1 INFO-VAX	Mon, 11 Aug 2003	Volume 2003 : Issue 442       Contents: Re: $asctim / Re: Calculating a delta time from two absolutes / Re: Calculating a delta time from two absolutes / Re: Calculating a delta time from two absolutes  Re: Changing group Re: Changing group Re: DSSI problem Re: DSSI problem! itrc - H.P. I.T. Resource Center. % Re: itrc - H.P. I.T. Resource Center.  Re: Netware is no VMS   Re: New kits of MySQL and Python* Re: patch corrupts product install utility* Re: patch corrupts product install utility* Re: patch corrupts product install utility* Re: patch corrupts product install utility Re: PDP-11 OS Release Dates  Re: PDP-11 OS Release Dates  Re: PDP-11 OS Release Dates  Re: PDP-11 OS Release Dates  Re: PDP-11 OS Release Dates  Re: PDP-11 OS Release Dates  Re: PDP-11 OS Release Dates  Re: PDP-11 OS Release Dates 9 Re: pgp 2.6.3ia multi05 finally running on OpenVMS V7.3-1 J Re: preventing DCL delete command from warning that a file does not  existP Re: preventing DCL delete command from warning that a file does not exist existe  Re: Sizes of logical name tables  Re: Sizes of logical name tables  Re: Sizes of logical name tables  Re: Sizes of logical name tables Re: VMS License audits  F ----------------------------------------------------------------------  % Date: Mon, 11 Aug 2003 10:49:30 -0400 + From: "Syltrem" <syltremzulu@videotron.com>  Subject: Re: $asctim4 Message-ID: <M1OZa.3183$Gf3.14640@tor-nn1.netcom.ca>  G In the manual: pay special attention to datatype and passing mechanism.    --   Syltrem    OpenVMS 7.3-1 + Oracle 8.1.7.4H http://pages.infinit.net/syltrem (OpenVMS related web site, en franais)% ---zulu is not in my email address--- F "David J. Dachtera" <djesys.nospam@fsi.net> a crit dans le message de  news:3F351AEA.653ED3E@fsi.net... > Eyal Sharabi Horwitz wrote:  > > E > > im having a bit of difficulty with the $asctim function, for some G > > reason, every once in a while it gives me a date that's not in it's  > > usual format,  > # > Could you be a bit more specific?  > ' > > which causes a crash in my program.  > > F > > is there a default return value, or a default error value for this
 > > function?  > G > Another poster cited the system services manual. Should prove useful,  > and can be found on line.  >  > --   > David J. Dachtera  > dba DJE Systems  > http://www.djesys.com/ > * > Unofficial Affordable OpenVMS Home Page:! > http://www.djesys.com/vms/soho/    ------------------------------  % Date: Mon, 11 Aug 2003 14:12:08 +0300 " From: Guy Peleg <guy.peleg@hp.com>8 Subject: Re: Calculating a delta time from two absolutes& Message-ID: <3F377A08.1F79998B@hp.com>   Jeff,   J With VMS V7.3-2 we will introduce a new lexical function F$DELTA_TIME. See the following example.   BLUSKY> a=f$time() BLUSKY> b=f$time() BLUSKY> sh sym a   A = "11-AUG-2003 14:10:26.26"  BLUSKY> sh sym b   B = "11-AUG-2003 14:10:30.05" * BLUSKY> write sys$output f$delta_time(a,b)    0 00:00:03.79 BLUSKY>   G What version of VMS do you use? I might be able to backport it for you.   	 Guy Peleg  OpenVMS Engineering      Jeff Cameron wrote:   F > Can anyone offer a DCL example of using the lexical functions F$TIMED > () and F$CVTIME to calculate the elapsed or delta time between twoI > absolute times without breaking down the days hours minutes and seconds B > And calculating a difference using a semi complicated algorithm? > 3 > Suppose I have a command procedure, and I need to C > find out how long a certain operation takes. Well I save the time G > before the action by $TIME1 = F$TIME(), then I do the action, and get > > the absolute time again from $TIME2 = F$TIME(). Now how do I9 > calculate the elapsed time between these two absolutes?  >  > Thank you in advance.    ------------------------------  % Date: Mon, 11 Aug 2003 10:47:48 -0400 + From: "Syltrem" <syltremzulu@videotron.com> 8 Subject: Re: Calculating a delta time from two absolutes4 Message-ID: <j0OZa.3182$Gf3.14681@tor-nn1.netcom.ca>    Is this what you're looking for?  C Won't work if the WAIT spans more than one one day. I can provide a K not-good-looking-but-functional routine that can count days if you need it.   " Hey! - Thanks for the sticker! :-)   $ before=f$cvt(,,"time") $ wait 0:0:5 $ after=f$cvt(,,"time") - $ elapsed=f$cvt(after + "-" + before,,"time")  $ sh sym elapsed --   Syltrem    OpenVMS 7.3-1 + Oracle 8.1.7.4H http://pages.infinit.net/syltrem (OpenVMS related web site, en franais)% ---zulu is not in my email address--- B "Jeff Cameron" <JCam90502@jcameron.com> a crit dans le message de, news:BB5B27D1.A875%JCam90502@jcameron.com...F > Can anyone offer a DCL example of using the lexical functions F$TIMED > () and F$CVTIME to calculate the elapsed or delta time between twoI > absolute times without breaking down the days hours minutes and seconds B > And calculating a difference using a semi complicated algorithm? > 3 > Suppose I have a command procedure, and I need to C > find out how long a certain operation takes. Well I save the time G > before the action by $TIME1 = F$TIME(), then I do the action, and get > > the absolute time again from $TIME2 = F$TIME(). Now how do I9 > calculate the elapsed time between these two absolutes?  >  > Thank you in advance.  >    ------------------------------    Date: 11 Aug 2003 09:50:35 -0700. From: al5vf03p02@sneakemail.com (William Webb)8 Subject: Re: Calculating a delta time from two absolutes= Message-ID: <d5ce4b06.0308110850.394ea29d@posting.google.com>   e Jeff Cameron <JCam90502@jcameron.com> wrote in message news:<BB5B27D1.A875%JCam90502@jcameron.com>... F > Can anyone offer a DCL example of using the lexical functions F$TIMED > () and F$CVTIME to calculate the elapsed or delta time between twoI > absolute times without breaking down the days hours minutes and seconds B > And calculating a difference using a semi complicated algorithm? > 3 > Suppose I have a command procedure, and I need to C > find out how long a certain operation takes. Well I save the time G > before the action by $TIME1 = F$TIME(), then I do the action, and get > > the absolute time again from $TIME2 = F$TIME(). Now how do I9 > calculate the elapsed time between these two absolutes?  >  > Thank you in advance.   F There's a DSNlink article about this with a non-trivial DCL procedure G that will do this for (I think) dates/times up to 9999 days difference.   9 I don't remember how it breaks things out for comparison.   > This is my correct email, right to left:  vog.spsu.liame@bbeww   ------------------------------  % Date: Mon, 11 Aug 2003 11:55:29 +0100 9 From: Alan Adams <alan.adams@orchard-way.freeserve.co.uk>  Subject: Re: Changing group ? Message-ID: <008f43204c.Alan.Adams@orchard-way.freeserve.co.uk>   2 In message <CAubBZU17J5H@eisner.encompasserve.org>F           koehler@eisner.nospam.encompasserve.org (Bob Koehler) wrote:  e > In article <agSYa.19808$qg3.1355009@twister.tampabay.rr.com>, "John N." <JNixon@cfl.rr.com> writes: O > > Will it be dangerous to give an in-house application permission to create a H > > captive account which would  execute a utility to change their GROUPN > > membership.  We have several uic groups set up which each have a differentL > > application version under testing.  We are looking for good ways for theP > > testers to jump from group to group without having several accounts for each	 > > user.  > 8 >    Don't use groups for this.  Use rights identifiers.  L That's OK if the application lets you. I suspect the application here may beL PROMIS. It works by using the GROUP logical name table to locate everything.L Change group - change database. It works, but it is not the way I would want to run VMS.      Alan   --  
 Alan Adams& alan.adams@orchard-way.freeserve.co.uk http://www.nckc.org.uk/    ------------------------------    Date: 11 Aug 2003 11:53:28 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: Changing group 3 Message-ID: <X26BLhHO2qEA@eisner.encompasserve.org>   { In article <008f43204c.Alan.Adams@orchard-way.freeserve.co.uk>, Alan Adams <alan.adams@orchard-way.freeserve.co.uk> writes: 4 > In message <CAubBZU17J5H@eisner.encompasserve.org>H >           koehler@eisner.nospam.encompasserve.org (Bob Koehler) wrote: > f >> In article <agSYa.19808$qg3.1355009@twister.tampabay.rr.com>, "John N." <JNixon@cfl.rr.com> writes:P >> > Will it be dangerous to give an in-house application permission to create aI >> > captive account which would  execute a utility to change their GROUP O >> > membership.  We have several uic groups set up which each have a different M >> > application version under testing.  We are looking for good ways for the Q >> > testers to jump from group to group without having several accounts for each 
 >> > user. >>  9 >>    Don't use groups for this.  Use rights identifiers.  > N > That's OK if the application lets you. I suspect the application here may beN > PROMIS. It works by using the GROUP logical name table to locate everything.N > Change group - change database. It works, but it is not the way I would want
 > to run VMS.   H    You can alter LNM$GROUP to point at something you have better controlH    over.  I've never dealt with PROMIS but it sounds like I'd try takingH    control of logical names away from it in this manner.  I really doubt>    they're looking up the group logical name table themselves.   ------------------------------  % Date: Mon, 11 Aug 2003 10:22:26 +0100 * From: Nic Clews <sendspamhere@[127.0.0.1]> Subject: Re: DSSI problem ' Message-ID: <bh7n4k$brs$1@lore.csc.com>    Thierry Dussuet wrote: > E > In article <87el04kmmz.fsf@prep.synonet.com>, Paul Repacholi wrote: = > >>> You didn't knock either of the terminators off did you? 9 > >> Do you know where the terminators are on a VAX 4000?  > > D > > On the silver panel for the CPU, there are 2 `Honda' connectors,A > > one at the bottom left corner, another a bit above and right. B > > These need aither a cable to a storage box or other system, orE > > a terminator. DSSI terminators have a largish window with a green  > > LED under them.  > H > Oh... I got the box with only one terminator on it.  IIRC, it was in aI > DSSI cluster before, so there must have been a cable and another VAX... B > But it worked like that for quite a long time, why did it changeE > suddenly?  Is there some way to let the VAX believe that there is a C > terminator on there, too? (the 'Honda' is on the upper right one)   F I believe that when the DSSI is unterminated there is a "greater risk" of the fuse popping.  E Assuming you have a terminator with a green light, if the green light   doesn't then the fuse is popped.  > There are a couple of types of terminator; one has a "largish"? translucent window behind which is a green LED, the other has a H rectangular green LED set into the end. These ends are opposite the plugG ends. There is also the "socket" version for pins, and the other socket 5 version which looks like a miniature centronics port.    --  ? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences  nclews at csc dot com    ------------------------------   Date: 11 Aug 2003 17:39:19 GMT, From: thierry@venus.family (Thierry Dussuet) Subject: Re: DSSI problem 6 Message-ID: <3f37d4c6$0$6194$5402220f@news.sunrise.ch>  T In article <bh7n4k$brs$1@lore.csc.com>, Nic Clews <sendspamhere@[127.0.0.1]> writes: >Thierry Dussuet wrote:  >>  F >> In article <87el04kmmz.fsf@prep.synonet.com>, Paul Repacholi wrote:> >> >>> You didn't knock either of the terminators off did you?: >> >> Do you know where the terminators are on a VAX 4000? >> >E >> > On the silver panel for the CPU, there are 2 `Honda' connectors, B >> > one at the bottom left corner, another a bit above and right.C >> > These need aither a cable to a storage box or other system, or F >> > a terminator. DSSI terminators have a largish window with a green >> > LED under them. >>  I >> Oh... I got the box with only one terminator on it.  IIRC, it was in a J >> DSSI cluster before, so there must have been a cable and another VAX...C >> But it worked like that for quite a long time, why did it change F >> suddenly?  Is there some way to let the VAX believe that there is aD >> terminator on there, too? (the 'Honda' is on the upper right one) > G >I believe that when the DSSI is unterminated there is a "greater risk"  >of the fuse popping.   < Is the fuse in the terminators?  Or somewhere on the system?  F >Assuming you have a terminator with a green light, if the green light! >doesn't then the fuse is popped.   @ No, the green light lights happily on.  At least the one left...  ? >There are a couple of types of terminator; one has a "largish" @ >translucent window behind which is a green LED, the other has aI >rectangular green LED set into the end. These ends are opposite the plug H >ends. There is also the "socket" version for pins, and the other socket6 >version which looks like a miniature centronics port.  < Hmm... here it is one with a largish window and a green LED.F Its light stays on the whole time, wherever I put it (on the 3 places)   Thierry    ------------------------------    Date: 11 Aug 2003 10:20:05 -0700, From: JimStrehlow@data911.com (Jim Strehlow)* Subject: itrc - H.P. I.T. Resource Center.= Message-ID: <4b6ec350.0308110920.727cf342@posting.google.com>   > I have mixed feelings about the new H.P. I.T. Resource Center.E Microsoft has had Windows Update to help administrators find patches. < There needed to be something to make things easier for us to administer OpenVMS patches. ; But H.P. is going to take away their FTP site in September. B I do not mind finding out about patches via some new tool at ITRC;E but I prefer using FTP on OpenVMS directly binary copying files to my  AlphaServer.  C H.P. will be forcing me to have to either "double transfer" from my B P.C. to OpenVMS, or I will have to go to a DECWindows terminal (inF another room away from my regular workspace), launch Mozilla (or other
 browser) etc.   C Since H.P. has been e-mailing me patch notifications, that has been . good enough for an "update notification tool."  % What is YOUR opinion of the new ITRC?   = Jim Strehlow, OpenVMS Systems Administrator, Alameda, CA, USA    ------------------------------  + Date: Mon, 11 Aug 2003 17:32:46 +0000 (UTC) 7 From: hamilton@Encompasserve.org (Bradford J. Hamilton) . Subject: Re: itrc - H.P. I.T. Resource Center./ Message-ID: <bh8jvu$p9i$1@grandcanyon.binc.net>   l In article <4b6ec350.0308110920.727cf342@posting.google.com>, JimStrehlow@data911.com (Jim Strehlow) writes:? !I have mixed feelings about the new H.P. I.T. Resource Center. F !Microsoft has had Windows Update to help administrators find patches.= !There needed to be something to make things easier for us to  !administer OpenVMS patches.< !But H.P. is going to take away their FTP site in September.C !I do not mind finding out about patches via some new tool at ITRC; F !but I prefer using FTP on OpenVMS directly binary copying files to my
 !AlphaServer.  ! D !H.P. will be forcing me to have to either "double transfer" from myC !P.C. to OpenVMS, or I will have to go to a DECWindows terminal (in G !another room away from my regular workspace), launch Mozilla (or other  !browser) etc.  O I share your concerns.  Is there any way to get feedback to the decision-making A folks, to see if they will still allow us to use FTP, if we wish?    ! D !Since H.P. has been e-mailing me patch notifications, that has been/ !good enough for an "update notification tool."  ! & !What is YOUR opinion of the new ITRC?  M I just used the browsing feature at ITRC - it's a little brain-dead (you have N to know what you are searching for in the patch tree before they will give youN a "match"), but at least they are still allowing access to patches back as farK as V6.2 on Alpha - I feared they would not allow access before the "newest" 	 releases.   O Can someone from HP find out how we can get in touch with the ITRC folks to let  our opinions be known?  N The ITRC (in general) is an OK idea.  The OpenVMS forum still has way too muchI "noise" from PC users looking to answer HW questions.  If the ITRC cannot M attract a critical mass of participants from here and EISNER::, I would still K regard c.o.v. and EISNER:: as my main sources of techinical information for  VMS.   ! > !Jim Strehlow, OpenVMS Systems Administrator, Alameda, CA, USA   ------------------------------    Date: 11 Aug 2003 14:25:57 +0200C From: vaxinf@chclu.chemie.uni-konstanz.de (Eberhard Heuser-Hofmann)  Subject: Re: Netware is no VMS- Message-ID: <3f378b55$1@news.uni-konstanz.de>   A In article <3F340B70.EF43F383@ost.cdrh.fda.gov>, Jonathan Boswell  <jsb@ost.cdrh.fda.gov> writes: |>jlsue wrote:G |>> What new features do you believe are needed above what's already in  |>the  |>> roadmap? |>K |>HPTC stuff such as MPI.  This was ported to T64 including message passing  |>overI |>shared memory on multiprocessor Alphas.  Nothing ever surfaced for VMS, J |>resulting in zero HP sales from my VMS-only shop this year.  Previously, |>I had H |>offered researchers impressive horsepower on ES40s running VMS, and we |>had 21E |>VMScluster nodes total.  But I'm the only one who has ever used the  |>cluster as aG |>compute engine, and I did that by spitting out as many separate batch  |>processes A |>as I have Alpha processors.  Our younger researchers all expect  |>MPI/PVM/OpenMPF |>or *something* that will speed up their jobs beyond that of a single
 |>processor.  I |>They are used to Beowulf cluster performance in academia that completes  |>their A |>jobs in minutes.  The same jobs takes all day on a single Alpha 
 |>processor.  H |>Needless to say, they are agitating for Linux on commodity processors. |> |> - JB  |>  G I'm sure MPI for VMS is portable. At least the core part. I've done the  portI of tcgmsg-message passing. I can show how to run programs on Tru64 alphas 6 +  OpenVMS alphas using this message passing library.    eberhard   ------------------------------    Date: 11 Aug 2003 07:23:21 -0700& From: jordan@ccs4vms.com (Rich Jordan)) Subject: Re: New kits of MySQL and Python = Message-ID: <cc5619f2.0308110623.5f52118b@posting.google.com>   W jf.pieronne@laposte.net wrote in message news:<bgvms3$ret$1@news-reader5.wanadoo.fr>... D > I have put online MySQL 4.0.14 which is the latest stable version. >  > Q > I have also put online a kit of Python 2.3 based on the first official release   > of Python 2.3.
 > ........ >  > Jean-Franois   
 Jean-Franois B      thanks for making the effort again.  I tried downloading thisA weekend but apparently the North American mirror is down, and the F Australian and your main site were running far far slower than even myF tiny IDSL pipe, so I held off.  Really looking forward to trying these out!   Rich Jordan    ------------------------------  # Date: Mon, 11 Aug 2003 12:39:00 GMT 3 From: hammond@not@peek.ssr.hp.com (Charlie Hammond) 3 Subject: Re: patch corrupts product install utility 1 Message-ID: <E7MZa.1948$kO1.126@news.cpqcorp.net>   x In article <4QsZa.19661$Vx2.10179887@newssvr28.news.prodigy.com>, "Phillip R Sobottke" <psobottke@ameritech.net> writes:I >I have version openvms version 7.3-1, and when installing the pcsi v0100 L >critical update.. it installs but then corrupts the product install utilityH >preventing the installation of additional patches.  Any ideas on how to >overcome this?   G If the information posted so far has not helped you, then pleas provide + some additional information.  Specifically:   = What happens to make you think that the utility is ocrrupted?   < What happens when you attempt to install additional patches?   --  J       Charlie Hammond -- Hewlett-Packard Company -- Ft Lauderdale  FL  USAF           (hammond@not@peek.ssr.hp.com -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------    Date: 11 Aug 2003 06:56:21 -0700( From: prsobottke@aep.com (Phil Sobottke)3 Subject: Re: patch corrupts product install utility < Message-ID: <5bcc5f24.0308110556.5c5e9de@posting.google.com>  n hamilton@Encompasserve.org (Bradford J. Hamilton) wrote in message news:<bh5quo$2aa$1@grandcanyon.binc.net>...z > In article <4QsZa.19661$Vx2.10179887@newssvr28.news.prodigy.com>, "Phillip R Sobottke" <psobottke@ameritech.net> writes:K > !I have version openvms version 7.3-1, and when installing the pcsi v0100 N > !critical update.. it installs but then corrupts the product install utilityJ > !preventing the installation of additional patches.  Any ideas on how to > !overcome this?  > F > I installed UPDATE after having installed PCSI V0100, and I have notL > encountered this problem.  Can you provide us with some more detail (errorP > messages)?  Please check the archives of this newsgroup (at Google) for recentO > (within the past month) discussions regarding the use of PCSI V0100, as well.  >    > ! } > !I tried using the command mentioned later in this string.. and logging off and back on, but still get the following error: 6  The previous versions of the images installed by this6     kit have been renamed to IMAGE_NAME.EXE_OLD, where/     IMAGE_NAME is the actual name of the image. 2 CHM002::SYSTEM> set comm/table=sys$share:dcltables$ CHM002::SYSTEM> product show history% %CLI-F-SYNTAX, error parsing 'DELETE' : -CLI-E-ENTNF, specified entity not found in command tables/ %TRACE-F-TRACEBACK, symbolic stack dump follows J   image    module    routine             line      rel PC           abs PCO                                             0 000000007AF9D4C8 000000007AF9D4C8 O                                             0 000000007AF9D3DC 000000007AF9D3DC O                                             0 000000007AF9C65C 000000007AF9C65C O                                             0 FFFFFFFF800885D4 FFFFFFFF800885D4 O                                             0 FFFFFFFF800885D4 FFFFFFFF800885D4 O  PCSI$MAIN                                  0 000000000003C9F8 000000000004C9F8 1  PCSI$MAIN  SPIU_COMMAND_VMS  process_DCL_command O                                         27589 000000000000971C 0000000000042A8C )  PCSI$MAIN  SPIU_COMMAND_VMS  process_DCL O                                         27899 000000000000A040 00000000000433B0 O  PCSI$MAIN  SPIU_COMMAND_VMS  main      30252 000000000000E6A0 0000000000047A10 O  PCSI$MAIN  SPIU_COMMAND_VMS  __main        0 0000000000000094 0000000000039404 O                                             0 FFFFFFFF8028563C FFFFFFFF8028563C  CHM002::SYSTEM>    ------------------------------   Date: 11 Aug 03 19:26:42 +0200) From: p_sture@elias.decus.ch (Paul Sture) 3 Subject: Re: patch corrupts product install utility ) Message-ID: <d7hG7S8Bydsu@elias.decus.ch>   i In article <bh5quo$2aa$1@grandcanyon.binc.net>, hamilton@Encompasserve.org (Bradford J. Hamilton) writes: z > In article <4QsZa.19661$Vx2.10179887@newssvr28.news.prodigy.com>, "Phillip R Sobottke" <psobottke@ameritech.net> writes:K > !I have version openvms version 7.3-1, and when installing the pcsi v0100 N > !critical update.. it installs but then corrupts the product install utilityJ > !preventing the installation of additional patches.  Any ideas on how to > !overcome this?  > F > I installed UPDATE after having installed PCSI V0100, and I have notL > encountered this problem.  Can you provide us with some more detail (errorP > messages)?  Please check the archives of this newsgroup (at Google) for recentO > (within the past month) discussions regarding the use of PCSI V0100, as well.  >     E I came across a similar problem. In my haste I rebooted, and realised D I should have tried logging in again as a first step. The reboot did: work, but I wish I had at least tried a fresh login first.   ------------------------------    Date: 11 Aug 2003 10:30:31 -0700( From: prsobottke@aep.com (Phil Sobottke)3 Subject: Re: patch corrupts product install utility = Message-ID: <5bcc5f24.0308110930.40b5da15@posting.google.com>   l hammond@not@peek.ssr.hp.com (Charlie Hammond) wrote in message news:<E7MZa.1948$kO1.126@news.cpqcorp.net>...z > In article <4QsZa.19661$Vx2.10179887@newssvr28.news.prodigy.com>, "Phillip R Sobottke" <psobottke@ameritech.net> writes:K > >I have version openvms version 7.3-1, and when installing the pcsi v0100 N > >critical update.. it installs but then corrupts the product install utilityJ > >preventing the installation of additional patches.  Any ideas on how to > >overcome this?  > I > If the information posted so far has not helped you, then pleas provide - > some additional information.  Specifically:  > ? > What happens to make you think that the utility is ocrrupted?  > > > What happens when you attempt to install additional patches?   >    > ! w > !I tried using the command mentioned in this string.. and logging off and back on, but still get the following error:    2 CHM002::SYSTEM> set comm/table=sys$share:dcltables$ CHM002::SYSTEM> product show history% %CLI-F-SYNTAX, error parsing 'DELETE' : -CLI-E-ENTNF, specified entity not found in command tables/ %TRACE-F-TRACEBACK, symbolic stack dump follows C   image    module    routine             line      rel PC            abs PC>                                             0 000000007AF9D4C8 000000007AF9D4C8>                                             0 000000007AF9D3DC 000000007AF9D3DC>                                             0 000000007AF9C65C 000000007AF9C65C>                                             0 FFFFFFFF800885D4 FFFFFFFF800885D4>                                             0 FFFFFFFF800885D4 FFFFFFFF800885D4>  PCSI$MAIN                                  0 000000000003C9F8 000000000004C9F81  PCSI$MAIN  SPIU_COMMAND_VMS  process_DCL_command >                                         27589 000000000000971C 0000000000042A8C)  PCSI$MAIN  SPIU_COMMAND_VMS  process_DCL >                                         27899 000000000000A040 00000000000433B0>  PCSI$MAIN  SPIU_COMMAND_VMS  main      30252 000000000000E6A0 0000000000047A10>  PCSI$MAIN  SPIU_COMMAND_VMS  __main        0 0000000000000094 0000000000039404>                                             0 FFFFFFFF8028563C FFFFFFFF8028563C CHM002::SYSTEM>   ? The same error shows up when trying to install patches (or just = showing the installed patches).  Interestingly enough, we are C installing on 8 machines of various vintages...and only the DS-10's  are giving us the problem.   ------------------------------  ! Date: Mon, 11 Aug 03 09:19:52 GMT  From: jmfbahciv@aol.com $ Subject: Re: PDP-11 OS Release Dates+ Message-ID: <bh7qrc$r02$2@bob.news.rcn.net>   / In article <bh5q7m$ddm$1@tabloid.uwaterloo.ca>, 2    dfevans@bcr10.uwaterloo.ca (David Evans) wrote:H >In article <bh58k6$mlg$2@bob.news.rcn.net>,  <jmfbahciv@aol.com> wrote:1 >>In article <bgtj2h$i34$1@tabloid.uwaterloo.ca>, 4 >>   dfevans@bcr10.uwaterloo.ca (David Evans) wrote:J >>>In article <bgtate$60n$4@bob.news.rcn.net>,  <jmfbahciv@aol.com> wrote:6 >>>>In article <3f3f0cd2.12775862@news.supernews.com>,' >>>>   ian@hammo.com (paramucho) wrote: I >>>>>One of the big problems with cramped spaces was that users would use E >>>>>up every available byte of user space available, thus increasing 8 >>>>>system size would render their programs inoperable. >>>>B >>>>This makes no sense.  That would be problem only if their code< >>>>(or data placement) was hardwired to absolute addresses. >>>> >>> G >>>  In my experiences in micro land, absolute addresses were the rule. H >>>Not many folks bothered with relocating loaders when you only had one@ >>>64K address space and something like the 6502 makes it pretty7 >>>irritating to write fully position-independant code.  >>G >>Wait a minute...this doesn't make any sense.  Code had to be location D >>independent if it can be loaded into core starting at any address. > I >  That was my point: frequently one could not load code into core at any I >address and expect it to work.  The assembler would be provided with the K >expected start address and resolve labels to absolute addresses.  As there I >was no proper link loader (loading the code simply consisted of stuffing I >the bytes in memory beginning at the proper address) no relocation could 	 >be done.   C Oh, gagmeverymuch.  No wonder there's code bloat and memory leakage A out there.  This means that a program had to know how much memory > it ever wanted to use before assembly.  There was no mechanism? to ask the monitor for more core (address space) if and when it ? needed it.  Most programmers would issue an over-inflated BLOCK > statement just to play it "safe" and never bother to do checks& for overflowing its own address space.   > - >>>>Users learned how about memory management - >>>>and the difference between code and data.  >>>> >>> B >>>  On weenie machines there was no such high-falutin excitement. >> >>Define weenie machines.    >> > ? >  I realise I may have misunderstood what you meant by "memory F >management." Things such as the KIM-1 or the Commodore 8-bit machinesF >(the micros that I have the most experience with) had no real concept0 >of physical vs. virtual addresses or what not.   / That doesn't matter.  It helps when it's there.    > .. While there wasF >sometimes memory management hardware of some sort, it wasn't the kind. >of thing that you could use to write a pager.  7 Even basic memory management at the program level isn't : being done if the program can't ask or give up core as it ; runs around doing its thing.  It helps if you have hardware @ assists, but you don't need it to do sensible memory management.   /BAH      ' Subtract a hundred and four for e-mail.    ------------------------------  % Date: Mon, 11 Aug 2003 07:05:39 -0500 ! From: rpw3@rpw3.org (Rob Warnock) $ Subject: Re: PDP-11 OS Release Dates2 Message-ID: <VrqdnYQY_sgOG6qiXTWc-g@speakeasy.net>  / David Evans <dfevans@bcr10.uwaterloo.ca> wrote:  +--------------- | <jmfbahciv@aol.com> wrote:= | >This sounds more like something that magically happened to " | >work rather than original code. | G |   No; it was a not uncommon technique to save placing things like LDA ? | #$00 in the mainline of your code.  The documentation for the 1 | subroutine I began above above might look like:  | 4 | Entry: $8000 Perform I/O to device specified in .A, |        $8001 Perform I/O to default device +---------------  B Even in PDP-10 land, having multiple entry points was not all thatC uncommon (though on the -10 one couldn't jump into the middle of an D instruction, of course). A quickly contrived (and therefore somewhat braindead) example:   0 	; To the memory location pointed to by t0, add:   	inc2mm: movei	t1,2		; 2 		skipa  	incmem: movei	t1,1		; 1) 	addmem:	addm	t1,(t0)		; the value in t1. 
 		popj	p,0  @ [Hmmm... Though now that I think about it, "jrst .+2" was faster than "skipa", wasn't it?]      -Rob   ----- ( Rob Warnock, PP-ASEL-IA		<rpw3@rpw3.org>( 627 26th Avenue			<URL:http://rpw3.org/>" San Mateo, CA 94403		(650)572-2607   ------------------------------  ! Date: Mon, 11 Aug 03 13:11:17 GMT  From: jmfbahciv@aol.com $ Subject: Re: PDP-11 OS Release Dates+ Message-ID: <bh88d7$qb3$1@bob.news.rcn.net>   / In article <bh87ia$gd6$1@tabloid.uwaterloo.ca>, 2    dfevans@bcr10.uwaterloo.ca (David Evans) wrote:H >In article <bh7qrc$r02$2@bob.news.rcn.net>,  <jmfbahciv@aol.com> wrote:E >>Oh, gagmeverymuch.  No wonder there's code bloat and memory leakage C >>out there.  This means that a program had to know how much memory ( >>it ever wanted to use before assembly. > H >  Well, if your machine had at most 64K RAM and only ran one program at6 >a time then this wasn't really a big stumbling block.   Sure it was!!!!      >  >>There was no mechanismA >>to ask the monitor for more core (address space) if and when it  >>needed it. > G >  There was no monitor per se on many machines.  There may have been a G >bunch of I/O subroutines that you could call (print a character to the / >printer, open a file on disk) but that was it.   @ And were all files limited to one length?  What do you do if you? have a file that is larger than available pre-allocated memory? # No wonder there exists a mess.  :-)    /BAH    ' Subtract a hundred and four for e-mail.    ------------------------------  + Date: Mon, 11 Aug 2003 14:00:42 +0000 (UTC) . From: dfevans@bcr10.uwaterloo.ca (David Evans)$ Subject: Re: PDP-11 OS Release Dates/ Message-ID: <bh87ia$gd6$1@tabloid.uwaterloo.ca>   G In article <bh7qrc$r02$2@bob.news.rcn.net>,  <jmfbahciv@aol.com> wrote:oD >Oh, gagmeverymuch.  No wonder there's code bloat and memory leakageB >out there.  This means that a program had to know how much memory' >it ever wanted to use before assembly.i  G   Well, if your machine had at most 64K RAM and only ran one program ato5 a time then this wasn't really a big stumbling block.r   >There was no mechanisms@ >to ask the monitor for more core (address space) if and when it >needed it.a  F   There was no monitor per se on many machines.  There may have been aF bunch of I/O subroutines that you could call (print a character to the. printer, open a file on disk) but that was it.   -- eM David Evans          (NeXTMail/MIME OK)             dfevans@bbcr.uwaterloo.ca M Ph.D. Candidate, Computer/Synth Junkie     http://bbcr.uwaterloo.ca/~dfevans/OM University of Waterloo         "Default is the value selected by the composer M Ontario, Canada           overridden by your command." - Roland TR-707 Manual    ------------------------------  % Date: Mon, 11 Aug 2003 09:28:33 -0600  From: Kevin Handy <kth@srv.net> $ Subject: Re: PDP-11 OS Release Dates9 Message-ID: <vONZa.6992$u44.5707@fe01.atl2.webusenet.com>    jmfbahciv@aol.com wrote:1 > In article <bh87ia$gd6$1@tabloid.uwaterloo.ca>,-4 >    dfevans@bcr10.uwaterloo.ca (David Evans) wrote: > I >>In article <bh7qrc$r02$2@bob.news.rcn.net>,  <jmfbahciv@aol.com> wrote:c >>F >>>Oh, gagmeverymuch.  No wonder there's code bloat and memory leakageD >>>out there.  This means that a program had to know how much memory) >>>it ever wanted to use before assembly.- >>H >> Well, if your machine had at most 64K RAM and only ran one program at7 >>a time then this wasn't really a big stumbling block.D >  >  > Sure it was!!!!o >  >  >  >>>There was no mechanismeB >>>to ask the monitor for more core (address space) if and when it
 >>>needed it.v >>G >> There was no monitor per se on many machines.  There may have been arH >>bunch of I/O subroutines that you could call (print a character to the0 >>printer, open a file on disk) but that was it. >  > B > And were all files limited to one length?  What do you do if youA > have a file that is larger than available pre-allocated memory?n% > No wonder there exists a mess.  :-)m > ; You start the program, and it gets ALL the memory. There is < NO Memory Management going on. If that isn't enough for your9 program, than you have to rewrite it to need less. If you < needed something to handle memory allocations, that was part$ of your program, not part of the OS.  = In CP/M, the 'OS' was in two parts. The command line handler,uA and the BDOS(?).  You never used the memory allocated to the BDOS9? (because it wouldn't work any more, not because of any hardwaree8 protection). If you needed the space used by the commandB line handler, you just used it then exited with the proper values, and the BDOS would reload it.Y   ------------------------------  + Date: Mon, 11 Aug 2003 15:28:27 +0000 (UTC)m. From: dfevans@bcr10.uwaterloo.ca (David Evans)$ Subject: Re: PDP-11 OS Release Dates/ Message-ID: <bh8cmr$irt$1@tabloid.uwaterloo.ca>n  G In article <bh88d7$qb3$1@bob.news.rcn.net>,  <jmfbahciv@aol.com> wrote:  >n* >And were all files limited to one length?  F   No; files were stored on floppy, and (at least in examples I'm awareF of) could be whatever number of disk blocks you wanted, subject to theG capacity of the disk.  Files could live on audio cassette tape as well,a9 but that was a major pain and I didn't like it very much.    >What do you do if you@ >have a file that is larger than available pre-allocated memory?  ?   You lose.  ;-)  Well, not really.  I regularly had files thatEG wouldn't fit in memory, so my program had to read in a piece, deal withoE it, and then write that piece back to disk.  It made it irritating to-I write a text editor that could handle files larger than available memory.uI I never embarked on such a project, but remember working out how it might " be done while driving the tractor.   -- eM David Evans          (NeXTMail/MIME OK)             dfevans@bbcr.uwaterloo.carM Ph.D. Candidate, Computer/Synth Junkie     http://bbcr.uwaterloo.ca/~dfevans/hM University of Waterloo         "Default is the value selected by the composereM Ontario, Canada           overridden by your command." - Roland TR-707 Manualf   ------------------------------    Date: 11 Aug 2003 09:26:13 -0700( From: Patrick Scheible <kkt@drizzle.com>$ Subject: Re: PDP-11 OS Release Dates) Message-ID: <tqmy8y088ka.fsf@drizzle.com>e   jmfbahciv@aol.com writes:o  1 > In article <bh5q7m$ddm$1@tabloid.uwaterloo.ca>,e4 >    dfevans@bcr10.uwaterloo.ca (David Evans) wrote:J > >In article <bh58k6$mlg$2@bob.news.rcn.net>,  <jmfbahciv@aol.com> wrote:3 > >>In article <bgtj2h$i34$1@tabloid.uwaterloo.ca>,n6 > >>   dfevans@bcr10.uwaterloo.ca (David Evans) wrote:L > >>>In article <bgtate$60n$4@bob.news.rcn.net>,  <jmfbahciv@aol.com> wrote:8 > >>>>In article <3f3f0cd2.12775862@news.supernews.com>,) > >>>>   ian@hammo.com (paramucho) wrote:iK > >>>>>One of the big problems with cramped spaces was that users would useIG > >>>>>up every available byte of user space available, thus increasingE: > >>>>>system size would render their programs inoperable. > >>>>D > >>>>This makes no sense.  That would be problem only if their code> > >>>>(or data placement) was hardwired to absolute addresses. > >>>> > >>>iI > >>>  In my experiences in micro land, absolute addresses were the rule.tJ > >>>Not many folks bothered with relocating loaders when you only had oneB > >>>64K address space and something like the 6502 makes it pretty9 > >>>irritating to write fully position-independant code.p > >>I > >>Wait a minute...this doesn't make any sense.  Code had to be locationsF > >>independent if it can be loaded into core starting at any address. > > K > >  That was my point: frequently one could not load code into core at anyiK > >address and expect it to work.  The assembler would be provided with the M > >expected start address and resolve labels to absolute addresses.  As thereeK > >was no proper link loader (loading the code simply consisted of stuffingtK > >the bytes in memory beginning at the proper address) no relocation couldb > >be done.t > E > Oh, gagmeverymuch.  No wonder there's code bloat and memory leakagebC > out there.  This means that a program had to know how much memoryt@ > it ever wanted to use before assembly.  There was no mechanismA > to ask the monitor for more core (address space) if and when itaA > needed it.  Most programmers would issue an over-inflated BLOCK-@ > statement just to play it "safe" and never bother to do checks( > for overflowing its own address space.    A On the weenie machines, there only one program running at any oneaD time.  That one program doesn't have to ask for more memory, it justB uses whatever it wants (other than avoiding the OS area, which wasB known ahead of time).  If it wanted more memory than there was, inB most cases it had to figure out how to store some of it on disk orB floppy all by itself without help from pager or the OS.  The OS is@ just some device drivers and just enough to stick the program in memory and jump to it.  F These were the bad old days as far as micros.  (On the other hand, youC didn't have horrible performance because too many other people were ' using the computer at the same time...)u   -- Patrick Scheiblet   ------------------------------    Date: 11 Aug 2003 11:50:31 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)s$ Subject: Re: PDP-11 OS Release Dates3 Message-ID: <D2JkWN3WpxD9@eisner.encompasserve.org>   E In article <bh88d7$qb3$1@bob.news.rcn.net>, jmfbahciv@aol.com writes:o  B > And were all files limited to one length?  What do you do if youA > have a file that is larger than available pre-allocated memory?-  F    You used files as files.  You read a record or two, processed that,A    and wrote a record or two to the output file.  You didn't needCE    a 256K of memory just because you had a file that took up the half     the floppy.   ------------------------------  % Date: Mon, 11 Aug 2003 07:51:05 +0100a* From: Nic Clews <sendspamhere@[127.0.0.1]>B Subject: Re: pgp 2.6.3ia multi05 finally running on OpenVMS V7.3-1' Message-ID: <bh7e8r$a45$1@lore.csc.com>    Gorazd Kikelj wrote: > M > I finaly manage to compile PGP 2.6.3ia multi05 on OpenVMS Alpha V7.3-1 withw
 > DECC 6.5 > G > Most hackery was in SYSTEM.C (as expected) and some in other modules.  > N > I put together a quick command procedure to compile all components and build > object library.e > 2 > I haven't time to test except pgp -h and pgp -g. > N > If somebody is interested, send me a private mail to gorazd dot kikelj at hp
 > dot com.  = I have built close versions, I say close because I discovered F distributions that would compile, seemingly very similar to those that) wouldn't. I'm not a C programmer however.t  F I will email, as I'll attempt to build the code on earlier versions of; VMS, and for VAX. I'll give a shout here if I'm successful.g   -- c? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Scienceso nclews at csc dot comn   ------------------------------  # Date: Mon, 11 Aug 2003 12:35:55 GMT 3 From: hammond@not@peek.ssr.hp.com (Charlie Hammond)rS Subject: Re: preventing DCL delete command from warning that a file does not  exist 1 Message-ID: <L4MZa.1947$kO1.879@news.cpqcorp.net>e  [ In article <3F345C17.9D25F15F@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:  >Charlie Hammond wrote:0 >> >o >> In article <e7fbc28d.0308080716.fa0a2e4@posting.google.com>, joseph_kanney@yahoo.com (Joseph Kanney) writes:e >> >hi!m >> >I >> >I'd like to make the DCL delete command behave like the unix "rm -f",fM >> >i.e., I don't want to see warnings if a file is not found. If a file does F >> >not exist, I want it to do nothing, but silently. Any hints on how >> >to do this? Thanks!t >> > >> >-- Joe Kanneya >> r >> Here is an idea.t >>   >> $ CREATE RM.COM1 >> IF F$SEARCH("''P1'") .NES. "" THEN DELETE 'P1'  >> [EXIT] <-- PRESS CTRL/Z >> $ RM = "@RM.COM"w >>  & >> Now use the sumbol RM as a command. > I >Ooohhh, careful there! I'm already seeing places where lines of DCL codeu8 >not preceded by a "$" are being rejected as image data.  2 Opps!  I agree that the "$" should not be omitted. I did so in error.   -- dJ       Charlie Hammond -- Hewlett-Packard Company -- Ft Lauderdale  FL  USAF           (hammond@not@peek.ssr.hp.com -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------  % Date: Mon, 11 Aug 2003 14:35:08 +0300i" From: Guy Peleg <guy.peleg@hp.com>Y Subject: Re: preventing DCL delete command from warning that a file does not exist existeh& Message-ID: <3F377F6C.23D67686@hp.com>  J  If this functionality is required by you I can build you a new DELETE.EXE that will accept  /SILENCE6 qualifier and will not report file not found messages.  E Please file a service call with your local support center and have ita escalated to VMS Engineering  	 Guy Peleg  OpenVMS Engineeringo   Joseph Kanney wrote:   > hi!- > G > I'd like to make the DCL delete command behave like the unix "rm -f",sK > i.e., I don't want to see warnings if a file is not found. If a file doessD > not exist, I want it to do nothing, but silently. Any hints on how > to do this? Thanks!i >: > -- Joe Kanneyt   ------------------------------  % Date: Mon, 11 Aug 2003 08:52:01 +0100h* From: Nic Clews <sendspamhere@[127.0.0.1]>) Subject: Re: Sizes of logical name tablesu' Message-ID: <bh7hr3$b7n$1@lore.csc.com>4   Jeff Cameron wrote:  >vL > I have an old utility (executable image) written by a company no longer inL > business called SIMPART. The company's name escapes me now which simulatesM > certain physical interactions in physics. It is aborting with a simple texta1 > message : "VMS error - Logical name table full"   D I think you'll find that the (dynamic) SYSGEN parameter PQL_DJTQUOTA7 (augmented by PQL_MJTQUOTA) is what you need to change.   A My suspicion is that the command procedure you refer to creates an. detached job where the UAF rules do not apply.   --  ? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciencese nclews at csc dot como   ------------------------------   Date: 11 Aug 03 06:45:39 PST From: mckinneyj@cpva.saic.comk) Subject: Re: Sizes of logical name tablesn( Message-ID: <1HiKaGmIYfmv@cpva.saic.com>  > In article <o4EZa.259839$BA.61002032@twister.columbus.rr.com>,0  Jack Patteeuw <jjpatteeuw@peoplepc.com> writes:1 >> On Sun, 10 Aug 2003 05:48:03 GMT, Jeff Cameront" >> <JCam90502@jcameron.com> wrote: >> t >> oM >>>I know the UAF quota value JTQuota governs the size of a job's job logical N >>>name table. What parameters or quotas govern the size of the process, group" >>>and system logical name tables. >  > snip >  >> o > Rob.Buxton@wcc.govt.nz wrote: D >> Try running Autogen to Testfiles, see if it wants to increase the >> LNMSHASHTBL Entry.h2 >> Or, try increasing this, needs a reboot though. >> m >> e > K > Rob has the answer, but here is a little background.  Admittedly this is  E > based on info from a V4.x internals book, but I don't think it has r > changed since then.  > H > First, Process Logical Names (LNM) are stored in the "context" of the ) > processs and JTQuota does control this.  >   F I suspect that this was intended to say that "JTQuota does not controlC this". JTQuota controls only the size of the job table. The processiC table has to fit into the process allocation region in P1 space and E the size is controlled by the SYSGEN parameter CTLPAGES. Changing thebE value of the paramter requires a reboot. If you're hitting this limitME then you'll wnat to increase the value of CTLPAGES (doubling probablyr isn't unreasonable).  H The shared name tables are all allocated from paged pool. As you've seenF the job table is limited (security, don't want a user to exhaust pool)H by UAF's JTQouta. Group and System tables require privilege to populate.  H You can MCR SYSGEN HELP LNMSHASHTBL to help determine if your issue is a% shared or private name table problem.   F > Second, **ALL** other LNMs, that can be referenced by more than one D > process, (ie. Job, Group, System, etc) are stored in **ONE** data I > structure whose initial size is controlled by LNMSHASHTBL (the S after uH > LNM is for "shared").  Initial size of any hash table is critical for  > fast lookups.t > J > The lookup is done by running the hash and going directly to that table I > entry.  Each hash entry is a sorted, link list so that lookup that can PG > fail early when the the size of the LNM being lookedup doesn't match.i > J > If the initial size is too small, hashes will "collide" and the entries I > will just be added to the link list.  The result is that lookup become  J > simple link list searches, very inefficient.  I don't remeber where the . > data for new shared LNM entries is obtained. > I > I actually won an award at "VAX Magic" (or was in ALL-IN-1 magic?) for cC > telling folks "How to get 10% of your system back" by increasing  I > LNMSHASHTBL.  A few years later Autogen, took my suggestion and it was  D > handled automatically.  (Ah yes, the good old days!  Too bad Un*x  > doesn't have LNMs !) > 6 > Finally, I think the error message is a Red Herring. >    ------------------------------  # Date: Mon, 11 Aug 2003 14:41:41 GMTH- From: "labadie" <tonari_no_tottoro@127.0.0.1>t) Subject: Re: Sizes of logical name tableso2 Message-ID: <FWNZa.1960$bR1.1286@news.cpqcorp.net>  * <mckinneyj@cpva.saic.com> wrote in message" news:1HiKaGmIYfmv@cpva.saic.com...@ > In article <o4EZa.259839$BA.61002032@twister.columbus.rr.com>,2 >  Jack Patteeuw <jjpatteeuw@peoplepc.com> writes:3 > >> On Sun, 10 Aug 2003 05:48:03 GMT, Jeff Cameron $ > >> <JCam90502@jcameron.com> wrote: > >> > >>G > >>>I know the UAF quota value JTQuota governs the size of a job's jobw logicaleJ > >>>name table. What parameters or quotas govern the size of the process, groupr$ > >>>and system logical name tables. > >  > > snip > >h > >>! > > Rob.Buxton@wcc.govt.nz wrote: F > >> Try running Autogen to Testfiles, see if it wants to increase the > >> LNMSHASHTBL Entry.:4 > >> Or, try increasing this, needs a reboot though. > >> > >> > >iL > > Rob has the answer, but here is a little background.  Admittedly this isF > > based on info from a V4.x internals book, but I don't think it has > > changed since then.  > >fI > > First, Process Logical Names (LNM) are stored in the "context" of thea+ > > processs and JTQuota does control this.i > >t >.H > I suspect that this was intended to say that "JTQuota does not controlE > this". JTQuota controls only the size of the job table. The process E > table has to fit into the process allocation region in P1 space and?G > the size is controlled by the SYSGEN parameter CTLPAGES. Changing the G > value of the paramter requires a reboot. If you're hitting this limit G > then you'll wnat to increase the value of CTLPAGES (doubling probablyi > isn't unreasonable).   Hello>   To see the Ctlpages usage, useL http://h18000.www1.hp.com/support/asktima/operating_systems/CY-1027624800-1. html  H DCL for OpenVMS - Example C - How To Use SHOW PROCESS/MEMORY For Another Processb     or  L http://h18000.www1.hp.com/support/asktima/operating_systems/CY-1021490401-1. html  D DCL for OpenVMS - How To Use SHOW PROCESS/MEMORY For Another Process   Regardsh   Grard   Volem rien foutre al paisc   ------------------------------   Date: 11 Aug 03 19:21:08 +0200) From: p_sture@elias.decus.ch (Paul Sture)e) Subject: Re: Sizes of logical name tables ) Message-ID: <pP8AHwdraVsd@elias.decus.ch>t  Z In article <3F365C57.30EB37B@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes: > Jeff Cameron wrote:t >> rM >> I know the UAF quota value JTQuota governs the size of a job's job logicalpN >> name table. What parameters or quotas govern the size of the process, group" >> and system logical name tables. >> -M >> I have an old utility (executable image) written by a company no longer in)M >> business called SIMPART. The company's name escapes me now which simulates5N >> certain physical interactions in physics. It is aborting with a simple text2 >> message : "VMS error - Logical name table full" >> lM >> When I run the image interactively, the error does not occur. When it runs N >> with all of it's other images and scripts in a multi-process job is when it
 >> aborts. >>  N >> I have already determined it is not the job table that is running out, as IN >> have increased the Jtquota for the account that it runs under, and modifiedM >> the supporting scripts and tested the image by itself, and it does not usee% >> the job tables to define logicals.e >> lL >> How can I determine which logical name table it is compalining about, and$ >> How do I increase the table size? >>  
 >> Thank you.m > I > I'll be interested (most curiosity, though) to see the answers to these E > questions. I've seen application implementations that use *TONS* of * > process-level logicals with no problems.  D I have seen the problem of a system running out of space for logicalN names. This was due to a problem with Oracle not deleting process logical nameG tables as part process rundown. I couldn't reproduce this on any of ouriI test of production systems, but did find about 4000 orphaned logical name $ tables on a develpment system today.  A The cure (which dates back to 1997 in our case) is a piece of DCLtB which lists all logical name tables which are named something like> TNS_ORA_blah_'pid', and deleting those where 'pid' refers to a non existant process.    > J > I can only suggest to remember that JTQuota is stated in bytes (per HELPH > ADD /JTQUOTA in AUTHORIZE). Try jacking it up to some outrageous valueE > like 65535 (64K) or (WSQUOTA * 512) and try running the test again.y > E > Of course, its always possible that the app.'s is returning a bogus1. > status value and thus a bogus error message. >  That could also be true.   ------------------------------  % Date: Mon, 11 Aug 2003 15:34:51 +0100g* From: Nic Clews <sendspamhere@[127.0.0.1]> Subject: Re: VMS License auditsa' Message-ID: <bh89ea$hg4$1@lore.csc.com>t  
 Rolona wrote:e > " > Any contact details for a quote? >   G I had a little bit of feedback on this line, send me a private email tov$ the address at the bottom of my sig.  B I should disclaim here, that there may well also be others on thisF newsgroup capable and as not to show favouritism, make yourself known!   -- i? Regards, Nic Clews a.k.a. Mr. CP Charges, CSC Computer Sciences  nclews at csc dot comh   ------------------------------   End of INFO-VAX 2003.442 ************************