (and Peter Weaver) suggest there was an8 invalid "set command" in the individual users login.com.    F "Alan Adams" <alan.adams@orchard-way.freeserve.co.uk> wrote in message9 news:d3ce81b84b.Alan.Adams@orchard-way.freeserve.co.uk...y+ > In message <JzcX9.201$84.74771@localhost>T9 >           "Jeff Barnes" <barnesjw@dfo-mpo.gc.ca> wrote:e >eI > > Part-time OpenVMS administrator seeking answer to difficult question.I > >DH > > A particular user receives the following error when trying to FTP to OpenVMSt# > > server 6.2-1H3 (Multinet V3.2).- > >-1 > > 520 %CDU-E-OPENIN, error opening !AS as input  > > 8 > > All assistance is appreciated.  Thank you in advance > >e > > Barnesjw@dfo-mpo.gc.ca > H > CDU is the Command Definition Utility, used when issuing a SET COMMAND command.? > It seems to be complaining about a file it is trying to open.eJ > Check the user's LOGIN.COM on the server, and while you are there, check any@H > log files the FTP server may have. With Multinet, I don't know whether thesey  > would be per user, or central. >dL > It may be that a required file doesn't allow access for an FTP connection, > but works for direct login.n >C > >  > >o > >h >n > -- > Alan Adams( > alan.adams@orchard-way.freeserve.co.uk > http://www.nckc.org.uk/n   ------------------------------  % Date: Wed, 22 Jan 2003 09:22:26 -0000e2 From: "Chris Sharman" <chris.sharman@sorry.nospam>/ Subject: Re: MX 5.3-2 ignoring vmsmail forwardsh4 Message-ID: <b0lns4$2h9$1$8302bc10@news.demon.co.uk>  ; "Peter LANGSTOEGER" <peter@langstoeger.at> wrote in messager- news:E7YW9.104910$TY.963915@news.chello.at...hF > In article <b0gk6f$poa$1$8302bc10@news.demon.co.uk>, "Chris Sharman"$ <chris.sharman@sorry.nospam> writes:( > >"Peter LANGSTOEGER"  wrote in message, news:TcwW9.67405$TY.611542@news.chello.at... > >> But I've another hint:  > >>? > >> MAIL> SET FORWARD/USER=JOHNNY.CASH SMTP%"CASH@domain.name"t > >tK > >Ours are all simple, local forwards, used purely to create aliases for a  vmsn > >username.G > >We might have john.cash, johnny.cash, jonathan.cash, all pointing toe cash.b >uJ > Sorry for being unclear. I do not use NAME_CONVERSION to do what you do.F > I use SET FORWARD. And I do this normally on the local machine only.	 > Maybe a  >l0 > MAIL> SET FORWARD/USER=JOHNNY.CASH SMTP%"CASH"2 > MAIL> SET FORWARD/USER=JONATHAN.CASH SMTP%"CASH" >hK > will do exactly the same (I can't remember why we decided to not omit thea; > domain names) but the magic portion is the SMTP% (or MX%)   & Sorry - I think maybe _I_ was unclear. We have (for inbound mail):3  $ MAIL> set forw/user=johnny.cash cash  E We used name_conversion.exe to do the reverse translation on outgoing0@ messages, so that mail sent by user cash would appear to be from@ johnny.cash@here.com (and any local CC's would be fixed up too).I For some reason the presence of name_conversion broke the inbound VMSmail $ forwarding after we upgraded to 5.3.   Chrise   ------------------------------  # Date: Wed, 22 Jan 2003 10:11:03 GMTd From: danco@ns2.pebble.org" Subject: Re: NIST Servers and ACTS0 Message-ID: <slrnb2sv8u.m0.danco@ns2.pebble.org>  @ yIn article <3E2D8AE0.524173A7@vl.videotron.ca>, JF Mezei wrote:. > re: using your own GPS receiver to get time. > L > How does it work with a GPS received ? Does one simply assume a particularN > transmission delay through the 4800 baud serial port if using NMEA sequences= > or slightly higher speeds if using a proprietary protocol ?h  F You use the one pulse per-second (1PPS) signal output by the GPS.  ForC instance, the Garmin GPS-16 HVS I have providing time service to myaG machine outputs a 1PPS signal aligned with the actual UTC second withinwD one microsecond.  The 1PPS signal is fed to the DCD line on a serialC port.  On the data lines of the same serial port is received a dataND stream indicating, among other things, what UTC date/time the second' indicated by the prior 1PPS signal was.c   - Danh   ------------------------------    Date: 23 Jan 2003 00:37:34 +0800, From: Paul Repacholi <prep@prep.synonet.com>" Subject: Re: NIST Servers and ACTS- Message-ID: <87el75b0oh.fsf@prep.synonet.com>s  2 JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes:  . > re: using your own GPS receiver to get time.  pB > With NTP or the dial-up services, the trip delay is measured and > time adjusted accordingly.  $A > How does it work with a GPS received ? Does one simply assume aeD > particular transmission delay through the 4800 baud serial port if; > using NMEA sequences or slightly higher speeds if using at > proprietary protocol ?  B You maesure the offset to a good chimer if you want to verify yourF instalation. You may then want to use a fudge to move the time a smallD amount to get the best accuracy. A 1PPS interface is the best way toD go, and if you have the right GPS, you can toggle an output line andD feed that back to the GPS. Then read out the GPS's time stamp on the? `event' and you have the round trip time for your interupt etc.r  C I you can only get an ASCII string output, then you must work out at! delay factor for your conditions.?  * With a non-bad rx you should get to 1usec.  / You should check out *your* GPS to ensure that:   C The 1PPS pulse is reliable. Esp for things like Power on, FW resetsrE and calibration, only one or 2 SVCs available. What does it do if you C lose the signal? Is it a TTL level signal? If so, you will need thek TTL-RS232 level converter.  A Is it UTC or GPS time? This will not be a issue for 99% of units.n   -- e< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.6@                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Wed, 22 Jan 2003 06:55:02 -0500>2 From: rdeininger@mindspring.com (Robert Deininger)Y Subject: Re: OpenVMS.org: Marvel article and HP's press release for Marveland Alpha Retai L Message-ID: <rdeininger-2201030655030001@user-2ive3rg.dialup.mindspring.com>  8 In article <3E2D8E73.99EA5668@vl.videotron.ca>, JF Mezei( <jfmezei.spamnot@vl.videotron.ca> wrote:    L >So yesterday,s press release about EV7 never made it to the NYSE site which1 >usually shows all the PC related press releases.n    Of course, EV7 isn't PC-related.  E I expect HP sent the press release to a wide range of press outlets. x! They, not HP, decide what to use.    ------------------------------  # Date: Wed, 22 Jan 2003 14:04:16 GMTy# From: "John Smith" <a@nonymous.com>tY Subject: Re: OpenVMS.org: Marvel article and HP's press release for Marveland Alpha RetaiaG Message-ID: <AxxX9.74407$ej1.1689@news02.bloor.is.net.cable.rogers.com>y  ? "Robert Deininger" <rdeininger@mindspring.com> wrote in messagerF news:rdeininger-2201030655030001@user-2ive3rg.dialup.mindspring.com...: > In article <3E2D8E73.99EA5668@vl.videotron.ca>, JF Mezei* > <jfmezei.spamnot@vl.videotron.ca> wrote: >, >vC > >So yesterday,s press release about EV7 never made it to the NYSEn
 site which3 > >usually shows all the PC related press releases.s > " > Of course, EV7 isn't PC-related. >aF > I expect HP sent the press release to a wide range of press outlets.# > They, not HP, decide what to use.     A If that is correct, which I believe it is, it only speaks volumeszD about how little importance HP Marketing attaches to these products.B Marketing should have been telling the newswires in advance of theC release that 'we have an extremely important announcement coming onND Monday and we'd really appreciate all the emphasis you can give it'.  @ Instead, we get http://search.news.yahoo.com/search/news?c=&p=HPD (launched from the Yahoo Technology section), which lists all recentE Yahoo articles with 'HP' in them. Included in such important articles E is news of a new printer ink plant in Israel, and HP's sponsorship ofa a marathon in Houston.  B If Marketing were really doing their job, ....... but I guess theyE just don't have any orders to do so, and probably nobody in marketinga) has the guts to do anything on their own.s  B HP can't afford to be too fussy about which customers run OpenVMS.D With customers being merged out of existence, or going bankrupt, theF pool of 'selected' customers is shrinking. Do you think that HP 'vets' who gets to run Linux?   ------------------------------  # Date: Wed, 22 Jan 2003 14:40:10 GMTt3 From: hammond@not@peek.ssr.hp.com (Charlie Hammond)b8 Subject: Re: Recovering from deleted .PCSI$DATABASE file0 Message-ID: <e3yX9.349$il2.304@news.cpqcorp.net>  0 In article <3E2E2B47.8A6104EA@vl.videotron.ca>, 2 JF Mezei <jfmezei.spamnot@vl.videotron.ca> writes:  N >Could someone with a similar system email the poor chap his .PCSI database so< >that he would have at least some basic stuff to work woth ?   NOT a good idea!  I It is extremely unlikely that the Polycenter Software Intallation (PCSI) sM utility database from one system would work correctly, if at all, on another iJ system.  And even less likley that it would correctly reflect the products* and options installed on the other system.  I Note that .PCSI$DATABASE files cannot be individually "swaped"; they have H internal structures that must be consistent among all the .PCSI$DATABASE" file that make up a PCSI database.  G If all you want to do is to get "some basic stuff to work", just rename K all the database files.  You can then rebuild the database at a later time. E (Although this may cause problems if the "basic stuff" you want to doc) includes installing additional products.)g      F Here is a suggestion: Before executing any PRODUCT INSTALL or REMOVE,  you can backup (and restore) 5       SYS$SYSTEM:*.PCSI$DATABASE   which is the PCSI database.t  < You might even keep the savset somewhere on the system disk.F If any of the database files gets lost or corrupted, this backup could" save you a lot of time and effort.   -- eI       Charlie Hammond -- Hewlett-Packard Company -- Pompano Beach  FL USA H        (hammond@not@peek.ppb.cpqcorp.net -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------    Date: 22 Jan 2003 23:07:16 +0800, From: Paul Repacholi <prep@prep.synonet.com>@ Subject: Re: Security threats seen big for 2003 ... not for VMS!- Message-ID: <87r8b5b4uz.fsf@prep.synonet.com>t  6 Doc.Cypher <Use-Author-Address-Header@[127.1]> writes:  D > 27th of March my killfile entry for Bob expires. I wonder if he'llA > have reached the top of the brain implant waiting list by then.e  ( ::note to self, don't get sick on 27-3::  - I wonder who the `lucky' recipient will be :)    -- d< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.s@                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------    Date: 22 Jan 2003 08:03:02 -0800. From: spamsink2001@yahoo.com (Alan E. Feldman)G Subject: Re: shadow-set consistency on simultaneous loss of all members = Message-ID: <b096a4ee.0301220803.705867e5@posting.google.com>t  | Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> wrote in message news:<01KRICYS3EAA96VR7Q@sysdev.deutsche-boerse.com>... [...] J > With non-shadowed single disks, I've had a power outage a few times and H > after a reboot everything was OK.  Could it be that a shadowed system G > disk is LESS robust in this respect, i.e. if the members  are in the /H > same box and power to that box fails, is it possible that, on reboot, = > the shadow set comes up in some sort of inconsistent state?a  D My experience is that the shadow set will go into the merge mode and> all will be fine. Performance will be degraded until the merge completes, however.o  A Actually, back arount 1995 or so, I was running a VAXcluster (VMSeE 5.4-1) and the room had a power outage. The file JBCSYSQUE.DAT became D corrupted, but I created a new one and all was fine after that. (The disks were not shadowed.)n   [...]h   Disclaimer: JMHO Alan E. Feldmani   ------------------------------  % Date: Wed, 22 Jan 2003 17:01:01 +0000c( From: Nic Clews <sendspamhere@127.0.0.1>G Subject: Re: shadow-set consistency on simultaneous loss of all memberst) Message-ID: <3E2ECE4D.4463F477@127.0.0.1>e   Phillip Helbig wrote:t >  ... I > With non-shadowed single disks, I've had a power outage a few times and?G > after a reboot everything was OK.  Could it be that a shadowed systemkF > disk is LESS robust in this respect, i.e. if the members  are in theG > same box and power to that box fails, is it possible that, on reboot,r= > the shadow set comes up in some sort of inconsistent state?a ...A  B The goal of volume shadowing is to have consistent data across allH members of the shadow set. It does NOT guarantee that the 'data'  is theH right data, that is up to your application. (Shadowing is only concerned# with LBNs not files, hence 'data'.)e  H Shadowing does attempt to get things right, and the clustering protocolsD try to stop corruption, merging as Alan describes is a 'best guess',G where a difference is found either the last known shadow master is usedLA as the source, or, in the case of a data read error, wherever the- successful read is from.  E As to corruption of a system disk, well I would say boot the one thattF boots. You'll not be able to boot a partial copy target with shadowingE enabled, so the worst case may be a conversational boot of one of theoB original system disk members of the shadow set, then reform. Still! quicker than a restore from tape.t  E There is a [STORAGE] article which discusses shadowing in some detaile! which is enlightening in DSN/WIS./ -- d? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciencese nclews at csc dot com    ------------------------------  + Date: Wed, 22 Jan 2003 19:40:33 +0100 (MET)g9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>cG Subject: Re: shadow-set consistency on simultaneous loss of all membersl; Message-ID: <01KRJPCSKVD296VR7Q@sysdev.deutsche-boerse.com>/  D > The goal of volume shadowing is to have consistent data across allJ > members of the shadow set. It does NOT guarantee that the 'data'  is theJ > right data, that is up to your application. (Shadowing is only concerned% > with LBNs not files, hence 'data'.)s  G I realise that a sudden power outage is not a good thing, and actually  G I'm a bit surprised that, the few times it has happened, there were no   ill effects.  J > Shadowing does attempt to get things right, and the clustering protocolsF > try to stop corruption, merging as Alan describes is a 'best guess',I > where a difference is found either the last known shadow master is usedaC > as the source, or, in the case of a data read error, wherever thea > successful read is from.  E If the disks were in different boxes, and only the power to one disk  G failed, then it should be obvious that, if there are differences, then t@ the non-failed disk should be taken as the valid source.  Thus, < obviously, a multi-member shadow set is better than a single non-shadowed disk.    @ So, if the power to both goes off simultaneously, will there be  differences?  H Let me rephrase my question: is there potentially more danger to losing F power to BOTH members of a shadow set simultaneously than to a single E non-shadowed disk?  If so, this would be an argument for putting the lG disks in different boxes.  (Of course, for non-system disks, I do this 1C anyway, so that the disk is still accessible if a single node or a wG single box fails.  With a system disk, if the node fails, then I don't nF need the disk anyway, thus the ONLY argument for putting members of a E system-disk shadow set in separate boxes (apart from guarding against>. controller failure) would be more robustness.)  G > As to corruption of a system disk, well I would say boot the one thath	 > boots.    F Imagine the power goes off while no-one is around, and the system has A AUTO_ACTION=BOOT.  I would want it to come back up automatically.    ------------------------------  # Date: Wed, 22 Jan 2003 07:22:22 GMT . From: peter@langstoeger.at (Peter LANGSTOEGER)6 Subject: Re: Standards for sending of VMS binary files4 Message-ID: <OErX9.145398$TY.1286287@news.chello.at>  [ In article <3E2E1729.AEE072D7@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:c >Arne Vajhj wrote:b >> JF Mezei wrote:Q >> > OK, in VMSmail, I can send an indexed file as /FOREIGN and it comes out witho3 >> > exactly the same attributes at the other hand.n >> >S >> > On a Macintosh, I can send a file as a MACbinary files which makes it possible Q >> > to restore the file at the other end totally intact with all its attributes.cO >> > (Some mailers will send a MAC file as 2 attachements, one for the resourcem( >> > fork, the other for the data fork). >> >R >> > Is there an implicit standard for packaging a file's attributes when the fileN >> > is encoded , for instance as base64 so that they could be restored at theP >> > other end by another VMS mail program, or just seen as binary data by other >> > platforms ? >> o >> No standard.o >>  : >> But I can think of two reasonable common and safe ways: >> 1)  ZIP "-V"r >> 2)  VMS_SHARE >  >3) MFTU
 >4) VMSHEX7 >5) Compressed .TLB's like we used to do on CompuServe?   A 6) MX (Matt Madison's SMTP server) does support MAIL>SEND/FOREIGNt 	(and MAIL>SEND/FOREIGN/TYPE=1).   -- e Peter "EPLAN" LANGSTOEGERl% Network and OpenVMS system specialistn E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  # Date: Wed, 22 Jan 2003 10:09:24 GMTt- From: bdc@world.std.com (Brian 'Jarai' Chase)rI Subject: Re: Storage network with VMS cluster and other OSses, ESS, EMC2? & Message-ID: <H941Jp.6nL@world.std.com>  . In article <3E2DA813.2060503@regenstrief.org>,< Gunther Schadow  <gschadowREMOVETHIS@regenstrief.org> wrote:   > [...] B > We have a quote from EMC Symmetrix on an 3930-50 frame with 6 TBF > of storage on 120 Drives. Has anyone heard anything wrong with that?@ > I heard rumors this was "old technology" and indeed on the EMCG > site I can only find 3930-18 or -36 but not -50. Don't know what thatu > means though.   D Well, for those trailing numbers (-18, -36, -50), I'd be inclined toH believe that those represent the drive sizes.  So, the 3930-18 would useI 18GB drives, the 3930-36 would use 36GB drives, and the 3930-50 would usetJ 50GB drives.  This would seem to be consistent with the notion that if youG multiply 120 x 50GB, you end up with 6TB of raw storage.  Remember thataF once you configure the storage, either RAID-5 or striped and mirrored,. you'll end up with up 50% less usable storage.  G My impression would be that the 3930-xx's may have been around for some H time if they were available in configs with 18GB drives.  You'll want toH make sure that the controllers used in the frames with larger disks haveF been enhanced proportionally to the sizes of the disks they support.  H I'm also a bit surprised that they don't have say, a 3930-72, since 72GB@ drives are sort of the norm for high end SCSI and FC these days.  J Another thing to keep in mind is how the additional storage space is goingH to affect your backups.  Adding a few TB of disk storage can result in aA jump of many tens of TB in your backups, depending on your backupuH policies.  It can also require the purchase of additional tape drives toH support the required data throughput. And then following on the heals ofH that, be sure that your network infrastructure will be able to cope withE the additional amount of data that needs to be backed up each night. hF My experiences have been that people tend to forget about those sorts 
 of things.  C > I heard that EMC has been in the OpenVMS market for a while, whenaA > IBM/ESS only stakes some first claims in this area. I read on atG > presentation abstract for DECUS Germany conference in April 2003 thatfE > ESS would only work through SCSI right now. That would be a bummer.hG > I don't think we would want to have a FC-to-SCSI bridge (HSG) hangingb? > there in the middle. Am I being unrealistic? Overly paranoid?k  H That sounds a little fishy to me.  I'm a bit surprised that they are notG supporting it with native FC yet.  Be sure you know what you're gettingaD into when you start to mix your storage and server vendors.  There'sC always the potential for finger pointing between companies wheneveryG problems arise.  When it's possible and relatively practical to do so, tB I'd recommend buying storage from the same vendor as your servers.  E > Anyone here have real experience with ESS? In our situation we haveCB > reasons to do ESS because our partners in the same computer roomF > have it (not on VMS) and they find maintenance service very good andD > like to have an integrated backup strategy (Tivoli). But, how muchF > of this backup strategy would work cross-filessystems? I would thinkD > unless it is a raw blockwise backup independent from file-systems,E > there is pretty little chance of a common backup mechanism for bothr0 > VMS and Unix file systems. But I may be naive.  G I know that Tivoli's TSM can support a wide variety of client platformsuE with all sort of different filesystem types. (or now it is IBM's TSM,dG apparently; a few years ago it was called IBM ADSM before they flung it1F off to Tivoli.  Now it seems to be back, directly under the control ofH IBM.)  It and VERITAS NetBackup are the two biggest players in the worldH of large scale backups.  Neither have an OpenVMS client in their currentH releases.  I seem to recall that TSM had one a number of years ago, but I I may be remembering incorrectly.  Having managed both a large TSM system1I on AIX and now a large NetBackup system on Solaris; I can say that TSM isuF by far the superior product from a purely technical standpoint, but it9 does have a few things about it that I really don't like.R  B My experiences with IBM, from the standpoint of their hardware andG support, have been positive.  Their support folks are really top notch,cJ and I never had any hassles getting something when I needed it in a timely@ manner.  Of course, it all tends to come at a rather high price.   -brian.3 --  F --- Brian Chase | bdc@world.std.com | http://world.std.com/~bdc/ ----->       IBM 1130 Club Motto:  "Computer People Are Happy People"   ------------------------------    Date: 23 Jan 2003 00:03:40 +0800, From: Paul Repacholi <prep@prep.synonet.com>I Subject: Re: Storage network with VMS cluster and other OSses, ESS, EMC2? - Message-ID: <87iswhb28z.fsf@prep.synonet.com>l  < Gunther Schadow <gschadowREMOVETHIS@regenstrief.org> writes:  ? > I don't think we would want to have a FC-to-SCSI bridge (HSG) = > hanging there in the middle. Am I being unrealistic? Overly. > paranoid?e  # They are ALL `FC-to-SCSI' bridges! W  5 Seek out and listen to peole who have EMC experience.o   -- u< Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.r@                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Wed, 22 Jan 2003 13:36:44 +0000d From: Roy Omond <Roy@Omond.net>sE Subject: TCP/IP Services v5.3 - how to increase the number of "mbufs"t) Message-ID: <3E2E9E6C.C7C36FE0@Omond.net>    Gentle colleagues,  , faced with the following problem (VMS 7.3-1)   $ tcpip sho verI  7   Compaq TCP/IP Services for OpenVMS Alpha Version V5.3y?   on a COMPAQ AlphaServer DS20E 666 MHzP running OpenVMS V7.3-1y   $ tcpip sho comm/mem 30197 mbufs in use:h"         65 mbufs allocated to data2         10003 mbufs allocated to socket structures8         20000 mbufs allocated to protocol control blocks3         31 mbufs allocated to routing table entriesi1         16 mbufs allocated to interface addresseso3         2 mbufs allocated to ip multicast addressesa:         2 mbufs allocated to interface multicast addresses8         6 mbufs allocated to network interface structure1         2 mbufs allocated to netisr thread queues",         5 mbufs allocated to Inet PCB queues4         1 mbufs allocated to rad specific structures2         4 mbufs allocated to OpenVMS miscellaneous8         1 mbufs allocated to OpenVMS Cluster Alias table9         1 mbufs allocated to OpenVMS Kernel VCI structure :         7 mbufs allocated to OpenVMS TCPIP Timer structure;         3 mbufs allocated to OpenVMS LAN VCI VCIB structure8<         1 mbufs allocated to OpenVMS LAN MCAST_REQ structure5         2 mbufs allocated to OpenVMS SELECT structurea,         1 mbufs allocated to OpenVMS ACP AQB/         1 mbufs allocated to OpenVMS ACP INETCB.A         5 mbufs allocated to OpenVMS ACP allocated SERV Structureo-         36 mbufs allocated to External buffer 6         1 mbufs allocated to OpenVMS VCI context block:         1 mbufs allocated to OpenVMS ACP IPCACHE Structure $ telnet 127.0.0.1- %TELNET-E-INETERROR, Internet interface error  -TELNET-I-INETCALL, socket()) -SYSTEM-F-EXQUOTA, process quota exceeded     G Seems to be a limit on the "mbufs allocated to protocol control blocks"R
 of 20,000.A In the good old days of UCX, it used to easy to adjust the "large  buffers" andG "small buffers" to accommodate a very busy system.  I'm damned if I cant find; out how to do this with the new "improved" interface, whose 
 documentationt% appears to leave a lot to be desired.u  ) Am I on the right track to be looking at:l   $ tcpip- TCPIP> sysconfig -q socket socket:2 sbcompress_threshold = 0D sb_max = 1048576                    <---- what's this a maximum of ? sobacklog_drops = 0O sobacklog_hiwat = 190a somaxconn = 1024 somaxconn_drops = 0c
 sominconn = 1s   TCPIP> sysconfig -q inet inet:4 icmp_rejectcodemask = 0f inifaddr_hsize = 32t
 ipdefttl = 64m ipdirected_broadcast = 0 ipforwarding = 0 ipfragttl = 60
 ipgateway = 0- ipport_userreserved = 655351 ipport_userreserved_min = 49152s ipqmaxlen = 1024 ipqs = 1 ipsendredirects = 1a ipsrcroute = 1 pmtu_decrease_intvl = 1200 pmtu_enabled = 1 pmtu_increase_intvl = 240r pmtu_rt_check_intvl = 20 subnetsarelocal = 1i tcbhashnum = 1 tcbhashsize = 512a tcbquicklisten = 1 tcp_compat_42 = 1% tcp_cwnd_segments = 2n tcp_dont_winscale = 0e tcp_keepalive_default = 1  tcp_keepcnt = 8. tcp_keepidle = 14400 tcp_keepinit = 150 tcp_keepintvl = 150l tcp_msl = 60 tcp_mssdflt = 536. tcpnodelack = 0  tcp_recvspace = 61440i tcp_rexmit_interval_min = 2A tcp_rexmtmax = 128 tcprexmtthresh = 3 tcp_rttdflt = 3o tcp_sendspace = 61440s tcp_ttl = 60 tcptwreorder = 0 tcp_urgent_42 = 1  udpcksum = 1 udp_recvspace = 420800 udp_sendspace = 9216 udp_ttl = 30 ovms_nobroadcastcheck = 0: ovms_printf_to_opcom = 1   TCPIP> sysconfig -q neti net: arpkillc = 1200t arpkilli = 180 arprefresh = 120 arpunicastrefresh = 0a ifqmaxlen = 1024 lo_devs = 1  lo_def_ip_mtu = 4096	 nslip = 0l    $ Where is all this stuff documented ?  @ What do I need to do to increase whatever it is that needs to be increased toF allow more connections ?  It's not the number of telnet sessions (this isF actually affecting a different service; the telnet session was just to capturemG the error message).  And the number of current sockets is more than thel& "somaxconn" above (1024) as seen from:   $ def/use sys$output bg. $ tcpip sho devp  $ sear/exact/noout/stat bg. "bg"  E Files searched:                 1       Buffered I/O count:         52E Records searched:            1189       Direct I/O count:           7nE Characters searched:        80814       Page faults:               30 H Records matched:             1188       Elapsed CPU time:  0 00:00:00.00  H Lines printed:                  0       Elapsed time:      0 00:00:00.07      # Any help would be much appreciated.s   Thanks in advance,  	 Roy Omondr Blue Bubble Ltd.   ------------------------------  % Date: Wed, 22 Jan 2003 15:04:07 +0100C0 From: "labadie" <en_trajectant_a_mort@127.0.0.1>I Subject: Re: TCP/IP Services v5.3 - how to increase the number of "mbufs"p0 Message-ID: <txxX9.347$Ym2.200@news.cpqcorp.net>  , "Roy Omond" <Roy@Omond.net> wrote in message# news:3E2E9E6C.C7C36FE0@Omond.net...  > $ tcpip sho comm/mem > 30197 mbufs in use:m4 >         10003 mbufs allocated to socket structures: >         20000 mbufs allocated to protocol control blocks' You have hit the limit of 9999 sockets.g	 $ ana/syse sda> tcpip sh inetcb/stat  and look ath act/max_socket1 do you have 2710 (10 000 in hex) for max_socket ?   7 may be there is a BG device leak, so ask your CSC a newt tcpip$internet_services.exeR   do$ $ @sys$startup:tcpip$define_commands
 $ netstat -an  Do you have many sockets in 	 TIME_WAITK  H Try to decrease tcp_msl (to 10 for example), in order to free BG devices
 more quickly.  $ sysconfig -r tcp_msl=10t   Regardse   Grard   ------------------------------  % Date: Wed, 22 Jan 2003 14:33:51 +0000s From: Roy Omond <Roy@Omond.net>-I Subject: Re: TCP/IP Services v5.3 - how to increase the number of "mbufs"n) Message-ID: <3E2EABCF.47A73EDB@Omond.net>a   labadie wrote:  . > "Roy Omond" <Roy@Omond.net> wrote in message% > news:3E2E9E6C.C7C36FE0@Omond.net...e > > $ tcpip sho comm/mem > > 30197 mbufs in use:f6 > >         10003 mbufs allocated to socket structures< > >         20000 mbufs allocated to protocol control blocks) > You have hit the limit of 9999 sockets.  > $ ana/sysI > sda> tcpip sh inetcb/statt
 > and look at  > act/max_socket3 > do you have 2710 (10 000 in hex) for max_socket ?l  7 Nope, nowhere near the 9999 sockets as can be seen fromo my original post:8   $ def/use sys$output bg. $ tcpip sho deve  $ sear/exact/noout/stat bg. "bg"  E Files searched:                 1       Buffered I/O count:         5aE Records searched:            1189       Direct I/O count:           7 E Characters searched:        80814       Page faults:               30gH Records matched:             1188       Elapsed CPU time:  0 00:00:00.00   Only 1188 sockets.  9 > may be there is a BG device leak, so ask your CSC a new  > tcpip$internet_services.exec  E Is this a known problem ?  Is there a new tcpip$internet_services.exen  which purports to address this ?   > do& > $ @sys$startup:tcpip$define_commands > $ netstat -ani > Do you have many sockets ina > TIME_WAIT  >pJ > Try to decrease tcp_msl (to 10 for example), in order to free BG devices > more quickly.  > $ sysconfig -r tcp_msl=10-  4 Thanks, I'll check this out when I get access again.  " Where *is* this stuff documented ?  I It's consistent;  when "20000 mbufs allocated to protocol control blocks"cC is reached, no more incoming (or outgoing) connections or even some A TCPIP internal commands are sucessful.  If one session does exit,sI reducing the 20,000 to, say, 19,998, then new connections are successful.i% Is this 20,000 hard coded somewhere ?d  	 Roy Omond  Blue Bubble Ltd.   ------------------------------  % Date: Wed, 22 Jan 2003 15:57:24 +0100t0 From: "labadie" <en_trajectant_a_mort@127.0.0.1>I Subject: Re: TCP/IP Services v5.3 - how to increase the number of "mbufs" 0 Message-ID: <pjyX9.350$_n2.236@news.cpqcorp.net>  , "Roy Omond" <Roy@Omond.net> wrote in message# news:3E2EABCF.47A73EDB@Omond.net...  > labadie wrote:J >  > Is this a known problem ?  Is there a new tcpip$internet_services.exe" > which purports to address this ?H There was at a moment a .exe_eco_u_a_v53 available at a moment, fixing a memory leak." I do not know if it is the latest.+ And you do not have the Eco 1 for Tcpip 5.3p  %  > Where *is* this stuff documented ?yI In the Tuning and troubleshooting guide for Tcpip, e.g. on this mirror of  the docsL http://www.pi-net.dyndns.org/docs/openvms0731/731final/6631/6631pro_index.ht ml    K > It's consistent;  when "20000 mbufs allocated to protocol control blocks" E > is reached, no more incoming (or outgoing) connections or even some C > TCPIP internal commands are sucessful.  If one session does exit,1K > reducing the 20,000 to, say, 19,998, then new connections are successful.l% > Is this 20,000 hard coded somewhere3 It seems :-(   Regards    Grard   ------------------------------  % Date: Wed, 22 Jan 2003 12:09:54 +0000s( From: Nic Clews <sendspamhere@127.0.0.1>2 Subject: Re: The HP marketing juggernaut grinds on) Message-ID: <3E2E8A12.46E08F70@127.0.0.1>4   John Smith wrote:w > D > Then clearly it is a case of their indexing not being updated on aE > timely basis. But if it isn't a 'front page' article on their site,mA > how else do you propose that one find it? It seems logical thatDF > searching for 'AlphaServer' (or a variant spelling thereof) would beE > more appropriate than wading through the endless results of printerED > and Wintel crap that's returned when one simply enters 'HP' as the
 > search key.    Blame the cache.  D With my own website I have interesting problems with caches and when@ they refresh from my site, and it can take several days. I don't: particularly want to put meta tags in to overload my host.  D But I do agree that their indexes should also be updated in a timely	 manner...    -- v? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences  nclews at csc dot com.   ------------------------------  # Date: Wed, 22 Jan 2003 13:44:24 GMTn# From: "John Smith" <a@nonymous.com>n2 Subject: Re: The HP marketing juggernaut grinds onH Message-ID: <YexX9.74264$ej1.39619@news02.bloor.is.net.cable.rogers.com>   Just a question....-  E Has anyone seen *ANY* print advertising for Marvel systems yet? ie. asD real ad that was paid for in a trade publication, not just a reprint of an HP press release?h  B Most companies time the placement of print advertising to coincide0 with the press release/launch of their products.  A How about any advertising for Marvel/OpenVMS in any form of median other than on the HP web site?  E Are there *any* plans to advertise Marvel/OpenVMS in mainstream trade B publications/ Wall Street Journal? One might think that any time aF computer company can advertise up to 2-300% performance improvement on; real-world customer apps they'd take full advantage of thatrE opportunity. After all, opportunities like that tend to come but once.> per decade or so. I suspect I know what the answer is already.   ------------------------------    Date: 22 Jan 2003 09:49:49 -0600+ From: young_r@encompasserve.org (Rob Young)a Subject: Versastor surfaces.3 Message-ID: <xdxDzq3sdRMK@eisner.encompasserve.org>H   	I'm glad for a few reasons.    4 	1)  They didn't succumb to the pressure to ship but 		did the right thing.  4 	2)  They didn't go for a proprietary locked-in HBA.  C 	3)   Headed to true distributed storage.  Your LBNs are out there,  		don't worry - be happy.u  
 	Downside:  > 	Will have to wait at least a year more (or so it appears) for* 	the full Versastor functionality to ship.  2 http://www.theregister.co.uk/content/63/28967.html  H For HP, Langeberg said: "We're very much leveraging all of the VersastorN technology. There's no trashing of code, or switcheroo here." According to HP,K the new versastor is neither in-band or out-of-band, but is a "distributed"- system.    				Roba   ------------------------------    Date: 22 Jan 2003 11:41:29 -0600+ From: young_r@encompasserve.org (Rob Young)n Subject: Re: Versastor surfaces 3 Message-ID: <3l8io8eJOVHF@eisner.encompasserve.org>   a In article <xdxDzq3sdRMK@eisner.encompasserve.org>, young_r@encompasserve.org (Rob Young) writes:  >    	Virtualization strategy:n  K ftp://ftp.compaq.com/pub/products/storageworks/HPVirtualizationStrategy.pdfs   	2 MByte file.   				Robg   ------------------------------    Date: 22 Jan 2003 08:47:48 -0800. From: spamsink2001@yahoo.com (Alan E. Feldman) Subject: Re: VMS Jokes= Message-ID: <b096a4ee.0301220847.19326081@posting.google.com>g  d "Rob Heyes" <rob.heyes@btopenworld.com> wrote in message news:<b0km87$37e$1@helle.btinternet.com>...M > I was wondering if anyone had any VMS jokes. Bit mad, I know, but you know.e >  > If not, to get us started... > 7 > "I re-grep ever going to UNIX when VMS is far better"n >  > Ba doom ching     E The (original?) VAXtrek series. There are others, but this is the onea> I am familiar with (the first three links on this page, i.e.):  1 http://terek-nor.rz.uni-mannheim.de/trek/vaxtrek/m    < SPR response expalaining why VMS treats 2000 as a leap year:  . http://www.gmi.edu/~jhuggins/humor/y2kdec.html  C There are other pages with this, use your favorite search engine toa
 find them.     Disclaimer: JMHO Alan E. Feldman    ------------------------------  # Date: Wed, 22 Jan 2003 11:12:06 GMT - From: bdc@world.std.com (Brian 'Jarai' Chase)M Subject: YA OpenVMS Newbie& Message-ID: <H944G7.3vo@world.std.com>  D Well, not a /complete/ newbie.  I had basic exposure to it as a userD during college.  We did Ada and MACRO development on a VAX 6420 (?) 2 I believe.  This was back in the VAX/VMS 5.x days.  H Anyway... I've got myself a lovely little (little?) pair of VAX 4000/300H (KA670, 64MB RAM, 2x 1GB RF72s) rackmount systems. I just brought one ofI them up and reset the SYSTEM password per the instructions of the OpenVMStI FAQ.  The system is running OpenVMS 6.1 as that is what was left on it.   J I've got OpenVMS 7.3-1 hobbyist media on order, and I've also got a set ofC 5.5-2H4 media for VAX.  I'm slowly getting familiar with the systemPG management aspects of it; it's not too hard and the online help is veryrG useful, though I do wish I could read the entire text of the help for agJ command without having traverse the nested layers of it.  I have, however,* gotten sidetracked playing with VAX MACRO.  E My initial goals are to get a basic handle on admining the systems.   D Following that, I'll work on getting them clustered over ethernet.     -brian.m -- 6F --- Brian Chase | bdc@world.std.com | http://world.std.com/~bdc/ ----->       IBM 1130 Club Motto:  "Computer People Are Happy People"   ------------------------------  % Date: Wed, 22 Jan 2003 12:09:42 -0000e* From: "Richard Brodie" <R.Brodie@rl.ac.uk> Subject: Re: YA OpenVMS Newbie, Message-ID: <b0m1m7$1de8@newton.cc.rl.ac.uk>  [ "Brian 'Jarai' Chase" <bdc@world.std.com> wrote in message news:H944G7.3vo@world.std.com...a  3 > ... it's not too hard and the online help is very5I > useful, though I do wish I could read the entire text of the help for a@L > command without having traverse the nested layers of it.  I have, however,, > gotten sidetracked playing with VAX MACRO.  E The 'markup' for the HELP isn't too bad to manage. Just take the texts4 module out of the library: see HELP LIBRARY/EXTRACT.  E > My initial goals are to get a basic handle on admining the systems.PD > Following that, I'll work on getting them clustered over ethernet.  * You get bonus points for a DSSI cluster ;)   ------------------------------  % Date: Wed, 22 Jan 2003 07:44:33 -0500 2 From: rdeininger@mindspring.com (Robert Deininger) Subject: Re: YA OpenVMS NewbieL Message-ID: <rdeininger-2201030744330001@user-2ive3rg.dialup.mindspring.com>  G In article <H944G7.3vo@world.std.com>, bdc@world.std.com (Brian 'Jarai'r
 Chase) wrote:n  E >Well, not a /complete/ newbie.  I had basic exposure to it as a usertE >during college.  We did Ada and MACRO development on a VAX 6420 (?)  3 >I believe.  This was back in the VAX/VMS 5.x days.y >oI >Anyway... I've got myself a lovely little (little?) pair of VAX 4000/300e4 >(KA670, 64MB RAM, 2x 1GB RF72s) rackmount systems.    Well, congratulations!  I The VAX 4000 pedestal systems are nicer than the rackmounts for home use, ? but only because they can double as end tables.  Either kind isr functional.e   >I just brought one ofJ >them up and reset the SYSTEM password per the instructions of the OpenVMSJ >FAQ.  The system is running OpenVMS 6.1 as that is what was left on it.  K >I've got OpenVMS 7.3-1 hobbyist media on order, and I've also got a set ofs >5.5-2H4 media for VAX.  /  I The VMS version for VAX is V7.3, not V7.3-1.  That's ok, don't panic whene you see it.r  I You'll need a CD drive.  That's not trivial on the typical VAX 4000, withnI no SCSI support.  If you already have DECnet installed on the system, yousF might find it easiest to obtain a dirt-cheap old alpha (or VAXstation)J system with a CD, and use DECnet and BACKUP to transfer your new VAX mediaH kit over to the VAX system.  Once you get the VAX and Alpha VMS versionsJ aligned, you can put all the machines in the same cluster, and they'll all" have access to each other's disks.  I You can get hardware to attach a CD drive to the VAX, but it isn't alwayssJ easy to find.  There are various other strategies for getting new VMS onto an old VAX.   , >I'm slowly getting familiar with the systemH >management aspects of it; it's not too hard and the online help is veryH >useful, though I do wish I could read the entire text of the help for a; >command without having traverse the nested layers of it.  w   $ HELP <command> * $ HELP/NOPAGE ...y $ HELP HELPe   >I have, however,d+ >gotten sidetracked playing with VAX MACRO.    That could take a while. :-)    F >My initial goals are to get a basic handle on admining the systems.  C >Following that, I'll work on getting them clustered over ethernet.l   Read:a OpenVMS User's Guide System Manager's Manual , System Management Utilities Reference Manual' Guide to OpenVMS Cluster Configurations. OpenVMS Clustering ManualEI (The manual names vary slightly over the years, and I don't remember themt	 exactly.)n  C Well, you can't really read all of those 7 manuals (some now have 2.' parts), but read the table of contents.   F And of course, obtain and read the FAQ.  But you've already done that.   ------------------------------  % Date: Wed, 22 Jan 2003 13:32:29 +0000f( From: Nic Clews <sendspamhere@127.0.0.1> Subject: Re: YA OpenVMS Newbie) Message-ID: <3E2E9D6D.28DB9A8A@127.0.0.1>e   Richard Brodie wrote:L > ] > "Brian 'Jarai' Chase" <bdc@world.std.com> wrote in message news:H944G7.3vo@world.std.com...  >  ...tG > > My initial goals are to get a basic handle on admining the systems.oF > > Following that, I'll work on getting them clustered over ethernet. > , > You get bonus points for a DSSI cluster ;)  $ - 20 points if you pop the pica fuse  D + 30 points if you have a quorum disk on a StorageWorks disk via HSD
 05/10/30 etc.p   :-)  --  ? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences' nclews at csc dot com-   ------------------------------   End of INFO-VAX 2003.044 ************************