1 INFO-VAX	Sun, 23 Feb 2003	Volume 2003 : Issue 105       Contents:P Re: =?ISO-8859-1?Q?Official_OpenVMS_name_for_the_?= =?ISO-8859-1?Q?Itanium=AE_reP =?iso-8859-1?Q?RE:_Official_OpenVMS_name_for_the_Itanium=AE_releases__and?=	=?is Re: Acrobat Viewer and OpenVMS( Re: Alphastation error (but running VMS) Dumping LAT  Re: Dumping LAT  Re: Dumping LAT  Re: Dumping LAT  Encompass Membership Re: EVA Courses ?  FA: DEC Wallhanging  Re: Googlism for OpenVMS Re: Googlism for OpenVMS Re: Hobbyist CD not recognized Re: Hobbyist CD not recognized Hobbyist CD not recognized( Re: How do I join the domain with Samba?P Re: HSxx shadowing/striping vis a vis hardware RAID - was RE: Volume set  acrossP Re: HSxx shadowing/striping vis a vis hardware RAID - was RE: Volume set across P Re: HSxx shadowing/striping vis a vis hardware RAID - was RE: Volume set across P Re: HSxx shadowing/striping vis a vis hardware RAID - was RE: Volumeset across R4 IBM Prophecy was (Re: IBM says AMD dead in 5yrs ...)H Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyH Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyH Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyH Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyH Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyH Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyH Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyH Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyD Re: Idea for installing VMS without a 512byte CD-ROM...Will it work?D Re: Idea for installing VMS without a 512byte CD-ROM...Will it work?D Re: Idea for installing VMS without a 512byte CD-ROM...Will it work?D Re: Idea for installing VMS without a 512byte CD-ROM...Will it work?" Re: Madison, successor to McKinleyB Re: memory errors and bugcheck dump (PROCGONE) with ALPHA 3000/600B Re: Migrating from PSI X-25 to another product (hardware/software)! Re: Official OpenVMS name for the P Re: Official OpenVMS name for the =?ISO-8859-1?Q?Itanium=AE?= releases	and  prodP Re: Official OpenVMS name for the =?iso-8859-1?Q?Itanium=AE?= releases	and  prodP Re: Official OpenVMS name for the =?iso-8859-1?Q?Itanium=AE?= releases  and  pro OpenVMS Hobbyist SPL - Update 4 Sun Prophecy was (Re: Sun and Opteron rumours again)L TCPIP, LAN, pseudointerface, SET ROUTE, SET HOST, ISDN, DSL: puzzles galore!+ TECO (was: Re: Upcoming VMS improvements ?)  Re: The End of OpenVMS Re: The End of OpenVMS Re: The End of OpenVMS Re: The End of OpenVMS Re: The End of OpenVMS Re: The End of OpenVMS Re: The End of OpenVMS Re: The End of OpenVMS Re: The End of OpenVMS Re: The End of OpenVMS Re: The End of OpenVMS Re: The End of OpenVMS Re: The End of OpenVMS Re: Upcoming VMS improvements ?  Re: Upcoming VMS improvements ?  Re: Upcoming VMS improvements ?  Re: Upcoming VMS improvements ?  Re: Upcoming VMS improvements ?   Re: VMS & Tru64 Disk I/O timings$ Re: VMSclusters vs. others' clusters$ Re: VMSclusters vs. others' clusters$ Re: VMSclusters vs. others' clusters$ Re: VMSclusters vs. others' clusters$ Re: VMSclusters vs. others' clusters> Re: What makes the VMS port to Itanium/IPF/whatever difficult? [DFG] Some questions  F ----------------------------------------------------------------------  + Date: Sun, 23 Feb 2003 05:49:32 +0000 (UTC) - From: bdc@world.std.com (Brian 'Jarai' Chase) Y Subject: Re: =?ISO-8859-1?Q?Official_OpenVMS_name_for_the_?= =?ISO-8859-1?Q?Itanium=AE_re ( Message-ID: <b39ndc$fao$1@pcls4.std.com>  = In article <857e9e41.0302211053.4c43b5c8@posting.google.com>, 2 Sue Skonetski <susan_skonetski@hotmail.com> wrote:  F > hp OpenVMS Industry Standard 64 conveys the true strategic directionG > for OpenVMS as a viable, robust, and evolving operating system on the * > industry standard 64-bit platform. [...]#   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^   + That claim is somewhat premature, isn't it?   G Apart from that, "hp OpenVMS Industry Standard 64" has to be the single H most awful official name for an operating system, ever.   HP may as wellH tape a sheet of paper on OpenVMS with the words "KICK ME" written on it.   -brian.  --  F --- Brian Chase | bdc@world.std.com | http://world.std.com/~bdc/ -----J Computers are useless.  They can only give you answers.  -- Pablo Picasso.   ------------------------------  % Date: Sat, 22 Feb 2003 12:36:55 -0800 # From: "Tom Linden" <tom@kednos.com> Y Subject: =?iso-8859-1?Q?RE:_Official_OpenVMS_name_for_the_Itanium=AE_releases__and?=	=?is 9 Message-ID: <CIEJLCMNHNNDLLOOGNJIMEJLGKAA.tom@kednos.com>    >-----Original Message----- 5 >From: Larry Kilgallen [mailto:Kilgallen@SpamCop.net] * >Sent: Saturday, February 22, 2003 4:23 AM >To: Info-VAX@Mvb.Saic.ComA >Subject: Re: Official OpenVMS name for the Itanium releases and 	 >products  >  > A >In article <22FEB200305273668@gerg.tamu.edu>, carl@gerg.tamu.edu  >(Carl Perkins) writes:  > L >> The problem is that everybody except a few unusually stupiud analysts andK >> various marketing people (and the people who picked this name) know that   >> it isn't "industry standard". > F >Unfortunately, your desire for a meaningful name has been outvoted byG >previous posters to the newsgroup who all said they want more coverage  >for VMS in the press. > G >Ad hominem attacks on marketing decisions are standard for this group, F >and ad hominem attacks on the press run a close second, but marketingG >is aimed at the press and a name like "industry standard" is what will G >appeal to the attention of a lot of people who never log into VMS (but   >have budget signoff authority). > C >The new name also works well with Alpha-huggers, who can decry the H >new "industry standard" (IPF) variant and point to technical advantagesF >of Alpha.  So long as those advantages are perceived, that is a validG >argument to use VMS on Alpha for some people.  But management at other E >companies will want the "industry standard" variant of VMS, which is  >also fine.  > G >Furthermore, the naming situation can lead to a lot of articles in the J >press about just what _is_ the difference between Alpha and IPF versions.D >Remember that the goal is to get coverage and thereby get exposure.B >Striving for technically meaningful trade press articles would be? >a separate campaign _not_ aimed at broadening interest in VMS.  > E >The question that should stick in the minds of the great unwashed is % >"Which flavor of VMS do you want ?".  > F Euphemisms are a useful marketing tool.  I imagine pre-owned BMWs sellD better than used ones, and OpenVMS is more appealing then VMS  (justI out of curiosity, was there anything differnet about VMS or license terms E from OpenVMS?)  I guess, however, it is fair to use the term Industry  Standard6 as long as you don't have to explain it too carefully.     --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.449 / Virus Database: 251 - Release Date: 1/27/2003   ------------------------------  % Date: Sat, 22 Feb 2003 10:19:50 +0100  From: Dirk Munk <munk@home.nl>' Subject: Re: Acrobat Viewer and OpenVMS 3 Message-ID: <RfH5a.27019$I6.4214213@zwoll1.home.nl>   P The Acrobat Java Viewer needs something extra to run on later Java versions. It F seems that small piece of software is part of a Mac Java distribution.  T Look in Google in the archive of this newsgroup, and you will find more information.     Rudolf Wingert wrote:  > Hello, > G > I did try to install the Acrobat Viewer under OpenVMS V7.3-1 and Java K > V1.3.1-5 and V1.4.0-1 without any success. If I did use the Fast VMS, the N > InstallAnyWhere window raise and the install hangs. If I did use the classicN > VM I did get an "execption in main" after the InstallAnyWhere window raised,M > then this window closed and the Acrobat Viewer install window is displayed. J > Then the installation hangs too. A call to TCSC Mnchen hsowed, that theI > Acrobat Viewer function only with Java V1.1.8-5. Does anybody know news 3 > about the Acrobat Viewer and newer Java versions?  >   > TIA and regards Rudolf Wingert >    ------------------------------  # Date: Sat, 22 Feb 2003 22:36:21 GMT - From: "John E. Malmberg" <wb8tyw@qsl.network> 1 Subject: Re: Alphastation error (but running VMS) = Message-ID: <FXS5a.13461$jR3.7015805@news1.news.adelphia.net>    Dave Pampreen wrote:N > A few months ago one of my 2 alphastation 250's stopped functioning.  I haveM > an error code that from the LED's on the back but it does not correspond to  > anything in the user guide.  >  > Anyone have any idea?  > L > The show the following, they switch back and forth (Looking at it from the > back, from left to right)  > First:= > 1111 1100  Console initialized (Final display if using ARC) 	 > Second:  > 1111 1011  ???????? # > And lastly then go back to first. ! > 1111 1001 Memory test 2 failure  > 3 > I've also swapped out the memory with no success.  >  > Any ideas??? > E > This machine boots VMS 7.2-1 and has the latest firmware installed.  >  > Dave( > e-mail dave.pampreen@us.add.gknplc.com  A I do not know, but you can check the OpenVMS FAQ to see if it is  C mentioned.  I also crossposted this to comp.os.vms as there may be  + someone there that knows what the LEDs are.   I The OpenVMS FAQ is available from a link at http://www.openvms.compaq.com    -John  wb8tyw@qsl.network Personal Opinion Only    ------------------------------  # Date: Sat, 22 Feb 2003 19:28:17 GMT # From: "John N." <JNixon@cfl.rr.com>  Subject: Dumping LAT< Message-ID: <lbQ5a.19610$Cv4.371124@twister.tampabay.rr.com>  L Our network guys are somewhat Unix and Windows literate, but have (nor want)J much VMS knowledge.  They want to get rid of  LAT and go to all IP.  TheseI are the same guys that took our many segmented LAN and "simplified" it by H conbining everything on one Lan segment, thereby sacrificing redundancy.K Now, if the one network segment glitches, my sattelite nodes CLUEXIT.  They  have agreed to fix this.  J But now I have to justify keeping LAT.  Other than saying "better trackingI of connection sources", or "better support for reverse operation ,such as E printing", I need some real arguments.  The word "better" is somewhat  subjective.   K Actually, I guess I should ask whether you guys think I should keep LAT, or K am I a dinosaur.  Should we migrate to an all-IP network?  We primarily use L LAT for incoming dialup connections.  They make a request for a service thatI is offered by several VMS systems.  We also do a fair amount of printing, ( but we also do some IP printing as well.  L Opinions are invited of course, but directions to appropriate literature may give my words more  validity.    Thanks.    ------------------------------  % Date: Sat, 22 Feb 2003 22:09:34 +0100 6 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= <arne@vajhoej.dk> Subject: Re: Dumping LAT) Message-ID: <3E57E70E.9090404@vajhoej.dk>    John N. wrote:N > Our network guys are somewhat Unix and Windows literate, but have (nor want)E > much VMS knowledge.  They want to get rid of  LAT and go to all IP.   L > But now I have to justify keeping LAT.  Other than saying "better trackingK > of connection sources", or "better support for reverse operation ,such as G > printing", I need some real arguments.  The word "better" is somewhat 
 > subjective.  > M > Actually, I guess I should ask whether you guys think I should keep LAT, or ; > am I a dinosaur.  Should we migrate to an all-IP network?   1 LAT is a very efficient protocol for some things.   , But I would not consider it that critical to replace it with IP.   5 Both terminal-servers and printers work fine with IP.   1 And by going IP you can also get through routers.    Arne   ------------------------------  % Date: Sat, 22 Feb 2003 16:02:34 -0600 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>  Subject: Re: Dumping LAT' Message-ID: <3E57F37A.5B4F07BB@fsi.net>    "John N." wrote: > N > Our network guys are somewhat Unix and Windows literate, but have (nor want)L > much VMS knowledge.  They want to get rid of  LAT and go to all IP.  TheseK > are the same guys that took our many segmented LAN and "simplified" it by J > conbining everything on one Lan segment, thereby sacrificing redundancy.M > Now, if the one network segment glitches, my sattelite nodes CLUEXIT.  They  > have agreed to fix this. > ) > But now I have to justify keeping LAT.    A Well, since they support your cluster, they are likely doing some D bridging of the associated packet types, or just bridging period, in: which case they'd have to intentionally block LAT packets.  F So long as they must bridge to support your cluster nodes, there is no+ justification for selectively blocking LAT.   D That said, the applications for LAT are getting rather, as they say,B "long in the tooth" (never understood that reference) and the gearE needed to support them is aging and becoming difficult to replace. If A for no other reason than supportability, perhaps IP would be more 
 advantageous.   @ If there's some LAT "application" that you absolutely cannot run. without, that'd be justification enough, IMHO.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sat, 22 Feb 2003 19:14:23 -0500 ' From: Howard S Shubs <howard@shubs.net>  Subject: Re: Dumping LAT< Message-ID: <howard-B66C57.19142322022003@enews.newsguy.com>  ' In article <3E57F37A.5B4F07BB@fsi.net>, 3  "David J. Dachtera" <djesys.nospam@fsi.net> wrote:   F > That said, the applications for LAT are getting rather, as they say,D > "long in the tooth" (never understood that reference) and the gear  E Rodents' teeth grow throughout their lives.  If they don't wear them  I down, they just get longer.  An old critter might not be able to chew on  ? enough stuff to wear down its teeth, so be "long in the tooth"?    --  4 Today, on Paper-view: The World Origami Championship   ------------------------------    Date: 22 Feb 2003 22:41:18 -0800/ From: robl@no-no-badpuppy.com (Robert Lawrence)  Subject: Encompass Membership = Message-ID: <eebf2e9b.0302222241.5f7078c6@posting.google.com>   = How long should it take for Encompass to approve an Associate A Membership?  I applied last week, but I have not heard from them.    Thanks.    ------------------------------  % Date: Sat, 22 Feb 2003 13:17:06 +0000 ( From: Nic Clews <sendspamhere@127.0.0.1> Subject: Re: EVA Courses ?) Message-ID: <3E577852.35E6C5A7@127.0.0.1>    Craig Cooke wrote: > - > The company I work for has recently aquired  >  > Two HSV110 Controllers. > 1 x Compaq SAN works Management Appliance II >  > 1 x SAN Switch 2/16-EL  @ Did they put them where the public trade counter used to be? ;-)  N > Is anyone aware of any course being run by any company that would help us to9 > configure / commission this EVA equipment effectively ?   H Of course, contact the supplier! The courses do have an OpenVMS bias but? cover all operating systems. It could almost make NT admins VMS $ converts, but that is another story.   --  ? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences  nclews at csc dot com    ------------------------------  % Date: Sat, 22 Feb 2003 15:00:49 -0500 3 From: "Homer J. Simpson" <hsimpson@burnsenergy.com>  Subject: FA: DEC Wallhanging: Message-ID: <DxQ5a.19623$HV.12361@fe07.atl2.webusenet.com>  1 Spruce up the office with this beautiful artwork! L http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=20135&item=3502868777   ------------------------------  % Date: Sat, 22 Feb 2003 15:29:06 -0600 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> ! Subject: Re: Googlism for OpenVMS ' Message-ID: <3E57EBA2.B3013A1D@fsi.net>    Baby Peanut wrote: > 6 > http://www.googlism.com/index.htm?ism=openvms&type=1  F Gives a blank window in Netscape 4, but reading the page source yields the secrets.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------    Date: 22 Feb 2003 15:41:54 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) ! Subject: Re: Googlism for OpenVMS 3 Message-ID: <BdYfin1AKcHS@eisner.encompasserve.org>   [ In article <3E57EBA2.B3013A1D@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:  > Baby Peanut wrote: >>  7 >> http://www.googlism.com/index.htm?ism=openvms&type=1  > H > Gives a blank window in Netscape 4, but reading the page source yields > the secrets.  D The secret it gives to me is that www.googlism.com is not interested1 in communicating to people with secured browsers.   G But from the original post, I gather that Baby Peanut is not interested G in communicating within this newsgroup; merely posting raw URLs without  comment is quite bad form.   ------------------------------  # Date: Sun, 23 Feb 2003 00:22:27 GMT . From: peter@langstoeger.at (Peter LANGSTOEGER)' Subject: Re: Hobbyist CD not recognized 3 Message-ID: <7vU5a.50204$Rb4.692051@news.chello.at>   a In article <eIS5a.305024$sV3.9859551@news3.calgary.shaw.ca>, Steve Young <steve@well.com> writes: O >  The CD drive works (tested it on another box), the CD is at least physically J >OK (I was able to make an image of it on a machine that doesn't talk ODS C >without media errors), so I am left wondering why this might be.     J You could install the ODS2 freeware (and the required ADAPTEC ACPI Driver)H on your WINTEL box and try to read your VMS CD there. At least it serves3 as a basic test where to look for the next error...    --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  # Date: Sun, 23 Feb 2003 01:43:10 GMT " From: Steve Young <steve@well.com>' Subject: Re: Hobbyist CD not recognized < Message-ID: <OGV5a.306082$sV3.9887655@news3.calgary.shaw.ca>  J   I know it's bad form to follow up on my own message, but I've discoveredE some additional information... if I put an audio CD in the drive, it  F shows up on show dev as an x mb CD, and I can use it to play audio CDsB under VMS.  When I try and mount a data CD under VMS though, I getB media offline.  It seems that just data cds are causing a problem.  G   I have parity turned off; 512 byte blocks turned on; I've tried using L both the termination the device comes with (by turning the terminator toggleI at the back to on), and using the external terminator that came with the   VAX.  	   Thanks,    Steve.   ------------------------------  # Date: Sat, 22 Feb 2003 22:19:54 GMT " From: Steve Young <steve@well.com># Subject: Hobbyist CD not recognized < Message-ID: <eIS5a.305024$sV3.9859551@news3.calgary.shaw.ca>     Hello everyone,   E   I have finally managed to get a CD-ROM drive that is recognized by  I my VAX without problems; a Pioneer DRM-604x 6-cd changer.  This has been  > confirmed by other members of this group as working fine with  VAXstation4000/VLCs.  L   In any case, I have the CD in slot 0, yet when I do a show dev, it doesn't
 recognize it:   P   DKA500       A/5/0     RODISK      ......       RM      WP    CD-ROM      2403  N   The CD drive works (tested it on another box), the CD is at least physicallyI OK (I was able to make an image of it on a machine that doesn't talk ODS  H without media errors), so I am left wondering why this might be.  Do anyI of you have suggestions? For completeness I have included the full output : of show config and show dev at the bottom of this message.  	   Thanks,    Steve.   >>> show config    KA48-A V1.4-386-V4.1 08-00-2B-BB-4E-8A  24MB   DEVNBR    DEVNAM           INFO 0 ------    --------    --------------------------      1         NVR    OK      2         LCG    OK+                        HR - 8 PLN FB - V1.2       3          DZ    OK      4       CACHE    OK      5         MEM    OK9                       24MB = S0/1=8MB, S2/3=8MB, S4/5=8MB       6         FPU    OK      7          IT    OK      8         SYS    OK      9          NI    OK     10        SCSI    OKP                       3-L0-RZ26L   5-L0-CD-ROM  5-L1-CD-ROM  5-L2-CD-ROM  5-L3-C( D-ROM  5-L4-CD-ROM  5-L5-CD-ROM  6-INITR     11         AUD    OK   >>> show dev  O   VMS/VMB      ADDR      DEVTYPE    NUMBYTES     RM/FX    WP    DEVNAM      REV O   -------      ----      -------    --------     -----    --    ------      ---     ESA0         08-00-2B-BB-4E-8AP   DKA300       A/3/0     DISK         1.05GB      FX            RZ26L       442DP   DKA500       A/5/0     RODISK      ......       RM      WP    CD-ROM      2403P   DKA501       A/5/1     RODISK      ......       RM      WP    CD-ROM      2403P   DKA502       A/5/2     RODISK      ......       RM      WP    CD-ROM      2403P   DKA503       A/5/3     RODISK      ......       RM      WP    CD-ROM      2403P   DKA504       A/5/4     RODISK      ......       RM      WP    CD-ROM      2403P   DKA505       A/5/5     RODISK      ......       RM      WP    CD-ROM      2403  ..HostID..    A/6       INITR   ------------------------------  # Date: Sat, 22 Feb 2003 21:02:25 GMT - From: "John E. Malmberg" <wb8tyw@qsl.network> 1 Subject: Re: How do I join the domain with Samba? = Message-ID: <BzR5a.13417$jR3.6995409@news1.news.adelphia.net>   	 jm wrote: @ > "John E. Malmberg" <wb8tyw@qsl.network> wrote in message news:6 > <B%Y4a.11820$jR3.6081369@news1.news.adelphia.net>... >  >  >>>[2003/02/19 09:24:04, 0] 6 >>>CMS_ROOT:[SAMBA.SOURCE.PASSDB]SMBPASSFILE.C;2:(287)H >>>  domain_client_validate: unable to open the machine account password
 >>>file for m ' >>>achine mymachine in domain mydomain. H >>>[2003/02/19 09:24:04, 0] CMS_ROOT:[SAMBA.SOURCE.LIB]UTIL_SEC.C;2:(52)B >>>  Failed to set uid privileges to (-1,3735552) now set to (0,0) >>>[2003/02/19 09:24:04, 0] 1 >>>CMS_ROOT:[SAMBA_VMS.SOURCE.LIB]UTIL.C;1:(2456)  >>>  PANIC: failed to set uid  >>J >>The panic is likely because it attempted to find a UAF account to match I >>the machine account.  There must be an account existing in the UAF for  ( >>the password lookup API's to function. >>K >>Also SAMBA seems to also test that the default directory for the account  G >>exists, and that the account has access to it before setting the UID.  >>A >>SAMBA also requires that the UIC's in the rightslist.dat match   >>sysuaf.dat usernames.  > E > Can you explain or point me to the file that describes what you are 5 > talking about?  What UAF account is it looking for.   F It is looking for one with the machine name in it.  All usernames are A looked up in the SYSUAF file or it's equivalent on UNIX before a  I password check is done.  At the time that 2.0.6 was current, the joining  F of a domain with a machine account was a new and experimental feature.  H >>>if you look, you will see that it is looking for a file in the formatG >>>xxx.xxx.xxx and I don't think OpenVMS supports that.  In the private F >>>directory, there is a file call mydomain__2Emymachine.mac, that was5 >>>created with the command "smbpasswd -j mydomain -r . >>>my_windows_2000_machine" domain controller. >>E >>The FRONTPORT library converts files from UNIX formats to an ODS-2  H >>format, as long as the resulting file ends up being in a 39.39 format.5 >>The first "." in the filename is converted to __2E.  >>- >>>The machine.sid is also not being created.  >>@ >>I am surprised that the machine.sid file is not being created. >   A > Actually, all that it is doing now is staying in RWAST forever.   F Is the SMBD in RWAST or NMBD in RWAST.  There was a bug in the TCP/IP F stack and CRTL that could produce this.  Make sure that you are up to H date on your ECOs if you can not upgrade.  I was only able to reproduce 8 this with a NT system that had a broken browser service.  B >>>Was I not supposed to join the domain this way?  How do I do it2 >>>properly?  Any help is appreciated.  Thank you. >>> ! >>>OpenVMS 7.2-1 and Samba 2.0.6.  >>J >>You may want to look into the 2.2.x port.  Search the SAMBA-VMS mailing K >>list archives for information on the download site.  It is a dynamic DNS  H >>location and is not always available.  So if it is not reachable, try K >>again a little later.  I have not done anything with the 2.2.x port, but  & >>it seems to be getting a lot of use. >   H Also be aware that SAMBA 2.0.7 and later is needed to properly interact 7 with newer versions of the Microsoft operating systems.    -John  wb8tyw@qsl.network Personal Opinion Only    ------------------------------  # Date: Sun, 23 Feb 2003 02:46:19 GMT + From: Ryan Moore <rmoore@rmoore.dyndns.org> Y Subject: Re: HSxx shadowing/striping vis a vis hardware RAID - was RE: Volume set  across = Message-ID: <Pine.LNX.4.50.0302221835330.26692-100000@jaipur>   J There's a bigger disadvantge than this.  All of your disks are in one rack$ connected to one set of controllers.  I Who's to say there won't be some power glitch that fries both controllers  or disks in the same rack?  C What if there were a software bug in the controllers such that both I controllers crash because of a cache corruption or something?  Or what if G one of the controllers locks the SCSI bus and dies?  What if a hardware $ failure occurs and it locks the bus?  I I know we've had a case where two HSJ redundant controllers pooped out on E us at once.  And others on this list have experienced the same thing.   : HBVS will save you from this type of catastrophic failure.  B In our case, our racks are right next to one another (connected toH separate power distribution units).  But in a better scenario, they'd be< separated in different buildings connected by fibre channel.  - On Sat, 22 Feb 2003, David J. Dachtera wrote: E > In *MY* mind, there are two major advantages to using RAID-1 in the  > HSx's instead of HBVS: >  >[....]  > F > The single disadvantage is that there is no supported way to split aH > mirror-set from within a batch (BACKUP) job. There once was with HSZs:   ------------------------------  % Date: Sat, 22 Feb 2003 18:39:17 -0500   From: John Santos <JOHN@egh.com>Y Subject: Re: HSxx shadowing/striping vis a vis hardware RAID - was RE: Volume set across  6 Message-ID: <1030222182611.26253A-100000@Ives.egh.com>  - On Sat, 22 Feb 2003, David J. Dachtera wrote:    > Daniel Allen wrote:  > >   > > > -----Original Message-----< > > > From: David J. Dachtera [mailto:djesys.nospam@fsi.net]. > > > Sent: Friday, February 21, 2003 10:47 PM > > > To: Info-VAX@Mvb.Saic.Com * > > > Subject: Re: Volume set across RAID5 > > >  > > >  > > > "henry g." wrote:  > > > >  > > > > Thanks David.  > > > > L > > > > This was a little complex, most of the engineers I spoke with didn'tI > > > > know the answer for certain... we all speculated but yours is the = > > > > first with actual hands-on experience in this matter.  > > > >  > > > > Thanks again!  > > >  > > > Any time.  > > > D > > > ...but that said, consider: HBVS *IS* RAID, RAID-1 to be exactN > > > (shadowing/mirroring). Shadow-sets can be bound into volume-sets. ShouldF > > > anything happen to a shadow-set (like lose all the members), the > > > volume-set goes *PIFT*.  > > > H > > > That's why I prefer using mirrored stripe-sets that are mounted asM > > > single-volume disk devices. Stripe for capacity, mirror for redundancy. G > > > Watch your FAILEDSET and SPARESET, and set your mirror-set member ( > > > replacement policy as appropriate. > >  > >         David, > > T > >         Your comment caught my eye and led me to ask myself "how does VMS (HSxx)R > >         volume shadowing and striping compare (performance/price/feature) withL > >         current hardware commodity RAID offerings?". I haven't had a VMSP > >         cluster with HS disks in a LONG time so I'm not current with today'sT > >         state of the art. If I understand your comment, you are suggesting usingR > >         HS based striping (RAID 0) and mirroring (RAID 1) to get big redundantQ > >         volumes. What is the advantage of that config to a commodity RAID box S > >         setup for RAID 0/1?. I've been sort of under the impression that modern J > >         HS controllers where basically cluster aware RAID controllers. > , > Well, that's as fair an assessment as any. > H > Disk striping for OpenVMS is rather out of vogue, but there's probably% > sites that still use it, somewhere.  > E > Disk shadowing for OpenVMS is still widely used, as this forum will 	 > attest.  > I > On HSx's (HSJ, HSZ, HSG), RAID usually refers to RAID-5. Otherwise, you F > as the SysAdmin elect whether to mirror stripe-sets or shadow (HBVS) > them.  > E > In *MY* mind, there are two major advantages to using RAID-1 in the  > HSx's instead of HBVS: > I > o Eliminates the shadow-set merge problem when a node exits the cluster C > ungracefully or fails to properly DISMOUNT a shadow-set during an  > orderly shutdown.  > H > o Eliminates the host I/O load associated with shadow-copy and -merge.E > For example, on a three-node Alpha 8400 cluster, shadow-copy of 8GB H > volumes was taking between five and eight hours based on what else wasG > happening in the cluster. Merges could runs for weeks. Mirror-copy on H > the HSJs took just short of 52 minutes fairly consistently when you do- > them one at a time per dual-redundant pair.  > F > The single disadvantage is that there is no supported way to split aH > mirror-set from within a batch (BACKUP) job. There once was with HSZs:G > HSZPAD$SCSITERM, the program behind the SET HOST/SCSI command. It can I > still be found on some older SPLs (then known as "Condist"'s) and rogue I > web sites (like mine), but is no longer supported (still works for HSGs G > AFAIK, we use it today). There are ways to do it on HSD, HSJ, HSZ and F > HSG, but they are all currently unsupported, and not without certainF > risks that can lead to downtime, but AFAIK, the risk of data loss isE > very small. Generally, stick to the HSx firmware doc. and you'll be  > fine.  > . > ...but, that's just me. YMMV, as they say...  B There is a 2nd, serious disadvantage to controller-based shadowing' (mirroring), if I understand correctly:   @ The controllers can only mirror disks that are local to them, so; all the mirror members are in the same box, which becomes a ? single point of failure.  (I realize you can use dual redundant B pairs of HSx's, and spread the mirrored disks across multiple SCSI= buses, and use dual power supplies fed from separate AC power B sources.)  But everything is in one box, if for example, something? catches fire or someone tips the box over into a missing raised @ floor tile[*], that's all she wrote.  (Of course, you can always= HBVS the mirror sets, but why not just HBVS the raw disks, or   make RAID-5 sets and HBVS them?)  = One of my customers is contemplating upgrading from two pairs < of HSJ40's w/shadowing to 2 pairs of HSZ70's with mirroring.> (Also moving from VAX to Alpha, and I guess they don't want to@ pay for Alpha shadowing licenses; this is a low-budget operationC and there getting all the equipment used or from internal surplus.) : I think they would be much better off sticking with volume2 shadowing, but don't know for sure.  Any comments?  > [*] This happened once at another site, to a rack full of Fuji; Eagles on an 11/70...  RSTS/E didn't have shadowing, but we 9 were doing our own homebrew shadowing in the app...  Lost 7 both copies, of course.  Good thing we had backups, and   transaction logs on other disks.   --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Sat, 22 Feb 2003 15:51:08 -0600 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> Y Subject: Re: HSxx shadowing/striping vis a vis hardware RAID - was RE: Volume set across  ' Message-ID: <3E57F0CC.59AA8D46@fsi.net>    Daniel Allen wrote:  >  > > -----Original Message-----: > > From: David J. Dachtera [mailto:djesys.nospam@fsi.net], > > Sent: Friday, February 21, 2003 10:47 PM > > To: Info-VAX@Mvb.Saic.Com ( > > Subject: Re: Volume set across RAID5 > >  > >  > > "henry g." wrote:  > > >  > > > Thanks David.  > > > J > > > This was a little complex, most of the engineers I spoke with didn'tG > > > know the answer for certain... we all speculated but yours is the ; > > > first with actual hands-on experience in this matter.  > > >  > > > Thanks again!  > > 
 > > Any time.  > > B > > ...but that said, consider: HBVS *IS* RAID, RAID-1 to be exactL > > (shadowing/mirroring). Shadow-sets can be bound into volume-sets. ShouldD > > anything happen to a shadow-set (like lose all the members), the > > volume-set goes *PIFT*.  > >aF > > That's why I prefer using mirrored stripe-sets that are mounted asK > > single-volume disk devices. Stripe for capacity, mirror for redundancy.hE > > Watch your FAILEDSET and SPARESET, and set your mirror-set member & > > replacement policy as appropriate. >  >         David, > R >         Your comment caught my eye and led me to ask myself "how does VMS (HSxx)P >         volume shadowing and striping compare (performance/price/feature) withJ >         current hardware commodity RAID offerings?". I haven't had a VMSN >         cluster with HS disks in a LONG time so I'm not current with today'sR >         state of the art. If I understand your comment, you are suggesting usingP >         HS based striping (RAID 0) and mirroring (RAID 1) to get big redundantO >         volumes. What is the advantage of that config to a commodity RAID boxaQ >         setup for RAID 0/1?. I've been sort of under the impression that modern H >         HS controllers where basically cluster aware RAID controllers.  * Well, that's as fair an assessment as any.  F Disk striping for OpenVMS is rather out of vogue, but there's probably# sites that still use it, somewhere.c  C Disk shadowing for OpenVMS is still widely used, as this forum willM attest.M  G On HSx's (HSJ, HSZ, HSG), RAID usually refers to RAID-5. Otherwise, younD as the SysAdmin elect whether to mirror stripe-sets or shadow (HBVS) them..  C In *MY* mind, there are two major advantages to using RAID-1 in theW HSx's instead of HBVS:  G o Eliminates the shadow-set merge problem when a node exits the clustersA ungracefully or fails to properly DISMOUNT a shadow-set during ans orderly shutdown.y  F o Eliminates the host I/O load associated with shadow-copy and -merge.C For example, on a three-node Alpha 8400 cluster, shadow-copy of 8GBfF volumes was taking between five and eight hours based on what else wasE happening in the cluster. Merges could runs for weeks. Mirror-copy onhF the HSJs took just short of 52 minutes fairly consistently when you do+ them one at a time per dual-redundant pair.s  D The single disadvantage is that there is no supported way to split aF mirror-set from within a batch (BACKUP) job. There once was with HSZs:E HSZPAD$SCSITERM, the program behind the SET HOST/SCSI command. It cannG still be found on some older SPLs (then known as "Condist"'s) and rogueeG web sites (like mine), but is no longer supported (still works for HSGsVE AFAIK, we use it today). There are ways to do it on HSD, HSJ, HSZ andVD HSG, but they are all currently unsupported, and not without certainD risks that can lead to downtime, but AFAIK, the risk of data loss isC very small. Generally, stick to the HSx firmware doc. and you'll beo fine.S  , ...but, that's just me. YMMV, as they say...   --   David J. DachteraO dba DJE Systemsc http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/e   ------------------------------  % Date: Sat, 22 Feb 2003 22:33:39 -0600e1 From: "David J. Dachtera" <djesys.nospam@fsi.net>eY Subject: Re: HSxx shadowing/striping vis a vis hardware RAID - was RE: Volumeset across R-' Message-ID: <3E584F23.FA8FF60A@fsi.net>0   Ryan Moore wrote:C > L > There's a bigger disadvantge than this.  All of your disks are in one rack& > connected to one set of controllers.   John and Ryan,  C In my experience, the likelihood of such a failure is low enough to0) justify mitigating the negatives of HBVS.    As I said, YMMV...  G Also, do not the newer controller support "remote copy" so a mirror setr, can be propagated to second controller pair?   -- b David J. Dachterap dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/i   ------------------------------  + Date: Sat, 22 Feb 2003 15:16:35 -0800 (PST)d. From: Fabio Cardoso <fabiopenvms@yahoo.com.br>= Subject: IBM Prophecy was (Re: IBM says AMD dead in 5yrs ...)e@ Message-ID: <20030222231635.51975.qmail@web20205.mail.yahoo.com>  C In my personal opinion the future of the IBM and AMD agreement willc= help in the Sun's aquisition by IBM. THE NEXT BIG FUSION !!!!aI As Sun wil begin to use Opterons, it opens an oportunity for acquisition.2N And IBM sold the Networking and HDs factories to open a easy way to do it....    RegardsE   FC i  , --- CBFalconer <cbfalconer@yahoo.com> wrote: > JF Mezei wrote:D > >  > > CBFalconer wrote:AF > > > I have been trying to come up with a firm, past or present, that > > > was simultaneously:p > > >  > > >   1. Profitable.& > > >   2. Considered a good employer.( > > >   3. Appreciated by its customers.- > > >   4. Appreciated by the general public. ) > > >   5. Not a closed family operation.a > > H > > The large pharmaceutical firms are generally extremely well regarded > > in all of the above. > ; > Not item 4.  Consider their general efforts to circumventtC > generics, and their pricing.  Compare drug cost in Canada and then > US.  >  > > ) > > IBM would probably also fit the bill.  > = > They used to be classified as Microsoft is now, but 1 and 2i@ > certainly applied.  Then fouling up 1 caused them to dump many= > employees, so 2 is gone.  Note my original qualification of  > 'simultaneously'.k >  > -- e> > Chuck F (cbfalconer@yahoo.com) (cbfalconer@worldnet.att.net)= >    Available for consulting/temporary embedded and systems.a< >    <http://cbfalconer.home.att.net>  USE worldnet address! >  >      =====t ========================== Fbio dos Santos Cardoso OpenVMS System Manager Rio de Janeiro - Brazild fabiopenvms@yahoo.com.br ==========================  2 __________________________________________________ Do you Yahoo!?2 Yahoo! Tax Center - forms, calculators, tips, more http://taxes.yahoo.com/c   ------------------------------  % Date: Sat, 22 Feb 2003 03:20:55 -0400 0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>Q Subject: Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolye. Message-ID: <3E572425.8667B58@vl.videotron.ca>   CBFalconer wrote:eB > I have been trying to come up with a firm, past or present, that > was simultaneously:  >  >   1. Profitable." >   2. Considered a good employer.$ >   3. Appreciated by its customers.) >   4. Appreciated by the general public.i% >   5. Not a closed family operation."  N The large pharmaceutical firms are generally extremely well regarded in all of
 the above.  % IBM would probably also fit the bill.e   ------------------------------  # Date: Sat, 22 Feb 2003 13:45:00 GMT ) From: Peter Flass <peter_flass@yahoo.com>wQ Subject: Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolye) Message-ID: <3E5754CA.AD09012A@yahoo.com>g   "J. Clarke" wrote:H > > I have been trying to come up with a firm, past or present, that was > > simultaneously:D > >s > >   1. Profitable.$ > >   2. Considered a good employer.& > >   3. Appreciated by its customers.B > >   4. Appreciated by the general public. 5. Not a closed family > >   operation. > >-J > > and (until the fire) the only one I could come up with was that fabricJ > > manufacturer in Lynn, Massachusetts.  Can't remember the name offhand.J > > Now it is missing qualification 1, but with any luck that will change. > . > Malden Mills.  But are they publicly traded? > E > There was a time when Delta Airlines met all of the above criteria.t   Southwest does now, AFAICT.    ------------------------------  + Date: Sat, 22 Feb 2003 15:16:13 +0000 (UTC)a8 From: hawk@slytherin.ds.psu.edu (Dr. Richard E. Hawkins)Q Subject: Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopoly - Message-ID: <b3847t$r88$4@f04n12.cac.psu.edu>d  ( In article <1bsmuhw4xm.fsf@cs.nmsu.edu>,+ Joe Pfeiffer  <pfeiffer@cs.nmsu.edu> wrote:a; >hawk@slytherin.ds.psu.edu (Dr. Richard E. Hawkins) writes:i  J >> It wasn't worth a new engine, but a client took it and did a crummy jobK >> on the rebuild (less than 1k miles!).  Then another mechanic needed some J >> work, and got it going--but warned me that it could fail any second, orK >> last for years.  A sound I had always took to be one of those many, manyrF >> clutches was actually the rods rattling (the entire time I had it!) >> under load.  G >What do you mean by "many, many clutches?"  You must be using the termtG >differently than I do; to me, a clutch is what goes between the enginec >and a manual transmission...    In a normal car, yes :)i  A This thing had several of those, run automatically, to engage andyH disengate the AC and God knows what else.  Reading the manual, every few= pages it would explain how another automatic clutch was used.f  H OTOH, properly tuned, that 3.8L V6 gave the vehicle a smoother ride than' the Mercedes 380SL of the same vintage.l   hawk -- 'K Richard E. Hawkins, Asst. Prof. of Economics    /"\   ASCII ribbon campaignhG dochawk@psu.edu  Smeal 178  (814) 375-4700      \ /   against HTML mail D These opinions will not be those of              X    and postings. 6 Penn State until it pays my retainer.           / \      ------------------------------  + Date: Sat, 22 Feb 2003 15:21:31 +0000 (UTC)k8 From: hawk@slytherin.ds.psu.edu (Dr. Richard E. Hawkins)Q Subject: Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyl- Message-ID: <b384hr$r88$7@f04n12.cac.psu.edu>-  ) In article <3E56F642.E4EF765F@yahoo.com>,m0 CBFalconer  <cbfalconer@worldnet.att.net> wrote:  >"Dr. Richard E. Hawkins" wrote:  B >> Someday . . . I will, uhh, liberate the ancient phone my fatherC >> has, and connect it to a small AI for a synthetic operator . . .   @ >In N.America at least one can normally expect pulse dialling toC >work.  However the phones with a magneto and a crank don't seem tov >be able to raise the operator.e  F That's what the AI is for :)  I tell it who I want, and then it uses aG modem to dial.  I'm assuming the phone actually connects to D/A and A/DeE converters rather than the phone systeThat's what the AI is for :)  ItG tell it who I want, and then it uses a modem to dial.  I'm assuming thegG phone actually connects to D/A and A/D converters rather than the phonee system.u  C >Not wishing to plow through 28 layers of menus, I always pretend I B >have a dial phone when accessing any automated attendant system. $ >Speaker phones are also handy here.   Oh, yes.   hawk -- eK Richard E. Hawkins, Asst. Prof. of Economics    /"\   ASCII ribbon campaigniG dochawk@psu.edu  Smeal 178  (814) 375-4700      \ /   against HTML mailoD These opinions will not be those of              X    and postings. 6 Penn State until it pays my retainer.           / \      ------------------------------  + Date: Sat, 22 Feb 2003 16:54:11 +0000 (UTC)98 From: hawk@slytherin.ds.psu.edu (Dr. Richard E. Hawkins)Q Subject: Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopoly - Message-ID: <b389vj$r8k$6@f04n12.cac.psu.edu>t  ) In article <3E578CFD.1D906891@yahoo.com>, 0 CBFalconer  <cbfalconer@worldnet.att.net> wrote: >JF Mezei wrote:  G >> The large pharmaceutical firms are generally extremely well regardedr >> in all of the above.f  : >Not item 4.  Consider their general efforts to circumventB >generics, and their pricing.  Compare drug cost in Canada and the >US.  D Huh?  Their pricing reflects the R&D costs involved.  If we paid theD same prices in the U.S.  for the drugs that the Canadian freeloaders- [1] do, they plain and simply wouldn't exist.u  D When there's a single price for all comers, it has to meet or exceedG *both* the average cost of production (including R&D), and the marginal F cost of production.  The U.S. price does both; the Canadians and otherF freeloaders only cover the second (and it makes economic sense for theG producer to sell to such thieves when they cover marginal costs, but itBC really amounts to a subsidy of Canadian buyers by the U.S. buyers).w   hawk  E [1] ok, the technical term is "free rider," but this is more accuratea --  K Richard E. Hawkins, Asst. Prof. of Economics    /"\   ASCII ribbon campaignaG dochawk@psu.edu  Smeal 178  (814) 375-4700      \ /   against HTML mailaD These opinions will not be those of              X    and postings. 6 Penn State until it pays my retainer.           / \      ------------------------------  % Date: Sat, 22 Feb 2003 14:53:33 -0400 0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>Q Subject: Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopoly / Message-ID: <3E57C72B.438658F0@vl.videotron.ca>    CBFalconer wrote:n; > Not item 4.  Consider their general efforts to circumvent C > generics, and their pricing.  Compare drug cost in Canada and thep > US.n  L But this is relatively recent. Pharmaceuticals used to be highly regarded asH quality companies that treated employees well, had nice "campuses", good@ cafeterias, loyal employees etc. (And a steady maker of profits)    = > They used to be classified as Microsoft is now, but 1 and 2A@ > certainly applied.  Then fouling up 1 caused them to dump many= > employees, so 2 is gone.  Note my original qualification ofJ > 'simultaneously'.w  @ But isn't IBM back to profitable, stable and good to employees ?   ------------------------------  # Date: Sun, 23 Feb 2003 02:49:14 GMT ' From: CBFalconer <cbfalconer@yahoo.com>SQ Subject: Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopoly2) Message-ID: <3E583141.626D56ED@yahoo.com><   "Dr. Richard E. Hawkins" wrote:i2 > CBFalconer  <cbfalconer@worldnet.att.net> wrote: > >JF Mezei wrote: > @ > >> The large pharmaceutical firms are generally extremely well" > >> regarded in all of the above. > < > >Not item 4.  Consider their general efforts to circumventD > >generics, and their pricing.  Compare drug cost in Canada and the > >US. > F > Huh?  Their pricing reflects the R&D costs involved.  If we paid theF > same prices in the U.S.  for the drugs that the Canadian freeloaders/ > [1] do, they plain and simply wouldn't exist.2  < Compare their R&D budgets and their advertising and lobbying< budgets.  The only god is Green and his name is the Almighty Dollah.5   -- @< Chuck F (cbfalconer@yahoo.com) (cbfalconer@worldnet.att.net);    Available for consulting/temporary embedded and systems.u:    <http://cbfalconer.home.att.net>  USE worldnet address!   ------------------------------  % Date: Sat, 22 Feb 2003 16:34:52 -0400 0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>Q Subject: Re: IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM monopolyp/ Message-ID: <3E57DEE8.CB2FC60C@vl.videotron.ca>S   "Dr. Richard E. Hawkins" wrote:wH > cost of production.  The U.S. price does both; the Canadians and otherH > freeloaders only cover the second (and it makes economic sense for theI > producer to sell to such thieves when they cover marginal costs, but it E > really amounts to a subsidy of Canadian buyers by the U.S. buyers).s  F Considering that the major pharmaceutical firsm all have major R&D andD production facilities in Canada which were heavily subsidized by the; governments in canada, your accusation is not well founded.S  M Fact is that the pharmaceuticals in the USA are taking advantage of the USA's L health care structure, knowing exactly just how much the insurance companiesH are capable of paying. Note that this applies also to the many companiesN headquartered outside the USA such as Bayer (germany) , Novartis (switzerland)L etc. Isn't Merck the one major remaining pharmaceutical that is based in the USA ?    ------------------------------  % Date: Sat, 22 Feb 2003 15:25:26 -0600r1 From: "David J. Dachtera" <djesys.nospam@fsi.net>uM Subject: Re: Idea for installing VMS without a 512byte CD-ROM...Will it work?a' Message-ID: <3E57EAC6.20510A4F@fsi.net>s   Hans Vlems wrote:p > [snip]: > Is the original VAX/VMS CD a bootable distribution then?   Typically, yes.h   -- t David J. Dachtera. dba DJE SystemsJ http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho//   ------------------------------    Date: 22 Feb 2003 15:30:19 -0800/ From: robl@no-no-badpuppy.com (Robert Lawrence)lM Subject: Re: Idea for installing VMS without a 512byte CD-ROM...Will it work?e< Message-ID: <eebf2e9b.0302221530.8980135@posting.google.com>  ` "Hans Vlems" <hvlems@iae.nl> wrote in message news:<b37t5f$1iln8i$1@ID-143435.news.dfncis.de>...@ > "Robert Lawrence" <robl@no-no-badpuppy.com> schreef in bericht9 > news:eebf2e9b.0302212235.21d31b41@posting.google.com...eG > > I have a friend that gave me a VAXstation 3100/M30, which should beeI > > delivered Monday. This machine has a hard drive expansion box, but noiG > > CD-ROM.  This machine also is missing an OS.  I tossed around a fewgE > > ideas and came up with the following method of intalling OpenVMS.iE > > (Keep in mind that my only exposure to VAX & VMS was as a user at3* > > university, so I am learning as I go.) > > F > > * Install Linux on a PC, with a SCSI host adpater and an extra 1GB > > SCSI drive.5H > > * From Linux do a: dd if=/dev/cdrom of=/dev/sda1  (copy VMS media to > > raw device)u? > > * Remove the drive, set the SCSI ID to that of a VAX CD-ROM J > > * Install SCSI hard drive in VAXstation and direct it to boot from the
 > > new drivea > > * Continue with VMS installt > >mF > > Will this work?  If I dump the CD contents byte for byte to a SCSIG > > hard drive on a PC, will the VAX be able to read and boot from that 
 > > drive? > : > Is the original VAX/VMS CD a bootable distribution then?  > From my understanding of the install guide, the hobbyist CD is! bootable (to stand-alone backup).y   ------------------------------    Date: 22 Feb 2003 19:53:45 -0800/ From: robl@no-no-badpuppy.com (Robert Lawrence) M Subject: Re: Idea for installing VMS without a 512byte CD-ROM...Will it work? = Message-ID: <eebf2e9b.0302221953.552f87dc@posting.google.com>g  ` "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message news:<3E57EB29.6F24C5AB@fsi.net>... > Robert Lawrence wrote: > > G > > I have a friend that gave me a VAXstation 3100/M30, which should beeI > > delivered Monday. This machine has a hard drive expansion box, but noaG > > CD-ROM.  This machine also is missing an OS.  I tossed around a fewgE > > ideas and came up with the following method of intalling OpenVMS. E > > (Keep in mind that my only exposure to VAX & VMS was as a user atn* > > university, so I am learning as I go.) > > F > > * Install Linux on a PC, with a SCSI host adpater and an extra 1GB > > SCSI drive.tH > > * From Linux do a: dd if=/dev/cdrom of=/dev/sda1  (copy VMS media to > > raw device)a? > > * Remove the drive, set the SCSI ID to that of a VAX CD-ROM J > > * Install SCSI hard drive in VAXstation and direct it to boot from the
 > > new driveh > > * Continue with VMS install  > > F > > Will this work?  If I dump the CD contents byte for byte to a SCSIG > > hard drive on a PC, will the VAX be able to read and boot from thatr
 > > drive? >  > Crapshoot, at best.o > J > This would be better: use another VMS machine and BACKUP/IMAGE the CD to@ > a hard disc, then stick the hard disc in the 3100 and boot it.  F Unfortunately, I only have an emulator (simh) and the 3100.  I have noB other VAX /VMS machine at my disposal. I am a few months away from/ having a large cluster in my home computer lab.i  A I had a SCSI CD-ROM drive, but it doesn't seem to have survived 7m years in storage.   C My VMS skills are rather weak and rusty.  I have never administeredeC VMS before, so this is all quite new to me.  I am learning as I go.    ------------------------------  % Date: Sat, 22 Feb 2003 15:27:05 -0600-1 From: "David J. Dachtera" <djesys.nospam@fsi.net>NM Subject: Re: Idea for installing VMS without a 512byte CD-ROM...Will it work? ' Message-ID: <3E57EB29.6F24C5AB@fsi.net>    Robert Lawrence wrote: > E > I have a friend that gave me a VAXstation 3100/M30, which should be G > delivered Monday. This machine has a hard drive expansion box, but no/E > CD-ROM.  This machine also is missing an OS.  I tossed around a few C > ideas and came up with the following method of intalling OpenVMS. C > (Keep in mind that my only exposure to VAX & VMS was as a user at ( > university, so I am learning as I go.) > D > * Install Linux on a PC, with a SCSI host adpater and an extra 1GB
 > SCSI drive.UF > * From Linux do a: dd if=/dev/cdrom of=/dev/sda1  (copy VMS media to
 > raw device)-= > * Remove the drive, set the SCSI ID to that of a VAX CD-ROMBH > * Install SCSI hard drive in VAXstation and direct it to boot from the > new driveD > * Continue with VMS installS > D > Will this work?  If I dump the CD contents byte for byte to a SCSIE > hard drive on a PC, will the VAX be able to read and boot from that  > drive?   Crapshoot, at best.   H This would be better: use another VMS machine and BACKUP/IMAGE the CD to> a hard disc, then stick the hard disc in the 3100 and boot it.   -- S David J. Dachtera  dba DJE SystemsR http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/s   ------------------------------  % Date: Sat, 22 Feb 2003 06:19:15 -0500 * From: "Bill Todd" <billtodd@metrocast.net>+ Subject: Re: Madison, successor to McKinleyd2 Message-ID: <9tmcnWpIKrUlwcqjXTWcoQ@metrocast.net>  4 "David Svensson" <icerq4a@spray.se> wrote in message7 news:734da31c.0302220145.751882b5@posting.google.com...s9 > "Bill Todd" <billtodd@metrocast.net> wrote in message >nA > > Well, sure.  Hell, if McKinley had been willing to settle forU SPARC-levelB> > > performance, it too could have consumed SPARC-level power. > E > You just can't put power into a processor and think it will perform E > better, it has to run too. A Sparc with 130W would not work at all._G > Power is a problem, but it is also impressive that a CPU actually cano > run at 130W!  L Since this discussion is about lower rather than higher power, the point you# think you're making is not obvious.a   >sF > > So when Deerfield rolls around, it'll perform a bit worse than the currentnH > > high-end McKinley (same clock speed, but Deerfield has only half the cache)K > > and use about half the power.  Unfortunately, as a lower-end entrant ity willL > > be competing with P4s, Xeons, Hammers, and the lower-end POWER4+ entriesK > > that will use about the same amount of power and run rings around it inh > > performance. >n% > Why would they run rings around it?m  G Because they already run at least half-rings around McKinley today, anduD tomorrow they will be faster while the 62W Deerfield will be slower.  A > Intel also said that a 1.5Ghz Deerfield will be made this year.v  B BFD:  this has no relevance to the 62W part under discussion here.   >a. > Those lower-end POWER4+ will also be slower.  H Since the current POWER4+ uses about the same power as the 62W DeerfieldK while out-performing McKinley, then either it will out-perform Deerfield bytG an even wider margin at its current power (a reasonable bet, I suspect,lK since the higher-end versions will be increasing their clock speed over theaK course of this year) or, if it is indeed slower than Deerfield, it will useh" a *lot* less power than Deerfield.    I have said this before, A > but the Power4 gains very much with it's off-chip large caches.f  E Fortunately, your assertion is subject to at least some verification.eL Unfortunately, doing so does not validate your assertion all that well, saveI possibly for SPECfp (which is certainly of interest to HPTC installationswK but which doesn't matter very much to the kinds of installations that bringr in the bulk of the revenue).  I A 1 GHz POWER4 (630) gets a SPECint2K_base score of 624.  1.1 GHz POWER4s I (655/670) get scores of 700/680.  1.3 GHz POWER4s (655/690) get scores of G 822, 804, and 790.  And the 1.45 GHz POWER4+ (650) gets a score of 909.o  L That's quite close to linear scaling with clock speed, which wouldn't be allJ that surprising except that the 1 GHz and 1.45 GHz systems have only 32 MBH of L3 off-chip cache, while the systems in the middle have four times asK much - so at least in that range the larger L3 seems to have no significanta effect whatsoever.  < > Without that large off-chip cache they are not impressive.  K Unless you can provide test results that back up that assertion, I'm afraidgF it falls into the realm of speculation.  My own impression is that useI off-chip cache is fairly central to the SMP architecture of the POWER4/4+eK platforms, and that since they need it anyway for that purpose they figured-L they might as well use a good amount even if it didn't significantly improveK performance (and the fact that they cut back the amount by a factor of 4 inrK POWER4+ lends credence to that impression).  You should also note that it's0H not fast SRAM (like EV68's) but eDRAM:  great for bulk, but a low slowerL (same philosophy PA-RISC will use in the 8800), so it may not much more thanD make up for the fact that POWER4/4+ only have about 1/4 as much fast. *on*-chip cache per processor as Itanic2 does.   - bill   ------------------------------  % Date: Sat, 22 Feb 2003 13:13:52 +0000c( From: Nic Clews <sendspamhere@127.0.0.1>K Subject: Re: memory errors and bugcheck dump (PROCGONE) with ALPHA 3000/600 ) Message-ID: <3E577790.F521C609@127.0.0.1>e   Robert Deininger wrote:e > L > In article <01KSPBQQUVTW9GVT85@sysdev.deutsche-boerse.com>, Phillip Helbig- > <HELBPHI@sysdev.deutsche-boerse.com> wrote:f > J > >> Are the memory test errors correctable or uncorrectable?  Once VMS is3 > >> up, you will ride right over corrected errors.t > >n( > >Once VMS is up, I notice no problems. > I > Correctable errors still result in error log entries.  There's no otherb$ > visible indication that I know of.  F True, the important thing to check is the date and time of occurrence.  > On some VAX systems, you can get the cache disabled, and it isA transparent (apart from a lack of speed) and reducing the ambientn, temperature is sometimes all that is needed.   -- -? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences0 nclews at csc dot com    ------------------------------  % Date: Sat, 22 Feb 2003 10:14:46 +0100D From: Dirk Munk <munk@home.nl>K Subject: Re: Migrating from PSI X-25 to another product (hardware/software),3 Message-ID: <6bH5a.26995$I6.4210496@zwoll1.home.nl>   Q The cheapest way is to buy a Alphaserver DS10 with a Sync card and X.25 software.n  Q It has all your present functionality, can work as a X.25 gateway or just on its bR own, whatever you like. We are going to replace our DECnis X.25 gateways this way.     Fabio Cardoso wrote: > Dear SirsD > A > We have at the company an old VAX 6520 with PSI X.25 installed.eD > The problem is, there is just one group of users - library - which2 > connect to other libraries worldwide using X.25. > > > Is there a way to  substitute a whole VAX + PSI by a router J > - I am imagining any DNPG hardware. The acess must be inbound x outbound >  > 	 > RegardsD >  > FC a >  >  > =====i > ========================== > Fbio dos Santos Cardoso > OpenVMS System Manager > Rio de Janeiro - Brazilo > fabiopenvms@yahoo.com.br > ========================== > 4 > __________________________________________________ > Do you Yahoo!?4 > Yahoo! Tax Center - forms, calculators, tips, more > http://taxes.yahoo.com/y   ------------------------------  # Date: Sat, 22 Feb 2003 15:53:59 GMTe# From: "John Smith" <a@nonymous.com>p* Subject: Re: Official OpenVMS name for theH Message-ID: <r2N5a.46807$Zr%.42487@news01.bloor.is.net.cable.rogers.com>  4 "Carl Perkins" <carl@gerg.tamu.edu> wrote in message' news:22FEB200304563857@gerg.tamu.edu... 7 > susan_skonetski@hotmail.com (Sue Skonetski) writes...n > }Official Name@ > }The following is the official name for OpenVMS on the ItaniumC > }platform. This is the full name of the product and to be used int any 1 > }title, header, and first mention in body text.  > } $ > }"hp OpenVMS Industry Standard 64" >n: > Once again confirming something everybody already knows: > C > The people who decide these things are complete and utter morons.s    C But as we all know, the 'hp' and 'Industry Standard 64' are silent.e   ------------------------------  % Date: Sat, 22 Feb 2003 17:30:19 -0500s( From: David Froble <davef@tsoft-inc.com>Y Subject: Re: Official OpenVMS name for the =?ISO-8859-1?Q?Itanium=AE?= releases	and  prod , Message-ID: <3E57F9FB.6060402@tsoft-inc.com>   John Smith wrote:r  < > "Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in message/ > news:3anfSArGYi7O@eisner.encompasserve.org...s > B >>In article <22FEB200305273668@gerg.tamu.edu>, carl@gerg.tamu.edu >> > (Carl Perkins) writes: > ? >>>The problem is that everybody except a few unusually stupiud  >>>  > analysts and > A >>>various marketing people (and the people who picked this name)y >>>a > know thatl >   >>>it isn't "industry standard". >>>lD >>Unfortunately, your desire for a meaningful name has been outvoted >> > by > ? >>previous posters to the newsgroup who all said they want more- >>
 > coverage >  >>for VMS in the press.v >>A >>Ad hominem attacks on marketing decisions are standard for thise >> > group, > = >>and ad hominem attacks on the press run a close second, butt >> > marketinge > C >>is aimed at the press and a name like "industry standard" is whath >> > will > C >>appeal to the attention of a lot of people who never log into VMS  >> > (but > ! >>have budget signoff authority).Z >>D >>The new name also works well with Alpha-huggers, who can decry the> >>new "industry standard" (IPF) variant and point to technical >> > advantages > A >>of Alpha.  So long as those advantages are perceived, that is a  >> > valid  > B >>argument to use VMS on Alpha for some people.  But management at >> > other  > F >>companies will want the "industry standard" variant of VMS, which is >>also fine. >>D >>Furthermore, the naming situation can lead to a lot of articles in >> > the  > A >>press about just what _is_ the difference between Alpha and IPFa >> > versions.a > E >>Remember that the goal is to get coverage and thereby get exposure. C >>Striving for technically meaningful trade press articles would bea@ >>a separate campaign _not_ aimed at broadening interest in VMS. >>F >>The question that should stick in the minds of the great unwashed is& >>"Which flavor of VMS do you want ?". >> > > > Which makes the case for keeping it simple....OpenVMS/Alpha,D > OpenVMS/IPF, OpenVMS/Opteron, OpenVMS/Power, OpenVMS/Sparc, etc...    O Ok, I doubt that I'll get many to disagree when I say that the new name sucks. i It's stupid, and such.  M Frankly, I don't care what they call it, as long as I get the product.  They  N don't even have to call it VMS, as long as the product is VMS.  The name will D not affect robustness, reliability, scalability, security, and such.  P If those at HP think that they'll get some mileage from the name, then fine, go L for it.  For the users of VMS, wouldn't it be better to stick to meaningful 9 issues than to waste bandwith arguing over a stupid name?i  Q That said, I have to observe that sticking the term 'industry standard' with the w9 albatrose that is IA-64 is a disservice to the term.  :-)p   Dave   ------------------------------  % Date: Sat, 22 Feb 2003 14:55:04 -0400i0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>Y Subject: Re: Official OpenVMS name for the =?iso-8859-1?Q?Itanium=AE?= releases	and  prod / Message-ID: <3E57C786.11EDA1B5@vl.videotron.ca>o   warren sander wrote: > L > The hp is lower case because of HP's rules about using hp in product names > etc.   I have no problem with that.  J > etc.. there about about 75 or so rules including for html you always use" > capital HP in the <title> tags..    N I feel for you.... if there are 75 rules that apply to just 2 letters, imagineL the number of rules that apply to the whole web site :-) :-) :-) :-) :-) :-)   ------------------------------    Date: 22 Feb 2003 06:22:43 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen):Y Subject: Re: Official OpenVMS name for the =?iso-8859-1?Q?Itanium=AE?= releases  and  pro 3 Message-ID: <3anfSArGYi7O@eisner.encompasserve.org>e  W In article <22FEB200305273668@gerg.tamu.edu>, carl@gerg.tamu.edu (Carl Perkins) writes:u  K > The problem is that everybody except a few unusually stupiud analysts and J > various marketing people (and the people who picked this name) know that > it isn't "industry standard".   E Unfortunately, your desire for a meaningful name has been outvoted by-F previous posters to the newsgroup who all said they want more coverage for VMS in the press.s  F Ad hominem attacks on marketing decisions are standard for this group,E and ad hominem attacks on the press run a close second, but marketingtF is aimed at the press and a name like "industry standard" is what willF appeal to the attention of a lot of people who never log into VMS (but have budget signoff authority).o  B The new name also works well with Alpha-huggers, who can decry theG new "industry standard" (IPF) variant and point to technical advantagesuE of Alpha.  So long as those advantages are perceived, that is a validnF argument to use VMS on Alpha for some people.  But management at otherD companies will want the "industry standard" variant of VMS, which is
 also fine.  F Furthermore, the naming situation can lead to a lot of articles in theI press about just what _is_ the difference between Alpha and IPF versions. C Remember that the goal is to get coverage and thereby get exposure.IA Striving for technically meaningful trade press articles would ben> a separate campaign _not_ aimed at broadening interest in VMS.  D The question that should stick in the minds of the great unwashed is$ "Which flavor of VMS do you want ?".   ------------------------------  % Date: Sat, 22 Feb 2003 17:14:27 -0600t1 From: "David J. Dachtera" <djesys.nospam@fsi.net>n& Subject: OpenVMS Hobbyist SPL - Update' Message-ID: <3E580453.2512EC65@fsi.net>s   Folks,  F I've been able to develop some DCL procedures that use the freeware LDH software to build CD images of the layered products licensed through the OpenVMS Hobbyist program.a   The kit can be found at:0 http://www.djesys.com/freeware/vms/hbyspl010.zip  2 Be sure to see the README.TXT file in the archive.  5 You'll find the pre-requisite LD software there also:h, http://www.djesys.com/freeware/vms/ld062.zip  > ...or there's a newer version at the OpenVMS Freeware CD site:A http://www.openvms.compaq.com/freeware/freeware50/ld063/ld063.zipn   P.S.: E I have not yet heard back from OpenVMS Management about producing ando, distributing a Hobbyist's SPL. Stay tuned...   -- m David J. Dachterad dba DJE Systemsr http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/b   ------------------------------  + Date: Sat, 22 Feb 2003 15:19:12 -0800 (PST) . From: Fabio Cardoso <fabiopenvms@yahoo.com.br>= Subject: Sun Prophecy was (Re: Sun and Opteron rumours again)e@ Message-ID: <20030222231912.41628.qmail@web20207.mail.yahoo.com>  4 As I said in another note the use of AMD processors < by IBM and Sun will help  in the Sun's acquisition by IBM ! < IBM loves  java too ! Dont forget ! Sun dont have a standard? database as IBM DB2. Sun have the most sold Unix in the market.k, Sun Solaris and IBM AIX will become Linux !    Regardst   FC s) --- Shane Smith <ssmith@icius.com> wrote:i* > http://www.theinquirer.org/?article=7894 > J > "INFORMED SOURCES called Tara tell us that when AMD launches its OpteronH > 64-bit processor on the 22nd of April, Sun Microsystems will certainly. > be introducing a product based on the chip." >  > OK, I'm intrigued. > @ >  #####   -----------------------------------------------------A > #-O-O-# | Arthur: "It's times like this I wish I'd listened   |eA > #  L  # | to what my Mother used to say." Ford: Why, what did |rA >  #===#  | she say?" Arthur: "I don't know, I didn't listen."  |o@ >   ###    ----------------------------------------------------- >      =====t ========================== Fbio dos Santos Cardoso OpenVMS System Manager Rio de Janeiro - Brazile fabiopenvms@yahoo.com.br ==========================  2 __________________________________________________ Do you Yahoo!?2 Yahoo! Tax Center - forms, calculators, tips, more http://taxes.yahoo.com/    ------------------------------    Date: 22 Feb 2003 17:51:34 -0800) From: helbig@multivax.de (Phillip Helbig)dU Subject: TCPIP, LAN, pseudointerface, SET ROUTE, SET HOST, ISDN, DSL: puzzles galore!y= Message-ID: <f019af2c.0302221751.7785376e@posting.google.com>i  A OK folks, the competition is open!  Prove you're a REAL VMS TCPIPeH network expert by explaining all the following puzzles!  There are a lotG of puzzling things here, some related, some unrelated.  With luck, I'lliD have all the answers by the end of the week!  (Note that I have beenH experimenting all day and am only including puzzling things which a) are1 reproducible and b) I really need to understand.)   H At location A I have a VMS machine with a 192.168.1.X address.  I have aF DSL connection.  All works fine.  In particular, I have set up the DSLG router to pass all incoming connections to the VMS machine.  When I log G in somewhere else via the DSL connection, I can log back in through the. DSL router.   F In short, all works as expected.  (Not that I am completely happy with> it---the default way to configure it is via a web browser, butH unfortunately it uses JavaScript and I don't have any browser on the VMSG machine which will operate properly with it (not sure on which side theBB problem is).  Fortunately, there is a COMPLETELY UNDOCUMENTED MENUH INTERFACE which can be accessed via telnet.  (One option here is to exit to a CLI!))e  G At location B, I have a VMS machine with a fixed routable IP address onoB a /29 subnet (i.e. I have 8 addresses, 5 usable since the network,H broadcast and address of the ISDN router are already taken).  The routerF does dial-OUT on demand and the ISP provides me with dial-IN on demandD (and also runs the DNS for my domain as well as providing a firewallF which filters out most unwanted connections before he calls my router,B thus saving me money).  This setup works fine and is actually fastD enough, at least most of the time.  However, a DSL flat rate is muchE cheaper, faster as well, and will allow both ISDN channels to be used-B for other things (not that I need them both that often, though one; option is to dial in to the ISDN router myself---in specialMG circumstances---so that I can connect completely internet-free from farg away if need be).7  A I want to have both the ISDN and the DSL connection working for aoH transitional time at location B.  (When all is working, a large fraction1 of the traffic will be me logged in at B from A.)   G The first problem is the fact that there are two networks.  This I havelE solved with a pseudointerface.  I can now communicate between the VMSlC machine on one network and the router on another.  This works fine.   G The first problem is the fact that there are two networks.  This I have E solved with a pseudointerface.  I can now communicate between the VMShC machine on one network and the router on another.  This works fine.o  C The documentation is, I think, wrong here since it gives an exampleJH using a NAME when as far as I can tell one must use an ADDRESS.  This is from the documentation CD:  9 TCPIP> SET CONFIGURATION INTERFACE interface /HOST=host - 1 _TCPIP> /NETWORK_MASK=mask /BROADCAST_MASK=b_maskrE    For example, to specify the pseudo-interface FFA0 on host KESTREL, F        with network mask 255.255.0.0 and broadcast mask to 128.30.0.0,
        enter:t   TCPIP> SET NOINTERFACE FFA0   C TCPIP> SET INTERFACE FFA0 /HOST=KESTREL /NETWORK_MASK=255.255.0.0 -i" _TCPIP> /BROADCAST_MASK=128.30.0.0  H I think this will only work if, instead of KESTREL, KESTREL's IP address is used.  E Another interesting point is that for the pseudointerface the networkwG mask shows up as that of the real interface, as does the broadcast mask G (should it be called an address instead of a mask?), regardless of whataH I specify (sometimes a warning occurs, but the pseudointerface is alwaysD created with the "proper" values for these two parameters).  (I alsoD notice that if I delete an interface and then recreate it, I have toG exit from TCPIP before recreating it; otherwise I get a QIO error.  Bugs or feature?)  F OK, regardless of things I don't understand and possible errors in the1 documentation, the pseuodinterface seems to work:o   TCPIP> sh interfacemB                                                            PacketsO Interface   IP_Addr         Network mask          Receive          Send     MTU,  O  WE0        213.252.154.202 255.255.255.248         46898         48816    1500uO  WEA0       192.168.1.202   255.255.255.248         46898         48816    1500tO  LO0        127.0.0.1       255.0.0.0               33003         33003       0a TCPIP> sh conf interface    Interface: WE0aO    IP_Addr: 213.252.154.202   NETWRK: 255.255.255.248   BRDCST: 213.252.154.207     Interface: WEA0O    IP_Addr: 192.168.1.202     NETWRK: 255.255.255.248   BRDCST: 213.252.154.207a    Interface: LO0u?    IP_Addr: 127.0.0.1         NETWRK: 255.0.0.0         BRDCST:i   TCPIP> sh interface/ful wea0  Interface: WEA0O    IP_Addr: 192.168.1.202     NETWRK: 255.255.255.248   BRDCST: 213.252.154.207"E                        Ethernet_Addr: 00-00-F8-21-B5-72    MTU:  1500       Flags: None5                                   RECEIVE        SEND 5    Packets                          46937       48854e5      Errors                             0           0-)    Collisions:                          0   ! TCPIP> sh conf interface/ful wea0d    Interface: WEA0O    IP_Addr: 192.168.1.202     NETWRK: 255.255.255.248   BRDCST: 213.252.154.207m?     C_Addr:                 C_NETWRK:                 C_BRDCST:t  	    Flags:2    Receive buffer:            0r TCPIP>  7 Like I say, the router and VMS machine can communicate.r  H Setting up the router to forward incoming stuff to the IP address of theC PSEUDOINTERFACE, nothing works.  I then gave the router a "real" IP G address, but still had it forward stuff to the pseudointerface.  Didn't.G work either.  I then set it to forward stuff to the real address of thetG VMS machine, and that worked.  (This is no surprise, since now---exceptaH for the fact that A has 192.168.1.X, i.e. non-externally-routable, and BD has "real" IP addresses---the situation at A and B is the same.)  MyA goal, of course, was to have the DSL and ISDN connections workingdG simultaneously and independently, to ease the transition after droppingnF the ISDN services of the ISP in favour of DSL.  While having them workH on the same network might allow them to work simultaneously, this is notH independent enough.  I NEED the ISDN on the real network (obviously) and< NEED the DSL on the non-routable network.  Can this be done?  C Another interesting problem is this:  I can use the ISDN or the DSL8E connection (with the DSL router on the "real" network) and everythingtC seems to work.  However, when one of the connections is active, the H other one is lame.  For example, if I log in from location A to locationE C (could be anywhere), I can log back in (or, presumably, log in from.G location D, again could be anywhere) if I use the corresponding addresssB ("real" address for ISDN, dynamic router WAN address for DSL), butF cannot get back in via the other connection.  In other words, if I logF in to location C via DSL, I cannot get back in via my publicly visibleF fixed IP address, which has worked for years and magically works againF as soon as I start using the ISDN again; if I log in to location C viaD ISDN, I cannot log back in via the dynamic WAN IP address of the DSLH router, even if I open a connection from the router to the outside world (which works!) by hand.S  E $ TCPIP SHOW HOST for the VMS machine only mentions the real address,mD not the pseudointerface address.  I thought perhaps such an entry isH needed (obviously, it is not created when the pseudointerface is (yes, IE also do SET CONFIG INTERF, not just SET INTERF, and have even stoppedtF and restarted TCPIP (and, in a fit of desperation, even rebooted threeE times: automatically, by hand, and with a power cycle!)---no change).r However, I cannot enter it:l   But it SHOULD work:    TCPIP> help set host examplesy    5          3.TCPIP> SET HOST MOA /ADDRESS = 128.33.33.9,  8            Establishes a second IP address for host MOA.  $ This is what I want to do, but I get  / %TCPIP-E-HOSTERROR, cannot process host request-D -TCPIP-I-DUPHOST, duplicate entry in hosts database (TCPIP$HOST) for <nodename>,  ignored  F Is this example only for the case that the second IP address is a real% one, as opposed to a pseudointerface?a  F (Another thing I tries was changing the subnet mask of the router fromD 255.255.255.0 to 255.255.255.238, i.e. like that of the VMS machine.F This doesn't work, even if I change its address from, say, 192.168.1.2H to, say, 192.168.1.203.  It doesn't appear to be possible to specify the broadcast for the router.)  ) I've also tried adding a route like this:   >    AN    192.168.1.0/24                        213.252.154.202  
 Doesn't help.g  
 Any ideas?   ------------------------------  # Date: Sat, 22 Feb 2003 16:47:50 GMTe% From: Roger Ivie <rivie@ridgenet.net>a4 Subject: TECO (was: Re: Upcoming VMS improvements ?)3 Message-ID: <slrnb5eedu.dfg.rivie@Stench.no.domain>e  J In article <l3Ka4QqhabyT@eisner.encompasserve.org>, Larry Kilgallen wrote: > H > That is the problem -- the commands are different from normal EDT use. > H > This issue is resolved by upgrading your editor to TECO which uses the3 > same commands for line mode and full screen mode.c  $ Great. Now I've got milk in my nose.  G While we're on the subject, are we going to get a native IA64 TECO withe OpenVMS/I64?   -- 2
 Roger Ivie rivie@ridgenet.net   ------------------------------  % Date: Sat, 22 Feb 2003 12:41:03 +0000 ( From: Nic Clews <sendspamhere@127.0.0.1> Subject: Re: The End of OpenVMSx) Message-ID: <3E576FDF.9CB6AF64@127.0.0.1>b   JF Mezei wrote:e >  > Nic Clews wrote:J > > Speed isn't and hasn't been an issue for OpenVMS. Do you want it fast,> > > or reliable? That's the OpenVMS difference, data is safer. > N > Alpha's higher speed and "clusterability" (real clusters as well as WildfireT > style arrangements) compensated for VMS' higher overhead to ensure data integrity.  B Granted, but the clusterability is taken as read for Itanium, and,F assuming you're familiar with the release plans, includes the adoption2 of InfiniBand as a CI. Go read up about InfiniBand? (http://www.infinibandta.org/) THEN come back and make the sameu
 statement.  L > But with VMS moving to an unimpressive chip, its performance will truly be> > lower than everyone else and this may start to matter a lot.  G Again, you need to take a closer look at the IPF. Check the pointers on-F the HP website leading to the Intel areas, and read up what the chip'sC features are. I can only assume from your statement that you're notcB familiar with the technical features of the chip. If you are, then> justify  "unimpressive". Arguments about speed will not count.  J > You need low prices and lots of marketing to sell a product that doesn't@ > perform the fastest. VMS nas neither low prices nor marketing.  D Well we do have an agreement here. However you get what you pay for.F Don't expect a Rolls Royce if you pay for a Mini. However there is theF point that if a Roller is made from the same components as a Mini, theE differential should be based on the "optional extras" that the Roller'H gives you, but priced enough to attract the Mini buyers. On this subjectB I've had some discussions, and some answers, but I can't and won't
 comment here.e   Marketing, shmarketing.    --  ? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciencese nclews at csc dot comc   ------------------------------  % Date: Sat, 22 Feb 2003 15:08:05 -0400a0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca> Subject: Re: The End of OpenVMSn/ Message-ID: <3E57CA92.D422512C@vl.videotron.ca>t   Bill Gunshannon wrote:G > of the system or the OS was classified information.  If DEC/COMPAQ/HP F > really cared there is nothing stopping them from publishing numbers,H > customer names and types of systems being run on VMS.  They choose not > to and rest is history.A  J Let's *assume* just for 1.5 seconds that Compaq had wanted to push VMS. ItG inherited a dwindling customer base thanks to Bob GQ Palmer. PublishingbM numbers which, each year, go down doesn't go too well. Remember that it takes5F a while to turn a tide around and if you have a sizeable proportion ofL customers who have already decided to retire VMS "as time progresses", then,I for many years after Palmer, you will still continue to feel the effects.-  N And because Compaq/HP have decided  that VMS is only high end, it relegates itM to low volume, similar to Tandem. In such volume, any individual sale is verydI important, but every loss hurts a lot. This means that you get much widert variations in growth/shinkage.  L Consider the case of MCI. I remember one of the posters here who kept sayingL how MCI had tried to boot VMS off but they had not been able to find anotherM system capable of replacing MCI. You can't really leverage this for marketing L "we are such a bad vendor that our customers work very hard at trying to getH rid of us and our only hope is that our product *still* has features not available elsewhere."   I So, if you limit your marketing to pointing to existing customers who are D actually growing their VMS systems (not the hardware, but adding newH applications to it) (eg: a real commitment to VMS), then you have a much smaller pool of customers.  K Lets face it, most customers have been forced to evaluate their position oneK VMS with regards to availability of Linux and even NT due to the "apparant"lL lower costs, and in many many cases, new applicatiosn are going to those newK platforms instead of VMS with VMS relegated to "legacy" applications (which-J may still be more important, but the glitz is now on the newer platforms).   ------------------------------  % Date: Sat, 22 Feb 2003 14:58:40 -0600.1 From: "David J. Dachtera" <djesys.nospam@fsi.net>2 Subject: Re: The End of OpenVMSv' Message-ID: <3E57E480.855717F8@fsi.net>r   Nic Clews wrote: > [snip] > As the news will spreadeA > inside their organization, VMS gets marketing that no amount oft > mailshots could achieve.  G What about outside their organization and into the trade media where itn= might be seen/heard by enough folks to actually do some good?a    > Is it enough to turn the tide?  # "...if that light's under a bushel,O/ BBBRRR!!! It's lost something kind of crucial." = ...from the song, "City of God" from the musical, "Godspell".g   > Time will tell.    Indeed.i   -- a David J. Dachteras dba DJE Systemse http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  # Date: Sat, 22 Feb 2003 23:53:10 GMTX# From: "John Smith" <a@nonymous.com>v Subject: Re: The End of OpenVMSbH Message-ID: <G3U5a.48972$Zr%.15278@news01.bloor.is.net.cable.rogers.com>  < "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message  news:3E57E34F.7092B12@fsi.net... > Nic Clews wrote: > >d > > Bill Gunshannon wrote: > > > / > > > In article <3E577653.4D680BE4@127.0.0.1>, 8 > > >         Nic Clews <sendspamhere@127.0.0.1> writes: > > > > John Smith wrote:d > > F > > > > Commercial and military reasons. Those that could say the most mustC > > > > remain silent, it is the way of the world. The examples aree	 unseen soaF > > > > the marketing has to be based on something else. I guess folks	 here havevB > > > > made their opinions heard in the surveys announced here... > > >i@ > > > Hardly.  Most military procurements are public information	 availableiD > > > through a simple(?) FOI request.  Especially when you consider thatF > > > they usually start as a public RFP published in the CBD.  I have neverdA > > > been at a government site, including the pentagon where thet
 brand name= > > > of the system or the OS was classified information.  If 
 DEC/COMPAQ/HPeA > > > really cared there is nothing stopping them from publishing0 numbers,A > > > customer names and types of systems being run on VMS.  Theyo
 choose not > > > to and rest is history.o > >i@ > > Bill, I wish I could say more, but I'd lose my job *and* get
 prosecuted >o > For what?g >e> > > into the bargain. Heck, I'm even nervous saying that much. >  > For the love of spit, why? >e > > > > F > > > > However, yesterday I was at a project closure meeting for some upgrade E > > > > work, hardware and software. They were so happy with it, thatp
 some stuff > > ...  > > > >  > > >oD > > > And will HP make this success public, in glowing details, in aD > > > media that someone outside the VMS faithful will actually see? > > F > > I'll be discussing this with our ambassador. They offered to write< > > something for our internal glossies. There are also some
 commercial > > sensitivities, >tE > Such as... (Hint: state he nature of the "sensitivity" in question,t notM+ > that actual solution used to address it.)d >,7 > > so one hopes that the correct conclusions are drawnS( > > without having to state the obvious. >/) > If it need be stated, it's not obvious.p >M     David,  D Just think about discussing the details of any work you have done orB may ever do in the future at the NSA, CIA, OSWR, LLNL, etc...  HowF long do you think it would be before you had a rather long-ish expenseD paid vacation in Marion, Ill. in a cell between those of John Walker and Aldrich Ames?m  * Now think UK-equivalents to all the above.   Is it obvious now?   ------------------------------  % Date: Sat, 22 Feb 2003 13:08:35 +0000H( From: Nic Clews <sendspamhere@127.0.0.1> Subject: Re: The End of OpenVMSa) Message-ID: <3E577653.4D680BE4@127.0.0.1>    John Smith wrote:t > ...p@ > A simple premise really. Why isn't it done? Why keep all VMS'sH > fabulous features and strengths a secret from companies the world-over, > that could really benefit from them today?  C Commercial and military reasons. Those that could say the most musttE remain silent, it is the way of the world. The examples are unseen sosH the marketing has to be based on something else. I guess folks here have: made their opinions heard in the surveys announced here...  F However, yesterday I was at a project closure meeting for some upgradeH work, hardware and software. They were so happy with it, that some stuffF they tried to do under NT, they are moving BACK to the VMS server, andD are heralding this as a major success story. As the news will spread? inside their organization, VMS gets marketing that no amount ofyG mailshots could achieve. Is it enough to turn the tide? Time will tell.eE Did I get my head out the door after the meeting? Just about... But IoF also felt embarrassed, because I was just doing my job, part of a team effort.a   -- t? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences4 nclews at csc dot com    ------------------------------  # Date: Sat, 22 Feb 2003 16:16:58 GMTd# From: "John Smith" <a@nonymous.com>  Subject: Re: The End of OpenVMSeH Message-ID: <_nN5a.47057$Zr%.41757@news01.bloor.is.net.cable.rogers.com>  5 "Nic Clews" <sendspamhere@127.0.0.1> wrote in messaget# news:3E578947.F7848E2B@127.0.0.1...  > Bill Gunshannon wrote: > >s- > > In article <3E577653.4D680BE4@127.0.0.1>,s6 > >         Nic Clews <sendspamhere@127.0.0.1> writes: > > > John Smith wrote:  >hD > > > Commercial and military reasons. Those that could say the most mustA > > > remain silent, it is the way of the world. The examples arev	 unseen sorD > > > the marketing has to be based on something else. I guess folks	 here haves@ > > > made their opinions heard in the surveys announced here... > >d> > > Hardly.  Most military procurements are public information	 availableEB > > through a simple(?) FOI request.  Especially when you consider thatD > > they usually start as a public RFP published in the CBD.  I have nevernE > > been at a government site, including the pentagon where the brandq name; > > of the system or the OS was classified information.  Ifl
 DEC/COMPAQ/HPn? > > really cared there is nothing stopping them from publishingL numbers,F > > customer names and types of systems being run on VMS.  They choose nott > > to and rest is history.s >s> > Bill, I wish I could say more, but I'd lose my job *and* get
 prosecuted< > into the bargain. Heck, I'm even nervous saying that much. >e > > >hD > > > However, yesterday I was at a project closure meeting for some upgrade C > > > work, hardware and software. They were so happy with it, thath
 some stuff > ...o > > >O > >4B > > And will HP make this success public, in glowing details, in aB > > media that someone outside the VMS faithful will actually see? >gD > I'll be discussing this with our ambassador. They offered to writeE > something for our internal glossies. There are also some commercial D > sensitivities, so one hopes that the correct conclusions are drawnB > without having to state the obvious. This is a reflection of the > technologies, not the people.u     Nic,  B I understand the particulars of the situation you find yourself inE regarding this specific installation, and you *have* said all you can-F . But there are MANY other non-military, non-intelligence, non-'black'D major users of VMS out there in the world whose stories parallel the2 one you alluded to that can be reported in detail.   HP chooses not to.  @ And the few occasions when they do, it is done in such a low-keyF manner that there's no traction. The success stories are buried in theE VMS section of their web site rather than featured prominently on therE www.hp.com home page and published widely in print media. Next time ae@ big catastrophe happens (natural or otherwise) where VMS systemsB stayed up, there should the a gonzo effort to push a TV commercial. spreading the gospel out the door in 48 hours.   But it isn't going to happen.   B Imaging the Rolling Stones song 'Satisfaction' without the openingD 'hook' that grabs you by the b*lls and carries you through the whole= song. It's that kind of raw response that HP has to go for ins marketing and positioning VMS.  D Not necessarily 'in your face' but more like Superman (tm) - 'FasterE than a speeding bullet. More powerful than a locomotive. Able to leap0C tall buildings in a single bound.' Everyone in the English-speakinga> world that has been exposed to this and can almost immediatelyC identify the context. VMS needs the same sort of thing done for it.i/ All the value proposition arguments can follow.X    D Hey HP - *if* IA-64 is going to become widely adopted, then what are) the real differentiators between vendors?o  $ Hint - it isn't the logo on the box.  # It's the operating system, stupid!!a  ( So market the freakin' operating system.   ------------------------------  % Date: Sat, 22 Feb 2003 22:03:59 -0500h2 From: rdeininger@mindspring.com (Robert Deininger) Subject: Re: The End of OpenVMSiL Message-ID: <rdeininger-2202032204000001@user-uinj4sp.dialup.mindspring.com>  H In article <_nN5a.47057$Zr%.41757@news01.bloor.is.net.cable.rogers.com>,$ "John Smith" <a@nonymous.com> wrote:    C >I understand the particulars of the situation you find yourself inoF >regarding this specific installation, and you *have* said all you canG >. But there are MANY other non-military, non-intelligence, non-'black'_E >major users of VMS out there in the world whose stories parallel the,3 >one you alluded to that can be reported in detail.  >- >HP chooses not to.-  K HP can't use success stories from other companies without their permission.c   ------------------------------  % Date: Sat, 22 Feb 2003 22:29:10 -0600Z1 From: "David J. Dachtera" <djesys.nospam@fsi.net>C Subject: Re: The End of OpenVMS ' Message-ID: <3E584E16.D70E6810@fsi.net>g   John Smith wrote:u > [snip]F > Just think about discussing the details of any work you have done orD > may ever do in the future at the NSA, CIA, OSWR, LLNL, etc...  HowH > long do you think it would be before you had a rather long-ish expenseF > paid vacation in Marion, Ill. in a cell between those of John Walker > and Aldrich Ames?e > , > Now think UK-equivalents to all the above. >  > Is it obvious now?  D No, because you can express such without revealing enough details to breach confidences.u   -- a David J. Dachterat dba DJE Systems- http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/>   ------------------------------    Date: 22 Feb 2003 20:24:36 -08001 From: susan_skonetski@hotmail.com (Sue Skonetski)0 Subject: Re: The End of OpenVMSo= Message-ID: <857e9e41.0302222024.35b94106@posting.google.com>s  
 Greetings,A I am on vacation but just could not keep away.  Please excuse the0	 spelling.e  E Just a couple of things about the new name and the old, which you maye. or may not know.  This is just an explanation.  C I do not really know the entire process on how names are chosen andtE just knowing some of the work that went into this name I have come toiF the conclusion that I dont ever want to either.  It is tied up with soF much legal stuff its amazing.  Legal departments in HP and the "other"D company are involved and all kinds of research has to be done, and a= large amount of signatures are required.  All the suggestionsnB mentioned in this note were given by either the Ambassadors or theC engineers or the customers, but there was always something (usually E legal).  A decision had to be made.  Everyone will still call it VMS,pD but there needed to be a way to differentiate between the platforms.  C Maybe not the folks on this newsgroup but many customers think that E VMS is for VAX, OpenVMS is for Alpha and I assume OpenVMS I64 will beaC thought of as the Intel box.  I am still surprised at the amount ofAB people that think the VMS vs OpenVMS, and I still get requests for+ presentations on VMS to OpenVMS migrations.r  @ Advertising - funny I have not seen an HPux advertisment either,F Everyone I know would love to do a VMS advertisment and we have gottenE some excellent suggestions from the newsgroup and enginnering and our E customers.  But if the company does not allow advertising for OS's wehE can not, we work for HP, and that is their rule.  It is not an optionv period.   E Regarding Nic's post about the excellent opportunity he has, Nic this D is great and if its ok with you I would like to use it as an OpenVMSE Pearl when I get back to the office.  Folks what Nic says is true, ine@ many cases an OS is considered a competititve advantage and manyD customers (not just governments) prefer to keep it quiet.  I can seeF their point, however I have a dream of one day getting a computer chip? that has "VMS inside" on it. (Hey I can dream)from one of thesen
 customers.  F We can ask why a company/government does not want to do a testimonial,> but what about where you work.  Does your company want to do a@ testimonial?  Also if possible can you let me know how much yourA company has purchased in the last year (if yes to either of these * questions please send email, do not post).   Warm Regards as always,o sue   r   We can talk or we can do  s "John Smith" <a@nonymous.com> wrote in message news:<G3U5a.48972$Zr%.15278@news01.bloor.is.net.cable.rogers.com>...o> > "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message" > news:3E57E34F.7092B12@fsi.net... > > Nic Clews wrote: > > >e > > > Bill Gunshannon wrote: > > > >a1 > > > > In article <3E577653.4D680BE4@127.0.0.1>,i: > > > >         Nic Clews <sendspamhere@127.0.0.1> writes: > > > > > John Smith wrote:- >  tH > > > > > Commercial and military reasons. Those that could say the most >  must E > > > > > remain silent, it is the way of the world. The examples area >  unseen soH > > > > > the marketing has to be based on something else. I guess folks >  here haveD > > > > > made their opinions heard in the surveys announced here... > > > >sB > > > > Hardly.  Most military procurements are public information >  availableF > > > > through a simple(?) FOI request.  Especially when you consider >  that)H > > > > they usually start as a public RFP published in the CBD.  I have >  neverC > > > > been at a government site, including the pentagon where thee
 >  brand name8? > > > > of the system or the OS was classified information.  If7 >  DEC/COMPAQ/HPC > > > > really cared there is nothing stopping them from publishing. >  numbers,lC > > > > customer names and types of systems being run on VMS.  Theyu
 >  choose nota > > > > to and rest is history.  > > >(B > > > Bill, I wish I could say more, but I'd lose my job *and* get
 >  prosecuted  > >t
 > > For what?e > >f@ > > > into the bargain. Heck, I'm even nervous saying that much. > >s > > For the love of spit, why? > >e	 > > > > >MH > > > > > However, yesterday I was at a project closure meeting for some
 >  upgradeG > > > > > work, hardware and software. They were so happy with it, thatn
 >  some stuff  >  ...	 > > > > >e > > > >dF > > > > And will HP make this success public, in glowing details, in aF > > > > media that someone outside the VMS faithful will actually see? > > >aH > > > I'll be discussing this with our ambassador. They offered to write> > > > something for our internal glossies. There are also some
 >  commercials > > > sensitivities, > > G > > Such as... (Hint: state he nature of the "sensitivity" in question,g >  not- > > that actual solution used to address it.)a > >a9 > > > so one hopes that the correct conclusions are drawnn* > > > without having to state the obvious. > > + > > If it need be stated, it's not obvious.w > >l >  >  > David, > F > Just think about discussing the details of any work you have done orD > may ever do in the future at the NSA, CIA, OSWR, LLNL, etc...  HowH > long do you think it would be before you had a rather long-ish expenseF > paid vacation in Marion, Ill. in a cell between those of John Walker > and Aldrich Ames?? > , > Now think UK-equivalents to all the above. >  > Is it obvious now?   ------------------------------    Date: 22 Feb 2003 22:09:30 -0800. From: spamsink2001@yahoo.com (Alan E. Feldman) Subject: Re: The End of OpenVMS'< Message-ID: <b096a4ee.0302222209.5bb5b36@posting.google.com>  f bill@cs.uofs.edu (Bill Gunshannon) wrote in message news:<b380fe$1j80pf$1@ID-135708.news.dfncis.de>... [...]h  F > Hardly.  Most military procurements are public information availableE > through a simple(?) FOI request.  Especially when you consider that H > they usually start as a public RFP published in the CBD.  I have neverH > been at a government site, including the pentagon where the brand nameG > of the system or the OS was classified information.  If DEC/COMPAQ/HP F > really cared there is nothing stopping them from publishing numbers,H > customer names and types of systems being run on VMS.  They choose not > to and rest is history.t  C Maybe they simply don't want to piss off their big customers. Maybe/2 they're just protecting the customers' privacy(s).  ? Maybe they could. But maybe they have other incentives not to. i   [...]    Disclaimer: JMHO Alan E. FeldmanS   ------------------------------  % Date: Sat, 22 Feb 2003 14:53:35 -060021 From: "David J. Dachtera" <djesys.nospam@fsi.net>s Subject: Re: The End of OpenVMSD& Message-ID: <3E57E34F.7092B12@fsi.net>   Nic Clews wrote: >  > Bill Gunshannon wrote: > > - > > In article <3E577653.4D680BE4@127.0.0.1>,a6 > >         Nic Clews <sendspamhere@127.0.0.1> writes: > > > John Smith wrote:I > I > > > Commercial and military reasons. Those that could say the most mustwK > > > remain silent, it is the way of the world. The examples are unseen soaN > > > the marketing has to be based on something else. I guess folks here have@ > > > made their opinions heard in the surveys announced here... > >uH > > Hardly.  Most military procurements are public information availableG > > through a simple(?) FOI request.  Especially when you consider thatiJ > > they usually start as a public RFP published in the CBD.  I have neverJ > > been at a government site, including the pentagon where the brand nameI > > of the system or the OS was classified information.  If DEC/COMPAQ/HP-H > > really cared there is nothing stopping them from publishing numbers,J > > customer names and types of systems being run on VMS.  They choose not > > to and rest is history.= > I > Bill, I wish I could say more, but I'd lose my job *and* get prosecuteda  	 For what?e  < > into the bargain. Heck, I'm even nervous saying that much.   For the love of spit, why?   > > > L > > > However, yesterday I was at a project closure meeting for some upgradeN > > > work, hardware and software. They were so happy with it, that some stuff > ...g > > >f > >2B > > And will HP make this success public, in glowing details, in aB > > media that someone outside the VMS faithful will actually see? > D > I'll be discussing this with our ambassador. They offered to writeE > something for our internal glossies. There are also some commercials > sensitivities,  G Such as... (Hint: state he nature of the "sensitivity" in question, not ) that actual solution used to address it.)a  5 > so one hopes that the correct conclusions are drawna& > without having to state the obvious.  ' If it need be stated, it's not obvious.f   > This is a reflection of thep > technologies, not the people.m  	 Huh???!!!n   -- e David J. Dachtera  dba DJE Systemse http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/s   ------------------------------  % Date: Sat, 22 Feb 2003 16:27:06 -0400o0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca> Subject: Re: The End of OpenVMSh/ Message-ID: <3E57DD17.993B6C22@vl.videotron.ca>e  C Actually, we should all be happy that the end of "openVMS" has beenn+ offficially announced in this newsgroup ;-)U  N It measn that hp is finally dropping the "OPEN" , hence the end of "openvms".  :-) :-) ;-)   L Note to VMS engineers: NOW is the time to drop the bloody open since you are5 reworking all of VMS for the port to that IA64 thing.,   ------------------------------  % Date: Sat, 22 Feb 2003 15:07:03 -0600 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>h Subject: Re: The End of OpenVMSs' Message-ID: <3E57E677.93C03DDE@fsi.net>n   Bill Gunshannon wrote: > [snip] > If DEC/COMPAQ/HPF > really cared there is nothing stopping them from publishing numbers,H > customer names and types of systems being run on VMS.  They choose not > to and rest is history.e  = VMS is not history quite yet. However, if the veil of secrecy F surrounding VMS is not lifted, VMS *WILL* be history very soon, and no; amount of pro-Itanic hype is going to be enough to save it.c  A Perhaps the greatest secret surrounding VMS is the reason for the H secrecy in the first place. There's no logical explanation for it. EveryG excuse proferred here in this forum has been repeatedly shot down quite  handsomely.n  H I'm currently negotiating with the powers-that-be to offer a cash rewardF for spotting a VMS ad in *ANY* print medium that can be authenticated.= Perhaps that will jar someone loose whose hands are currentlyn5 super-glued to their posterior over marketing of VMS.e   -- h David J. Dachterat dba DJE Systemse http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/M   ------------------------------  % Date: Sat, 22 Feb 2003 07:18:59 -0500 2 From: rdeininger@mindspring.com (Robert Deininger)( Subject: Re: Upcoming VMS improvements ?L Message-ID: <rdeininger-2202030718590001@user-uinj4lr.dialup.mindspring.com>  > In article <G8ednVIlc4atTsujXTWcpQ@metrocast.net>, "Bill Todd" <billtodd@metrocast.net> wrote:_  M >> I doubt enhancements to TPU will get higher priority than, say, a new fileL
 >> system. > D >And since plans for a new file system seem to have pretty much beenK >languishing for over three years now, with no obvious progress (not even aEM >mention on the roadmap through early 2005), well, draw your own conclusions.h  H What makes you think the file system has been languishing?  Just because nothing has shipped?   ------------------------------  % Date: Sat, 22 Feb 2003 15:18:19 -0600t1 From: "David J. Dachtera" <djesys.nospam@fsi.net>o( Subject: Re: Upcoming VMS improvements ?' Message-ID: <3E57E91B.999EE1ED@fsi.net>m   Robert Deininger wrote:t > @ > In article <G8ednVIlc4atTsujXTWcpQ@metrocast.net>, "Bill Todd"! > <billtodd@metrocast.net> wrote:i > O > >> I doubt enhancements to TPU will get higher priority than, say, a new fileo > >> system. > >tF > >And since plans for a new file system seem to have pretty much beenM > >languishing for over three years now, with no obvious progress (not even arO > >mention on the roadmap through early 2005), well, draw your own conclusions.  > J > What makes you think the file system has been languishing?  Just because > nothing has shipped?  . Probably that whole "veil of secrecy" thing...   --   David J. Dachtera  dba DJE SystemsW http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sat, 22 Feb 2003 20:22:51 -0500i* From: "Bill Todd" <billtodd@metrocast.net>( Subject: Re: Upcoming VMS improvements ?2 Message-ID: <uK6cnVngUdjvv8WjXTWcpg@metrocast.net>  ? "Robert Deininger" <rdeininger@mindspring.com> wrote in message5F news:rdeininger-2202030718590001@user-uinj4lr.dialup.mindspring.com...@ > In article <G8ednVIlc4atTsujXTWcpQ@metrocast.net>, "Bill Todd"! > <billtodd@metrocast.net> wrote:  >WJ > >> I doubt enhancements to TPU will get higher priority than, say, a new file > >> system. > >7F > >And since plans for a new file system seem to have pretty much beenK > >languishing for over three years now, with no obvious progress (not even  aEB > >mention on the roadmap through early 2005), well, draw your own conclusions. > & > What makes you think the file system  I I assume you mean the new file system, since otherwise your post would be4 irrelevant.1  $  has been languishing?  Just because > nothing has shipped?  = Just to tick off a few of the reasons off the top of my head:   H 1.  When I visited ZK to discuss it, it had no useful budget, no definedE commitment, and only a single developer associated with it (I suspectU part-time).n  D 2.  That developer didn't think that significant enhancements to theG existing Scotland code would be funded:  it was clearly on a very shortn shoestring.r  E 3.  Indeed, in the three years since, it hasn't shipped (or even beenh rumored to be on the way).  G 4.  And, as noted, it doesn't appear anywhere in the future VMS roadmapl (through early 2005).   I Now, I suppose some people might be entirely satisfied with that state oft% affairs; me, I call it 'languishing'.k   - bill   ------------------------------  # Date: Sat, 22 Feb 2003 17:36:01 GMT 4 From: Tim Llewellyn <tim.llewellyn@blueyonder.co.uk>( Subject: Re: Upcoming VMS improvements ?0 Message-ID: <3E57B501.FFCD7E44@blueyonder.co.uk>   Ken Robinson wrote:     > K > I don't know about that... My main computer at home right now is a laptopkI > running XP. I telnet into Encompasserve using Tera Term Pro and can use M > both Notes and EVE fine.  I just have to remember to hit the "Num Lock" keye@ > while I'm in Notes and that "Shift-F6" is the "Do" Key in EVE. >    & ^B gets you a "do" prompt also in EVE.   Regards,   -- a tim.llewellyn@blueyonder.co.uk n  H * PLEASE NOTE tim.llewellyn@cableinet.co.uk address is NO LONGER VALID *   ------------------------------  % Date: Sat, 22 Feb 2003 16:24:43 -0400 0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>( Subject: Re: Upcoming VMS improvements ?/ Message-ID: <3E57DC88.552CF736@vl.videotron.ca>D   Tim Llewellyn wrote:( > ^B gets you a "do" prompt also in EVE.  C And also acts as a command recall with successive uses of <ctrl> B.a   ------------------------------  % Date: Sat, 22 Feb 2003 17:20:10 -0500b( From: David Froble <davef@tsoft-inc.com>) Subject: Re: VMS & Tru64 Disk I/O timings * Message-ID: <3E57F79A.10706@tsoft-inc.com>  N There are many things that *CAN* be done to improve disk I/O.  There are also * some rather basic things controlling such.  O One issue that is so hard to plan for is transfer sizes to/from disk.  Today's  Q drives will screen you from much of this problem.  RAID and such will do more to eP distance you from this issue.  The best you can hope for is that the OS will do M things rather decently.  However, when changing RMS system defaults, keep in oM mind that storage on disk is (to outside observation) in 512 byte blocks, or pQ multiples of such.  Consider for example a disk subsystem that can transfer data bL in chunks of 16 blocks.  Setting a buffer size of 17 blocks would require 2 M transfers, the second being a single block.  Bad for everyone on the system. cQ So, when setting RMS system defaults, stay with buffersizes of some power of 2 * lN blocks.  Note that I'm not that intimate with today's storage, and things may M have changed to the point that the above is now meaningless.  Still, I doubt r- that some things changed without some reason..  M Another consideration is the diminishing return on larger buffer sizes.  Say  P your original situation was one buffer of 512 bytes.  For sequential access, if M you increase the size to 2 blocks, 1024 bytes, then you've cut the number of kO reads in half.  By definition, you cannot exceed this single gain in reduction sP of I/Os.  You can of course reduce the I/O count further.  The consideration is N don't go overboard on increasing buffer sizes, because you can easily reach a O point of diminishing returns, and even worse should you commit too many needed nL resources to buffer space.  The same considerations exist for the number of Q buffers, however, up to a point multiple buffers is a good thing, for sequential   and random I/O.   Q I once had a customer who misunderstood a parameter, and set the minimum working dM set size to some hugh number.  I soon got a call about the system being REAL  K slow.  Know what you're setting, and why, and increase parameters in small   steps, not hugh chunks.i   Dave     Dave Gudewicz wrote:   > Thanks > --	 > Dave...e > J > Such is the human race.  Often it does seem such a pity that Noah didn't > miss the boat. > -----Mark Twain6 > H > "Hein van den Heuvel" <hein_netscape@eps.zko.dec.com> wrote in message+ > news:3E5525CA.532F9865@eps.zko.dec.com...  >  >> >>Dave Gudewicz wrote: >> >> >>>Hein, >>>MB >>>Thanks for joining in here.  Now a question.  If the RMS sysgen >>>i > parameters > J >>>are left to default values, which are mostly zero, and a program is notF >>>written to take advantage of any of the RMS stuff mentioned in this >>> 	 > thread,a > > >>>can one safely assume that mostly zero is what they'll get? >>> L >>Close. Zero translates to 1 buffer of 16 blocks (8KB). Not much these days >> > huh? > F >>If (and only if) you have WBH enabled (Cobol default we'd give you a >> > whopping 2 > 
 >>buffers! >>K >>While VMS recognized a long time ago those those params were probably way.I >>to conservative the believe was that a wholesale change could cause too- >> > much >  >>disturbance.F >>There is change though. As Kerry already indicated,  the latstes VMS >>
 > versions > 
 >>has severaloI >>new features, notably 'forcing WBH', large buffers, and Query locks fort >> > indexs >  >>files all withoute >>program changes. >> >> >>>If yes, thennI >>>**in general** is it a good idea to bump these sysgen parameters up toV >>>s > getr > % >>>an I/O performance boost from VMS?v >>>a: >>Yes, go for it: SET RMS/SYS/SEQ/BLOCK=64/BUF=4/EXTE=1000 >>and SET RMS/SYS/IND/BUF=20 >> >> >>+ >>>How might this bump affect programs that 7 >>>are written to take advantage of these RMS features?l >>>lK >>Well, if they hardcoded an improvement from 1*8KB buf to 2*16KB then theynL >>would (unfortunately?) not benefit from a news 4*32KB default as suggested >>above. >>J >>With a perfect read-ahead implementation you really only need 2 buffers. >>	 > But RMSC >  >>doesI >>not try to 'keep ahead'. It just reads-ahead when it runs out ofdata in- >> > the- >  >>current buffers.J >>If have measured signifincant improvementes goiing from 1 to 2 to 4, but >> > very >  >>minor K >>further improvements going from 4 to say 8. Also, the multi buffer really  >> > only >  >>helpsrH >>when RAH and/ro WBH is active, so no need to get carried away too much >> >>J >>>And if one uses an external disk controller, HS*nn, VMS thinks its done >>>m > with > I >>>writes when the controller (not the disk) tells it.  So the data isn't  >>>. > on > K >>>the platter/s until the controller gets around to it and VMS goes on itsh" >>>merry way thinking all is well. >>>r9 >>Correct. Writeback caching in the controller will work.hI >>It does just not on direct-attached storage. (well, it is not supportedo >> > there, >  >>but it can be done).K >>The data will be 'out of the system', over the wire, and the process willu >> > only > 
 >>continueJ >>when the controller indicates cussesfull receipt, but yes that is before >>	 > it hitse >  >>the disk itself. >> >>hth, >> >>Hein.o >> >> >>>--i
 >>>Dave... >>>dJ >>>It is noble to teach oneself, but still nobler to teach others-and less >>>trouble.i >>>-----Mark Twain >>>s6 >>>"hein" <hein_news@eps.zko.dec.com> wrote in message+ >>>news:cmF4a.41$OV2.35@news.cpqcorp.net...M >>> ? >>>>"Lyndon Bartels" <lbartels@pressenter.com> wrote in messagel, >>>>news:3E52959D.5F56EC4D@pressenter.com... >>>> >>>>>Lyndon Bartels wrote: >>>>>tA >>>>>On my 500au, I ran 6 tests changing DIOLM, BIOLM, and BYTLM.n: >>>>>These tests ran from 1394 seconds up to 1580 seconds. >>>>>eC >>>>As per my previous reply, nice experiment, but a waste of time.uE >>>>Very, very, few, programs know how to honor these process limits.  >>>> > Backup >  >>>>being one of the few. E >>>>For normal program it suffices to have enough. More will not help  >>>> > more.o >  >>>>I >>>>>I've been reading the performance management manual, and I've prettyiE >>>>>much concluded that the computer isn't CPU bound. Memory doesn'ti >>>>>m > look > E >>>>>bad, but I haven't got to that part of the manual yet. I/O, bothM# >>>>>buffered and direct, are high.( >>>>> E >>>>Fix the buffered IO by pre-allocating and/or using large extents.s >>>>E >>>>>Now... the question I have.... Is how do I figure out how/why to. >>>>> 	 > correcta > E >>>>>the I/O bind. My gut tells me that the I/O rate is high, but the- >>>>>5	 > size is7 > G >>>>>too small. The computer is spending more time keeping track of thei >>>>>r > I/Os > F >>>>>than actually doing the I/Os.... If I could make the size larger, >>>>>o > then > $ >>>>>the number might be reduced.... >>>>>eG >>>>The RMS knob for this is MULTI BLOCK COUNT = RAB$B_MBC ( 0 - 127 ).  >>>> >>>>hth, >>>>     Hein. >>>> >>>> >>>> >  >    ------------------------------  % Date: Sat, 22 Feb 2003 14:51:18 -0400h0 From: JF Mezei <jfmezei.spamnot@vl.videotron.ca>- Subject: Re: VMSclusters vs. others' clustersa/ Message-ID: <3E57C6A5.EEA90F1D@vl.videotron.ca>e  E > > So what are the particular features of VMSclusters that sets themo8 > > apart from the rest, esp. Unix and Windows systems?   I Cluster interconnects.  VMS can cluster via ethernet and a whole gamut ofRL other types of connections. Ethernet allows you to cluster your workstationsN which greatly reduces management since you can work with a common system imageN with a common set of software etc etc, and during lunch hour, you can make useD of available computing on those iddle workstation to run batch jobs.  M The "up to 96 nodes" tells you that there is a great amount of flexibility ineK how you put together your clusters. Others let you cluster 2 nodes togethero; with a particular interconnect that limited maximum length.T  G Of course, the distributed lock manager really makes the disk and othert, software operate seamlessly on the cluster.   N Drives can connect via MSCP or via some hadrware interconnect (SCSI, CI, fibre etc).   L Host Based Volume shadowing provides for far more robust data integrity in aN cluster (especially multi site) since it uses MSCP for "transport" and MSCP isK fully "aware" of cluster quorum. If you cluster disk arrays using their owntK intersite links, the arrays are far less aware of quorum of the cluster andt you may get data corruption.   ------------------------------    Date: 22 Feb 2003 15:30:20 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen)r- Subject: Re: VMSclusters vs. others' clustersT3 Message-ID: <BauZrpntrrwg@eisner.encompasserve.org>i  T In article <3E577A63.65B10CBE@127.0.0.1>, Nic Clews <sendspamhere@127.0.0.1> writes:  B > I'm arguing the toss about the title for it "OpenVMS Clustering:H > Foundations of Reliability and Scalability", but I'm being badgered toF > lose the "OpenVMS" bit so that a wider audience books, attends, then > weep as they leave.w >  > Opinions?    I vote with the badgers.  F Please come on tour to the US also - like HP World, Atlanta in August.C If your boss knows the US, they will realize that Atlanta in AugustS is not a boondoggle.   ------------------------------  % Date: Sat, 22 Feb 2003 17:39:03 -0500 ( From: David Froble <davef@tsoft-inc.com>- Subject: Re: VMSclusters vs. others' clusterse, Message-ID: <3E57FC07.6080506@tsoft-inc.com>   Alan E. Feldman wrote:   > Hello fellow VMS-ers,h > E > Occasionally the topic of clusters comes up at work and I'd like torE > say what makes VMSclusters so much better than others'. I know that/D > VMSclusters go back to the 80's and that VMSclusters are the best. > D > But I know very little about what others' clusters really do. I'veF > heard that Google has a cluster of thousands of Linux boxes. Is that' > true? How would VMS make that better?o > C > So what are the particular features of VMSclusters that sets themJE > apart from the rest, esp. Unix and Windows systems? If readers givebH > good answers here I'll be well equipped to tell my coworkers about it.    Q One basic concept of VMS clustering is that the 'logical' filesystem is the core eO of the cluster.  You can have 2 or more computers accessing the filesystem, at tQ the same time, with the DLM providing lock granularity down to the record level. oQ   I don't know much about IBM's Sysplex, but I don't know of any other so called l cluster with this capability.o  Q There's much more to VMS clusters than this, but from my perspective this is the   core of VMS clusters..   Dave   ------------------------------  % Date: Sat, 22 Feb 2003 13:25:55 +0000e( From: Nic Clews <sendspamhere@127.0.0.1>- Subject: Re: VMSclusters vs. others' clusters,) Message-ID: <3E577A63.65B10CBE@127.0.0.1>n   "Alan E. Feldman" wrote:  E > Occasionally the topic of clusters comes up at work and I'd like toiE > say what makes VMSclusters so much better than others'. I know thataD > VMSclusters go back to the 80's and that VMSclusters are the best. > D > But I know very little about what others' clusters really do. I'veF > heard that Google has a cluster of thousands of Linux boxes. Is that' > true? How would VMS make that better?a > C > So what are the particular features of VMSclusters that sets themaE > apart from the rest, esp. Unix and Windows systems? If readers givesH > good answers here I'll be well equipped to tell my coworkers about it.  D I'm doing a UK based user event on this subject, why not come along?  G I made it my business to understand others' so called clusters, currentZH "technology". While they are making improvements, it still takes a whileC for me to stop laughing. I wasn't really going to include it in the D agenda, but if there is public demand, I could do a compare/contrastD with a number of technologies. The "popular" ones are as you suspect? pretty hopeless, but some of the more proprietary offerings areT@ interesting, but the definition and restrictions also need to be understood.o  @ I'm arguing the toss about the title for it "OpenVMS Clustering:F Foundations of Reliability and Scalability", but I'm being badgered toD lose the "OpenVMS" bit so that a wider audience books, attends, then weep as they leave.r  	 Opinions?r -- n? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciencese nclews at csc dot com    ------------------------------  % Date: Sat, 22 Feb 2003 08:19:10 -0700 $ From: Lee Y T Mah <lytmah@cha.ab.ca>- Subject: Re: VMSclusters vs. others' clusterst) Message-ID: <3E5794EE.694C2A66@cha.ab.ca>i  A Consider a hypothetical VMScluster comprised of five nodes spreadtF across three separate sites (2 nodes, 2 nodes, 1 node).  Each node has; one vote.  Two of the sites have mirrored drives using HBVSo (host-based volume shadowing).  C From any one node, users can access any device mounted clusterwide.n  D The following can be done while still leaving the cluster up 24x365:  D   You can lose any combination of two nodes (scheduled/unscheduled).   You can lose any one site.=   You can perform VMS upgrades/updates to one node at a time.4D     (Theoretically, you could migrate from 5.5-2 to 7.3-1 this way.)@   Upgrades can be performed on redundant hardware (nodes, disks, controllers...).       "Alan E. Feldman" wrote:   > Hello fellow VMS-ers,  >eE > Occasionally the topic of clusters comes up at work and I'd like tosE > say what makes VMSclusters so much better than others'. I know thatuD > VMSclusters go back to the 80's and that VMSclusters are the best. >sD > But I know very little about what others' clusters really do. I'veF > heard that Google has a cluster of thousands of Linux boxes. Is that' > true? How would VMS make that better?  > C > So what are the particular features of VMSclusters that sets themfE > apart from the rest, esp. Unix and Windows systems? If readers give H > good answers here I'll be well equipped to tell my coworkers about it. >i	 > Thanks.e >o > Alan E. Feldmanh   -- Lee   5 L Y T Mah                    Email:  lytmah@cha.ab.cab   ------------------------------  % Date: Sat, 22 Feb 2003 20:12:23 +0100y6 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= <arne@vajhoej.dk>G Subject: Re: What makes the VMS port to Itanium/IPF/whatever difficult?w) Message-ID: <3E57CB97.7000302@vajhoej.dk>h   Alan E. Feldman wrote:E > A Unix Admin co-worker asked me what the big deal was about portingtG > VMS to Itanium. He says that NetBSD has been ported to numerous chips'H > and that it's no big deal. I told him that I didn't really know. WouldG > any of the chip experts reading this care to comment? What is it thatf > makes this port difficult?  . VMS was never intended for multi-platform and:5    - uses several constructs which was present in the ;      original VAX hardware and need to be emulated on others      architectures:    - VMS still has a lot of Macro-32 and Bliss code (while      NetBSD is probably 100% C)a  = Furthermore: how many architectures hs that guy ported NetBSD @ to ?  The fasct that NetBSD is available for many platforms does not say that it was easy !   Arne   ------------------------------  # Date: Sun, 23 Feb 2003 03:40:31 GMTo. From: peter@langstoeger.at (Peter LANGSTOEGER) Subject: [DFG] Some questionsa3 Message-ID: <PoX5a.50966$Rb4.707615@news.chello.at>-  H The posting with the possible MAIL problem in DEFRAG V2.7 reminded me of* some other problems I've seen with DEFRAG.    D 1) DEFRAG always translates Filenames (Option File, Logfile, ...) atH Script define time to their real physical names. I find it very annoyingB to have logicals in VMS and can't use the flexibility they provide@ because of this obvious misbehaviour of DEFRAG (Vx.y up to V2.7)    G 2) I don't really understand why the DEFRAGger needs so many privilegesaH (BYPASS, CMKRNL, DETACH, EXQUOTA, NETMBX, SHARE, SYSGBL, SYSLCK, SYSPRV,E TMPMBX, and WORLD) for even showing scripts (especially BYPASS, whichrJ I've haven't set as /DEFPRIV of my own user). It is ok that the defraggingE process itself might need them, but I think the DEFRAG SHOW/NOVOLUME,hI DEFRAG VOLUME/SCRIPT and DEFRAG MODIFY should not need that much of them.e    J 3) Opening (and Closing) the Full Volume Map (REPORT Volume Fragmentation)B in DFG$DW.EXE brings every time 3 lines with the following Warning  D X Toolkit Warning: Null child passed to XtUnmanageChildren	(at open)C X Toolkit Warning: null child passed to XtManageChildren	(at close)e  4 I don't see real errors, only this annoying lines...    I 4) Installing DEFRAG creates every time (regardless if it already exists),N the file DFG$DATABASE:DFG$MAIL.DIS with only one line "nodename::installuser".L Despite the fact, that it should not create what already exists (and despiteM the fact, that one can override the content of said file during installation) K it is annoying that this mailaddress contains the nodename which might leadiM to an undeliverable mail (eg.when the installnode of a cluster is temporarily- down).    I 5) I don't know the current status (because I no longer use HSM) but manyiL months ago it was neccessary to force the DEFRAGger to _not_ unshelve files.3 I did it with the PROLOGUE Command File containing:g  @ $ SET PROCESS/NOAUTO_UNSHELVE/IDENT='F$GETJPI (0, "MASTER_PID")' $ SHOW ERROR $ DEFRAG SHOW DFG$DISK/VOLUMEu  K It might be fixed now, but maybe it is not. Maybe someone can test/comment.x    H 6) If I check my DEFRAGger (/FULL) logfile I always find many many filesF marked as "OPEN" which I'm sure aren't touched by anyone at this time.F Most of them are .DIR files at "dormant file placement" time, but IIRC
 some are not.l  H Is this normal, that directory files (eg. SYS$EXAMPLES:) are really open and therefore not moveable ?   TBC (to be continued)    -- m Peter "EPLAN" LANGSTOEGERi% Network and OpenVMS system specialistn E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------   End of INFO-VAX 2003.105 ************************MS,pD but ther.
 <<< PASV? >>> 227 Entering passive mode; use PORT (198,151,12,104,12,6) <<< RETR 2003_080.txtY >>> 150 ASCII retrieve of /disk$misc/decus/info-vax/2003_080.txt (33118 bytes) started.; >>> 226 Transfer completed.  32100 (8) bytes transferred.
 <<< PASV? >>> 227 Entering passive mode; use PORT (198,151,12,104,12,7) <<< RETR 2003_081.txtY >>> 150 ASCII retrieve of /disk$misc/decus/info-vax/2003_081.txt (32650 bytes) started.; >>> 226 Transfer completed.  31506 (8) bytes transferred.
 <<< PASV? >>> 227 Entering passive mode; use PORT (198,151,12,104,12,8) <<< RETR 2003_082.txtY >>> 150 ASCII retrieve of /disk$misc/decus/info-vax/2003_082.txt (90008 bytes) started.; >>> 226 Transfer completed.  88328 (8) bytes transferred.
 <<< PASV? >>> 227 Entering passive mode; use PORT (198,151,12,104,12,9) <<< RETR 2003_083.txtY >>> 150 ASCII retrieve of /disk$misc/decus/info-vax/2003_083.txt (83050 bytes) started.; >>> 226 Transfer completed.  81984 (8) bytes transferred.
 <<< PASV@ >>> 227 Entering passive mode; use PORT (198,151,12,104,12,11) <<< RETR 2003_084.txtZ >>> 150 ASCII retrieve of /disk$misc/decus/info-vax/2003_084.txt (113066 bytes) started.< >>> 226 Transfer completed.  111108 (8) bytes transferred.
 <<< PASV@ >>> 227 Entering passive mode; use PORT (198,151,12,104,12,12) <<< RETR 2003_085.txtZ >>> 150 ASCII retrieve of /disk$misc/decus/info-vax/2003_085.txt (126086 bytes) started.< >>> 226 Transfer completed.  124532 (8) bytes transferred.
 <<< PASV@ >>> 227 Entering passive mode; use PORT (198,151,12,104,12,13) <<< RETR 2003_086.txtY >>> 150 ASCII retrieve of /disk$misc/decus/info-vax/2003_086.txt (75694 bytes) started.; >>> 226 Transfer completed.  73936 (8) bytes transferred.
 <<< PASV@ >>> 227 Entering passive mode; use PORT (198,151,12,104,12,14) <<< RETR 2003_087.txtY >>> 150 ASCII retrieve of /disk$misc/decus/info-vax/2003_087.txt (60652 bytes) started.; >>> 226 Transfer completed.  59554 (8) bytes transferred.
 <<< PASV@ >>> 227 Entering passive mode; use PORT (198,151,12,104,12,15) <<< RETR 2003_088.txtZ >>> 150 ASCII retrieve of /disk$misc/decus/info-vax/2003_088.txt (117740 bytes) started.< >>> 226 Transfer completed.  115440 (8) bytes transferred.
 <<< PASVA >>> 227 Entering passive mode; use PORT (198,151,12,104,12,150)< <<< RETR 2003_089.txt>Z >>> 150 ASCII retrieve of /disk$misc/decus/info-vax/2003_089.txt (111958 bytes) started.< >>> 226 Transfer completed.  110029 (8) bytes transferred.
 <<< PASVA >>> 227 Entering passive mode; use PORT (198,151,12,104,12,181)  <<< RETR 2003_090.txt Y >>> 150 ASCII retrieve of /disk$misc/decus/info-vax/2003_090.txt (80596 bytes) started. ; >>> 226 Transfer completed.  79323 (8) bytes transferred. 
 <<< PASVA >>> 227 Entering passive mode; use PORT (198,151,12,104,12,182)  <<< RETR 2003_091.txt Y >>> 150 ASCII retrieve of /disk$misc/decus/info-vax/2003_091.txt (75230 bytes) started. ; >>> 226 Transfer completed