1 INFO-VAX	Sun, 15 Sep 2002	Volume 2002 : Issue 510       Contents:! Re: How to debug PRODUCT INSTAL ?  Re: IMAP performance Re: IMPORTANT SECURITY ALERT Re: IMPORTANT SECURITY ALERT@ Re: lexicals (was: RE: SET PROTECTION (SET SECURITY/PROTECTION)"@ Re: lexicals (was: RE: SET PROTECTION (SET SECURITY/PROTECTION)"P Re: license costs (was: RE: Microsoft reeling from hack attacks, VMS  just yawns@ Re: Preinstalled hobby VMS distribution on resold hardware query@ Re: Preinstalled hobby VMS distribution on resold hardware query% Re: Saveset distribution over the web % Re: Saveset distribution over the web % Re: Saveset distribution over the web % Re: Saveset distribution over the web 4 Re: Seeking Senior VMS System Administrator Position Re: Sybase and VMS Re: Sybase and VMS  Re: TCPIP 5.3 bug in VAX install  Re: TCPIP 5.3 bug in VAX install" Re: TPU Error using VMS from Linux" Re: TPU Error using VMS from Linux	 WTC cruft 
 Re: WTC cruft   F ----------------------------------------------------------------------  % Date: Sun, 15 Sep 2002 08:27:26 +0200 - From: Didier Morandi <Didier.Morandi@Free.fr> * Subject: Re: How to debug PRODUCT INSTAL ?' Message-ID: <3D84284B.6A6025A9@Free.fr>    JF Mezei wrote:  >  ../..  > 8 > execute install "@pcsi$source:[sysmgr]cleanup_ucx.com" >            remove ""- >            source [sysmgr]cleanup_ucx.com ;  >  > still executes for me.  G Fyi, I found that this procedure is referred to in the following files:   ( 4477Pro_005.HTML OpenVMS Cluster SystemsJ 5952pro_003.htm POLYCENTER Software Installation Utility Developer's Guide 5952pro_005.htm & 6489Pro_037.HTML OpenVMS User's Manual 6489Pro_038.HTML 6489Pro_index.HTMLJ 6523pro_006.html DIGITAL TCP/IP Services for OpenVMS Concepts and PlanningO 6524profile_001.html DIGITAL TCP/IP Services f/O Installation and Configuration C 6526profile_005.html DIGITAL TCP/IP Services for OpenVMS Management  6526profile_007.html 6526profile_008.html 6526profile_009.html 6526profile_021.html 6526profile_index.html; 6551.HTML DIGITAL TCP/IP Services for OpenVMS Release Notes 
 6551_001.HTML 
 6551_002.HTML 
 6551_005.HTML  adduser.com  decw$startup.com lan$populate.comO OVMS_MIG_ENVIRON.PDF Migrating an Environment from VMS/VAX to Alpha AA-QSBTA-LE  tcpip$cleanup.com  tcpip$cleanup_helplib.com  tcpip$inet_shutdown.com  tcpip$inet_startup.com tcpip$nfs_server_startup.com tcpip$pwip_shutdown.com  tcpip$service_setup.com  tcpip$smtp_shutdown.com  tcpip$smtp_startup.com tcpip$snmp_shutdown.com  tcpip$snmp_startup.com tcpip$ucp_startup.com  tcpip_mgnt_cmd_ref.pdf tcpip050.release_notes ucx$config.com ucx$dnfc_shutdown.com  ucx$nfs_startup.com  ucx$smtp_config.com  ucx$startup.com   O Last, you'll find in http://Didier.Morandi.Free.fr/pcsi/ some more stuff if you  are interested in PCSI:   I PCSI.HLP                15-Sep-2002 08:16     7k  help file for PCSI_MENU H PCSI_BUILD_HELP.COM     15-Sep-2002 08:16     1k  to build the help fileK PCSI_BUILD_KIT.COM      15-Sep-2002 08:16     2k  an example to build a kit N PCSI_DEV_GUIDE.DOC      15-Sep-2002 08:17   503k  the Dev guide in Word formatK PCSI_MENU.COM           15-Sep-2002 08:16    28k  a "return and voil" menu E PCSI_MENU_FAQ.TXT       15-Sep-2002 08:17     4k  the faq of the menu N PCSI_WIZ.COM            15-Sep-2002 08:18     5k  a tool to debug kit building= PCSI_WIZ2.COM           15-Sep-2002 08:19     1k  same part 2 = PCSI_WIZ3.COM           15-Sep-2002 08:19     1k  same part 3 ; POLYCENTER_INFO.TXT     15-Sep-2002 08:19     5k  my Primer   G The PCSI_MENU tool is a menu driven utility to build a kit without PCSI O knowledge. On line help is included with three levels of display (none, medium, K full). The TOTO demo in the primer has been fully generated with this tool.   L The philosophy of the tool is the (former?) DEC philosophy. You hit <RETURN>> when you do not know and the job is successfully done for you.  
 Enjoy (I did)    D.   ------------------------------  % Date: Sun, 15 Sep 2002 18:10:26 +0200 @ From: Arne =?ISO-8859-1?Q?Vajh=F8j?= <arne.vajhoej@mail.tele.dk> Subject: Re: IMAP performance + Message-ID: <3D84B0F2.9040405@mail.tele.dk>    JF Mezei wrote:   C  > Well, I know IMAP isn't for me. Yesterday, I sent myself ( MAIL   /FOREIGN ) a  > 7000 block .SWF file.  >A  > It took forever to get a list of messages (that one, I do not   understand), butH  > when i tried to open the email, it took about half an hour before it  started C  > to download, and the IMAP process grew to take up all available  
 memory (about F  > 14000 pages) and outswapped most other processes, eating up 60% of 
 cpu duringI  > that time. At times, the process was even in COMO state ( i had never   seen
  > that !)  >3  > And then the download started and was very slow.   0 Sounds to me as if UCS 5.3 IMAP is not ready for serious usage yet !    Arne   ------------------------------    Date: 15 Sep 2002 19:51:15 +0800, From: Paul Repacholi <prep@prep.synonet.com>% Subject: Re: IMPORTANT SECURITY ALERT - Message-ID: <87elbva2to.fsf@prep.synonet.com>    system@SendSpamHere.ORG writes:   > > In article <3D80BEAF.1974F9E@mindspring.com>, Atlant Schmidt' > <atlantnospam@mindspring.com> writes:  > >"David J. Dachtera" wrote:   0 > >That could, of course, describe another world( > >leader, one who is *KNOWN* to possess- > >Nuclear, excuse me, "Nukular", Biological, , > >and Chemical Weapons of Mass Destruction.   > You have someone in mind?     = Well, the Butcher of Beirut, `l'idiot', or puppydog Blair all 
 come to mind.    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  # Date: Sun, 15 Sep 2002 15:18:43 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> % Subject: Re: IMPORTANT SECURITY ALERT ' Message-ID: <3D84AA4A.DA3012BC@fsi.net>    Paul Repacholi wrote:  > ! > system@SendSpamHere.ORG writes:  > @ > > In article <3D80BEAF.1974F9E@mindspring.com>, Atlant Schmidt) > > <atlantnospam@mindspring.com> writes:  > > >"David J. Dachtera" wrote:   ; Another case of misattribution - this quote is from Atlant.   2 > > >That could, of course, describe another world* > > >leader, one who is *KNOWN* to possess/ > > >Nuclear, excuse me, "Nukular", Biological, . > > >and Chemical Weapons of Mass Destruction. >  > > You have someone in mind?  > ? > Well, the Butcher of Beirut, `l'idiot', or puppydog Blair all  > come to mind.   G My nomme d'guerre(sp?) for "The Butcher of Baghdad" was "Sadly Insane".    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------   Date: 15 Sep 02 09:44:31 +0200) From: p_sture@elias.decus.ch (Paul Sture) I Subject: Re: lexicals (was: RE: SET PROTECTION (SET SECURITY/PROTECTION)" ) Message-ID: <kr3gJPCrz7V3@elias.decus.ch>   \ In article <3D83A80D.2ECE304A@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: > Arne Vajhj wrote:1 >> How can F$FAO support something DCL does not ?  > F > If F$FAO lexical simply calls SYS$FAO system routine, then it should4 > implicitely support all of the SYS$FAO constructs. > K > However, it is obvious that since DCL doesn't have quandwords or floating C > points, it would not be able to supply such arguments to SYS$FAO.   ? But IIRC it does support the formatting of quadword dates. I am . pretty sure I have an example of that at work. __
 Paul Sture Switzerland    ------------------------------  % Date: Sun, 15 Sep 2002 18:10:37 +0200 @ From: Arne =?ISO-8859-1?Q?Vajh=F8j?= <arne.vajhoej@mail.tele.dk>I Subject: Re: lexicals (was: RE: SET PROTECTION (SET SECURITY/PROTECTION)" + Message-ID: <3D84B0FD.6090603@mail.tele.dk>    David J. Dachtera wrote:    > Arne Vajhj wrote:   >>David J. Dachtera wrote: A  >> > I should think that F$FAO would support anything that $FAO  
 supports, no? K  >> > ...regardless of whether DCL supports the associated data(type(s))...   >> 1  >>How can F$FAO support something DCL does not ?   >J  > Simple: the F$xxx() routines are nothing more than DCL-level interfacesK  > to the underlying system services. They perform no filtering nor do they D  > impose any other limitations. The services themselves may support@  > features that DCL does not provide access to, but that is the%  > coincidence of it, not the design.      ????  : 1) DCL is not a language using VMS calling standard, which>      means that DCL will need to convert from DCL variable and!      arguments to a SYS$FAO call.   = 2) DCL does not have floating point support => floating point B      is not tested => floating point is not supported (it could be$      an unsupported feature though).   Arne   ------------------------------    Date: 15 Sep 2002 18:53:15 +0800, From: Paul Repacholi <prep@prep.synonet.com>Y Subject: Re: license costs (was: RE: Microsoft reeling from hack attacks, VMS  just yawns - Message-ID: <87n0qja5ic.fsf@prep.synonet.com>   ( jlsue <jlsuexxxz@screaminet.com> writes:  E > But the competitor, low-end products (especially Windows) do not go B > through the exhaustive qualification that VMS does for supported > platforms.   ::bang::bang::hello wall::  F If you have to do it for ONE customer, then it is free from then on...# the money is gone, spent, ka-pisch.   F But the savings in support overhead and time will roll in for a decadeE or more. If you can also boost sales by compeditivly pricing a better ? product, then you win all around. But only if you get the deal.    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Sun, 15 Sep 2002 09:41:40 +0200 - From: Didier Morandi <Didier.Morandi@Free.fr> I Subject: Re: Preinstalled hobby VMS distribution on resold hardware query ' Message-ID: <3D8439B2.6438FE51@Free.fr>    Rich Jordan wrote: > G > Is it legal to sell/resell hardware, in this case without an existing H > commercial (transferable) license, and include a fresh install (on topH > of a hard erase of the disk) of VMS (DECnet? TCPIP?), but with no PAKsB > installed, hobbyist or otherwise?  Such a system would boot to aD > username prompt on the display (or on the console serial port) andI > 'prove' operational status, and perhaps save someone from having to dig J > up media that is apparently not always easy to get, even though they canI > get the hobbyist license PAKs themselves. It would not really be useful G > until they got licenses but I could put the latest releases on it, to J > the benefit of a possible purchasing hobbyist user without media access.  " Well, the answer to me is obvious.  O As long as it is not sold, it is yours and it is legal to have "your" (actually M HP's) software with your licenses installed. When the purchase is done, it is P still HP software without license for the purchaser. I do not see why s/he wouldD erase the disk and wait for receiving the Hobbyist CD to rebuild it.   My personal advice is YES.   D.   ------------------------------  % Date: Sun, 15 Sep 2002 08:13:38 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)I Subject: Re: Preinstalled hobby VMS distribution on resold hardware query J Message-ID: <rdeininger-1509020813380001@1cust15.tnt2.nashua.nh.da.uu.net>  ; In article <uo7p3ar5n47l19@corp.supernews.com>, Rich Jordan  <duodec@speakeasy.net> wrote:   F >I'm probably selling a VAXstation 3100m76 GPX on Ebay in the not too E >distant future.  Since purchase it has been used only with hobbyist  H >licenses.  As it will go complete (sans monitor/keyboard) including an I >RZ26L, and it could be nice for the potential buyer to get a 'bootable'  @ >system, and since I didn't find an answer in an admittedly not B >painstaking google search, or on the Q website license section... > G >Is it legal to sell/resell hardware, in this case without an existing  H >commercial (transferable) license, and include a fresh install (on top H >of a hard erase of the disk) of VMS (DECnet? TCPIP?), but with no PAKs B >installed, hobbyist or otherwise?  Such a system would boot to a D >username prompt on the display (or on the console serial port) and I >'prove' operational status, and perhaps save someone from having to dig  J >up media that is apparently not always easy to get, even though they can I >get the hobbyist license PAKs themselves. It would not really be useful  G >until they got licenses but I could put the latest releases on it, to  I >the benefit of a possible purchasing hobbyist user without media access.   H This seems perfectly ok to me.  I have seen a number of VMS systems soldF on ebay  with the OS installed "for demonstration purposes" or similarH wording.  You need to make it very clear in the auction description thatH there is no license included.  You can add links to Encompass and to the1 hobbyist license page to make it more attractive.   F Once the sale closes, you could e-mail the system serial number to theJ buyer, who could obtain license PAKs immediately (assuming hobbyist use). 1 The system would be ready to use when it arrived.   J This method would make the auction much more attractive to a new VMS user.   ------------------------------   Date: 15 Sep 02 09:57:43 +0200) From: p_sture@elias.decus.ch (Paul Sture) . Subject: Re: Saveset distribution over the web) Message-ID: <X4K7DwALgIZR@elias.decus.ch>   c In article <1032044729.420698@ananke.eclipse.net.uk>, "Andy Proctor" <aproctor@hotmail.com> writes: K > Tom has helped me out by E mail here, I have a great solution now thanks!  >  > Andy > ( Care to share your new found knowledge?  __
 Paul Sture Switzerland    ------------------------------  % Date: Sun, 15 Sep 2002 08:48:28 +0100 = From: "Colin Butcher" <colinDOT.butcherAT@xdeltaDOT.coDOT.uk> . Subject: Re: Saveset distribution over the web@ Message-ID: <bXWg9.48199$rd2.45687@news-binary.blueyonder.co.uk>  I 1) Make a proper install kit - PCSI (preferred) or VMSINSTAL from all the  pieces you need to deliver 2) Test it thoroughly I 3) Compress it using ZIP on VMS with the "-V" option to preserve VMS file G attributes, or use the 'self expanding' mechanism as used for VMS patch , distribution (see Freeware CDs for ZIP etc.)% 4) Make it available for FTP download    -- Hope this helps. Cheers, Colin. ' (colinDOT.butcherAT@xdeltaDOT.coDOT.uk)     6 "Andy Proctor" <aproctor@hotmail.com> wrote in message/ news:1032040960.921194@ananke.eclipse.net.uk...  > All,L > I hope I could grab a little of your thoughts on web based distribution ofL > VMS Software. For our customers currently we ship them a DAT tape with theL > 90MB or so of our software (Runs on VMS 6.2-1h3 and above, for Alpha only)F > as 5 savesets on the tape, they are installed using VMSINSTAL. The 5K > savesets are created with the BACKUP utility as I guess you would expect, @ > and they are specific portions of the software, one is all the executables,H > one documents, one library routines etc etc. My questions is about web based K > distribution over the web. I would like to be able to put these up on our D > ftp site for download, but what would be the best format for this? > My thoughts are:I > 1. Put them up as they are after a binary ftp from my VMS server to our  ftp I > site (run with Win2k) and allow the customers to simply binary ftp them  fromL > us, and binary ftp to their system, then install..we would create a scriptG > to install the software. BUT would this corrupt the VMS format of the  files?K > 2. As above but use ZIP to ZIP the files prior to putting on the web..not @ > for any compression advantage, but to prevent file corruption. > L > What are your thoughts please? I know the .exe's should be ok on their own5 > anyway. Is there another, better way of doing this?  > - > Any comments welcome as it will help me! :) 	 > Regards  >  > Andy Proctor >  >    ------------------------------  # Date: Sun, 15 Sep 2002 15:01:39 GMT 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> . Subject: Re: Saveset distribution over the web' Message-ID: <3D84A64A.EAF49138@fsi.net>    JF Mezei wrote:  > P > You can simply provide a simple command procedure that uses SET FILE/ATTRIB to7 > set the transfered savesets to the proper attributes.  > O > Also, I am not sure if this still works, but I *think* at one point you could O > specify /BLOCKSIZE in the backup command and this essentially sets the record N > lenbth of the save set file. If you set it to 512, the  system receiving theP > file would create a binary file with fixed 512 attribute (for TYPE IMAGE file)P > and that would let BACKUP use the file out of the box. Haven't tried this in a > long time though.     The minimum /BLAOCKSIZE is 2048.  O > Digital often uses a utility that is self executable. That is transfered as a M > binary file, created by default as a 512 fixed file which is executable and J > the small program extracts itself into the backup saveset with the rightB > attributes. I believe this file now compresses its data as well.  G I have web pages explaining how to make self-extracting "ZIP" archives:   http://www.djesys.com/unzip.html4 http://www.djesys.com/vms/support/zipunzip/index.htm5 http://www.djesys.com/vms/support/zipunzip/sld047.htm    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sun, 15 Sep 2002 18:10:20 +0200 @ From: Arne =?ISO-8859-1?Q?Vajh=F8j?= <arne.vajhoej@mail.tele.dk>. Subject: Re: Saveset distribution over the web+ Message-ID: <3D84B0EC.9090803@mail.tele.dk>    Andy Proctor wrote:   >  > I hope I could grab a little of your thoughts on web based  distribution of E  > VMS Software. For our customers currently we ship them a DAT tape   with theH  > 90MB or so of our software (Runs on VMS 6.2-1h3 and above, for Alpha  only) G  > as 5 savesets on the tape, they are installed using VMSINSTAL. The 5 L  > savesets are created with the BACKUP utility as I guess you would expect,B  > and they are specific portions of the software, one is all the  executables,F  > one documents, one library routines etc etc. My questions is about 	 web based L  > distribution over the web. I would like to be able to put these up on ourE  > ftp site for download, but what would be the best format for this?   > My thoughts are: G  > 1. Put them up as they are after a binary ftp from my VMS server to   our ftp F  > site (run with Win2k) and allow the customers to simply binary ftp 	 them from G  > us, and binary ftp to their system, then install..we would create a   scriptI  > to install the software. BUT would this corrupt the VMS format of the   files?L  > 2. As above but use ZIP to ZIP the files prior to putting on the web..notA  > for any compression advantage, but to prevent file corruption.   >D  > What are your thoughts please? I know the .exe's should be ok on 	 their own 6  > anyway. Is there another, better way of doing this?  & ZIP is used by many with good results.   Backup save sets means trouble.    Arne   ------------------------------   Date: 15 Sep 02 09:55:38 +0200) From: p_sture@elias.decus.ch (Paul Sture) = Subject: Re: Seeking Senior VMS System Administrator Position ) Message-ID: <5JpO2frST7xo@elias.decus.ch>   o In article <ac4e3b24.0209141413.40c3f946@posting.google.com>, chinachowchow@mailcity.com (Rick Nickles) writes:  > Thanks so much!! > F > Please let me know if you have other ideas! As far as geography I amF > open.  I realize that I will probably need to relocate, contract, orG > work from home somehow.  There isn't much in this area.  I would like G > to make sure that wherever the work has a good work market so that if D > something happens I can find another place to go without uprooting > over and over. > 	 > thanks!  > Rick >   E But you still didn't answer the question. You never know, there could F be a potential employer close by, that you don't know about. I mentionI this because a few years ago I stumbled across a local job with a company B who I had assumed was pure IBM, but turned out to run VMS as well,; simply because I mentioned my location in a newsgroup post.     j > Kilgallen@SpamCop.net (Larry Kilgallen) wrote in message news:<Ewq47tZ5sOcT@eisner.encompasserve.org>...n >> In article <M7vg9.15992$Wa.914882@twister.southeast.rr.com>, "Kenneth Farmer" <kfarmer@openvms.org> writes:Q >> > You may want to post this message in the OpenVMS.org resume forum.  It can't 
 >> > hurt. >> >  / >> > http://www.openvms.org/phorum/list.php?f=3  >>  = >> It also would not hurt to indicate a geography preference.  >>     <snip> __
 Paul Sture Switzerland    ------------------------------    Date: 15 Sep 2002 20:03:51 +0800, From: Paul Repacholi <prep@prep.synonet.com> Subject: Re: Sybase and VMS - Message-ID: <878z23a28o.fsf@prep.synonet.com>   B "Antony Wardle" <antony.wardle@nospammmmm.optusnet.com.au> writes:  - > I hear that sybase and VMS isn't supported.     > anyone know if it works?  ; Well, if you look back, you may remember Boeing looking for C VMS/Sybase people to go to Geralton. Security clearance required...   / So I think we can put a tick next to `works' :)    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  + Date: Sun, 15 Sep 2002 15:59:44 +0000 (UTC) + From: david20@alpha1.mdx.ac.uk (David Webb)  Subject: Re: Sybase and VMS + Message-ID: <am2apg$rde$1@aquila.mdx.ac.uk>   [ In article <3D828E3E.208E6191@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:  >David Webb wrote: >>  ^ >> In article <3D8145AE.26854403@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes: >> >Arne Vajhj wrote: >> >>  >> >> Jakob Erber wrote: >> >> R >> >> > The ending support of sybase for VMS is the reason for us to leave VMS. WeS >> >> > cannot live with an unsupported DB and migration is easier than changing to  >> >> > oracle.  >> >> A >> >> I think that many has choosen to migrate from Sybase/VMS to , >> >> Sybase/non-VMS instead of Oracle/VMS ! >> >H >> >I forwarded this info. to Sue Skonetski, Cc: Mark Gorham and RichardB >> >Marcello with the subject line "More Endangered VMS accounts". >> > >> >-- >>  L >> Probably too late. Most sites running Sybase on VMS have probably alreadyL >> been forced to move. Our Library system was running Sybase based softwareK >> on VMS. It's now running the latest version of that software on Solaris.  > F >I know, but that's the "nature of the beast". By the time a post or a? >rpess release appears, the damage is done and difficult if not  >impossible to undo. >   M Sorry you misunderstand. This is too late because the decision to drop Sybase L on VMS and all the press releases were released over 2 years ago. I'm prettyF sure the dropping of Sybase support on VMS has featured in a number ofH previous posts on comp.os.vms much nearer the date when the decision was publicised.     D >All the more reason to keep a fire going under them to try to avoid. >losing the few ISVs and partners that remain. >   2 Won't disagree. VMS can't afford to lose any ISVs.     
 David Webb VMS and Unix team leader CCSS Middlesex University   >--  >David J. Dachtera >dba DJE Systems >http://www.djesys.com/  > ) >Unofficial Affordable OpenVMS Home Page:   >http://www.djesys.com/vms/soho/   ------------------------------  % Date: Sun, 15 Sep 2002 09:51:36 +0200 - From: Didier Morandi <Didier.Morandi@Free.fr> ) Subject: Re: TCPIP 5.3 bug in VAX install ' Message-ID: <3D843C05.28A07A6F@Free.fr>    JF Mezei wrote:  >  > TCPIP 5.3-18 > 9 > On VAX, it executes a procedure called cleanup_ucx.com.  > L > That procedure tests for the existence of 2 files. If neither exists, then > the procedure exists. C > If one exists, the procedure continues and assumes UCX is present  > H > 1- it defines SYS$OUTPUT ***AND SYS$ERROR *** to NL:, preventing errorL > messages to be sent back to the main process and displayed when a user has! > specified /LOG/TRACE. BAD BOY !   K No. A PCSI subprocess has no way to write to sys$output which just does not O exist unless it is specified (i.e. $ spawn/nowait/output=...). The PCSI utility P defines sys$output and sys$error to the null device to avoid errors generated by- DCL trying to write to a non-existent device.   K > 2- One of the two files checked is actually supplied by the TCPIP 5.3 kit K > [SYSEXE]UCX$UCP.EXE. The file date from march 2002, and contains 0 blocks O > (eg, just a place holder). Therefore, the IF statement is alwasy true and the K > procedure never exits and always assumes an old UCX kit was installed and  > needs to be deleted.  K These UCX files are placeholders as you said to ease TCP/IP v4 (aka UCX) to M TCP/IP v5 transition and should *never* be deleted. If you have a look at the D TCP/IP doc, you will find a paragraph talking about this in details.  N Looks like this test was wrongly coded. To me, it may be a bug which should beL reported to HP VMS Engineering (remember that these guys are not supposed to read this forum).   L > Because of the presence of ucx$ucp.exe , the cleanup$ucx.com continued andJ > tried to delete a whole bunch of old UCX files which never existed on my	 > system.  > M > What I do not quite understand is why the instal succeeded on my other node N > which was also upgraded from 5.0-9 to 5.3-18.  It also has the empty UXC$UCPK > file, so the cleanup_ucx.com would have also been called. Perhaps when it I > ran, it ended after having succesfully performed some action, hence the O > $STATUS would be OK, whilst on the second node, execution stopped perhaps due I > to an on error then exit of something, which would yield a $STATUS that  > indicated a failure.  9 Looks strange. Two identical systems with two behaviours?    D. --  2   ------------------------------------------------2 MORANDI Consultants  http://Didier.Morandi.Free.fr0   19 chemin de la Butte, 31400 Toulouse, France.2 Tel.: +33 (0)6 7983 6418 - Fax: +33 (0)5 6154 19282 OpenVMS, APPLE, Computer Security, Migration plans2 --------------------------------------------------   ------------------------------  % Date: Sun, 15 Sep 2002 05:12:22 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> ) Subject: Re: TCPIP 5.3 bug in VAX install , Message-ID: <3D844EF5.1C5E570A@videotron.ca>   Didier Morandi wrote: M > No. A PCSI subprocess has no way to write to sys$output which just does not Q > exist unless it is specified (i.e. $ spawn/nowait/output=...). The PCSI utility R > defines sys$output and sys$error to the null device to avoid errors generated by/ > DCL trying to write to a non-existent device.   J The develop manual (thanks for pointing it to me) states that the detachedG subprocess has its input and output "connected" to the PCSI process via N mailboxes. It is the PCSI process which filters all output from the subprocessL unless /TRACE is activated at which point only vms-style message are allowed to make it to the output.   M However, in the TCPIP installation files, the subprocess sprcifically defines Q sys$output and sys$error to NL: so that the master process has nothing to filter.     ; > Looks strange. Two identical systems with two behaviours?   N Gets worse. TCPIP 5.0 worked fine on my both vaxes. 5.3 works fine on my 3100.J But it fails on my microvax II (the TCPIP$INETACP process gets created butG crashes the system after having created the ethernet interface.) In all L fairness, I don't think it ever got started with the defined slip interface.L perhaps this is it. Hard to debug because there is no input or output to the ACP.    L Interestingly, I SET PROC/SUSP of the batch job once the ACP was created andB the ACP lived happily, and a few minutes later, when I resumed theM TCPIP$STARTUP batch job, INETACP crashed the system. I'll have to investigate  further when I wake up.    ------------------------------   Date: 15 Sep 2002 15:51:58 GMT/ From: Thomas Dickey <dickey@saltmine.radix.net> + Subject: Re: TPU Error using VMS from Linux * Message-ID: <am2aau$j87$2@news1.radix.net>  , Marty Kuhrt <kuhrt@encompasserve.org> wrote:x > In article <20020913103027.24191.qmail@nym.alias.net>, Doc.Cypher <Use-Author-Supplied-Address-Header@[127.1]> writes:5 >> On 13 Sep 2002, hoonig@ifrance.com (hoonig) wrote:  >>>Hi, >>> E >>>I'm trying to edit a file in OpenVMS. I'm connected to the OpenVMS 4 >>>server by telnet, using kConsole(xterm) in Linux. ...   H > Also, try issuing a $ SHOW TERM command when you get on the VMS systemG > and see if it is a known terminal type.  The Device_Type: field from  G > that output will let you know of the terminal is known.  If it is not C > known, then you can set it to a known terminal type by issuing a  G > $ SET TERM/DEV=VT100 command.  This should satisfy the %TPU-E-NONANSI E > problem, but you may still have problems with key mapping.  If so,  D > check out the recent thread on key mapping in comp.os.vms through  > Google newsgroup search.  H The key mapping stuff cited was for xterm (konsole has some customizable6 tables, but I've seldom seen anyone comment on those).  I XFree86 xterm seems to do the right thing w/o a lot of work.  Here's what  I see from show term:   > Terminal: _TNA40:     Device_Type: VT200_Series  Owner: DICKEYB Remote Port Info: Host: pcp741060pcs.reston01.va.comcas Port: 1087  B    Input:    9600     LFfill:  0      Width:  80      Parity: None6    Output:   9600     CRfill:  0      Page:   24         Terminal Characteristics: E    Interactive        Echo               Type_ahead         No Escape ?    Hostsync           TTsync             Lowercase          Tab D    Wrap               Scope              No Remote          EightbitC    Broadcast          No Readsync        No Form            Fulldup B    No Modem           No Local_echo      No Autobaud        HangupE    No Brdcstmbx       No DMA             No Altypeahd       Set_speed G    No Commsync        Line Editing       Overstrike editing No Fallback F    No Dialup          No Secure server   No Disconnect      No PasthruH    No Syspassword     No SIXEL Graphics  No Soft Characters Printer portI    Numeric Keypad     ANSI_CRT           No Regis           No Block_mode D    Advanced_video     Edit_mode          DEC_CRT            DEC_CRT2F    No DEC_CRT3        No DEC_CRT4        No DEC_CRT5        Ansi_Color    VMS Style Input  N (don't try to set konsole to vt200, etc., since it only aspires to be a vt100)   --  = Thomas E. Dickey <dickey@radix.net> <dickey@herndon4.his.com>  http://dickey.his.com  ftp://dickey.his.com   ------------------------------   Date: 15 Sep 2002 16:05:16 GMT/ From: Thomas Dickey <dickey@saltmine.radix.net> + Subject: Re: TPU Error using VMS from Linux * Message-ID: <am2b3s$k5m$1@news1.radix.net>  . JF Mezei <jfmezei.spamnot@videotron.ca> wrote: > hoonig wrote: F >> The problem is i cannot use edit/tpu under VMS with this terminal I >> have this VMS error:  >>  5 >> %TPU-E-NONANSICRT, SYS$INPUT must be supported CRT   P > $ SET TERMINAL/INQUIRE will cause VMS to ask your terminal to identify itself.M > If this isn't done, VMS assume you are a very basic TTY device, which can't  > support full screen editors.  J > if SET TERMINAL/INQUIRE fails, you can force a particular terminal type: > $SET TERM/DEVICE=VT300 > or > $SET TERM/DEVICE=VT200  M but in the case of konsole, it is doomed to failure, since konsole implements M at most about 2/3 of vt100 functionality (enough to run vi, but a frustrating ! experience is guaranteed on vms).   K > If, upon invocation of TPU, you get garbage, exit ( <CTRL-Z> ) and issue:  > 	SET TERM/NOEIGHTBIT  K > This will force TPU to use the slower 7 bit escape sequences. (some older T > connections were in 7 bit mode and 8 bit characters had their 8th bit snipped off)  K konsole doesn't implement that (at least not when I checked 5 minutes ago).O  9 The XFree86 xterm supports ANSI color and VT220 emulation: There's an faq atP1 	http://invisible-island.net/xterm/xterm.faq.htmlN! 	ftp://invisible-island.net/xtermU   see also# 	http://invisible-island.net/vttest " 	ftp://invisible-island.net/vttest   -- a= Thomas E. Dickey <dickey@radix.net> <dickey@herndon4.his.com>o http://dickey.his.com  ftp://dickey.his.com   ------------------------------    Date: 15 Sep 2002 19:42:12 +0800, From: Paul Repacholi <prep@prep.synonet.com> Subject: WTC cruft0 Message-ID: <87it17a38r.fsf_-_@prep.synonet.com>  / JF Mezei <jfmezei.spamnot@videotron.ca> writes:   E > As far as the WTC site, I will go as far as stating that a memorial > > must be built elsewhere. You don't want tens of thousands ofC > commuters leaving the Path or subway stations to be reminded that D > they are walking through the graves where 3000 people were crushedC > to death. You want the new site to be itself, not a shadow of the-? > old site, and you want to make it attractive to new tennants, ; > workers and shops. If you insist on leaving the site as a<D > rememberance of the castrastrophe, then you might as well make the@ > whole thing a park with the path and subways stations moved to( > ensure they are not in the grave site.  < With Battery Park oposite, that should be no problem at all.  # BTW, where does the name come from?    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda.S@                                              West Australia 6076. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  # Date: Sun, 15 Sep 2002 15:16:13 GMTn From: system@SendSpamHere.ORGT Subject: Re: WTC cruft0 Message-ID: <00A14031.C3DFD5C1@SendSpamHere.ORG>  _ In article <87it17a38r.fsf_-_@prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes:6 {...snip..}t$ >BTW, where does the name come from?  G Perhaps it's just a Noo Yo'kuh pronunciation of barratry, a name chosen5J to honor all of the shyster frivolous litigators in the NYC metro area. ;)  I Or, perhaps, it was just wishful thinking that Battery would be the only XG criminal offense practiced in the area saving homicide, sexual assault,cH robbery and myriad other popular criminal pastimes for other NYC parks,  streets and alleyways.   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COM$            .5   "Well my son, life is like a beanstalk, isn't it?" c   ------------------------------   End of INFO-VAX 2002.510 ************************