1 INFO-VAX	Sun, 02 May 2004	Volume 2004 : Issue 242       Contents:# Re: 11:05 PM Encompass Wake Up Call # Re: 11:05 PM Encompass Wake Up Call  Re: Asdk the Wizard ? 5 Re: Determining who in a cluster has the lock manager 5 Re: Determining who in a cluster has the lock manager 5 Re: Determining who in a cluster has the lock manager ( Re: Disk De-Frag utilities, suggestions?( Re: Disk De-Frag utilities, suggestions?) Garbled font in DECW Mail and root window > Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...> Re: How to turn linux into VMS - memory refresher for Dave ...$ Infoserver client on VMS 7.3 VAX?!?!( Re: Infoserver client on VMS 7.3 VAX?!?! Re: Newbye quetionJ Re: non-system but priviledged account OK for installing layered products?F non-system but priviledged account OK for installing layered products?P Re: non-system but priviledged account OK for installing layered products? produP Re: non-system but priviledged account OK for installing layered products? produP Re: non-system but priviledged account OK for installing layered products? produ+ Re: System organization preference question  Re: telnet in batch " Re: Tomcat / Java / process quotas) Re: VMS opportunities HP misses every day ) Re: VMS opportunities HP misses every day 9 Wanted: Various anonymous FTP servers for limited testing   F ----------------------------------------------------------------------  % Date: Sat, 01 May 2004 21:53:59 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>, Subject: Re: 11:05 PM Encompass Wake Up Call6 Message-ID: <409462C7.453B1B3A@NeOaSrPtAhMlNiOnWk.net>   tutor wrote: >  > Ok, > >    Did anyone else receive a call last night from ENCOMPASS? > ; > I received a 11:05 pm recorded message from Encompass....  > D > Not sure how they got my number. I've never paid for membership...
 > Oh wait.- > Maybe that's why they call AFTER 11 pm !!!!  > ? > If you have paid membership, they call during NORMAL BUSINESS  > hours????    Kenneth Farmer wrote:  > , > I'm a member.  That didn't seem to matter. > N > I got the call at around 11:00 PM EST.  Everyone was asleep.  I must admit II > was a little pissed about that.  I was in such a daze I forgot about it  > until I saw your message.  > " > 11:00 PM, who thunk that one up!   Robert Deininger wrote:  > B > I got the call at 11 PM also.  Asking me to take a survey, IIRC.   Tom Simpson wrote: > L > Yes.  I got the Encompass call at 11:20 pm Thursday night.  I was not real > thrilled, as I am & > an early-riser and was sound asleep.    G Suggestion: Cross-post to comp.decus.org as I have done with his reply.    (No - they didn't call me.)    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sat, 01 May 2004 22:02:15 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>, Subject: Re: 11:05 PM Encompass Wake Up Call6 Message-ID: <409464B7.F0BB548F@NeOaSrPtAhMlNiOnWk.net>   "David J. Dachtera" wrote: >  > tutor wrote: > >  > > Ok, @ > >    Did anyone else receive a call last night from ENCOMPASS? > > = > > I received a 11:05 pm recorded message from Encompass....  > > F > > Not sure how they got my number. I've never paid for membership... > > Oh wait./ > > Maybe that's why they call AFTER 11 pm !!!!  > > A > > If you have paid membership, they call during NORMAL BUSINESS 
 > > hours????  >  > Kenneth Farmer wrote:  > > . > > I'm a member.  That didn't seem to matter. > > P > > I got the call at around 11:00 PM EST.  Everyone was asleep.  I must admit IK > > was a little pissed about that.  I was in such a daze I forgot about it  > > until I saw your message.  > > $ > > 11:00 PM, who thunk that one up! >  > Robert Deininger wrote:  > > D > > I got the call at 11 PM also.  Asking me to take a survey, IIRC. >  > Tom Simpson wrote: > > N > > Yes.  I got the Encompass call at 11:20 pm Thursday night.  I was not real > > thrilled, as I am ( > > an early-riser and was sound asleep. > I > Suggestion: Cross-post to comp.decus.org as I have done with his reply.   # That's comp.org.decus, naturally...    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sat, 01 May 2004 21:39:06 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> Subject: Re: Asdk the Wizard ?6 Message-ID: <40945F4A.231C8098@NeOaSrPtAhMlNiOnWk.net>   Tom Linden wrote:  >  >   -----Original Message-----. >   From: Hoff Hoffman [mailto:hoff@hp.nospam]* >   Sent: Thursday, April 29, 2004 2:59 PM >   To: Info-VAX@Mvb.Saic.Com " >   Subject: Re: Asdk the Wizard ? > B >   In article <NDEMLKKEBOIFBMJLCECIGEIADCAA.tom@kednos.com>, "Tom$ >   Linden" <tom@kednos.com> writes:? >   :When a FAQ says to read discussions referenced by numbers, @ >   :how do you access those?  Certainly would nice if they were >   :links.  > & >     Yes, it certainly would be nice. > D >     Some folks still don't have operating systems with mouse-based, >     cut-and-paste support, after all.  :-) > 5 >     BTW, as for the FAQ, are you volunteering?  :-)  > @ > I wonder if you could make tool to parse the text and make URL > substitutions. > For example, from 8876:  > G >   Please also consider reading through at least some of the following  >   existing discussions:  > D >     (546), (2312), (2407), (2631), (2696), (3202), (3280), (5173),$ >     (6447), (7504), likely others. > H > upon encountering the comma delimited list of parentheses encapsulated	 > number, ) > extract the number and build the string * > "http://h71000.www7.hp.com/wiz_546.html"K > number make the substitution (after validating that the number is a valid  > reference) >  > (546)  for= > <A HREF="http://h71000.www7.hp.com/wiz_546.html"> (546)</A>  > M > in the text. Of course, this assumes that all the pages behave the same and N > that one can recognize the references.  How many pages are there?  Are thereM > any such entries  longer than 32K characters each (so I could put an entire  > page into a PL/I string)?   G I though XML was supposed to be the "thingamabob to do the job" in such @ cases? Create your own tag, and then when that tagged texts gets+ rendered, insert the appropriate HTML link.   H I guy I once worked for used to talk about "how many angels can you haveD dancing on the head of pin". I never understood what he meant, but I. think his meaning might apply to XML. Dunno...   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sun, 02 May 2004 01:31:20 +0800 , From: Paul Repacholi <prep@prep.synonet.com>> Subject: Re: Determining who in a cluster has the lock manager- Message-ID: <87brl8awbb.fsf@prep.synonet.com>   1 rcyoung@aliconsultants.com (Robert Young) writes:   F > Some databases on VMS utilize the lock manager in a cluster setting.E > If the manager migrates to a "less powerful" node, the database can % > suffer severe performance problems.   D > I am wondering how to tell which node is the "current home" of theE > cluster lock manager so that if performance problems arise, one can D > readily "eliminate" the lock manager migration issue from the list% > of possible sources of the problem.   F > I believe that by setting PE1 in sysgen to "0" it will "prevent" theD > manager from migrating, but is there another way to tell "where it3 > is now" if migration is allowed, or has occurred?   K There is no node running the lock manager, they ALL run it. What can happen F is that the lock tree you are interested in can get mastered on a nodeJ that does just enough locking to keep a hold of the tree, when you really 1 want it to move to the main users of those locks.   F You can look at a RSB and LKB and see which node holds the master copy with SDA or the like.   B In ther freeware stuff you should be able to find a copy of KeithsC stuff on this, and the talks on how to get lock trees to re-master.  May only be in an older one.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  $ Date: Sat, 1 May 2004 20:25:47 +0200, From: "Hans Vlems" <hvlems.dotweg@zonnet.nl>> Subject: Re: Determining who in a cluster has the lock manager9 Message-ID: <c70q77$gv96s$1@ID-143435.news.uni-berlin.de>   > "Robert Young" <rcyoung@aliconsultants.com> schreef in bericht7 news:91437ce6.0404301650.62079150@posting.google.com... F > Some databases on VMS utilize the lock manager in a cluster setting.E > If the manager migrates to a "less powerful" node, the database can % > suffer severe performance problems.  > E > I am wondering  how to tell which node is the "current home" of the E > cluster lock manager so that if performance problems arise, one can G > readily "eliminate" the lock manager migration issue from the list of " > possible sources of the problem. > F > I believe that by setting PE1 in sysgen to "0" it will "prevent" theG > manager from migrating, but is there another way to tell "where it is 0 > now" if migration is allowed, or has occurred?  B LOCKDIRWT will tell you whether a node has more chance of owning a
 particularK lock. Higher values indicate a, errr, reluctancy to release lock ownership. 7 There's no such thing as one lock owner in the cluster.  Hans   ------------------------------   Date: 1 May 2004 17:59:03 -0700 / From: rcyoung@aliconsultants.com (Robert Young) > Subject: Re: Determining who in a cluster has the lock manager= Message-ID: <91437ce6.0405011659.48b90d88@posting.google.com>   F Thanks to veryone for helping me out on this. I suppose more properly,> I should have asked how can you tell which node is handling  aB specific  lock. However, I understand what you are saying.  Thanks again    ------------------------------  % Date: Sat, 01 May 2004 12:39:13 -0700 + From: "Barry Treahy, Jr." <Treahy@MMaz.com> 1 Subject: Re: Disk De-Frag utilities, suggestions? ' Message-ID: <4093FCE0.9060301@MMaz.com>    Beach Runner wrote:   H > If you go to 7.3-1 and use XFC (requires patches) or 7.3-2 disk cache  > programs become ( > obsolete, and it's free and supported. >  > Just set vcc_flags =2   F Disk Cache and Disk Defragmentation software are not one in the same, @ and their use is not exclusive, but rather mutually beneficial..   Barry    --    > Barry Treahy, Jr                       E-mail: Treahy@MMaz.com> Midwest Microwave                          Phone: 480/314-1320> Vice President & CIO                         FAX: 480/661-7028                            ------------------------------  % Date: Sat, 01 May 2004 21:23:31 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>1 Subject: Re: Disk De-Frag utilities, suggestions? 5 Message-ID: <40945BA3.3B64EF9@NeOaSrPtAhMlNiOnWk.net>    "Barry Treahy, Jr." wrote: >  > Beach Runner wrote:  > I > > If you go to 7.3-1 and use XFC (requires patches) or 7.3-2 disk cache  > > programs become * > > obsolete, and it's free and supported. > >  > > Just set vcc_flags =2  > G > Disk Cache and Disk Defragmentation software are not one in the same, B > and their use is not exclusive, but rather mutually beneficial..  A Yes. Where I'm at now, caching is what keeps us alive, to a large 1 extent. VCC on VMS, writeback cache on the HSx's.   > On the backend systems, freespace fragmentation tends to be anA irritation. I do have one Radnet machine that does file re-org.'s E regularly and it suffers both file and freespace fragmentation. Files F get fragmented as they extend, new files are created fragmented due to freespace fragmentation.   A never-ending battle...   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  # Date: Sun, 02 May 2004 00:00:36 GMT ( From: Alder <PGDEHMKOKIMD@spammotel.com>2 Subject: Garbled font in DECW Mail and root window* Message-ID: <ESWkc.4329$LA4.1898@edtnps84>   VMS 7.3 with DECWindows 1-2.6   I I run the CDE window system remotely using an X server on a PC.  Lately,  H I've made a number of configuration changes to various DECW$*.DAT files E and created some display problems (annoyances, really) I can't trace.   C I now have garbled text in a couple of places.  By garbled, I mean  > portions of almost all the letters are the same colour as the < background, giving the letters a "spilled rice grains" look.  H I see this effect in the text displayed in the blue window that appears G after login (i.e., "Starting the New Desktop for OpenVMS"), and in the  F DECwindows Mail application.  In the latter case, the garbled text is B confined to folder names in the upper pane under "Folders in MAIL < directory".  The listing of mails in the lower pane is fine.  6 I'd appreciate any advise on how to troubleshoot this.   Thanks,    Alder    ------------------------------  % Date: Sun, 02 May 2004 01:24:06 +0800 , From: Paul Repacholi <prep@prep.synonet.com>G Subject: Re: How to turn linux into VMS - memory refresher for Dave ... - Message-ID: <87fzakawnd.fsf@prep.synonet.com>   0 jealousxmp@aol.commonplace (jealous xmp) writes:  > >>wrong ... VMS has all the services that everything else has,> >>telnet, ethernet, SSH, webservers, FTP ... a buffer overflow> >>is a buffer overflow, only on VMS these exploits are twarted: >>with that nasty "ACCESS VIOLATION" error time after time  E > There are things like W^X, PAX, gr-security, SE linux, stack guard, D > format guard, and more that are helping linux and bsd.  With unix,C > there are trusted systems like Trusted HP-UX and Trusted Solaris. F > Unix can also use chrooting which helps a bit.  And trusted variants1 > mean the local root exploits are not as useful.   B Your `Trusted <foo>' unix variants get you to almost the point a aF standard do nothing VMS new install. Any VMS user is almost `chrooted'$ as a result of the different design.  A > As I understand it, the usual stack overflows won't work on VMS E > because of the non-executable stack.  Not sure about heap overflows A > or format bugs.  I've seen a DCL script exploited on a web site  > before though.  D > There is/was a hardened (trusted even?) form of VMS.  I don't know> > the details and how this compares to some of the trusted and > hardened unices.  9 SEVMS adds nondiscressional access controls and MKS stuff B mainly. Expensive and ugly to live with unless you really have to.  @ The biggest difference is if you go to a trusted foo system, youE lose code and read-only-data sharing, so as to prevent covert channel A signaling. So your memory requirment will go up, and can send you C into heavy paging. VMS has code segments set no read/exececute only $ so it is not impacted quite so much.  C Stack overflows etc `work', but you don't gain anything you already F have. Rather than busting into a root process, you get the last chance% error handler and a big dose of FOAD.      --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  * Date: Sat, 1 May 2004 17:45:32 +0000 (UTC) From: david20@alpha2.mdx.ac.ukG Subject: Re: How to turn linux into VMS - memory refresher for Dave ... ) Message-ID: <c70nns$bp4$1@news.mdx.ac.uk>   Y In article <c70d6q$4vp$1@news.cybercity.dk>, "Karsten Nyblad" <nospam@nospam.com> writes:  > , ><david20@alpha2.mdx.ac.uk> wrote in message$ >news:c6r9k8$gmb$1@news.mdx.ac.uk...I >> Independently of their defintions the virus companies tend to call any  >> malware a virus.  > J >True, however, I think the viruses at large were code attaching itself to> >DOS programs.  Macro viruses in, e.g., Word files came later. > F >> However your definition of a virus (by example of getting a user to >installH >> and run a DCL command procedure which appends code to any DCL commandH >> procedure the user has write access to and then requiring the user to >lowerL >> permissions and tell his friends to run it - in order for it to spread to3 >> other accounts) is a totally useless definition. M >> Such a "virus" could be written for any computer system which had either a L >> scripting language or compiler and which allowed the user to grant others >> access to his files.  >>H >I get the impression that you think that the victims know that they runM >viral code.  Of course they don't.  The first victim on the system will copy K >a command procedure from, e.g., the Internet, because it seems like it can G >do something that the victim wants done.  He does not realise that the M >command procedure is infected with a virus.  He tries it and all his command I >procedures become infected, including those used by friends, colleagues, J >etc.   All this has happened without the victim realising that he has letL >malicious code at large on the system.  The friends and colleagues continueH >to use the now infected command procedures of the first victim.  If theI >friends and colleagues make command procedures available to others, then % >those others may become victims too.  > E >Such a virus can infect quite a number of user accounts in the right K >environment of users with some but limited computer knowledge.  I remember M >reading abort two professors discussing whether og not one of the professors M >could attack the other professors Unix account.  The first professor place a K >virus infected copy of ls in a directory, that was on the other professors H >path.  The other professor noticed that ls was a little bit slower thanM >usual and realised what had happened.  The virus had already made a to digit 3 >number of copies and was spreading to other users.  >   M Placing a user written program with the name of a system command like ls in a N directory such as the path of another user (preferably root)is an old standardO on Unix. Also once a hacker has gained root access they usually replace lots of H the system commands such as ls with their own versions. This activity is. generally known as planting trojaned programs.H All I'm arguing with is your defintion of a virus. I do not consider anyO program which a user has to explicitly run to be a virus. I would only consider M it a virus if any human action needed to spread it was otherwise innocuous eg K opening a file, putting a floppy in a drive. Explicitly running code is not P innocuous. Tricking someone into explicitly running code I would consider to be ' tricking someone into running a Trojan.   F Your definition appears to be that if a program makes copies of itselfL irrespective of how stupid an action a user needs to take in order for it to spread then it is a virus.  I Which would mean that you would consider the "Irish Virus" I mentioned in N another post to be a virus. The user would have to be really really stupid to H follow it's "honour system" instructions but if they did it would indeed spread.     
 David Webb VMS and Unix team leader CCSS Middlesex University         >Karsten Nyblad   >ibpit1202 at sneakemail dot com >  >    ------------------------------  % Date: Sat, 01 May 2004 21:16:26 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>G Subject: Re: How to turn linux into VMS - memory refresher for Dave ... 6 Message-ID: <409459FA.A26A087D@NeOaSrPtAhMlNiOnWk.net>   Glenn Everhart wrote:  > F > I suspect many of our younger colleagues in IT shops have no conceptB > what running on a secure platform is like, and that they must beD > taught lessons from the ground up. Notice the occasional trolls byA > folks who know Windows or some unixoid flavors, but are used to A > being surrounded by systems where most every process is running  > in highly privileged states. >  > [snip] > E > Those who recall V4.0 will recall the numerous brand new holes that C > got poked in it at once, which led to more rewriting/redesign and A > impressive security strength. VMS got strong that way by having C > been to the wars. Look hard and some of the old bullet wounds are  > visible...   <rant>G What gets me is how little the ISVs seem to know about VMS security. It H almost seems an annoyance. They'll tell you that (such and such) processA needs (x) privilege, when what really needs to be done is to have @ volume, directory and file "permissions" (UN*X term, I know) set	 properly.   F There is RARELY a need for a general user to hold CMKRNL privilege, orD SYSPRV, not to mention others, but I can't tell you how many of them> actually do just to eliminate "insufficent privilege or object protection error" and others.   + Even Cerner is not immune to this syndrome.   G If IT auditors were at all clueful in this area, HIPAA compliance would G be many times the headache it is now. As it is, they seem to think that G password lifetimes and password change dates are all that is important.  </rant>    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sat, 01 May 2004 21:19:06 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>G Subject: Re: How to turn linux into VMS - memory refresher for Dave ... 6 Message-ID: <40945A9A.7DCC34C4@NeOaSrPtAhMlNiOnWk.net>   Nic Clews wrote: >  > Paul Sture wrote:  > >  > > Karsten Nyblad wrote: L > > > "Andrew Harrison" <andrew_._remove_harrison@su_n.com> wrote in message0 > > > news:c6ntl3$nvf$1@new-usenet.uk.sun.com... > > > A > > >>So far in this thread we have had Bob K making claims about > > > >>UNIX authentication which are incorrect and Bob C makingA > > >>claims about the relatie Security of OpenVMS vs UNIX namely ? > > >>that OpenVMS cannot be infected with virii which are also  > > >>incorrect. > ...  > > L > > Oh, but as I remember it, VMS did have security logs with V4.0. Not diskK > > logs as they later became, because the concept then was to log messages  > > to a hardcopy terminal.  > J > Actually I know a couple of sites that up until quite recently still ranE > with the LA120 paper consoles, religiously stowing away the logs as 
 > "evidence".  > D > Version 3? version 2? Not sure when the systems were installed but9 > Paul's right that the "electronic" log is a moot point.  > H > We're back in the days when VMS systems were hackers (easier) targets,I > 1200 baud modems were starting to become commonplace, and is of the era F > that the current phreak information is based. Windows were something > that let the Sun in.  C As recently as summer of 2000, I was at a shop where they STILL had C VAXes (no production Alphas) using LA120s for OPCOM terminals - the D consoles were connected to a PCM box (that was the only Alpha in the shop).   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  * Date: Sat, 1 May 2004 19:05:57 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)- Subject: Infoserver client on VMS 7.3 VAX?!?! $ Message-ID: <c70sel$s7j$1@online.de>  ) The following products will be processed:   
   ESS V1.0    3         Beginning installation of ESS V1.0 at 21:03   6 %VMSINSTAL-I-RESTORE, Restoring product save set A ...F %VMSINSTAL-I-RELMOVED, Product's release notes have been moved to SYS$ HELP. : * Do you want to run the IVP after the installation [YES]?D %VMSINSTAL-E-EXPIRED, This product has expired after VMS version 5.3I %VMSINSTAL-E-EXPIRED, Please contact your local sales representative for   this pr  oduct.> %ESS-E-BADVMS, This kit requires VMS versions V5.0 thru V5.3-2> %VMSINSTAL-E-INSFAIL, The installation of ESS V1.0 has failed.  D What am I missing?  This product comes on the latest layered-productC CDs.  Can I really not install it?  (I'm assuming it is covered by   the hobbyist license.)   ------------------------------  % Date: Sat, 01 May 2004 21:58:36 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>1 Subject: Re: Infoserver client on VMS 7.3 VAX?!?! 6 Message-ID: <409463DC.78629463@NeOaSrPtAhMlNiOnWk.net>  / Phillip Helbig---remove CLOTHES to reply wrote:  > + > The following products will be processed:  >  >   ESS V1.0 > 5 >         Beginning installation of ESS V1.0 at 21:03  > 8 > %VMSINSTAL-I-RESTORE, Restoring product save set A ...H > %VMSINSTAL-I-RELMOVED, Product's release notes have been moved to SYS$ > HELP. < > * Do you want to run the IVP after the installation [YES]?F > %VMSINSTAL-E-EXPIRED, This product has expired after VMS version 5.3J > %VMSINSTAL-E-EXPIRED, Please contact your local sales representative for	 > this pr  > oduct.@ > %ESS-E-BADVMS, This kit requires VMS versions V5.0 thru V5.3-2@ > %VMSINSTAL-E-INSFAIL, The installation of ESS V1.0 has failed. > F > What am I missing?  This product comes on the latest layered-productD > CDs.  Can I really not install it?  (I'm assuming it is covered by > the hobbyist license.)  G You shouldn't need to install it, AFAIK; it is now bundled with VMS, if  I'm not too badly mistaken.    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sat, 01 May 2004 21:29:18 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> Subject: Re: Newbye quetion 6 Message-ID: <40945CFE.FA4FCB67@NeOaSrPtAhMlNiOnWk.net>   John Dziurlaj wrote: >  > "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> wrote in message news:<40904FE4.EBE54270@NeOaSrPtAhMlNiOnWk.net>...# > > jose luis fernandez diaz wrote:  > > > 	 > > > Hi,  > > > ; > > > Can anyone give a the definition and the relation of:  > > >  > > > - OSF  > > > - OpenVMS 
 > > > - Tru64 
 > > > - Alpha  > >  > > Homework question? > > < > > Well, at least OpenVMS is mentioned. That's some hope...@ > I am not so sure on how much hope we have anymore. The word inF > academia is that Linux is *IT*. Any and all other operating systems,F > like BSD and VMS are shunned, for fear that if other CS majors writeC > software for anything else, that resources will become to watered F > down, and thus hurt the great *LINUX* superiority. Most believe that$ > this stuff is old hat. Sad. [snip]  > Rather sounds like my philosophy: One reason why I've resistedE cross-training on M$, UN*X etc. is the concept of skills dilution. By > VMS standards (DCL, RMS, security, etc.) UN*X and M$ are quiteH primitive. By having to deal with stone axes and carved-wood spears, oneA quickly forgets how to use power tools and other highly-developed  refinements.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  * Date: Sat, 1 May 2004 19:00:49 +0000 (UTC)% From: Dan Foster <usenet@evilphb.org> S Subject: Re: non-system but priviledged account OK for installing layered products? 6 Message-ID: <slrnc97sul.9er.usenet@gaia.roc2.gblx.net>  G In article <c70rpe$rb1$1@online.de>, Phillip Helbig---remove CLOTHES to . reply <helbig@astro.multiCLOTHESvax.de> wrote:G > The documentation on the layered-product CDs talks about "you must be G > logged in to the system account".  Presumably, an account with enough B > privs would be OK as well, or is there the possibility that someA > installed files will have that account as the owner, instead of 	 > system?   G Well, it isn't necessarily just a concern with privileges... sometimes, D some install procedures may be interested in ensuring it has the the) (process) quotas afforded to user SYSTEM.   , So to answer your question is: "it depends."  A It will probably usually work ok when trying it from a privileged F non-SYSTEM account. If it doesn't, you may need to login as SYSTEM and retry the installation.    -Dan   ------------------------------  * Date: Sat, 1 May 2004 18:54:38 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)O Subject: non-system but priviledged account OK for installing layered products? $ Message-ID: <c70rpe$rb1$1@online.de>  F The documentation on the layered-product CDs talks about "you must be F logged in to the system account".  Presumably, an account with enough A privs would be OK as well, or is there the possibility that some  G installed files will have that account as the owner, instead of system?    ------------------------------  # Date: Sat, 01 May 2004 19:48:47 GMT 1 From: Michael Austin <maustin@firstdbasource.com> Y Subject: Re: non-system but priviledged account OK for installing layered products? produ 2 Message-ID: <4093FF1D.CE2E464D@firstdbasource.com>  F it also is depending on the layered product you are trying to install.G IBM's MQSeries will ONLY install, startup and/or shutdown with the user 2 SYSTEM. MQSeries is much like the old DECMessageQ.  D There is a way to actually become system using an old program calledI SETUSER.  It came in handy when maintaining very large numbers of systems I and accounts.  It was available on both VAX and Alpha.  The Alpha version F actually changed all of your logicals (like sys$login), so I used thisJ command proc to be able to return to my original username.  At this point,A I don't recall where I got the setuser kit... but it is out there  somewhere...  " $mcr sysuser:[maustin]setuser 'p1' $x = f$getsyi("nodename") C $y = f$extract(0,7,f$edit(f$getjpi("","USERNAME"),"TRIM,COMPRESS"))  $z =J f$extract(f$length(f$directory())-8,f$length(f$directory()),f$directory())/ $       terminal = f$getjpi(0,"TERMINAL") - ":" , $       set process /name="''y'-''terminal'" $set prompt="''x'>''y'>''z'$ "  $ It also does several other things...5 1) change the process name (not really important) and ! 2 changes the prompt so you know:       a) what node you are on      b) what your username is .      c) last 8 characters of your directory...
      example:   ( ALPHA1>SYSTEM>[SYSMGR]$ show sym setuser)   SETUSER == "@SYSUSER:[maustin]Setuser " ' ALPHA1>MAUSTIN>MAUSTIN]$ setuser system  Username:  SYSTEM  Directory: SYS$SYSROOT:[SYSMGR]  UIC:       [1,4] Account:   SYSTEM   I ALPHA1>SYSTEM>[SYSMGR]$@login     !! pick up any additional user-specific  logicals and/or symbols.      / Phillip Helbig---remove CLOTHES to reply wrote:   G > The documentation on the layered-product CDs talks about "you must be G > logged in to the system account".  Presumably, an account with enough B > privs would be OK as well, or is there the possibility that someI > installed files will have that account as the owner, instead of system?    ------------------------------  % Date: Sun, 02 May 2004 01:50:58 +0200 9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> Y Subject: Re: non-system but priviledged account OK for installing layered products? produ ' Message-ID: <409437E2.58E99373@aaa.com>    Michael Austin wrote:  > F > There is a way to actually become system using an old program called
 > SETUSER.   Or even better (IMHO) :   < http://vms.process.com/scripts/fileserv/fileserv.com?HGLOGIN  ) Creates a virt term, then attach to that. / Runs all system and process login COMs, so your 1 process is just as if actualy logged in using the  username: prompt.   ' You need a bunch of privs, of course...   	 Jan-Erik.    ------------------------------  % Date: Sat, 01 May 2004 21:57:06 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>Y Subject: Re: non-system but priviledged account OK for installing layered products? produ 6 Message-ID: <40946382.84E5FE7B@NeOaSrPtAhMlNiOnWk.net>   Michael Austin wrote:  > H > it also is depending on the layered product you are trying to install.I > IBM's MQSeries will ONLY install, startup and/or shutdown with the user14 > SYSTEM. MQSeries is much like the old DECMessageQ.  > I was given to understand that MQseries *IS* DECmessageq, someF generations removed, just as SLS is descended from an early version of TapeSys.   -- l David J. Dachterai dba DJE Systemsi http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sun, 02 May 2004 07:35:18 +0200o* From: Paul Sture <nospam@sture.homeip.net>4 Subject: Re: System organization preference question9 Message-ID: <c721ao$h7p3j$1@ID-132135.news.uni-berlin.de>    Main, Kerry wrote: >>-----Original Message-----4 >>From: Paul Sture [mailto:nospam@sture.homeip.net]  >>Sent: April 30, 2004 10:17 PMb >>To: Info-VAX@Mvb.Saic.Com 6 >>Subject: Re: System organization preference question >> >>David M Smith wrote: >>4 >>>On Thu, 29 Apr 2004 13:21:44 -0700, Jack Fortune  >> >><jfortune@uoregon.edu> wrote:o >> >>>o@ >>>>While I'm sure there is no _right_ or _wrong_ way to manage  >> >>a cluster, >>6 >>>>I'm curious as to what approach others have taken. >>>>B >>>>The main differences concern the use of the SYSTEM accound andG >>>>SYS$MANAGER directory. At my former job, I hardly ever logged in aseH >>>>SYSTEM and never ran any batch jobs under the SYSTEM username. Also,? >>>>we did not put any site-specific command procedures in any - >>
 >>SYS$MANAGER- >>F >>>>directory (except for SYSTARTUP_VMS.COM, SYPAGSWPFILES.COM ,etc.). >>>>B >>>>Here, there are many batch jobs running under SYSTEM, and mostE >>>>site-specific command files used to run the systems are typicallyS >>>>located in SYS$MANAGER.  >>>>- >>>>How have others configured their cluster?  >>>n >>>f >  > [snip ..]  >  > ; >>On the subject of knowing what is "ours", I've worked at   >>several places oF >>where production stuff includes a short abbreviation of the company H >>name, and I've found that very useful. For example, if the company is = >>call AB Corp, I'd have logicals and associated directories w >>called ABC_*. G >>Good naming conventions for logicals, directories and procedures can := >>also save a lot of time when it comes to problem diagnosis.e >>= >>I also make sure that "key" startup stuff is on the system i >>disk so that a@ >>I can get at it from a minimal (vs standalone, but standalone  >>as well)  A >>boot. The definition of "key" here is obviously site dependant.  >> >  >  >  > Paul,  > " > Re: company names in logicals .. > F > Fwiw, in this day of mergers, acquisitions and name changes, I wouldI > likely stay away from company names in logicals and any types of thingseG > like directories. I now have seen internal systems with DEC, Digital,D? > Compaq, CPQ, HP etc references and directories all over them.A >r  F A good point Kerry, but the aim here is to provide some naming scheme / which makes "what is ours" easily identifiable.2    H > My preference these days is to keep everything as rock solid simple asF > possible while at the same time allowing one to grow the std easily.H > System names like SYS001, SYS002...SYSxxx etc, disk labels like APP001 > e.g. disk$app001:[HR] etc. >   H LOL, SYSnnn is a naming scheme I have used for system, layered software  and pagefile disks.s  I > Its not as exotic as some naming stds, but imho, simplifying IT stds aseI > much as possible will have long term benefits. Server HW, company namesRC > and OS versions can change, but the IT resource naming std should + > continue with as few changes as possible.t >   
 Ayup KISS.  G > Course, there is no better way to start a long discussion than to trya4 > and resolve which is the best resource naming std. >    I'll shut up now :-)   ------------------------------  % Date: Sat, 01 May 2004 21:32:03 -0500a@ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> Subject: Re: telnet in batch5 Message-ID: <40945DA3.D4CA862@NeOaSrPtAhMlNiOnWk.net>e   Frank da Cruz wrote: > G > On 2004-04-29, norm.raphael@metso.com <norm.raphael@metso.com> wrote:s > :iJ > : When I try to access ftp://kermit.columbia.edu with anonymous login or > : click the appropriaten> > : download from the decvms binaries I am not allowed access. > : What will fix that?n > :lE > You are probably behind a firewall that's blocking ftp connections..F > The Kermit ftp site has been serving Kermit software to the InternetE > (and before that, the ARPANET) for 20 years, but everything is more2G > complicated now, what with firewalls, NATs, VPNs, and whatnot.  I canW/ > assure you that the ftp site works just fine.D > @ > The first place you might want to look is in your PC's network? > configuration.  Do you have some kind of "personal firewall"?h  C More than likely, his IP address as it appears at Columbia does notoF back-translate; so, the server rejects the connection. In that aspect,E his case would likely be the rule rather than the exception, which ism unfortunate.   --   David J. Dachteras dba DJE Systemso http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/n   ------------------------------  % Date: Sat, 01 May 2004 21:42:52 -0500l@ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>+ Subject: Re: Tomcat / Java / process quotas 6 Message-ID: <4094602C.46541E29@NeOaSrPtAhMlNiOnWk.net>   "Main, Kerry" wrote: >  > > -----Original Message-----5 > > From: Chuck Chopp [mailto:ChuckChopp@rtfmcsi.com]e! > > Sent: April 28, 2004 10:52 PMe > > To: Info-VAX@Mvb.Saic.Com + > > Subject: Tomcat / Java / process quotas2 > >4< > > I've looked over the FAQ for Java on OpenVMS Alpha and I > > don't see any.I > > recommendations regarding how to properly estimate the process quotast? > > [bytlim, pgflquo & fillm in particular] that need to be sett > > for the accountmG > > that runs the Tomcat server.  I'm trying to run some sample servlet A > > applications, but after the initial JSPs are displayed and it: > > is time toA > > interact with the actual servlet itself, I'm finding that thea > > servlet simply@ > > dies and I get a 503 error from Tomcat.  I'm not sure if the > > sample servletI > > is simply broken [e.g. coded wrong] or if some process quota is beingsA > > exceeded.  The Tomcat logs directory has log files in it, butX > > they are empty/ > > so there's nothing of use there to look at.? > >t9 > > Is there any sort of additional information regardingr > > analyzing the actual= > > servlet code, combined with the estimated # of concurrent  > > client sessionsa? > > that would allow me to more accurately estimate the process, > > quota valuesA > > that will be needed for Tomcat to run my servlet application?b > > = > > I am using Java v1.3.1-17 & the CSWS_JAVA v2.1 on OpenVMSM > > Alpha v7.3-1 [with$ > > all required patches installed]. > >- > >- > > TIA, > > 	 > > Chuck  > > -- > > Chuck Choppm > >l< > > ChuckChopp (at) rtfmcsi (dot) com http://www.rtfmcsi.com > > D > > RTFM Consulting Services Inc.     864 801 2795 voice & voicemail6 > > 103 Autumn Hill Road              864 801 2774 fax > > Greer, SC  29651 > >o0 > > Do not send me unsolicited commercial email. > >  >  > Chuck, > I > Fwiw, I prefer using the free Availability Manager utility for not onlyaH > monitoring process quota's, but also dynamically updating them in realH > time. That is something one can not typically do with the numerous DCL > programs kicking around. >  > Reference:@ > http://h71000.www7.hp.com/openvms/products/availman/index.html  F Well, the deal with the DCL proc.'s is not so much about monitoring as it is about notification.g  + One of those items "that Digital forgot..."    -- s David J. Dachterai dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/h   ------------------------------  % Date: Sat, 01 May 2004 21:49:28 -0500t@ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>2 Subject: Re: VMS opportunities HP misses every day6 Message-ID: <409461B8.84DA1657@NeOaSrPtAhMlNiOnWk.net>   John Smith wrote:A > [snip]  < E-mail me privately about rights to use this stuff, will ya?   -- t David J. Dachteraa dba DJE Systemso http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sun, 02 May 2004 07:55:54 +0200 * From: Paul Sture <nospam@sture.homeip.net>2 Subject: Re: VMS opportunities HP misses every day9 Message-ID: <c722hc$hdhu4$1@ID-132135.news.uni-berlin.de>s   John Smith wrote:C   <snip for brevity>   > 	 > <TV ad>s > H > Multiple tekkie types going nuts downloading and patching server after	 > server.2 > H > Big Boss talking to IT head: "This is crazy. It's costing us a fortuneH > having all these people patching everything all there time. Our serverL > infrastructure is killing our business. I don't care what it takes, fix it	 > now!!!"r >  > 2 weeks later....0 > L > Big Boss walks into glass house. Server farm is gone. One rack standing. AE > few tech guys playing frisbie in the empty space in the glass houseu > I > Big Boss to IT head: "What's going on? Microsoft fix up it's act, huh?"w > H > IT head: "Actually no. We fixed our server infrastructure permanently.N > Reliability is way up, patches are non-existent; viruses & worms are a thingL > of the past. We laid off 50% of our staff. All of our web and mail serversM > have been switched over to OpenVMS from HP. All our file servers too. Piece" > of cake."e > I > Big Boss: "Great work, Griswold. You can expect a big bonus this year."r > K > Voiceover: "OpenVMS from HP has been providing enterprise-class security,mN > throughput, and reliability for the world's largest corporations for over 25; > years. Now available at a price every company can afford.e > = > Call 1-800-OPENVMS or visit us online at www.hp.com/openvms-H > (also www.hp.com/reliability , www.hp.com/security , www.hp.com/24x7 ) > 
 > </TV ad>, > (c) 2004, John Smith. All Rights Reserved. >  >  > 	 > <TV ad>-  @ Here's some raw material for your imagination to work with John.  H And to lend it all some official credibility, "As Per Recommendation of # Her Majesty's Secret Services, MI5"h  ( http://www.mi5.gov.uk/output/Page55.html   "ELECTRONIC ATTACK   ...h Malicious software  G The techniques and effects of malicious software (e.g. viruses, worms,  G Trojans) are as variable as they are widely known.  The growing use of eI email, of interconnectivity between systems, of external contractors and wD of remote access (i.e. for home-working) mean that virus infections I spread ever more widely and rapidly.  Most dangerous are those concealed  B in legitimate software or inserted into systems during production.  E <Not a problem if you have VMS. James Bond references as appropriate>a  " Malicious modification of hardware  F Computer hardware can be modified so as to mount or permit an attack. E This is normally done at the point of manufacture or supply prior to eF installation, though it could also be done during maintenance visits. H The purpose of such modifications would be to allow a subsequent attack I to be made, possibly by remote activation.  The level and sophistication >E of the attack would depend partly on the location of manufacture and 5I supply; the threat is greater from those systems made, supplied from, or e maintained from overseas.     E <All the more reason to pick a supplier who has a longstanding track  8 record of rigorous testing before products are released>  % <Ed: Ooh, can we say offshoring? ;-)>  ...k   WHAT YOU CAN DOr  D Each attack is particular.  Therefore advance advice is necessarily  general and procedural.a  G      * Acquire your systems from reputable manufacturers and suppliers.t8        Cheaper options may be expensive in the long run.   ...etc."   ------------------------------  * Date: Sat, 1 May 2004 23:07:56 -0500 (CDT) From: sms@antinode.orgB Subject: Wanted: Various anonymous FTP servers for limited testing) Message-ID: <04050123075616@antinode.org>a  F    I was thinking about porting a more current version (1.9.1) of WgetG to VMS.  A quick code inspection suggests that the Wget developers haveyF a rather narrow view of VMS FTP server DIR listing formats.  (Or maybe I'm stupid.)  G    In the hope of producing some more accomodating code, I'd like to doaC a bit of exploration of VMS FTP servers other than the one in TCPIPcE V5.3, without having to install them all here.  I thought that a goodA set would include these:  /    Multinet (VMS-style and UNIX-style listings)s1    TCPware (if different from Multinet, options?)     HGFTP (options?)   E If you are running one (or more) of these and don't mind it getting alF little harmless anonymous exercise, please send me a message.  PostingH to the newsgroup would probably reduce the traffic in duplicates.  Also,H Windows NT servers with variant formats would be educational as well, if4 anyone knows where to find those.  Thanks, I assume.  H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode-org     Saint Paul  MN  55105-2547m   ------------------------------   End of INFO-VAX 2004.242 ************************er pane is fine.  6 I'd appreciate any advise on how to troubleshoot this.   Thanks,    Alder    ------------------------------  % Date: Sun, 02 May 2004 01:24:06 +0800 , From: Paul Repacholi <prep@prep.synonet.com>G Subject: Re: How to turn linux into VMS - memory refresher for Dave ... - Message-ID: <87fzakawnd.fsf@prep.synonet.comf1޻$v-o,g
N#iF	_g>sS9rET
4Ibr=L^&g'#,Ōuk&F	`фr
,B&+:##rTsQ>NԖݎWPJ}Y)Iͤt)-E@#7Scq7N6j?45G]d#GKU@|5e_yV\'eYh~kؽ
wުl6)y8抱{gj
GCe\HMIѰ?x#xjKIG{B}d^UZA=ʟ{wDcS
5crN~K^מ0}9luIn/̟@whOy_Aُ1\\ |,x`qJ#hZsu ĠJNB/!ӺurJ ;g_V:+Z X6;J>.PܷE<̑Hlh	5hwtrĠT__6Ell
;܊u=>,l&5`t1vC:RPmll!360!3yGqJ*ХSFJI6=1ٌkV°^F5`)Xb*Nڌ쿜͢O
`21%M3ꊺnL5`U`UE­Jg;q|':.ڜ69<FXF_fwv-}tg'@$y{llDCuߘBk.Gg%Yu
v_!Jܤ 5>MSrCqW7$SUi>U"pmYݷ%OCAH[6_w>YHIv/OZd^lSKwvZl6r'VE*~As
h	; Z0/!:UMn#`|nXf-La(ڙ}RMod@c+u
dG3П)fŃݡސ7k8|,|0Hv|'@sIۅ{c4^YʟZ
6J39V"XX*f	ӫ𑂑u4GYԳKW@g)=-`i
|o	B@JtD@	m_N1_5W!g)%?(iee A<pn  r!,o,FAR#C,Yc{8Rfhv1wi$f78F+ )R/CW$UHLdZ%ہ[	Bx	aIJAБ	CGܧ5[pY9|fnҁ0Uj҅Tq  Thd>(b
c-gPo|? =(AdOG虞2| %?g{?>ߖ'χ޿
}'e=[2[̓p>%|:R)¿_<b3?O)NFRp~E7eo4\^BF/{8֘pT^2B,NCFOs
`<FCBdA h*\)
(ץB},M;a	z݌%nȦ3ŝKKߌ}>wgQ{$U%p	֟E9̋jJ^dH9G9۲u{cNV'XmK
Z`{~ݖ%+8Fĳ't57t__GEh%/"ѕSNM'+`R0NdU=ڻ_λx0ێ-qA(y!-`^o9`M*2/2Z3\2AzSv`BamH@
l!1ej['fu:֩#i^4HA	)%o;_]ɋ=ؼXW>r{n2kS1*xS-\5U;3mONi6W܀bLysrrp=K+
x	D(xf
4xBQ9(',=r]{}gf.W5})1'Tt+SAH1-&Y~foR"Cc72mQEZ
iUUX!JVam[8U/&9Fo1
!3!s1!7r15#yr-_! 5 V-\LaT3)RbkbŤ:h1e5P[/kb
800+`c<=\LE