1 INFO-VAX	Sat, 08 Jun 2002	Volume 2002 : Issue 315       Contents:4 Re: A dvdwrite(r)-Program: save 4.3 GB on a DVD-R(W)B Re: Carly as a speaker (was - RE: Carly was here in ZKO yesterday)B Re: Carly as a speaker (was - RE: Carly was here in ZKO yesterday)B Re: Carly as a speaker (was - RE: Carly was here in ZKO yesterday)# Re: Carly was here in ZKO yesterday # Re: Carly was here in ZKO yesterday  Re: cluster wide logical names Re: Could linux become VMS? 2 Re: Data redundancy options for 3000 foot distance# Re: DEFINE/LOG/TABLE=LNM$SYSCLUSTER # Re: DEFINE/LOG/TABLE=LNM$SYSCLUSTER P Re: ES-40 slower than AS-4100 was Re: Running Test and Production systems on difP Re: ES-40 slower than AS-4100 was Re: Running Test and Production systems on dif) Re: For all you hobbyists: IDE on SCSI !! ) Re: For all you hobbyists: IDE on SCSI !! = Re: How to put my VMS box on a BroadBand Connection (part II) $ Re: HP Alphaservers and Lan Consoles& Re: Mark Gorham's Beer Bash in Reading& Re: Mark Gorham's Beer Bash in Reading& Re: Mark Gorham's Beer Bash in Reading& Re: Mark Gorham's Beer Bash in Reading& Re: Mark Gorham's Beer Bash in Reading Re: Newsreader for OpenVMS 7.2 Re: No new Alpha sales Re: Open Letter to HP = open source porting (was Re: Carly was here in ZKO yesterday) A Re: open source porting (was Re: Carly was here in ZKO yesterday) G Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading) G Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading) G Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading) G Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading) G Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading) G Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading) G Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading) P Porting Open Source to OpenVMS - was: Re: Mark Gorham's Beer Bash in Reading in  RE: SAN shadowing question. SSH v2 (was Re: OpenVMS, Volume Desktop OS...)2 Re: SSH v2 (was Re: OpenVMS, Volume Desktop OS...) Re: STARTUP stops after boot
 Re: Test post 
 Re: Test post P Re: The Press and the IA-64 Port (was Re: Another analyst says VMS port won't beP Re: The Press and the IA-64 Port (was Re: Another analyst says VMS port won't beP Re: The Press and the IA-64 Port (was Re: Another analyst says VMS port won't be Re: VMS Monitoring a User > Win32/VMS porting (was Re: Mark Gorham's Beer Bash in Reading)  F ----------------------------------------------------------------------  # Date: Sat, 08 Jun 2002 02:52:01 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> = Subject: Re: A dvdwrite(r)-Program: save 4.3 GB on a DVD-R(W) ' Message-ID: <3D0174FA.9C170DB3@fsi.net>    Eberhard Heuser-Hofmann wrote: > = > In article <3D0024A0.7E494BD6@fsi.net>, "David J. Dachtera" ! > <djesys.nospam@fsi.net> writes:  > |>Larry Kilgallen wrote: > |>> @ > |>> In article <3D001ADF.D514318@fsi.net>, "David J. Dachtera"# > |><djesys.nospam@fsi.net> writes:  > |>> > Larry Kilgallen wrote: > |>> >>6 > |>> >> In article <3cff1e60.0@news.uni-konstanz.de>,I > |>vaxinf@chclu.chemie.uni-konstanz.de (Eberhard Heuser-Hofmann) writes:  > |>> >>H > |>> >> > I want to release my first version of my dvdwrite program for > |>OpenVMS.
 > |>> >> >3 > |>> >> > Get informations via the following page: 
 > |>> >> >5 > |>> >> > http://134.34.7.25/~dvdwrite/dvdwrite.html  > |>> >>I > |>> >> Ok, looking at that I think I understand how I could write a 4GB  > |>> >> ODS-2 volume on DVD.  > |>> >>K > |>> >> But how would I read files back from that DVD onto VMS ?  VMS does ( > |>> >> not have DVD support, does it ? > |>> > J > |>> > I think it might, in so far as it has CD-ROM support, it stands toG > |>> > reason that ODS on DVD is reasonable to expect to work. I can't  > |>believe I > |>> > he'd post the recording stuff without being able to read what was  > |>> > written. > |>> K > |>> For CDROM there are VMS drivers, but I don't know about that for DVD.  > |>I > |>I am not aware of any CD-ROM specific drivers other than DKDRIVER for D > |>SCSI disks (mag. or optical) and DQDRIVER for IDE disks (mag. orF > |>optical), and VMS support of ISO-9660 (for non-ODS media). Can you > |>enlighten me?  > |> > |>-- > |>David J. Dachtera  > |>dba DJE Systems  > |>http://www.djesys.com/ > |>, > |>Unofficial Affordable OpenVMS Home Page:# > |>http://www.djesys.com/vms/soho/  > |> > E > The dkdriver needs a drive that read/writes 512-byte blocks per I/O C > and some additional features in the ROM part of the drive so that D > the dkdriver gets reasonable answers form the drive when VMS comesD > up. SCSI is a poorly defined standard. I'm sure Glenn Everhardt is > able to tell you much more...  > ( > The dqdriver supports two "protocols": > ATAPI = CDROM/DVD  > and > > IDE= cheap PC-disks. The IDE support is poor: no DMA = slow.  E So, there's no driver specifically for CD-ROM or DVD, or any specific  make/model of such?    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Fri, 07 Jun 2002 12:55:01 -0700 ' From: David Mathog <mathog@caltech.edu> K Subject: Re: Carly as a speaker (was - RE: Carly was here in ZKO yesterday) + Message-ID: <3D010F95.BEAE587F@caltech.edu>    "Bradford J. Hamilton" wrote:  >  > Thanks, Sue!!!  : > I was *very* impressed with Carly's ability as a speaker  ; She could orate like Jesse Jackson, sing like Jesse Norman, 9 and run like Jesse Owens and it _still_ wouldn't indicate 8 that she had the ability to manage a large multinational corporation.   Regards,   David Mathog mathog@caltech.edu   ------------------------------  % Date: Fri, 07 Jun 2002 16:57:17 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> K Subject: Re: Carly as a speaker (was - RE: Carly was here in ZKO yesterday) , Message-ID: <3D011E29.88D4DEC5@videotron.ca>   David Mathog wrote: = > She could orate like Jesse Jackson, sing like Jesse Norman, ; > and run like Jesse Owens and it _still_ wouldn't indicate : > that she had the ability to manage a large multinational > corporation.    N In all fairness, the CEO of a company MUST be a good leader, and must have theK skills to communicate and convince the troops. And I believe that Carly has L that.  Her problem is that her vision doesn't match what the troops want, so0 her job of convincing the troops is much harder.  K Microsoft is full of weenies who think that Bill gates is god, so when Bill L gates speak, they act on it as if it were the ten commandments. But the moreK mature companies have grown up adult employees who may not see their CEO as ; GOD, so "GOD" must work much harder to convince the troops.    ------------------------------  # Date: Fri, 07 Jun 2002 23:07:39 GMT L From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")K Subject: Re: Carly as a speaker (was - RE: Carly was here in ZKO yesterday) 8 Message-ID: <00A0F1C5.E912CE78@SSRL04.SLAC.STANFORD.EDU>  \ In article <3D011E29.88D4DEC5@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: >David Mathog wrote:> >> She could orate like Jesse Jackson, sing like Jesse Norman,< >> and run like Jesse Owens and it _still_ wouldn't indicate; >> that she had the ability to manage a large multinational  >> corporation.  >  > O >In all fairness, the CEO of a company MUST be a good leader, and must have the L >skills to communicate and convince the troops. And I believe that Carly hasM >that.  Her problem is that her vision doesn't match what the troops want, so 1 >her job of convincing the troops is much harder.  > L >Microsoft is full of weenies who think that Bill gates is god, so when BillM >gates speak, they act on it as if it were the ten commandments. But the more L >mature companies have grown up adult employees who may not see their CEO as< >GOD, so "GOD" must work much harder to convince the troops.  H I have a friend who works at Microsoft.  (His Colorado-based company wasK acquired by them and he ended up working in Redmond, although reporting to  M Hyderabad.)  He says general employee satisfaction levels tend to track quite  closely with the stock price.     I My not-very-well-substantiated impression is that the average MS employee L doesn't think BG is God, or always right, or any of that stuff, but (a) likeL working on stuff that actually gets sold and used and (b) like the idea that4 their stock options might make them actually rich.    L At upper management levels they may vie to be as rapacious as Bill & SteveB;L at lower levels they're just trying to make deadlines, keep their jobs, and  get more stock.    -- Alan     O =============================================================================== 0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056 M  Physical mail to: SSRL -- SLAC BIN 69, PO BOX 4349, STANFORD, CA  94309-0210 O ===============================================================================    ------------------------------   Date: 7 Jun 2002 17:45:22 GMT 1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon) , Subject: Re: Carly was here in ZKO yesterday+ Message-ID: <adqrfi$5su$1@info.cs.uofs.edu>   = In article <d7791aa1.0206070838.2d1bc998@posting.google.com>, +  bob@instantwhip.com (Bob Ceculski) writes: n |> "Andrew Dodd" <Andrew.J.Dodd@HP.spamfreecom> wrote in message news:<Cn_L8.2$pB1.127099@news.cpqcorp.net>...K |> > My reading, and I admit I may be looking through weird spectacles, was 
 |> > simply:-  |> >  O |> > If a VMS customer makes a decision to move to UNIX, make sure they move to  |> > HP-UX.  |> >   |>  I |> why would a vms customer ever want to move to unix, except for reasons  |> of stupidity?    D How about because their vendor has decided to stop suppoorting theirF operation onf VMS any longer??  How about because that same vendor hasC been pushing them to get off VMS for more than a few years already.   J |>                You can port and run almost anything on vms that you canG |> on unix, and with real security, clustering, and the fastest jvm ...   F Let us all know when you have OpenOffice (or any real Office Suite) upD and running on VMS, Bob.  Like the man with the better mousetrap the. world will be beating a path to your doorstep.   billH (Who has seriously looked at what it takes to port software from Unix to VMS!!)   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  # Date: Sat, 08 Jun 2002 03:05:38 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> , Subject: Re: Carly was here in ZKO yesterday' Message-ID: <3D01782B.CAE9528F@fsi.net>    Dave Gudewicz wrote: >  > JF,  > L > When you wrote to Scott Stallard on this issue, what was his reply to you? > You did write didn't you?  > J > If you didn't write then........  but you are entitled to your opinions.  H I've not written only to Scott Stallard, but I did Cc him in an exchangeE earlier this week. No Read Receipt as requested, no direct reply. Sue  did respond, however.    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Fri, 07 Jun 2002 20:56:41 +0200 ) From: Bart Zorn <B.Zorn@xs4all.nospam.nl> ' Subject: Re: cluster wide logical names / Message-ID: <3D0101E9.6050802@xs4all.nospam.nl>    Phillip Helbig wrote: J >>On Alpha VMS 7.2-1, is it possible to create a cluster-wide logical name5 >>table that is accessible only to a specific group?   >  >  > Yes. > > > As you have probably noticed, this is a VERY useful feature. >  >  >>If so, how?    >  >  > See below. >  > I >>From what I see, a cluster-wide logical name table is accessible to all  >>users on the cluster.  >  > 7 > Yes (though you could perhaps change this with ACLs).  > H > Solution: create a table with the LNM$SYSCLUSTER as the parent table, ? > and give the group the appropriate rights to the child table.   H This is definitely WRONG! it should be LNM$CLUSTER_TABLE, or LNM$CLUSTER  I > Note that there at least WAS a mistake in the HELP with regard to this  K > (IIRC, CREATE/NAME_TABLE): SYSNAM and SYSPRV were swapped in the HELP at  K > one place---it said one of the privs is needed, but in fact the other is   > required.  >   	 Bart Zorn    ------------------------------  $ Date: Fri, 7 Jun 2002 18:02:22 -0700% From: "RLC" <rlc@magicalsoftware.com> $ Subject: Re: Could linux become VMS?/ Message-ID: <ug2es84sk9gi55@corp.supernews.com>   L > How about: Lets try to change feature xyz. WHAT? NO SOURCE! DONE WITH VMS!  L Actually, source listings have been available practically since day 1 of VMS	 at a VERY L reasonable price.  Most commercial versions of UNIX price source licenses to
 commercialL developers VERY high.  I have been changing VMS functionallity for ??? (well let's just say< since the late 70s) using source listings to figure out how.   Just my opinion,   Rick...    ------------------------------   Date: 7 Jun 2002 11:44:08 -0700 1 From: keithparris_NOSPAM@yahoo.com (Keith Parris) ; Subject: Re: Data redundancy options for 3000 foot distance = Message-ID: <cf15391e.0206071044.1eda9fa5@posting.google.com>   k jordan@ccs4vms.com (Rich Jordan) wrote in message news:<cc5619f2.0205311312.6803db8b@posting.google.com>... F > Two sites, about 3000' apart, single mode dark fiber, multiple pairsD > available, with connecting equipment to be determined based on the > following need.  > H > Looking for a 'primary' site with a DS20e box and local storage, and aH > 'backup' site with a DS10 and equivalent storage, running OpenVMS V7.3F > (or newer) and applications that are mostly based on RMS  and OracleB > Codasyl DBMS (used to be DEC's).  The backup site will be a 'hotF > standby', not used unless the main goes down, and in that case would1 > still see less load (hence the smaller system).   C You may also find that you can easily run some portion of the total E workload at the secondary site, use some of its CPU and I/O capacity, > and thus not totally waste it.  This could also provide better? response times to users of the equipment at the primary site by  offloading some of the work.   > Ideal would beG > 100% same time data consistency across the link, but price may decide H > on a 'slower' method that might require some manner of recovery on the" > backup site before it goes live.  ? You already have the most-expensive part of a disaster-tolerant A cluster, the inter-site link.  And your distance is so small that C inter-site latency due to speed of light is a non-issue, so you can @ use synchronous replication (as Volume Shadowing does) without aF distance-related performance penalty and have the safety of no loss of= transactions between sites.  So why settle for anything less?     > This is not a large data site;E > initial plans for a standalone server had 6 each 18GB spindles, and 1 > actual current data usage is around 24GB total.   D SANs are "hot stuff" right now, but with only 6x18 GB, it's probablyB overkill to use a SAN at this point.  A single StorageWorks shelf,D attached to a local SCSI controller, would hold that many disks.  IfB you need to stripe across disks for performance, you could do that with host-based RAID software.  D If you want a controller (for write-back caching for performance, orC for controller-based mirroring within each site so you're protected F against disk failure even after you lose a site), an HSZ would be more) cost-effective than an HSG plus switches.   : > Depending on the capabilities and bandwidth of the fiber? > communications, one option might be to cluster and use volume  > shadowing.  C That would be my recommendation.  Volume Shadowing does synchronous B writes, so you don't have to worry about losing transactions.  AnyD application which runs on VMS then gets disaster tolerance by virtue: of running in such a configuration.  And there's no manualC intervention involved in storage failover, as there is with DRM (15v minutes to do the failover).  F With dark fiber, you can have either Gigabit Ethernet or FDDI speeds. E It shouldn't take more than a few hours at most to shadow-copy six 18r GB disks across that.e  < > That leaves us with a 'symmetrical' 2 node cluster withoutH > redundant quorum disk access (since you can't shadow the quorum disk),E > so the 'backup' system has to be a 'slave' node with fewer votes inpD > order to have a safe/survivable cluster where the primary node can- > survive the loss of the backup site/server.l  F You can use DECamds on a standalone VMS system or Availability ManagerF on VMS or even a PC to recover from a quorum loss and continue without
 rebooting.  C Based on my experience, I will contend that you have the votes idea ? exactly backwards.  If you think about it, it is the (typicallyrE un-manned) backup site that you probably want to have enough votes tosA continue without human intervention, rather than the primary sitea> (which is typically manned and so a human is around to DO that intervention).  D > Our lowball option is have the two sites standalone with a nightly3 > 'batch update' of the backup site.  Much cheaper.   > You mean cheaper because of not having to buy Cluster Software( licenses?  How else would it be cheaper?   > Alternatives wouldG > involve significant modifications to existing programs and procedures G > to (perhaps) involve some kind of journaling that could be applied to ( > the backup at more frequent intervals.   Yuck.    > I do not believe an optionB > that involves significant software overhauls will be considered.   Makes sense.  C > I do not have experience with fiber channel or equivalent storageaG > technologies.  Assuming appropriate equipment is available to connectaE > a SAN across the fiber, what types of equipment would we be lookingsG > for?  I've used HSZ controllers on standalone and shared-bus systems, E > but never the HSG units; do they provide any capability to 'mirror'n > across the distance involved?T  D As others have noted, Fibre Channel can run across fiber for a 3000'E distance.  You would need a second pair of fibers, in addition to theh? pair used for cluster (SCS) traffic over Gigabit Ethernet, Fast ; Ethernet, or FDDI, since FC cannot (yet) carry SCS traffic.n  E You can use DRM to mirror between HSGs across Fibre Channel.  Licensen" cost may be an issue in your case.   C > Any recommendation on good reading material to get up to speed on-G > (OpenVMS compatible) fiber storage would also be appreciated.  Once IeC > get the shadowing items corrected (earlier post) I'll be perusingeE > compaq.com anyway, but it'd be nice to get targetted on the meat ofr
 > the matter.m  A There's a lot of good info on the HP/Compaq web site, including aaE white paper on using Fibre Channel in disaster-tolerant VMS clusters.:  D There some introductory material on disaster-tolerant clusters in my3 'VMS Clusters: Advanced Concepts' seminar slides atrW http://www.geocities.com/keithparris/decus_presentations/f2001_vmsclusters_advanced.ppt>: ----------------------------------------------------------: Keith Parris | parris <at> DECUServe <dot> decus <dot> org   ------------------------------  % Date: Fri, 07 Jun 2002 20:54:28 +0200t) From: Bart Zorn <B.Zorn@xs4all.nospam.nl>>, Subject: Re: DEFINE/LOG/TABLE=LNM$SYSCLUSTER/ Message-ID: <3D010164.4000202@xs4all.nospam.nl>a   Phillip Helbig wrote:nJ > I see that (OpenVMS V7.3) DEFINE/LOG/TABLE=LNM$SYSCLUSTER only gives me H > the message %DCL-I-SUPERSEDE when the actual value of the logical has I > changed, i.e. if I issue the command again with the same value for the  I > definition, I don't get the message.  In another table, however, I get n6 > the message whether or not the value changes or not. > D > Why the different behaviour for LNM$SYSCLUSTER?  Is it documented  > somewhere?   > D > While I'm on the subject, why not DEFINE/CLUSTER as a synonym for F > DEFINE/TABLE=LNM$SYSCLUSTER in line with /PROCESS, /GROUP, /JOB and 
 > /SYSTEM?  7 I am not sure about the first question of your posting.D  G W.r.t. the second, I believe that implementing a /CLUSTER qualifier to rF DEFINE, ASSIGN and DEASSIGN, would require a change to the meaning of C the /SYSTEM qualifier. Currently, /SYSTEM means /TABLE=LNM$SYSTEM.  ? LNM$SYSTEM is a search list consisting of LNM$SYSTEM_TABLE and . LNM$SYSCLUSTER.   0 /CLUSTER should (in my opinion) be an alias for ; /TABLE=LNM$SYSCLUSTER_TABLE. This is partly a duplicate of   /TABLE=LNM$SYSTEM.  I I did a presentation about all this in Lyon. If you can't get the sheets S, from the conference site, send me an e-mail.  	 Bart Zornl   ------------------------------  # Date: Sat, 08 Jun 2002 02:48:18 GMTp1 From: "David J. Dachtera" <djesys.nospam@fsi.net> , Subject: Re: DEFINE/LOG/TABLE=LNM$SYSCLUSTER' Message-ID: <3D01741A.82747120@fsi.net>e   Bart Zorn wrote: >  > Phillip Helbig wrote: K > > I see that (OpenVMS V7.3) DEFINE/LOG/TABLE=LNM$SYSCLUSTER only gives mewI > > the message %DCL-I-SUPERSEDE when the actual value of the logical hasyJ > > changed, i.e. if I issue the command again with the same value for theJ > > definition, I don't get the message.  In another table, however, I get8 > > the message whether or not the value changes or not. > >eE > > Why the different behaviour for LNM$SYSCLUSTER?  Is it documentedo > > somewhere? > >uE > > While I'm on the subject, why not DEFINE/CLUSTER as a synonym forhG > > DEFINE/TABLE=LNM$SYSCLUSTER in line with /PROCESS, /GROUP, /JOB andJ > > /SYSTEM? > 9 > I am not sure about the first question of your posting.a > H > W.r.t. the second, I believe that implementing a /CLUSTER qualifier toG > DEFINE, ASSIGN and DEASSIGN, would require a change to the meaning of7D > the /SYSTEM qualifier. Currently, /SYSTEM means /TABLE=LNM$SYSTEM.@ > LNM$SYSTEM is a search list consisting of LNM$SYSTEM_TABLE and > LNM$SYSCLUSTER.1  F *BUT*: notice that LNM$SYSTEM_TABLE is the first element in the search& list. That's the first one you'll hit.  1 > /CLUSTER should (in my opinion) be an alias fori > /TABLE=LNM$SYSCLUSTER_TABLE.  @ Perhaps some change to some appropriate code somewhere could add< /CLUSTER as meaning the same as /TABLE=LNM$SYSCLUSTER_TABLE.   > This is partly a duplicate ofo > /TABLE=LNM$SYSTEM.  F Well, yes, partly, but not entirely. DEFINE/SYSTEM puts the lnm in theE LNM$SYSTEM_TABLE where it should be. I suppose it's possible, but I'mhD not sure it's likely, that mucking around with the LNM$SYSTEM searchG list would allow /SYSTEM to do the same as /TABLE=LNM$SYSCLUSTER_TABLE.m4 I don't have the resources to test it at the moment.   -- e David J. Dachterat dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/a   ------------------------------  % Date: Sat, 08 Jun 2002 01:39:28 +0200A' From: Brass Christof <welcome@spam.not>jY Subject: Re: ES-40 slower than AS-4100 was Re: Running Test and Production systems on difF( Message-ID: <3D014430.43B8BFFC@spam.not>   Fabio Cardoso wrote: >  > Paul > - > The tests were based in comparisons betweeni5 > the production system (AS-4100) and the new machines > ES-40.6 > The development people tested some queries and Cobol3 > programs. The performance was slower in the ES-40b2 > than the AS-4100. The AS-4100 was with 300 users	 > online. 4 > The ES-40 was just with the developm team. The RDB6 > database was in the EMC discs... before the database6 > was installed in the local disks (36 GB and Symbions  > 895 controller) without cache. > 	 > Regardsw >  > FC   Stupid idea?  3 If database queries are involved the response time  6 might depend on the cache state of the database, i.e. 5 some queries might be faster if sent to a production f5 database than if sent to a database with empty cache.o   > 3 > --- Paul Repacholi <prep@prep.synonet.com> wrote: 4 > > Fabio Cardoso <fabiopenvms@yahoo.com.br> writes: > >e2 > > > We are trying to discover why our ES-40 with > > OpenVMS 7.2-1h1 /m6 > > > Oracle RDB 7.0-31 and LP-8000 connected to a EMC > > storage (Fibre5 > > > Channel) with zero users is slower than the oldf > > machine AS-4100 with, > > > OpenVMS 7.2-1H1 and Oracle RDB 7.0-31. > >t6 > > What do you mean by `zero users' and `slower'? How > > did you do what . > > ever you did to determine that it is slow? > >f > > --2 > > Paul Repacholi                               1 > > Crescent Rd.,t > > +61 (08) 9257-1001 > > Kalamunda.5 > >                                              Wests > > Australia 60762 > > Raw, Cooked or Well-done, it's all half baked.4 > > EPIC, The Architecture of the future, always has > > been, always will be.A >  > ====== > ========================== > Fbio dos Santos Cardoso > OpenVMS System Manager > Rio de Janeiro - BrazilA > fabiopenvms@yahoo.com.br > ========================== > 4 > __________________________________________________ > Do You Yahoo!?2 > Yahoo! - Official partner of 2002 FIFA World Cup > http://fifaworldcup.yahoo.com=   -- =7 moc dot slupofni at ssarb - please reverse the sequence-   ------------------------------  # Date: Fri, 07 Jun 2002 23:50:42 GMTB6 From: "Andy Bustamante" <a_c_bustamante@earthlink.net>Y Subject: Re: ES-40 slower than AS-4100 was Re: Running Test and Production systems on difbA Message-ID: <mFbM8.1320$Pv2.833@newsread2.prod.itd.earthlink.net>t  K Many good questions already out and this may or may not apply to RDB but in_H my environment, the "first query" will be slowest since it requires hardJ faults to build global sections.  Following queries which use that alreadyK populated global section will run very noticeably faster.  Is this possiblyH in your environment?  I Second question, have you reviewed ECO's, particularly for fiber channel.> There are updates out there.   --   Andy Bustamanted( remove the ascii-95's to reply by e-mail      ; "Fabio Cardoso" <fabiopenvms@yahoo.com.br> wrote in messagen: news:20020606163444.59157.qmail@web20202.mail.yahoo.com... > Good Luck  > . > We are trying to discover why our ES-40 with1 > OpenVMS 7.2-1h1 / Oracle RDB 7.0-31 and LP-8000y2 > connected to a EMC storage  (Fibre Channel) with3 > zero users is slower than the old machine AS-4100i- > with OpenVMS 7.2-1H1 and Oracle RDB 7.0-31.w > 1 > The last 30 days were test days... our ES-40 isa. > not working fine  as we expected under OVMS.7 > We are planning to consolidadte 20 old VAX/AS serversM4 > in 4 ES-40. The first one (mine) was the first one4 > to be installed. No sucess. Compaq  and Oracle .BR3 > are researching here but until now we dont have an1 > solution. If someone here as a ES-40 / ORDB 7.0|! > configuration please email me !> >  > 	 > Regards- >- > FC. > --- Tom Williams <twilliams@gscc.com> wrote:4 > > We're looking into migrating our environments to > > ES40s, and are2 > > planning on having each box contain two Galaxy > > instances: one5 > > production, one test. Each instance will have its  > > own sets of disks,6 > > separate UAFs, etc. Assuming the instances are set > > up correctly, is7 > > there anything that might happen on the test systemi > > (hardware or6 > > software) that could affect the production system? > > I realize I'mh4 > > being vague, but I don't feel comfortable having > > production in thew5 > > same cabinet as test.  All comments, pro and con,t > > are welcome. > >  > > Thanks in advancep > >  > > Tom Williams >w >e > =====m > ========================== > Fbio dos Santos Cardoso > OpenVMS System Manager > Rio de Janeiro - Brazils > fabiopenvms@yahoo.com.br > ========================== >e4 > __________________________________________________ > Do You Yahoo!?2 > Yahoo! - Official partner of 2002 FIFA World Cup > http://fifaworldcup.yahoo.com  >t   ------------------------------   Date: 7 Jun 2002 13:22:28 -0700y1 From: keithparris_NOSPAM@yahoo.com (Keith Parris)t2 Subject: Re: For all you hobbyists: IDE on SCSI !!= Message-ID: <cf15391e.0206071222.3684f621@posting.google.com>i  L Dirk Munk <munk@home.nl> wrote in message news:<3D006A87.8070203@home.nl>...N > I see no reason why you can't shadow two disks, VMS thinks it sees two SCSI  > disks.  E Dirk, would you mind checking to see if it supports the SCSI ReadLongcC and WriteLong commands?  There's a discussion of this in the VolumeBF Shadowing for OpenVMS Manual; section 4.9.4.1 in the 7.3 version shows= how to use SDA to check if a SCSI disk supports the simulatednD Forced-Error operations or not (if it does NOT support READL/WRITEL," then it will display "nofe" in theB "Characteristics:" section of an SDA> SHOW DEVICE DKAxx: display).: ----------------------------------------------------------: Keith Parris | parris <at> DECUServe <dot> decus <dot> org   ------------------------------  % Date: Fri, 07 Jun 2002 23:33:27 +0200t From: Dirk Munk <munk@home.nl>2 Subject: Re: For all you hobbyists: IDE on SCSI !!& Message-ID: <3D0126A7.9040702@home.nl>   These are the characteristics:     I/O data structuresr ------------------- P MNKALP$DKA200                                  WDC WD80 0JB-00CRA1 UCB: 818BBE00  M Device status:   18021810 online,valid,unload,lcl_valid,exfunc_supp,fast_patheC Characteristics: 1C4D4408 dir,qsvbl,fod,shr,avl,mnt,elg,idv,odv,rnd-4                   25010201 clu,nnm,nlt,scsi,nofe,dtn  P Owner UIC [000001,000004]   Operation count     611927   ORB address    818C0600Q        PID        00000000   Error count              1   DDB address    81873C80tP Alloc. lock ID   01000259   Reference count          1   DDT address    8155EBA0P Alloc. class            0   Online count             1   SUD address    818C0500P Class/Type          01/36   Retry cnt/max        16/16   VCB address    81AA4380P Def. buf. size        512   BOFF              00000000   CRB address    81873D00P DEVDEPEND        42406060   Byte count        00002000   I/O wait queue 818BBE6C6 DEVDEPND2        00000000   SVAPTE            81B229086 DEVDEPND3        01000001   DEVSTS            00000004 FLCK index             3A. DLCK address     81873D80  Preferred CPUDB  81814000u Preferred CPUID        00n    Q So it seems it does not support Forced Error operations. Now I have to check how oP serious that is for shadow operations. It doesn't matter for me because I'm not / going to shadow anyway, but it is good to know.n       Keith Parris wrote:.N > Dirk Munk <munk@home.nl> wrote in message news:<3D006A87.8070203@home.nl>... > N >>I see no reason why you can't shadow two disks, VMS thinks it sees two SCSI  >>disks. >  > G > Dirk, would you mind checking to see if it supports the SCSI ReadLonghE > and WriteLong commands?  There's a discussion of this in the VolumelH > Shadowing for OpenVMS Manual; section 4.9.4.1 in the 7.3 version shows? > how to use SDA to check if a SCSI disk supports the simulated F > Forced-Error operations or not (if it does NOT support READL/WRITEL,$ > then it will display "nofe" in theD > "Characteristics:" section of an SDA> SHOW DEVICE DKAxx: display).< > ----------------------------------------------------------< > Keith Parris | parris <at> DECUServe <dot> decus <dot> org   ------------------------------  # Date: Sat, 08 Jun 2002 03:00:41 GMTn1 From: "David J. Dachtera" <djesys.nospam@fsi.net>lF Subject: Re: How to put my VMS box on a BroadBand Connection (part II)' Message-ID: <3D017701.F289C524@fsi.net>    Didier Morandi wrote:l >   > This is exactly what happened. > L > I reinstalled IPNetRouter, followed the automatic configuration procedure,Q > removed all options (NAT, PPPoE, masquerading and such) and Bob is my uncle :-)o  H The idea seems to be to think of the broadband modem/router as servicingG your home LAN rather than the "Whinebloze" paradigm of having that gearo! service a single computer system.o  H Due to a lack of available braodband options, I'm still using a 3com 56KF LANmodem. It connects to two five-port hubs: one down here in my room,< one upstairs where my wife and daughter use their computers.  D If broadband where available, the LANmodem would get replaced with aB Netgear FR314 which would in turn connect to the ADSL modem, cable. modem, broadband transceiver(?) or whatever...   -- i David J. Dachterai dba DJE Systemsn http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/n   ------------------------------  % Date: Fri, 07 Jun 2002 21:58:02 -0400r2 From: rdeininger@mindspring.com (Robert Deininger)- Subject: Re: HP Alphaservers and Lan Consoles K Message-ID: <rdeininger-0706022158020001@11cust89.tnt2.nashua.nh.da.uu.net>   F In article <20020607152432.37112.qmail@web20205.mail.yahoo.com>, Fabio) Cardoso <fabiopenvms@yahoo.com.br> wrote:   6 >The Lan Console is based in a board (GSP) with serial6 >an lan ports. I installed HP-UX in two L-1000 servers3 >and this board was default in the config. When the=( >Itanium servers from HP become availble7 >(HP-UX/OpenVMS) may be it still as a default hardware.n  H Well, I'm not familiar with the way the HP classic servers do it, but soE far it sounds a lot like a terminal server connected to the console. t5 That's been available since forever with VMS systems..   ------------------------------   Date: 7 Jun 2002 17:47:44 GMTc1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon)0/ Subject: Re: Mark Gorham's Beer Bash in Reading + Message-ID: <adqrk0$5su$2@info.cs.uofs.edu>E  B In article <PE5M8.154233$%o.14725670@bin3.nnrp.aus1.giganews.com>,-  "Bill Todd" <billtodd@metrocast.net> writes:d |> dA |> "Bill Gunshannon" <bill@triangle.cs.uofs.edu> wrote in messagey) |> news:adqcii$30it$2@info.cs.uofs.edu...n |>   |> ... |> eJ |> > This is the second message I have seen today that hinted that portingF |> > software from Unix to VMS was somehow trivial.  Being involved inI |> > trying to sell the idea of doing exactly that, as academic projects,eK |> > to faculty and students here I can assure it is not the case.  PortingrJ |> > even trivial programs is difficult and porting something the size andL |> > complexity of StarOffice is probably on a par with the FreeVMS project. |> nJ |> While the port would likely not be trivial, it would surprise me if theI |> effort already expended porting Star Office to Windows (especially NT)c- |> didn't reduce its magnitude significantly.C  I The underlying kernel of Win2K is a lot closer to Unix than it is to VMS.SH I would bet the infamous "fork()" call is in there.  Anybody want to bet against me?  :-)   bill   -- iJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   b   ------------------------------  $ Date: Fri, 7 Jun 2002 18:44:13 +00002 From: John Eisenschmidt <jweisen@eisenschmidt.org>/ Subject: Re: Mark Gorham's Beer Bash in Readingm4 Message-ID: <20020607184413.C19828@eisenschmidt.org>   --GPJrCs/72TxItFYR* Content-Type: text/plain; charset=us-ascii Content-Disposition: inline + Content-Transfer-Encoding: quoted-printableh  L Unless the Voices are Mistaken, Bill Gunshannon (bill@triangle.cs.uofs.edu)=  Wrote:sD > In article <PE5M8.154233$%o.14725670@bin3.nnrp.aus1.giganews.com>,/ >  "Bill Todd" <billtodd@metrocast.net> writes:o > |>=20hC > |> "Bill Gunshannon" <bill@triangle.cs.uofs.edu> wrote in message + > |> news:adqcii$30it$2@info.cs.uofs.edu...  > |>=20i > |> ... > |>=20eL > |> > This is the second message I have seen today that hinted that portingH > |> > software from Unix to VMS was somehow trivial.  Being involved inK > |> > trying to sell the idea of doing exactly that, as academic projects,eL > |> > to faculty and students here I can assure it is not the case.  Porti= ngL > |> > even trivial programs is difficult and porting something the size andL > |> > complexity of StarOffice is probably on a par with the FreeVMS proje= ct.  > |>=20-L > |> While the port would likely not be trivial, it would surprise me if theK > |> effort already expended porting Star Office to Windows (especially NT)S/ > |> didn't reduce its magnitude significantly.@ >=20K > The underlying kernel of Win2K is a lot closer to Unix than it is to VMS.7J > I would bet the infamous "fork()" call is in there.  Anybody want to bet > against me?  :-) >=20 > bill  L Well, I was ready to bet with you Bill - I even went into the MSDN to find =B the proof for your statement, and then I found we were both wrong.  L The following info is from the MSDN, and I'm sure a very large software com=L pany on the American Left Coast holds the Trademark, Copyright, Copyleft, P=9 atent, and first born child of the following information:l   Port from UNIX to Win32r   =2E..s  L One of the largest areas of difference is in the process model. UNIX has fo=L rk; Win32 does not. Depending on the use of fork and the code base, Win32 h=L as two APIs that can be used: CreateProcess and CreateThread. A UNIX applic=L ation that forks multiple copies of itself can be reworked in Win32 to have=L  either multiple processes or a single process with multiple threads. If mu=L ltiple processes are used, there are multiple methods of IPC that can be us=L ed to communicate between the processes (and perhaps to update the code and=L  data of the new process to be like the parent, if the functionality that f=L ork provides is needed). SeeCreating a New Interprocess (IPC) Message Class.  L And there you have it - Windows is "as broken" as OpenVMS. Thank David Cutl= er kids! "Thanks David Cutler".6   > --=20PL > Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesF > bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. > University of Scranton   |E > Scranton, Pennsylvania   |         #include <std.disclaimer.h>  =20    --=201/ John W. Eisenschmidt <jweisen@eisenschmidt.org> 6  Homepage URL    | http://www.eisenschmidt.org/jweisenL  PGP Public Key  | http://www.eisenschmidt.org/jweisen/misc/jeisenschmidt.a= scD  PGP Fingerprint | 5F9B F916 5AD1 3295 CF99 BC1E 1F97 E6A3 37E3 BEF2  L FOO MANE PADME HUM: "Our first obligation is to keep the FOO counters turni= ng."   --GPJrCs/72TxItFYR' Content-Type: application/pgp-signature  Content-Disposition: inline    -----BEGIN PGP SIGNATURE-----  Version: GnuPG v1.0.6 (OpenBSD)n* Comment: For info see http://www.gnupg.org  @ iD8DBQE9AP79H5fmozfjvvIRAmLYAJ9FNwNHWK+KjTlem38fFG4vHr/DZgCglram /8CFBaAoz4zk9lN4A/SnmkI= =WqjZ  -----END PGP SIGNATURE-----a   --GPJrCs/72TxItFYR--   ------------------------------  # Date: Fri, 07 Jun 2002 18:51:42 GMT * From: "Bill Todd" <billtodd@metrocast.net>/ Subject: Re: Mark Gorham's Beer Bash in Reading B Message-ID: <2h7M8.154571$%o.14795516@bin3.nnrp.aus1.giganews.com>  > "Bill Gunshannon" <bill@triangle.cs.uofs.edu> wrote in message% news:adqrk0$5su$2@info.cs.uofs.edu...    ...   K > The underlying kernel of Win2K is a lot closer to Unix than it is to VMS.i  L With all due respect, Bill, you don't have a clue in this matter.  You couldL educate yourself by searching for 'Cutler' in the c.o.v. archives on Google.  J > I would bet the infamous "fork()" call is in there.  Anybody want to bet > against me?  :-)  J I'd take a small amount of that, since "Inside Windows 2000" (3rd edition)L states, on page 299, that fork is implemented in the POSIX subsystem (ratherI than being accessible from the normal Win32 environment) - though it doesS? make use of underlying copy-on-write memory-management support.2  I (Of course, IIRC 'fork' existed in RSX.  But it wasn't quite the same...)e   - bill   ------------------------------  $ Date: Fri, 7 Jun 2002 14:40:46 -05001 From: "Dave Gudewicz" <david.gudewicz@abbott.com>n/ Subject: Re: Mark Gorham's Beer Bash in Readings1 Message-ID: <adr2ao$1b4$1@fizban.pprd.abbott.com>r  L Not betting but I did hear at a DECUS (oops Encompass) LUG meeting this week; that fork() is coming real soon to a  VMS theater near you.l   -- Dave...o  L Let me make the superstitions of a nation and I care not who makes its laws. -----Mark Twain   > "Bill Gunshannon" <bill@triangle.cs.uofs.edu> wrote in message% news:adqrk0$5su$2@info.cs.uofs.edu...hD > In article <PE5M8.154233$%o.14725670@bin3.nnrp.aus1.giganews.com>,/ >  "Bill Todd" <billtodd@metrocast.net> writes:a > |>C > |> "Bill Gunshannon" <bill@triangle.cs.uofs.edu> wrote in messagem+ > |> news:adqcii$30it$2@info.cs.uofs.edu...f > |> > |> ... > |>L > |> > This is the second message I have seen today that hinted that portingH > |> > software from Unix to VMS was somehow trivial.  Being involved inK > |> > trying to sell the idea of doing exactly that, as academic projects, D > |> > to faculty and students here I can assure it is not the case. Porting(L > |> > even trivial programs is difficult and porting something the size andE > |> > complexity of StarOffice is probably on a par with the FreeVMSl project. > |>L > |> While the port would likely not be trivial, it would surprise me if theK > |> effort already expended porting Star Office to Windows (especially NT)a/ > |> didn't reduce its magnitude significantly.t >tK > The underlying kernel of Win2K is a lot closer to Unix than it is to VMS.sJ > I would bet the infamous "fork()" call is in there.  Anybody want to bet > against me?  :-) >a > bill >b > --L > Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesF > bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. > University of Scranton   |@ > Scranton, Pennsylvania   |         #include <std.disclaimer.h>   ------------------------------   Date: 7 Jun 2002 14:43:19 -0600e+ From: young_r@encompasserve.org (Rob Young)t/ Subject: Re: Mark Gorham's Beer Bash in Reading 3 Message-ID: <0R8D5559Pfxl@eisner.encompasserve.org>i  o In article <2h7M8.154571$%o.14795516@bin3.nnrp.aus1.giganews.com>, "Bill Todd" <billtodd@metrocast.net> writes:  > @ > "Bill Gunshannon" <bill@triangle.cs.uofs.edu> wrote in message' > news:adqrk0$5su$2@info.cs.uofs.edu...  >  > ...e > L >> The underlying kernel of Win2K is a lot closer to Unix than it is to VMS. > N > With all due respect, Bill, you don't have a clue in this matter.  You couldN > educate yourself by searching for 'Cutler' in the c.o.v. archives on Google. >   ' 	This urban legend keeps resurfacing.  r  G 	One of the top kernel hackers (outside of OS developers) did a classicy@ 	post out on this a few years ago.  A real keeper.  It should be 	in a FAQ somewhere.  : From: Jamie Hanrahan, Kernel Mode Systems (jeh@cmkrnl.com)" Subject: Re: FX!32, affinity etc.  Newsgroups: comp.os.vmsh Date: 1997/05/28 l  e  ( In article <338A488D.6FC7@videotron.ca>,1  jfmezei <"[nospam]jfmezei"@videotron.ca> writes:S) > David Cathey (Remove MX to mail) wrote:wF >> culmulated into OpenVMS.  The sad thing is Cutler had to capitulateA >> to a Windows assimilation of his theoretically sound base O/S.oB >> You'd think he would have taken the tenets of procedure calling >> standards with him as well. > D > I think it is quite pretentious of the VMSer to think that Cutler C > had so much design authority over NT and that he would have taken  > his VMS knowledge with him.   F Not pretentitious at all.  We're simply aware of facts you apparently  haven't heard about.    F From the internals point of view there is utterly no question that NT  is VMS re-implemented.    F > Cutler was told to replace DOS with a real kernel over which Windows: > could run, over which teh windows API could run etc etc.  D This is off-topic, but you're incorrect here also.  Originally thereC wasn't even going to be a Win32 API - 16-bit Windows had not gotten-B all that popular when NT was conceived.  NT was originally to be aD follow-on to OS/2, and a cooperative effort with IBM.  But somewhereF along the way Windows got pretty popular, IBM and MS split the beast, 7 IBM getting the OS/2 parts and MS keeping the NT parts.L  J > If you look at the PSION PDA operating system (called EPOC), you'll alsoA > find many many similarities with VMS. Event Flags, Interprocess J > Mailboxes, shared memory between processes, process priorities, and evenG > a utility (SPY) which is the equivalent to SHOW SYS. Its IO system iso? > similar to VMS (an equivalent to $ASSIGN with the device name A > determining which driver to use, and $QIO which is more or less % > independant of the device itself.).   B That's out at the UI and API level.  We're talking about internal  similarities.  m  A > NT is an WINDOWS operating system with modern operating systemscA > services. They were implemented with the Windows API mentality. 5 > Stop thinking that NT is VMS with WINDOWS above it.i  C No, not "VMS with Windows above it", but a VMS-derived design with E" Windows above it, most certainly.   H > NT differs from DOS in that it has real operating system features, but+ > the later are not the exclusivity of VMS.L  C The "real operating system features" you speak of are at the UI andi; API level.  They are not the reasons we consider NT to be a 0 reimplementation of VMS at the internal level.    
 How about:  C The scheduler.  (process scheduler in VMS, thread scheduler in NT) 1C 32 scheduling priorities, divided into the "real-time" (16-31) and RE "variable" (0-15) priority ranges.  identical preemption at ready by n> higher-priority threads; identical quantum and priority boost E implementations; identical CPU starvation avoidance mechanism to get  B out of priority inversion situations; a null thread for each CPU;  etc., etc.    6 Memory management.  0-7FFFFFFF is per-process, mostly C user-mode-accessible only; 80000000-FFFFFFFF is systemwide, mostly  C kernel-accessible only.  Functionally identical implementations of o paging vs. swapping. i  B I/O.  I could write a book (in fact, I am), but briefly, IRPs are E IRPs, UCBs are "device objects", CRBs are "controller objects", ADPs o= are "adapter objects", FDT routines are "dispatch routines", ,G EXE$QIODRVPKT is IoStartPacket, StartIO routines are StartIO routines, sC fork routines are DPC routines, ASTs are APCs... etc., etc., etc., m etc., etc.    F Interrupt handling.  32 levels of interrupts (some simulated but this E is nevertheless the way the code is written).  IPLs on VMS, IRQLs on BG NT.  In order:  Passive level, APC (AST) Level, Dispatch (fork) level, BB then the IO hardware interrupts, then some "hardware maintenance" E functions like the hardware timer, IPI, power fail notification, and .$ HIGH_LEVEL to block all interrupts.   > Face it, JF, you're wrong.  Worse, you are writing not just in? misunderstanding but in ignorance of the facts.  Please go readeE _Showstopper_ and _Inside Windows NT_ (Custer) before opining furtherp on this subject. E  6 	--- Jamie Hanrahan, Kernel Mode Systems, San Diego CAB         Internet: jeh@cmkrnl.com (JH645)  CompuServe: 74140,2055  O drivers, internals, networks, applications, and training for VMS and Windows NTtD NT driver FAQ, links, and other information:  http://www.cmkrnl.com/   ------------------------------  # Date: Fri, 07 Jun 2002 20:00:31 GMTm. From: peter@langstoeger.at (Peter LANGSTOEGER)' Subject: Re: Newsreader for OpenVMS 7.2A5 Message-ID: <zh8M8.146819$305.2010668@news.chello.at>/  e In article <slyL8.23$xB2.57@news-srv1.fmr.com>, sy18889@rabbit.fmr.com (Bradford J. Hamilton) writes:<g >In article <LSxL8.125583$305.1640999@news.chello.at>, peter@langstoeger.at (Peter LANGSTOEGER) writes: J >>NEWSRDR does not support XOVER. It also handles postings via the articleG >>numbers and not via the Msg-IDs. Many NNTPSERVERs out there are sillyiJ >>enough to renumber the postings at random times so that you then may getH >>already read postings as new or unread postings as read (maybe I run a; >>NNTPSERVER myself to avoid this - I'm sick of this crap).s > 4 >Forgive my asking a silly question - what is XOVER?  6 1.) I've to correct myself. NEWSRDR does support XOVERJ At least if NEWSRDR_NO_XOVER is defined to FALSE or not defined at all ;-)  N 2.) XOVER is a NNTP command to get a quick OVERview of the articles (headers).> XOVER is said to have lower performance hits on the NNTPSERVER> (Syntax is XOVER [article-range]) cause of better cache usage. I can't really comment on that.a  : >I've noticed some of the symptoms you've mentioned above.  L Better would be, how to fix them without switching to another newsreader ;-)  G >>NEWSRDR has also a problem with the NEWSRC (if you choose to use it). G >>Reading threadwise marks some articles read, leaving big holes in theEL >>numbers list. Writing the NEWSRC file does work (making really big lines -I >>you know, one line per newsgroup), reading it afterwards doesn't always  >>(because lines are too long).0 >>G >>Workaround is to not use NEWSRC or not use READ/THREAD or not let theh4 >>list of unseen postings grow beyond low hundreds).> >>But no such problems with the native NEWSRDR_PROFILE.NRPF... >0C >I don't use the NEWSRC now, but I noticed problems when I used it.b) >I'll have to investigate using the .NPRFC  @ NEWSRDR does automatically use NEWSRDR_PROFILE (SYS$LOGIN:.NRPF)A The NEWSRC can be used _in addition_ as a way to synchronize witheC another Newsclient (like MXRN). The NEWSRC file (if set) is read by/K starting NEWSRDR (to update NEWSRDR_PROFILE) and written on NEWSRDR's exit.t  I >>NEWSRDR does also not cache article headers so that a READ/THREAD scansrK >>through all the article headers on the server AGAIN and AGAIN. NNTPSERVER M >>admins do not like this (cause of server performance and access statistics)4< >>and YOU also won't like this (it could get REALLY boring). >c >Thanks for the warning.  B Again, I've to correct me. NEWSRDR has some header cache built in./ NEWSRDR_HEADER_CACHE_SIZE is used to adjust it.iG But, I so far haven't found out how to use it better. All I see here isi9 that NEWSRDR does only fetch[/cache] 10 articles at once.yE Obviously not good enough for an overview of threads if you have overT 1000 articles unseen...   0 NEWSRDR does also use HEAD/BODY and not ARTICLE,N which also doesn't make your NNTPSERVER admin happy. At least not my previous.  M >I wish to use only a "character cell" interface, for reasons of security andtP >bandwidth - can you or others recommend a newsreader with those charateristics?  B I know of BULLETIN (used it about a decade ago), LYNX (use it onlyE for HTTP, but so far not for NNTP), NEWSRDR (my favorite) and [AFAIK]eI SLRN (the installation is still on my ToDoList) which meet this criteria.r There may be others as well...   -- t Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialista E-mail  peter@langstoeger.atP A-1030 VIENNA  AUSTRIA              I'm looking for (a) Network _and_ VMS Job(s)   ------------------------------   Date: 7 Jun 2002 14:51:30 -0700i( From: bob@instantwhip.com (Bob Ceculski) Subject: Re: No new Alpha salesh= Message-ID: <d7791aa1.0206071351.278f524b@posting.google.com>h   Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> wrote in message news:<adqg4m$73h$1@new-usenet.uk.sun.com>...  > F > I agree entirely, hence why I used the SAP results to illustrate whyA > Robs claims that you will need more SPARC CPU's than Alphas are  > BS.- > N > > When we did the SAP BM that was the latest version, adjust it by 70% and aP > > per CPU measure comes out just about the same, 50 users per cpu on GS, 53 onO > > F15k. Then we could have a discussion about the shape of the tail-off curveaM > > and the effect of the GS with 1001MHz CPUs with bigger caches which couldaK > > well put the GS back up there on a per CPU basis. I don,t, can't, claimhK > > parity with the F15k at its top end. What I wanted to show, and I stillnH > > believe that I did, was that the GS series are not the total failureO > > technically that Andrew claims. The only competitor I really worry about issG > > IBM with their p690 - anything else can be beaten - even with a GS.t > >  >  l@ > But sadly none of the none SPECint/SPECfp results show this as= > I have illustrated. There are no performance numbers of theoD > type you seem to prefer which show the GS320/160 to be competitiveB > against, Sun, HP or IBM, you should be afraid of all of them and > not just IBM.M > = > And this is not just a criticism that can be applied to ther9 > GS series, the Turbolasers suffered in exactly the samee8 > way. Great SPECint and SPECfp numbers below average on8 > the kind of apps benchmarks that you appear to prefer. > 9 > In both the Turbolaser and WildFires there were obviouso6 > inadeqacies in the system design. The TurboLaser had6 > a relatively slow and long latency interconnect when; > compared with other competing systems and having to tradei7 > CPU's/Memory and I/O in configs ment that most of the : > claims made by Digital for the boxes would have required; > two 8400 backplanes to be bolted together to be realised.  > 6 > The WildFire problems have been discussed at length. > ; > Of course on the plus side being bought by HP removes onen > of the competitors to fear.  > K > > As to TPC-H - TPC counsel against comparing differing database sizes. IhI > > can't remember the date of the GS 300GB benchmark, and I think it got L > > withdrawn. It's yet another benchmark where it would be nice to have theN > > same benchmark on the same day and from current price lists - but it ain't > > going to happen. > >  > 6 > They do, but there is in practice a good correlation9 > between say 300 GB and 1 TB results for the same systemt9 > and this makes it a fairly safe bet to compare 300 withd > 1000 GB results. > 	 > Regardsd > Andrew Harrisone  B of course all of this means nothing when EV7 is in the picture ...< then Andrew, you and IBM and all the rest get blown away ...   ------------------------------   Date: 7 Jun 2002 14:48:11 -0700D( From: bob@instantwhip.com (Bob Ceculski) Subject: Re: Open Letter to HP= Message-ID: <d7791aa1.0206071348.6c59713c@posting.google.com>    Andrew Harrison SUNUK Consultancy <andrew_nospam.harrison_remove_this@sun#.com> wrote in message news:<adqeka$6jq$1@new-usenet.uk.sun.com>...  >  > E > As I said earlier, revenues are down but in our sector Sun is doingoI > better than the rest. And its the sector that HP are targetting becauseeF > it has the pottential for higher margins and better access to anuityD > business. Why do you think HP wanted to develop IA-64 it wasn't to > dominate the desktop.r >  > 	 > Regards  >  > Andrew Harrisonu  $ How are you doing in the VMS sector?   ------------------------------  % Date: Fri, 07 Jun 2002 13:23:43 -0500GC From: "Craig A. Berry" <craig.berry@nospam.SignalTreeSolutions.com>rF Subject: open source porting (was Re: Carly was here in ZKO yesterday)H Message-ID: <craig.berry-152C0B.13234207062002@news.directvinternet.com>   In article a? <HM4M8.245482$t8_.177891@news01.bloor.is.net.cable.rogers.com>,e%  "John Smith" <a@nonymous.com> wrote:   M > HP should take some of its staff and port of the top 100 PD apps that don'tdB > currently have a VMS port. It shows commitment, it increases theN > availability of apps/tools, it helps negates statements from Sun/IBM, etc...A > about app/tool availability on VMS. And it isn't overly costly.f  A This is a terrific idea but one that would require some cultural sE readjustment for OpenVMS engineering (not for the engineers, but for oC the managers who allocate resources).  Most of the ports they make  G available they either grab something others have done (Perl) or do all tE the work themselves (Mozilla, Apache).  The only collaborative thing WF I've seen done is available at <http://gnv.sourceforge.net>, but work A on this proceeds very slowly, probably because the engineers who rH participate are in the same boat we are, i.e., trying to fit in porting  in their spare time.  H A commitment of more resources not only to improving the infrastructure C for porting (which seems to be happening slowly but surely) but to ,E actually get a suite of ports done would be nice.  The benefits seem tD obvious to customers, but since the impact on any potential revenue E stream is indirect it might be a hard sell.  Relationships with ISVs  B could also get prickly.  Would Oracle, Mimer, and Intersystems be D pleased if HP committed resources to porting MySQL and PostgreSQL?  F None of this means it shouldn't be done, I'm just saying I don't know : how to make the case in terms that would get it to happen.   ------------------------------  % Date: Fri, 07 Jun 2002 18:08:59 -0400E( From: David Froble <davef@tsoft-inc.com>J Subject: Re: open source porting (was Re: Carly was here in ZKO yesterday), Message-ID: <3D012EFB.6010808@tsoft-inc.com>   Craig A. Berry wrote::  
 > In article oA > <HM4M8.245482$t8_.177891@news01.bloor.is.net.cable.rogers.com>,e' >  "John Smith" <a@nonymous.com> wrote:p >  > M >>HP should take some of its staff and port of the top 100 PD apps that don't>B >>currently have a VMS port. It shows commitment, it increases theN >>availability of apps/tools, it helps negates statements from Sun/IBM, etc...A >>about app/tool availability on VMS. And it isn't overly costly.  >> > C > This is a terrific idea but one that would require some cultural rG > readjustment for OpenVMS engineering (not for the engineers, but for iE > the managers who allocate resources).  Most of the ports they make fI > available they either grab something others have done (Perl) or do all aG > the work themselves (Mozilla, Apache).  The only collaborative thing  H > I've seen done is available at <http://gnv.sourceforge.net>, but work C > on this proceeds very slowly, probably because the engineers who nJ > participate are in the same boat we are, i.e., trying to fit in porting  > in their spare time. > J > A commitment of more resources not only to improving the infrastructure E > for porting (which seems to be happening slowly but surely) but to -G > actually get a suite of ports done would be nice.  The benefits seem pF > obvious to customers, but since the impact on any potential revenue G > stream is indirect it might be a hard sell.  Relationships with ISVs hD > could also get prickly.  Would Oracle, Mimer, and Intersystems be F > pleased if HP committed resources to porting MySQL and PostgreSQL?  H > None of this means it shouldn't be done, I'm just saying I don't know < > how to make the case in terms that would get it to happen. >     O Well, I have an idea that just may work.  Have a list of prefered products for iM porting to VMS.  Anybody who wants to sign up for one of the products, would .Q port the product, and upon delivery of the ported code to HP, and possibly after  Q independant verification of it working, they would be rewarded with DS10 system, e or other suitable reward..  N Not too expensive for HP.  They get the work done without ISVs having much to J complain about.  They don't have to commit resources to the project.  The P biggest problem would be to determine suitable rewards for each job.  Some will T be small, some large.  Some would be more desirable, some only marginally desirable.  O There's also the issue of the quality of the work.  Then again, that issue may d" exist before the port begins.  :-)  P Some variations may be loan of some resources to do the job.  Could be over the J internet, so HP doesn't have to hand over hardware until the work is done.  Q I'm guessing all products will be implemented in C.  That leaves me out, I won't wA touch the language.  Might be coaxed to re-implement it in BASIC.u   Dave   ------------------------------  * Date: Fri, 7 Jun 2002 18:09:37 +0000 (UTC) From: david20@alpha2.mdx.ac.ukP Subject: Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading)+ Message-ID: <adqst1$am3$1@aquila.mdx.ac.uk>n  e In article <adqole$t07$1@fizban.pprd.abbott.com>, "Dave Gudewicz" <david.gudewicz@abbott.com> writes:c >I'm no security expert, but...a >tM >I've read where SSH is a simple solution to a comlex problem.  That's OK butrD >I've also read that Kerberos is *really* a better solution, but itsK >complexity keeps people away from it.  Kerberos is in TCP/IP services v5.3  >t  N If that were the case then its usually better to support the simpler solution.K History suggests that complex solutions X400 mail, X500 directories usuallyPI lose out to what appear to be simpler solutions SMTP, LDAP (even if laterrM those solutions have the complexity added back in as has happened with LDAP).   L The Kerberos server has been available for quite a few months now. HopefullyK (and I haven't looked at the specs for TCP/IP services v5.3 yet) it has nownJ enabled its TELNET, FTP etc to work with that (or other) kerberos servers.  K However Kerberos and SSH are really tackling different although in somewaysiK related problems. Kerberos provides a mechanism for a single authentication M domain for a group of systems. Once the authentication has been completed theaO actual data transfered between the user and the applications he/she is using ont+ remote systems may or may not be encrypted. J SSH provides end-to-end encryption between systems running SSH servers andO clients in the form of  Telnet/R-commands, FTP replacements. There is no singlefK authentication domain you just use your username and password on the remoteiN ssh server system as you would if connecting with telnet - the only differenceL is that the username, password and subsequent traffic is not passed in clear but is encrypted.aM Kerberos requires all systems in a domain to talk to special Kerberos systems6I in those domains to obtain authorisation credentials to be used to accessdO services in that domain. Trust relationships can also be setup between domains.aM SSH requires no third-party systems to be involved in the exchange - just then ssh client and the ssh server.  H VMS does have a public domain implementation of SSH version 1 server andA client due to the valued work of David Jones and Richard Levitte.mJ I believe Multinet and TCPWARE will shortly (if they do not already) have " version 2 SSH clients and servers.L Dec TCPIP services though, despite frequent requests, seems not to have any ' firm plans for providing SSH version 2.eN (Note. I believe Multinet and Tcpware have had kerberised Telnet etc services K for a number of years so that they could participate in Unix based kerberos 	 domains).e  O SSH version 1 is becoming more difficult to use because it has been replaced by K version 2. Unix users have all moved to using version 2 because of security I holes in version 1 (which were generally buffer overflows on Unix systemssK implementations of SSH version 1). A pure version 2 client will not talk totI a version 1 server. Also the SSH FTP replacement protocol has changed in e SSH version 2.  M The work involved in converting the public domain VMS SSH version 1 server torK SSH version 2 is fairly substantial and the author does not (as far as I amc, aware) have any plans to upgrade the server.  
 David Webb VMS and Unix team leader CCSS Middlesex University   >--. >Dave... >mM >Let me make the superstitions of a nation and I care not who makes its laws.m >-----Mark Twain > , ><david20@alpha1.mdx.ac.uk> wrote in message& >news:adqdpj$5kp$1@aquila.mdx.ac.uk...; >> In article <adq91e$bjj$1@mordred.cc.jyu.fi>, Osmo Kujala/! ><kujala@tukki.cc.jyu.fi> writes:-- >> >Bob Ceculski <bob@instantwhip.com> wrote:3; >> >> Osmo Kujala <kujala@tukki.cc.jyu.fi> wrote in messagec) >news:<adnmh9$4f3$1@mordred.cc.jyu.fi>...- >> >....L >> >>> services more reliable and easier to maintain. BUT when OpenVMS is soL >> >>> neglected and kept secret, I would never even suggest to increase its >use.1 >> >' >> >>  ... who cares if its a secret...oG >> >They who choose. I'm only VMS system manager here. VMS was our maint >systemdJ >> >once. DEC got us convinced to go other way (mainly by overpricing, butH >> >saying it directly too.) Now we clearly are not in VMS niche. (Hint: >> >university)hI >> >One little example of the neglect: OpenVMS doesn't reach our securityeC >> >standards. There must be SSH level II. (I know it can be hunteds >> >somewhere.)  >> > >> >Osmo >> > >>I >> Yes this seems to be a major problem with Dec TCPIP services. Althought >lots ofH >> people ask for things like SSH version 2 it takes forever for them to >appear.M >> I just hope this isn't like the request for XDMCP support - Something likeh >6 >> years wait. >>
 >> David WebbE >> VMS and Unix team leaderl >> CCSS  >> Middlesex Universityg >  >a   ------------------------------  $ Date: Fri, 7 Jun 2002 14:36:26 -05001 From: "Dave Gudewicz" <david.gudewicz@abbott.com>mP Subject: Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading)1 Message-ID: <adr22k$1ap$1@fizban.pprd.abbott.com>h  L David you are correct and thanks for the detailed descriptions.  Hope othersJ here will benefir from them.  We are using SSH on some of our VMS systems.E And on a related note, the new Kermit 95 release supports a number oflK different security related things including SSH v2 and Kerberos and  ......H   -- Dave...   L Let me make the superstitions of a nation and I care not who makes its laws. -----Mark Twaine  + <david20@alpha2.mdx.ac.uk> wrote in messageN% news:adqst1$am3$1@aquila.mdx.ac.uk...IC > In article <adqole$t07$1@fizban.pprd.abbott.com>, "Dave Gudewicz"s# <david.gudewicz@abbott.com> writes:Y! > >I'm no security expert, but...s > >nK > >I've read where SSH is a simple solution to a comlex problem.  That's OK  butCF > >I've also read that Kerberos is *really* a better solution, but itsH > >complexity keeps people away from it.  Kerberos is in TCP/IP services v5.3 > >t > F > If that were the case then its usually better to support the simpler	 solution.IE > History suggests that complex solutions X400 mail, X500 directoriesa usually K > lose out to what appear to be simpler solutions SMTP, LDAP (even if latercH > those solutions have the complexity added back in as has happened with LDAP). >eD > The Kerberos server has been available for quite a few months now.	 Hopefully'I > (and I haven't looked at the specs for TCP/IP services v5.3 yet) it hasy nowhL > enabled its TELNET, FTP etc to work with that (or other) kerberos servers. >aD > However Kerberos and SSH are really tackling different although in someways> > related problems. Kerberos provides a mechanism for a single authenticationK > domain for a group of systems. Once the authentication has been completede the H > actual data transfered between the user and the applications he/she is using on- > remote systems may or may not be encrypted.bL > SSH provides end-to-end encryption between systems running SSH servers andJ > clients in the form of  Telnet/R-commands, FTP replacements. There is no singleF > authentication domain you just use your username and password on the remoteE > ssh server system as you would if connecting with telnet - the onlyt
 differenceH > is that the username, password and subsequent traffic is not passed in clear  > but is encrypted.-G > Kerberos requires all systems in a domain to talk to special KerberosB systemskK > in those domains to obtain authorisation credentials to be used to access4H > services in that domain. Trust relationships can also be setup between domains.K > SSH requires no third-party systems to be involved in the exchange - justo the   > ssh client and the ssh server. > J > VMS does have a public domain implementation of SSH version 1 server andC > client due to the valued work of David Jones and Richard Levitte.MK > I believe Multinet and TCPWARE will shortly (if they do not already) havee$ > version 2 SSH clients and servers.I > Dec TCPIP services though, despite frequent requests, seems not to havep anyt) > firm plans for providing SSH version 2.hF > (Note. I believe Multinet and Tcpware have had kerberised Telnet etc servicesD > for a number of years so that they could participate in Unix based kerberos > domains).W > E > SSH version 1 is becoming more difficult to use because it has been  replaced bytD > version 2. Unix users have all moved to using version 2 because of securityK > holes in version 1 (which were generally buffer overflows on Unix systems J > implementations of SSH version 1). A pure version 2 client will not talk toJ > a version 1 server. Also the SSH FTP replacement protocol has changed in > SSH version 2. >oL > The work involved in converting the public domain VMS SSH version 1 server toJ > SSH version 2 is fairly substantial and the author does not (as far as I am. > aware) have any plans to upgrade the server. >r > David Webb > VMS and Unix team leader > CCSS > Middlesex University >S > >--u
 > >Dave... > >gI > >Let me make the superstitions of a nation and I care not who makes itst laws.u > >-----Mark Twain > >e. > ><david20@alpha1.mdx.ac.uk> wrote in message( > >news:adqdpj$5kp$1@aquila.mdx.ac.uk...= > >> In article <adq91e$bjj$1@mordred.cc.jyu.fi>, Osmo Kujalao# > ><kujala@tukki.cc.jyu.fi> writes: / > >> >Bob Ceculski <bob@instantwhip.com> wrote:o= > >> >> Osmo Kujala <kujala@tukki.cc.jyu.fi> wrote in messager+ > >news:<adnmh9$4f3$1@mordred.cc.jyu.fi>...e
 > >> >....K > >> >>> services more reliable and easier to maintain. BUT when OpenVMS isc soJ > >> >>> neglected and kept secret, I would never even suggest to increase its  > >use.  > >> >) > >> >>  ... who cares if its a secret...nI > >> >They who choose. I'm only VMS system manager here. VMS was our main,	 > >systemkL > >> >once. DEC got us convinced to go other way (mainly by overpricing, butJ > >> >saying it directly too.) Now we clearly are not in VMS niche. (Hint: > >> >university)oK > >> >One little example of the neglect: OpenVMS doesn't reach our securityiE > >> >standards. There must be SSH level II. (I know it can be huntedh > >> >somewhere.)y > >> >
 > >> >Osmo > >> > > >>K > >> Yes this seems to be a major problem with Dec TCPIP services. Although 
 > >lots ofJ > >> people ask for things like SSH version 2 it takes forever for them to
 > >appear.J > >> I just hope this isn't like the request for XDMCP support - Something like > >6 > >> years wait. > >> > >> David WebbR > >> VMS and Unix team leaderD	 > >> CCSS. > >> Middlesex University. > >e > >    ------------------------------  % Date: Fri, 07 Jun 2002 16:45:48 -0400u- From: JF Mezei <jfmezei.spamnot@videotron.ca>-P Subject: Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading), Message-ID: <3D011B79.6516718B@videotron.ca>   re: SSH   L Ok, just a sanity check here. Is unencrypted telnet really that dangerous in terms of risk ?e  K Are there really people at ISPs who put traces on lines to what what people  are typing ?  L Can a jacker in China "listen in" on a Telnet session I am having between my& host in Canada and a host in the USA ?  I I am trying to understandthe type of spying that would be done to see the"  contents of your telnet session.  N If SSH is so important, how come SMTP still runs over unencrypted lines ? DoesJ Bill gates encrypt the emails he sends to Carly, threathening to raise the' price of Windows if Carly markets VMS ?1   ------------------------------   Date: 7 Jun 2002 14:45:01 -0700o( From: bob@instantwhip.com (Bob Ceculski)P Subject: Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading)= Message-ID: <d7791aa1.0206071345.15bdf42f@posting.google.com>o  ^ Osmo Kujala <kujala@tukki.cc.jyu.fi> wrote in message news:<adq91e$bjj$1@mordred.cc.jyu.fi>...+ > Bob Ceculski <bob@instantwhip.com> wrote:ib > > Osmo Kujala <kujala@tukki.cc.jyu.fi> wrote in message news:<adnmh9$4f3$1@mordred.cc.jyu.fi>... >  ...K > >> services more reliable and easier to maintain. BUT when OpenVMS is so eP > >> neglected and kept secret, I would never even suggest to increase its use.  >  t% > >  ... who cares if its a secret...cL > They who choose. I'm only VMS system manager here. VMS was our main systemH > once. DEC got us convinced to go other way (mainly by overpricing, butF > saying it directly too.) Now we clearly are not in VMS niche. (Hint:
 > university)tG > One little example of the neglect: OpenVMS doesn't reach our securitytB > standards. There must be SSH level II. (I know it can be hunted 
 > somewhere.)  >  > Osmo  H ssh 2 is now on multinet and soon to be on tcpware Ip stacks for vms ...   ------------------------------   Date: 7 Jun 2002 14:46:11 -0700 ( From: bob@instantwhip.com (Bob Ceculski)P Subject: Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading)= Message-ID: <d7791aa1.0206071346.6fc10a3f@posting.google.com>s  Q david20@alpha1.mdx.ac.uk wrote in message news:<adqdpj$5kp$1@aquila.mdx.ac.uk>...o[ > In article <adq91e$bjj$1@mordred.cc.jyu.fi>, Osmo Kujala <kujala@tukki.cc.jyu.fi> writes: , > >Bob Ceculski <bob@instantwhip.com> wrote:c > >> Osmo Kujala <kujala@tukki.cc.jyu.fi> wrote in message news:<adnmh9$4f3$1@mordred.cc.jyu.fi>...e >  .... L > >>> services more reliable and easier to maintain. BUT when OpenVMS is so Q > >>> neglected and kept secret, I would never even suggest to increase its use. " >  S& > >>  ... who cares if its a secret...M > >They who choose. I'm only VMS system manager here. VMS was our main system I > >once. DEC got us convinced to go other way (mainly by overpricing, butaG > >saying it directly too.) Now we clearly are not in VMS niche. (Hint:> > >university)H > >One little example of the neglect: OpenVMS doesn't reach our securityC > >standards. There must be SSH level II. (I know it can be hunted f > >somewhere.) > >g > >Osmo  > >n > P > Yes this seems to be a major problem with Dec TCPIP services. Although lots ofO > people ask for things like SSH version 2 it takes forever for them to appear. N > I just hope this isn't like the request for XDMCP support - Something like 6
 > years wait.v >  > David Webb > VMS and Unix team leader > CCSS > Middlesex University  E SSH 2 has been on multinet since beginning of year and will now be on  TCPware ...i   ------------------------------   Date: 7 Jun 2002 22:39:20 GMTm2 From: "Zane H. Healy" <healyzh@shell1.aracnet.com>P Subject: Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading)+ Message-ID: <adrcmo0dsr@enews3.newsguy.com>u  . JF Mezei <jfmezei.spamnot@videotron.ca> wrote:	 > re: SSH   N > Ok, just a sanity check here. Is unencrypted telnet really that dangerous in > terms of risk ?i  J A few years ago I was playing around with some network monitoring softwareL (publically available software for Linux) on my home network.  I was rather M shocked to find that I could basically point it at one packet, and from then oN on I could see anything that was typed/displayed in a specific Telnet session.  G So, yes, it really is that dangerous.  Unfortunatly a lot of people are ; still forced to use Telnet.  IIRC, another problem is POP3.,   			Zaneo   ------------------------------  % Date: Fri, 07 Jun 2002 21:15:08 -0400t- From: JF Mezei <jfmezei.spamnot@videotron.ca>hP Subject: Re: OpenVMS, Volume Desktop OS (Re: Mark Gorham's Beer Bash in Reading), Message-ID: <3D015A9C.2A89BBAF@videotron.ca>   "Zane H. Healy" wrote:L > A few years ago I was playing around with some network monitoring softwareM > (publically available software for Linux) on my home network.  I was rathereN > shocked to find that I could basically point it at one packet, and from thenP > on I could see anything that was typed/displayed in a specific Telnet session.  M But on a LAN when you can run such traces, you have access to a hell of a lote more than telnet data.  I My question pertains to the internet itself. What are the odds of someonetJ monitoring your telnet packets between your LAN and the destination LAN ?   M Also, assume you manage the network backbone. You have access to all the dataiK that transits in the backbone. But when you telnet to the router that feeds0L one departmental lan, someone with an ethermon on that LAN will not see yourN telnet session to the router and hence won't be able to capture the managementH passord for the router. But if you do this from a de]artmental lan, thenF anyone on that lan will be able to ethermon you and see what you type.  G Are ISPs and internet backbine networks really so untrustworthy that ane7 average telnet session is considered such a high risk ?   N I agree that you may not want to use unencrypted telnet for high value or highL security stuff. But would it be fair to state that a telnet session over theH internet has a lesser risk of being intercepted than a paper letter sent through the postal system ?e   ------------------------------  # Date: Fri, 07 Jun 2002 18:43:40 GMTD- From: "John E. Malmberg" <wb8tyw@qsl.network>mY Subject: Porting Open Source to OpenVMS - was: Re: Mark Gorham's Beer Bash in Reading in e* Message-ID: <3D00FBC2.3080700@qsl.network>   Bill Gunshannon wrote: > K > The underlying kernel of Win2K is a lot closer to Unix than it is to VMS.rJ > I would bet the infamous "fork()" call is in there.  Anybody want to bet > against me?  :-)  > The only system that I have MSDN on no longer seems to have a I functioning C drive, so I can not directly make a search on all the APIs.h  G However the cross platform open source code I have seen indicates that eD Windows NT does not have a native fork() call.  There do seem to be 0 third party applications that provide it though.  H What NT has is a set of spawn*() calls.  And the OpenOffice source uses F these instead of fork() in dmake.exe program that is used to build it.   -John  wb8tyw@qsl.network Personal Opinion Onlyr   ------------------------------  $ Date: Fri, 7 Jun 2002 22:00:20 +0200 From: zessin@decus.deh# Subject: RE: SAN shadowing questionr* Message-ID: <00A0F1F7.2DE832AF.3@decus.de>   Norm wrote:r6 > I came out of a meeting yesterday with the following3 > (excuse me if I mix OpenVMS and SAN terminology):n >e6 > If I have a remote VMScluster with AlphaServers with5 > 2 KGPSA's each connected to a Brocade Switch Fabrica5 > to 2 HSG80's connected to storage at each location, 1 > I can configure a storage-based shadowset _on ar3 > single fabric_ with a member disk in EMA-box A ono6 > one site and a member disk in EMA-box B on the other9 > site, and present this as a single LUN to Alphaservers.a  7 I need a picture of the fabric layouts to really under-s stand that.e  6 > I had understood that storage-based shadowset member& > disks had to be in the same EMA-box.  7 A 'shadowset' of physical disk drives within an EMA-boxl is called a 'mirror-set'.k  8 You cannot 'shadow' the 'units'/LUNs of 2 EMA-boxes with4 controller-based software. The DRM (Data Replication6 Manager) software for the HSG80 array controllers does 'replication'.  7 That means that the hosts see the HSG units (LUNs) from 9 one site (the initiator) and read from/write to them. Allh7 write data is then copied by the HSG80 array controllerl7 modules to units (LUNs) at the other site (the target).l7 Those units do exist on the storage array, but must nota be seen by the hosts!n  4 If there is a failure in the initiator site you need5 to do a manual DRM failover, which means (among other 0 things) giving the hosts access to these units -. you *must* understand your storage very well!!0 (I do storage workshops and for the DRM workshop0 alone I calculate between 3 to 5 days. It really3 takes time + practice to get a 'feeling' for this.)n   > What is correct?  5 Sorry, but we're just not using the same language andx1 have a different understanding of your situation..3 We first need to make sure that we're talking aboutj the same thing(s).   -- a
 Uwe Zessin   ------------------------------  % Date: Fri, 07 Jun 2002 13:46:51 -0500oC From: "Craig A. Berry" <craig.berry@nospam.SignalTreeSolutions.com>$7 Subject: SSH v2 (was Re: OpenVMS, Volume Desktop OS...)rH Message-ID: <craig.berry-DE7718.13465107062002@news.directvinternet.com>  E In article <adqst1$am3$1@aquila.mdx.ac.uk>, david20@alpha2.mdx.ac.uk   wrote:  L > I believe Multinet and TCPWARE will shortly (if they do not already) have $ > version 2 SSH clients and servers.  G v2 is shipping in Multinet.  It looks like it should be in TCPWare any   day now.  N > Dec TCPIP services though, despite frequent requests, seems not to have any ) > firm plans for providing SSH version 2.V  D A year ago they had no plans.  Now they have plans, but perhaps not B much more than that yet.  Hoff said in this forum a few weeks ago:    I   We have OpenSSL in the V7.3-1 release (which is itself in field test), tI   and we expect to have a version of SSH available and incorporated in a r   release after V7.3-1.j   For the full article, seer  eI <http://groups.google.com/groups?hl=en&lr=&selm=CTjx8.5%24M65.222244%40cab cnews.cac.cpqcorp.net>  / I hope "after" doesn't mean 2-3 years from now.a   ------------------------------  # Date: Sat, 08 Jun 2002 00:58:20 GMT0# From: "John Smith" <a@nonymous.com>c; Subject: Re: SSH v2 (was Re: OpenVMS, Volume Desktop OS...)MJ Message-ID: <MEcM8.215402$ah_.154147@news01.bloor.is.net.cable.rogers.com>  F "Craig A. Berry" <craig.berry@nospam.SignalTreeSolutions.com> wrote inJ message news:craig.berry-DE7718.13465107062002@news.directvinternet.com...F > In article <adqst1$am3$1@aquila.mdx.ac.uk>, david20@alpha2.mdx.ac.uk > wrote: >oH > > I believe Multinet and TCPWARE will shortly (if they do not already) have& > > version 2 SSH clients and servers. >iH > v2 is shipping in Multinet.  It looks like it should be in TCPWare any
 > day now. >.K > > Dec TCPIP services though, despite frequent requests, seems not to havet any<+ > > firm plans for providing SSH version 2.u >iE > A year ago they had no plans.  Now they have plans, but perhaps notbD > much more than that yet.  Hoff said in this forum a few weeks ago: >b >.J >   We have OpenSSL in the V7.3-1 release (which is itself in field test),J >   and we expect to have a version of SSH available and incorporated in a >   release after V7.3-1.  >  > For the full article, sees >uK > <http://groups.google.com/groups?hl=en&lr=&selm=CTjx8.5%24M65.222244%40cae > cnews.cac.cpqcorp.net> >i1 > I hope "after" doesn't mean 2-3 years from now.b    D Therein lies some of the perception problems with VMS by prospectiveG customers and even within the existing VMS user-base. SSL, SSH2, etc...sD often are *very* late in availability compared to unix platforms. InL fairness, it isn't always clear what will be adopted as the defacto standardF vs. a de jure standard, and that can account for some of the delays in, making services like these available on VMS.  G Still, in a VMS community that is stagnant to dwindling, this has to bepH expected. And that too is another one of those 'self-fulfilling' actionsG that increase the rate of downward spiral. That's another reason why HP L can't market VMS just to 'selected' accounts in selected industries, most of whom are already VMS customers.t  7 Sorry to come back to this again, but it's all related.r   ------------------------------  * Date: Fri, 7 Jun 2002 20:27:01 +0000 (UTC)' From: Thomas Dzubin <dzubint@vcn.bc.ca>i% Subject: Re: STARTUP stops after bootd) Message-ID: <adr4ul$ie6$1@luna.vcn.bc.ca>e  9 Charlie Hammond <hammond@not@peek.ppb.cpqcorp.net> wrote:nK > Caution: Compaq warns against changing WRITESYSPARAMS.  (And one supposes F > HP does too -- <smile>)  So, do this cautiously.  You should NOT set" > WRITESYSPARAMS to 0 permanently.  G Wow, that's bizzare.  Digital Equipment Corporation ALSO warned againstaD it!   So we have three major computer companies: DEC, Compaq, and HP ... ALL warning against this.u# OK, I'll make sure I never do this.o   :-) 5 (sorry, I couldn't help it...it's a Friday afternoon)r   ------------------------------   Date: 8 Jun 2002 01:56:37 GMTs% From: Jacqueline <003564M@nyp.edu.sg>a Subject: Re: Test post4 Message-ID: <adro8l$3sr$593@clematis.singnet.com.sg>   Jacqueline Wang         o        )C                                                                    o.                                                        +                                             >                                                                    7                                                        i   <                                                             C                                                                                                 a              K                                                                            b                        y                    i6                                                                -                                              e6                                                       0                                                 B                                                                   G                                                                        t               /                                                 1                                                  g@                                                                 O                                                                                 G                                                                        q$                                     <                                                             	          n(                                         F                                                                       =                                                               -                                               
              d                            e                                                       >!                                  n
           O                                                                                >
           9                                                          tM                                                                               L                                                                                                   >                                                           e                   8                                                         P                                                                                 D                                                                     ,                                                                                  e3                                                    >                m                       +                                            -;                                                            0                                               
              RH                                                                         !                                  rD                                                                     B                                                                                                   Jacqueline Wangt   ------------------------------   Date: 8 Jun 2002 01:56:37 GMTl) From: Jacqueline <JACWANG@singnet.com.sg>  Subject: Re: Test post4 Message-ID: <adro8l$3sr$594@clematis.singnet.com.sg>   Jacqueline WangtB                                                                   8                                                               g  #                                    lE                                                                                                   -                                                   5                                                      eK                                                                            2                             	                                    <                                                             4                                                                                             F                                                                                             n                               *                                                                      K                                                                            y	          .	          M        m                w2                                                   =                                                              tH                                                                         '                                        n;                                                            I9                                                                                    O                                                                                 &                                       =                                                              e2                                                                             6M                                                                              g  e!                                  .L                                                                                                            k                     I                                                                          a               5                                                      b5                                                      >                e                              Jacqueline Wangs   ------------------------------   Date: 7 Jun 2002 14:47:52 -0700h1 From: nothome@spammers.are.scum (Malcolm Dunnett) Y Subject: Re: The Press and the IA-64 Port (was Re: Another analyst says VMS port won't bea- Message-ID: <EdmpSJcNpYMF@malvm7.mala.bc.ca.>e  C In article <LcNL8.146831$Kp.13865068@bin7.nnrp.aus1.giganews.com>, e/    "Bill Todd" <billtodd@metrocast.net> writes:s > J > So his comment needs a bit of firming up.  Its real core seems to be theI > ability to purchase near-identical hardware/OS combinations from *many* L > different suppliers such that they can be readily mixed and matched, whichL > pretty much limits hardware to x86 and limits OSs to whatever runs on x86,G > biggies being Windows, Linux, *BSD, and - if indeed many Solaris ISVs N > support their applications on x86 - Solaris.  Itanic may or may not become aN > similarly widely-supported hardware platform (with similarly wide-spread ISV' > support) at some point in the future.2 > K    So once the IPF version of VMS becomes available ( and assuming multipleeJ manufacturers build Itanic boxes ) it won't be proprietary anymore, right?G Unless, of course, HP chooses to keep it proprietary by restricting its.  availability to only HP servers.   ------------------------------  # Date: Fri, 07 Jun 2002 21:58:16 GMT * From: "Bill Todd" <billtodd@metrocast.net>Y Subject: Re: The Press and the IA-64 Port (was Re: Another analyst says VMS port won't bexB Message-ID: <Y%9M8.159066$Kp.14881658@bin7.nnrp.aus1.giganews.com>  > "Malcolm Dunnett" <nothome@spammers.are.scum> wrote in message' news:EdmpSJcNpYMF@malvm7.mala.bc.ca....cD > In article <LcNL8.146831$Kp.13865068@bin7.nnrp.aus1.giganews.com>,1 >    "Bill Todd" <billtodd@metrocast.net> writes:n > >IL > > So his comment needs a bit of firming up.  Its real core seems to be theK > > ability to purchase near-identical hardware/OS combinations from *many*eH > > different suppliers such that they can be readily mixed and matched, whichiI > > pretty much limits hardware to x86 and limits OSs to whatever runs on  x86,I > > biggies being Windows, Linux, *BSD, and - if indeed many Solaris ISVsaG > > support their applications on x86 - Solaris.  Itanic may or may nots become aL > > similarly widely-supported hardware platform (with similarly wide-spread ISVn) > > support) at some point in the future.y > > D >    So once the IPF version of VMS becomes available ( and assuming multipleL > manufacturers build Itanic boxes ) it won't be proprietary anymore, right?I > Unless, of course, HP chooses to keep it proprietary by restricting itst" > availability to only HP servers.  G I suppose that would be one interpretation, but you'd have to ask CurlyaK whether it matches his (since I suspect if you asked him about Solaris he'd  call it proprietary).    - bill   ------------------------------  % Date: Fri, 07 Jun 2002 18:08:13 -0400s- From: JF Mezei <jfmezei.spamnot@videotron.ca> Y Subject: Re: The Press and the IA-64 Port (was Re: Another analyst says VMS port won't be , Message-ID: <3D012EC5.29C4628F@videotron.ca>   Malcolm Dunnett wrote:M >    So once the IPF version of VMS becomes available ( and assuming multipleeL > manufacturers build Itanic boxes ) it won't be proprietary anymore, right?I > Unless, of course, HP chooses to keep it proprietary by restricting itsu" > availability to only HP servers.  L I *believe* that the philosophy is that VMS will only be certified to run onM specific models built by HP. I do not believe there is a commitment to ensure-; that VMS will be certified to run on all HP IA64 computers..  M And it goes to say that it is very unlikely that VMS will be certified to runRN on other manufacturer's IA64 computers, if other manufacturers choose to build such.   H There will be interesting business practices issues with regards to someK company buying its hardware from another vendor and VMS from HP, wanting ton, have full support of its VMS infrastructure.  N In the end, it is very likely that VMS, like NSK, will remain very proprietaryN on HP hardware and support, even if hobbyists might be able to run VMS on some/ other manufacturer's hardware (if such exists).e   ------------------------------  % Date: Fri, 07 Jun 2002 17:01:08 -0400b1 From: Michael Austin <maustin@firstdbasource.com>." Subject: Re: VMS Monitoring a User1 Message-ID: <3D011F14.22430CC@firstdbasource.com>t   warren sander wrote: > M > You could try getcmd ftp://ftp.process.com/vms-freeware/fileserv/getcmd.zipn8 > which will dump the command recall buffer of a process > 	 > -warren0 > --D > ------------------------------------------------------------------8 > Warren Sander                        OpenVMS MarketingI > Hewlett-Packard Company              Work:  warren.sander@remove.hp.compN > 200 Forest Street MR01-3/J1          Personal: sander@remove.ma.ultranet.com5 > Marlboro, MA 01752                   (508) 467-4875h7 >    My opinions are my own and I only speak for myselfa. >          Read http://www.openvms.compaq.com/D > ------------------------------------------------------------------ > 3 > "Ab" <spammitplease@yahoo.co.in> wrote in messagee8 > news:9f100812.0206070001.9ffd7ac@posting.google.com...
 > > Hi AllG > >   We have OpenVMS 6.2 .. want to monitor a user for the commands hes
 > > executes.gG > >  We have a problem with some privileged user logging in and killingaI > > other logins (other users logged in are automatically logged out). We'E > > want to monitor that user and create logs of the commands that helH > > executes with time information. Is there any VMS utility that allows > > us to do this. > >  Thanks in Advance > >t > >   Ab  > just a question, but would that product work after-the-fact inH otherwords after "that" user uas logged off or is it only current users?  D if it is current process only you can set a lo*gout symbol that willG quietly execute the getcmd procedure for all users with the appropriate  privs. -- n Regards,  7 Michael Austin            Registered Linux User #261163u7 First DBA Source, Inc.    http://www.firstdbasource.comg Sr. Consultant 704-947-1089 (Office)  704-236-4377 (Mobile)s   ------------------------------  % Date: Fri, 07 Jun 2002 14:56:56 -0500 C From: "Craig A. Berry" <craig.berry@nospam.SignalTreeSolutions.com>gG Subject: Win32/VMS porting (was Re: Mark Gorham's Beer Bash in Reading) H Message-ID: <craig.berry-8107E0.14565607062002@news.directvinternet.com>  + In article <adqrk0$5su$2@info.cs.uofs.edu>,r3  bill@triangle.cs.uofs.edu (Bill Gunshannon) wrote:s  D > In article <PE5M8.154233$%o.14725670@bin3.nnrp.aus1.giganews.com>,/ >  "Bill Todd" <billtodd@metrocast.net> writes:m > |> TC > |> "Bill Gunshannon" <bill@triangle.cs.uofs.edu> wrote in messageO+ > |> news:adqcii$30it$2@info.cs.uofs.edu...1 > |> 2L > |> > This is the second message I have seen today that hinted that portingH > |> > software from Unix to VMS was somehow trivial.  Being involved inK > |> > trying to sell the idea of doing exactly that, as academic projects, M > |> > to faculty and students here I can assure it is not the case.  PortingoL > |> > even trivial programs is difficult and porting something the size andN > |> > complexity of StarOffice is probably on a par with the FreeVMS project. > |> wL > |> While the port would likely not be trivial, it would surprise me if theK > |> effort already expended porting Star Office to Windows (especially NT)i/ > |> didn't reduce its magnitude significantly.v > K > The underlying kernel of Win2K is a lot closer to Unix than it is to VMS.   E I assume that's a joke since the intellectual roots of the NT kernel =- are well known and frequently discussed here.c  J > I would bet the infamous "fork()" call is in there.  Anybody want to bet > against me?  :-)  F As others have said, there is no native fork() in the Win32 API.  The H Perl port has a pseudo-fork implemented with threads that was developed A by ActiveState and funded by Microsoft.  On any system with Perl  3 installed, type the following for more information:o   $ perldoc perlfork  F From what I've seen with the Perl port, porting from Unix to Win32 is H no harder or easier than porting from Unix to VMS.  When a new chunk of B Unix-developed code comes down the pipe, the amount of stuff that C breaks on Win32 is approximately equal to the amount of stuff that 2H breaks on VMS.  It's just that there are *lots* more people involved in B anything related to Win32.  There is an entire company devoted to D porting and supporting Perl on Win32 (though ActiveState does other @ things now).  There are extremely adept engineers whose day job F consists of making sure that free software runs properly on Win32 and < developing associated add-on products and support services.   G There is no question that porting is hard for all but the most trivial HH bits of software and (Star|Open)Office is not trivial.  Supposedly OVMS F Engineering had a look at StarOffice at some point and decided it was B just too much code to consider porting (I forget how many million H lines).  But I think that was before the Mozilla port and GNV and other ? Unixy infrastructure activities, so it should be somewhat more wG manageable now.  There's no question it can be done if someone devotes t the resources to it.   ------------------------------   End of INFO-VAX 2002.315 ************************