1 INFO-VAX	Sun, 17 Dec 2000	Volume 2000 : Issue 703       Contents:C Re: %INSTALL-I-NONSHRADR, image installed ignoring '/SHARE=ADDRESS'  Re: %MOUNT-F-DIFVOLMNT again Re: Decserver 700 ) FREE & EASY PICTURES! NO BULL SHIT!  5566 & How to determine current license usage* Re: How to determine current license usage
 Re: Max Users   Re: patch question (7.2-1 ALPHA) Re: puzzled by patches Re: puzzled by patches Re: puzzled by patches Re: puzzled by patches Re: RDB and a web browser  Re: RIP LN03, 1987-2000  Re: soon enough?C Re: standalone backup, disk-saveset image backups, and cluster size C Re: standalone backup, disk-saveset image backups, and cluster size  Re: UCX Bug  Re: UCX Bug  Re: UCX Bug - Re: VMS tools for Linux: EDT, TPU, TECO, LAT? & Re: when is SYS$NODE normally defined?& Re: when is SYS$NODE normally defined?& Re: when is SYS$NODE normally defined?& Re: when is SYS$NODE normally defined?& Re: when is SYS$NODE normally defined?& Re: when is SYS$NODE normally defined?  F ----------------------------------------------------------------------  % Date: Sun, 17 Dec 2000 10:55:43 +0100   From: Paul Sture <paul@sture.ch>L Subject: Re: %INSTALL-I-NONSHRADR, image installed ignoring '/SHARE=ADDRESS'+ Message-ID: <VA.000001e4.0f481a84@sture.ch>   D In article <91fr4t$bn1$2@info.service.rug.nl>, Phillip Helbig wrote:, > From: helbig@astro.rug.nl (Phillip Helbig) > Newsgroups: comp.os.vms N > Subject: Re: %INSTALL-I-NONSHRADR, image installed ignoring '/SHARE=ADDRESS'  > Date: 16 Dec 2000 13:36:29 GMT > 8 > In article <3A3AA9D6.7A3B4964@rcn.com>, Jonas Lindholm > <jlindholm@rcn.com> writes:  > R > > Even if you enable all privileges before you run VMSINSTAL it will run withoutR > > bypass because VMSINSTALL is disabling that privilege during the installation. > H > OK.  That might explain it.  Presumably, this is a security measure.  I > Note that I explicitly turned on SET PROC/PRIV=ALL and was not relying  K > on something VMSINSTAL does to give me the privs if needed.  OK, I guess  H > I should always log in as SYSTEM.  It seems to me, though, that there J > are some legitimate reasons why one might want to install stuff from an C > account other than SYSTEM, and why this might not have all privs   > enabled by default.  > K > I have done this many times in the past.  VMSINSTAL says I am not logged  8 > in as SYSTEM etc, but this was never a problem before. > G The VAX and Alpha VMSINSTAL procedures differ, but both have the lines:       $if .not. vmi$debug*    $then vmi$saved_privs = f$setprv("all")     $vmi$ == f$setprv("nobypass")	    $endif    Both also do a set uic [1,4].   S FWIW, I too have run many installations under my own user name without problems. I  S prefer it that way whenever there is more than one person responsible for managing  H a system, as the user name appears in vmsinstal.history or product show 
 history/full.   M At work we use VMSINSTAL for in house applications, so developers and system  N testers are also using it. Again it's nice to be able to audit the user names.   ___ 
 Paul Sture Switzerland    ------------------------------   Date: 17 Dec 2000 07:16:06 GMT* From: helbig@astro.rug.nl (Phillip Helbig)% Subject: Re: %MOUNT-F-DIFVOLMNT again . Message-ID: <91hp7m$sr6$4@info.service.rug.nl>  A In article <3A3C38F9.5B770E08@earthlink.net>, "David J. Dachtera" & <djesys.nospam@earthlink.net> writes:    > Phillip Helbig wrote:  > > A > > I'm waiting to see if patches will make this problem go away.  > * > I thought there were patches for this...  @ Yes, I just hadn't installed them.  Now I have, and all is fine.  J > > As I mentioned, I have a .COM file which mounts disks /CLUSTER if they> > > are hosted by the node in question and /SYSTEM otherwise.  > I > Not sure why you'd want to make that distinction, or take that specific % > action under that condition, but...   G The idea is that each node executes this file when it comes up.  If it  H hosts the disk, it mounts it /CLUSTER, so all can see it.  If not, then H it is already mounted on all the other nodes, so it is less overhead to  just do a mount /SYSTEM.   > > Until thisG > > error goes away (or even after---seems like a good idea), I want to F > > avoid this error causing the procedure (and thus mounting of other5 > > disks) to fail.  For some strange reason, neither  > > ' > >    $  ON SEVERE_ERROR THEN CONTINUE  > G > I think you want ON WARNING. The THEN clause you set takes effect for H > the specified severity OR GREATER (check the doc. if you don't believe > me).   I first had ON ERROR.    > > or > >  > >    $  SET NOON > >  > > has any effect.  > E > I have trouble believing that  - unless there is a subsequent error H > (especially a missing label - always fatal, as far as I can tell) that > causes the proc. to fail.   H I was really surprised as well!  However, the disk is patched, there is 2 no problem and I have more important things to do!  E > The key there is "subsequent". SET NOON remains in effect after any H > error. "ON condition" does not - you must reset the ON action you want3 > after each occurrence. This usually works for me:  > " > $ ON WARNING THEN GOSUB RESET_ON > 	. > 	. > 	. > $RESET_ON:" > $ ON WARNING THEN GOSUB RESET_ON
 > $ RETURN  F Right, but I DIDN'T reset it, so it COULDN'T have failed---but it did!  J > Remember also that the current ON processing is effective at the currentB > procedure depth only - it is *NOT* inherited by proc.'s that are
 > invoked.  & Right, but that was not an issue here.   ------------------------------  % Date: Sun, 17 Dec 2000 15:54:47 +0100 ( From: Bernd Eckstein <B.Eckstein@cli.de> Subject: Re: Decserver 700& Message-ID: <3A3CD3B7.F66E87A6@cli.de>   Robert Greene wrote: > N > Is this is a good place to ask a question about configuring a decserver 700? > K > How do I configure the decserver 700 to request it's load (WWENG2) from a  > specific IP node?  > C BOOTP, like DHCP, uses broadcasts to find an usable bootserver. Per 
 definitionE IP-Routers are borders to broadcast-domains. To rosolve your problem,  yourF routers must have an bootp-relais setup, something lika an "proxy" for the D bootp type of broadcasts. It depends on your router. Recent ones may7 satisfy your need. Tkae a look into the routers manual.    --  G Dem Phnomen eines Computerabsturzes wohnt ein stumpfer Starrsinn inne, A ein Starrsinn, im Vergleich zu dem die Bockigkeit eines Esels von G nahezu Einsteinscher Weisheit zu sein scheint.          (Stanislav Lem)   ( Mit freundlichen Gruessen / Best regardsC B.Eckstein, CLI GmbH - mailto:B.Eckstein@cli.de - http://www.cli.de C Matthiashofstr. 28, D-52064 Aachen - Fon: +49 241 47051-0, Fax: -89    ------------------------------   Date: 17 Dec 2000 12:23:37 GMT From: rdvkpy@worldex.com2 Subject: FREE & EASY PICTURES! NO BULL SHIT!  5566/ Message-ID: <91ib89$fqd$14851@venus.telepac.pt>    FREE PICTURE #1 6 http://www.cybererotica.com/ad/rrr-in.html?rrr_60624/A   FREE FUNY PICTURE #26 http://www.cybererotica.com/ad/rrr-in.html?rrr_60624/B   FREE FUNY VIDEO #16 http://www.cybererotica.com/ad/rrr-in.html?rrr_60624/C   FREE EXXTRA PICTURE!!!!!!!! 6 http://www.cybererotica.com/ad/rrr-in.html?rrr_60624/XT iswilipvzbeegiwujcidkevqxptqenebifxqdfuksbzfptsofegyfxedmdqqfcizfjfniyuynvqzwvbeslvz   ------------------------------  % Date: Sun, 17 Dec 2000 12:13:57 +0100 > From: "Jean-Franois Marchal" <jean-francois.marchal@x9000.fr>/ Subject: How to determine current license usage . Message-ID: <91i6t2$enp$1@reader1.imaginet.fr>   Bonjour  tous  A I would need to know what usage is made of some "concurrent user" < licenses, in order to prevent a "license exceeded ..." error= (eg for LSE, or some compilers), and buy some additional ones ! if near from the licensed number.    Is it possible ?   Cordialement Jean-Franois Marchal  X9000 - LYON (FR)    ------------------------------  # Date: Sun, 17 Dec 2000 11:44:16 GMT ( From: Terry Kennedy <terry@gate.tmk.com>3 Subject: Re: How to determine current license usage ' Message-ID: <G5pn9s.6p9@spcuna.spc.edu>   @ "Jean-Francois Marchal" <jean-francois.marchal@x9000.fr> writes:C > I would need to know what usage is made of some "concurrent user" > > licenses, in order to prevent a "license exceeded ..." error? > (eg for LSE, or some compilers), and buy some additional ones # > if near from the licensed number.  >  > Is it possible ?  J   On recent(-ish) versions of VMS (it's in V7.2-1 and not in V5.5-2H4, notK that that's a useful piece of information 8-) you can use the "show license  /charge" DCL command:    7_SERVER::$ sh lice/usage   O View of loaded licenses from node SERVER                17-DEC-2000 06:41:36.53   > ------- Product ID --------   ---- Unit usage information ----> Product            Producer       Loaded  Allocated  AvailableE ACMS               DEC        Unlimited license, no usage information > BLAH               BLAHBLAH            5          0          5  4         Terry Kennedy             http://www.tmk.com5         terry@tmk.com             Jersey City, NJ USA    ------------------------------  % Date: Sun, 17 Dec 2000 11:54:35 +0100   From: Paul Sture <paul@sture.ch> Subject: Re: Max Users+ Message-ID: <VA.000001e7.0f7dffe5@sture.ch>   1 In article <91d9av$ss8$1@nnrp1.deja.com>,  wrote:  > From: pasmith@ppg.com  > Newsgroups: comp.os.vms  > Subject: Max Users% > Date: Fri, 15 Dec 2000 14:20:20 GMT  > G > I'm running VMS 7.1 on an Alpha 2100.  Is there a way I can tell what D > was the maximum number of users that has been logged in since last@ > reboot? I know I have had at least 70 users at one time and myH > interactive login is set to 112 (The command 'Set login /inter=112' is? > the last thing in a script run on reboot).  There has been an G > occasional user indicating that they cannot connect.  These users are E > coming in through telnet.  I have run the command 'UCX SHOW SERVICE G > TELNET/ALL' and it indicates that the telnet maximum users was 60 out  > of a 70 limit." > Any suggestions will be helpful. >  > B I've just come across this posting from 1998, which might help ;-)   From - Fri Nov 13 15:08:17 1998  From: KAPLANSKYM@NCCCOT.AGR.CA Newsgroups: comp.os.vms + Subject: Re: Max number of concurrent users       G I have used the following 'brute force and ignorance method' procedure   for G many years to report the largest number of concurrent users.  It is run C as part of my daily housecleaning/bookkeeping procedure.  Of course F accounting has to be enabled and we create a new accounting file everyH day so there is only one day's worth of data to deal with, otherwise the" sort step would not work properly.     Marvin Kaplansky Agriculture Canada      ; $! Calculate maximum number of concurrent interactive users 7 $! Optional parameter overrides current accounting file  $!
 $ set noon4 $ accounting/full/process=interactive/type=process -    /output=maxusers.tmp 'p1'9 $ search/output=maxusers.tmp maxusers.tmp;-1 start,finish 5 $ sort/key=(pos=57,size=23) maxusers.tmp maxusers.tmp  $ open/read infile maxusers.tmp  $ count = 0 
 $ maximum = 0  $  $loop:$ $ read/end_of_file=over infile inrec? $ if f$extract(37,5,inrec) .eqs. "Start" then count = count + 1 @ $ if f$extract(37,6,inrec) .eqs. "Finish" then count = count - 1, $ if count .ge. maximum then maximum = count $ goto loop  $  $over: $ write sys$output "" E $ write sys$output "Maximum concurrent interactive users: ''maximum'"  $ write sys$output ""  $ close infile $ delete maxusers.tmp;*      ___ 
 Paul Sture Switzerland    ------------------------------   Date: 17 Dec 2000 07:17:25 GMT* From: helbig@astro.rug.nl (Phillip Helbig)) Subject: Re: patch question (7.2-1 ALPHA) . Message-ID: <91hpa5$sr6$5@info.service.rug.nl>  A In article <3A3C3BBD.787A9816@earthlink.net>, "David J. Dachtera" & <djesys.nospam@earthlink.net> writes:   . > > There is VMS721_MOUNT96-V0100 but it says: > > # > > 2  KITS SUPERSEDED BY THIS KIT:  > >  > >      None. > >  > > What gives?  >  > You'll find it as: > 4 > dec-axpvms-vms721_mount96-v0100--4.pcsi-dcx_axpexe >  > at   > > > ftp://ftp.service.digital.com/patches/public/vms/axp/v7.2-1/ > G > The shorter names (minus the PCSI "noise") are there for convenience, B > but don't include the kits themselves. I guess we're supposed to; > extrapolate to get the complete name (I hate that, FWIW).   F OK, I guessed right!  :-)  I find it strange that most things show up ) more than once in the FTP directories....    ------------------------------  % Date: Sun, 17 Dec 2000 10:55:45 +0100   From: Paul Sture <paul@sture.ch> Subject: Re: puzzled by patches + Message-ID: <VA.000001e5.0f4820c6@sture.ch>   D In article <91halm$p4f$1@info.service.rug.nl>, Phillip Helbig wrote:  H > I tried to install the newest TCPIP ECO for VAX, but it fails since I H > need a newer PCSI.  However, I couldn't find this in the obvious place > ? >    ftp://ftp1.service.digital.com/patches/public/vms/vax/v7.2  > ? Several of us have been waiting quite a while for it to appear. G IIRC someone suggested contacting CSC, but that's not a lot of use for   hobbyist users.   & > Also, for CC on ALPHA, the patch at  > E >    ftp://ftp1.service.digital.com/patches/public/vms/axp/v7.2/c/6.2  > J > fails because it says that it can't find saveset B.  Is there something 
 > missing? > H > Note that the latter is under 7.2, not 7.2-1.  However, I assume that G > the compiler patch should be applied if it is running under 7.2-1 as   > well---right?  >   K I assumed that since it wasn't in the V7.2-1 directory tree, I didn't need   it.   I > In general, shouldn't the directories for each operatingsystem versions E > contain directories for ALL compilers etc which can run under those I > operating-system versions?  In general, this seems to be the case, but  F > as mentioned above for C (and also for ADA) it seems to be that not 0 > everything is visible everywhere it should be. >    ___ 
 Paul Sture Switzerland    ------------------------------   Date: 17 Dec 2000 13:00:03 GMT* From: helbig@astro.rug.nl (Phillip Helbig) Subject: Re: puzzled by patches . Message-ID: <91idcj$583$2@info.service.rug.nl>  F In article <VA.000001e5.0f4820c6@sture.ch>, Paul Sture <paul@sture.ch> writes:   ( > > Also, for CC on ALPHA, the patch at  > > G > >    ftp://ftp1.service.digital.com/patches/public/vms/axp/v7.2/c/6.2  > > L > > fails because it says that it can't find saveset B.  Is there something  > > missing? > > J > > Note that the latter is under 7.2, not 7.2-1.  However, I assume that I > > the compiler patch should be applied if it is running under 7.2-1 as   > > well---right?  > M > I assumed that since it wasn't in the V7.2-1 directory tree, I didn't need   > it.    How sure are you about that?  K > > In general, shouldn't the directories for each operatingsystem versions G > > contain directories for ALL compilers etc which can run under those K > > operating-system versions?  In general, this seems to be the case, but  H > > as mentioned above for C (and also for ADA) it seems to be that not 2 > > everything is visible everywhere it should be.   ------------------------------  % Date: Sun, 17 Dec 2000 15:15:14 +0100   From: Paul Sture <paul@sture.ch> Subject: Re: puzzled by patches + Message-ID: <VA.000001e9.1035b257@sture.ch>   D In article <91idcj$583$2@info.service.rug.nl>, Phillip Helbig wrote:  H > In article <VA.000001e5.0f4820c6@sture.ch>, Paul Sture <paul@sture.ch>
 > writes:  > * > > > Also, for CC on ALPHA, the patch at  > > > I > > >    ftp://ftp1.service.digital.com/patches/public/vms/axp/v7.2/c/6.2  > > > N > > > fails because it says that it can't find saveset B.  Is there something  > > > missing? > > > L > > > Note that the latter is under 7.2, not 7.2-1.  However, I assume that K > > > the compiler patch should be applied if it is running under 7.2-1 as   > > > well---right?  > > O > > I assumed that since it wasn't in the V7.2-1 directory tree, I didn't need   > > it.  >  > How sure are you about that? >  Let's just check...    $ cc/ver) Compaq C V6.2-008 on OpenVMS Alpha V7.2-1    And the README at I http://ftp1.support.compaq.com/public/vms/axp/v7.2/c/6.2/ccae04062.README     Problems addressed in CCAE04062:  @ o  When the return statement of an old-style function definition;    contained a recursive call, the compiler would sometimes :    try to replace the recursive call with a loop constructB    that did not correctly intialize all the loop control elements.@   [First corrected in V6.2A ECO kit 4, compiler ident V6.2-009.]  , So it looks like my assumption was wrong :-(  A OK, off to download it and see if I can reproduce your problem...z ___y
 Paul Sture Switzerlandp   ------------------------------  % Date: Sun, 17 Dec 2000 16:12:15 +0100g  From: Paul Sture <paul@sture.ch> Subject: Re: puzzled by patchess+ Message-ID: <VA.000001ea.1069e512@sture.ch>g  D In article <91halm$p4f$1@info.service.rug.nl>, Phillip Helbig wrote:, > From: helbig@astro.rug.nl (Phillip Helbig) > Newsgroups: comp.os.vmsa > Subject: puzzled by patchesO  > Date: 17 Dec 2000 03:07:34 GMT > H > I tried to install the newest TCPIP ECO for VAX, but it fails since I H > need a newer PCSI.  However, I couldn't find this in the obvious place > ? >    ftp://ftp1.service.digital.com/patches/public/vms/vax/v7.2  > & > Also, for CC on ALPHA, the patch at  > E >    ftp://ftp1.service.digital.com/patches/public/vms/axp/v7.2/c/6.2s > J > fails because it says that it can't find saveset B.  Is there something 
 > missing? >1K Sorted. Accept the default of NO for the question about providing manuals. r  H How did I find out? I extracted KITINSTAL.COM from saveset A, and found  the following code:s   $   IF  (cc$full_docs) $   THEN& $       VMI$CALLBACK RESTORE_SAVESET B    H > Note that the latter is under 7.2, not 7.2-1.  However, I assume that G > the compiler patch should be applied if it is running under 7.2-1 as L > well---right?o > I > In general, shouldn't the directories for each operatingsystem versionslE > contain directories for ALL compilers etc which can run under thosetI > operating-system versions?  In general, this seems to be the case, but nF > as mentioned above for C (and also for ADA) it seems to be that not 0 > everything is visible everywhere it should be. >ME Having explored these, I'll agree. It would be much clearer that way.h ___ 
 Paul Sture Switzerlandf   ------------------------------  % Date: Sun, 17 Dec 2000 08:09:34 -0500i2 From: norm lastovica <norman.lastovica@oracle.com>" Subject: Re: RDB and a web browser* Message-ID: <3A3CBB0E.3A8EC060@oracle.com>  E access to oracle rdb via a web interface is commonly and easily done.cG use the rdb web agent.  checkout the rdb web site at www.oracle.com/rdbh or contact oracle directly.(   Antony Wardle wrote: > % > Anyone managed to get a web browsero" > changing their rdb database yet? > - > Im interested in setting up an easy way fort( > someone to be able to change things in% > an RDB database without knowing anyp > sql  > ' > I have heard of Attunity, but that isp > about all. >  > cheers >  > antony   -- r> norman lastovica / oracle rdb engineering / usa / 610.696.4685   ------------------------------  # Date: Sun, 17 Dec 2000 11:40:44 GMT ( From: Terry Kennedy <terry@gate.tmk.com>  Subject: Re: RIP LN03, 1987-2000' Message-ID: <G5pn3x.6oI@spcuna.spc.edu>2  / JF Mezei <jfmezei.spamnot@videotron.ca> writes:pB > This trusted heavy box had printed 124923 pages in its lifetime.  K   Heavy as in weight or heavy as in duty? I have printers that do that pageDK count in a month. Even the HP 5SiMX is rated at 100K (or is it 120K) pages/g month.  P > I had been told that there was a concrete boock in it to provide the vibrationL > resistance. But alas, after tearing its guts out, I found out this was notM > true... The main motor is pretty heavy though. And boy, there were a lot ofu > screws in that puppy.y  K   Many of the laser printers of that era used Canon engines. A friend had aeM DEC 2100-something (I think that's the model) which was the same Canon enginer as the HP IIP.  I   Parts Now! (http://www.partsnowllc.com) has all of the common parts forI3 this, omitting only the DEC-unique formatter card.    4         Terry Kennedy             http://www.tmk.com5         terry@tmk.com             Jersey City, NJ USA    ------------------------------   Date: 17 Dec 2000 07:18:22 GMT* From: helbig@astro.rug.nl (Phillip Helbig) Subject: Re: soon enough?e. Message-ID: <91hpbu$sr6$6@info.service.rug.nl>  C In article <91hhr0.bl.1@hans.myfqdn.de>, Hans.Bachner@altavista.netR (Hans Bachner) writes:    K > >      o  The problem commonly known as the "Y2038 bug" has  been  fixed.h > >cK > >It seems that the Y2k problem was not as bad as feared (or everyone got i7 > >everything fixed on time); what about the Y2038 bug?  > L > The Unix timebase (and so the related functions in the C Runtime Library) K > measure time by counting the seconds since 1-Jan 1970. They use a 32-bit  J > signed integer to store the time, which will go negative on 19-Jan 2038 E > after 03:14:07 a.m. It then will display 13-Dec 1901 - a Friday ;-)e > G > The above mentioned ECO modifies the time related functions of the C  I > Runtime Library to use an unsigned integer for storing the time, which  - > should make it safe until sometime in 2106.i  F Right.  I was just wondering whether all platforms have this fixed by # now, or is VMS a few decades ahead?G   ------------------------------   Date: 17 Dec 2000 07:09:25 GMT* From: helbig@astro.rug.nl (Phillip Helbig)L Subject: Re: standalone backup, disk-saveset image backups, and cluster size. Message-ID: <91hor5$sr6$2@info.service.rug.nl>  A In article <3A3C3412.99741635@earthlink.net>, "David J. Dachtera" & <djesys.nospam@earthlink.net> writes:    > > A non-saveset backup, i.e. > > * > >    $  BACK/LOG/VER/IMAGE DKA0: DKA700: > >  > > works fine.e > > / > > Is this a limitation of standalone backup? r > # > To the best of my knowledge, yes.k   OK.t  J > > I want to use standalone backup since DKA0:is a system disk and I wantH > > a clean copy.  I want to avoid removing it and connecting it and theH > > target disk to another machine, though I guess I'll have to do it if > > it's the only way. > F > Are there any non-system disks present that could be backed up, thenJ > make a /IMAGE copy to it of the system disk? You could then boot minimumH > from the system disk copy and make a saveset backup of the live system > disk in non-standalone.    Right, but not enough disks.  G > Another option might be to do a minimal startup from the usual systemiG > disk and then backup the system disk /IGNORE=INTERLOCK since the onlyWI > volatile files open will be things that you don't really care that muchsF > about (PAGE, SWAP, Error log) - the queue manager won't start up andJ > neither will most of the other stuff that causes potential corruption of > files.   OK.-  K > > It is NOT an option to do a non-disk-saveset backup, i.e. make an exactd > > copy of the disk.  >  > Why?  = I want more than one backup on that disk.  So, first, I did aaH non-saveset backup.  I'll use this on another system.  In the top-level F directory, I also put some save-set backups (including the one of the @ VAX system disk mentioned above).  Due to space, I'm ZIPping it.  I > Well, technically, a /IMAGE restore - regardless of the source - is notrF > an *EXACT* copy, since fragmentation of both files and free space isA > effectively eliminated. FIDs are preserved, but actual physical- > locations of extents are not.1  ( Right---I DO want /IMAGE, not /PHYSICAL.  J > The only advantage I've ever seen to backing up to a saveset rather thanI > disk-to-disk /IMAGE is that the latter doesn't preserve ACLs where diskuJ > to saveset to disk does preserve ACLs. I always kinda wondered if *THAT* > was a bug in BACKUP...   Probably....   ------------------------------   Date: 17 Dec 2000 07:11:09 GMT* From: helbig@astro.rug.nl (Phillip Helbig)L Subject: Re: standalone backup, disk-saveset image backups, and cluster size. Message-ID: <91houd$sr6$3@info.service.rug.nl>  C In article <91hfp2.bl.1@hans.myfqdn.de>, Hans.Bachner@altavista.net. (Hans Bachner) writes: w  - > Phillip Helbig (helbig@astro.rug.nl) wrote:  >  > >I tried to do > > > > >   $  BACK/LOG/VER/IMAGE DKA0: DKA700:[000000]TEST.BCK/SAVE > > L > >I get the error message that the target disk cluster size must be 1.  It J > >is 3 on the source disk (and 4 on the target disk---these are just the - > >defaults which I see no reason to change).  > L > You need to mount DKA700 first (not /FOREIGN !). Otherwise, BACKUP mounts J > the disk /FOREIGN and tries to create a so-called "sequential-disk save J > set". More details can be found in section 10.5.4 of the OpenVMS System  > Manager's Manual.h  ) Right, true, but that was not my problem.n  I > This kind of save sets made sense back in the days of relatively small -I > removable disks and allowed multivolume backups to disk media. I'm not oH > aware of anyone still using it these days (but then, I'm sure someone 	 > does!).   ? Aren't there still thousands of PDP production systems running?    ------------------------------  % Date: Sun, 17 Dec 2000 08:11:02 -0500e2 From: norm lastovica <norman.lastovica@oracle.com> Subject: Re: UCX Bug* Message-ID: <3A3CBB66.5C424B1C@oracle.com>   MrSignor wrote:e >  > Greetings, > L >    I have also confirmed the UCX bug posted by moi_is_me on 14/15 to these > two newsgroups > M >    how do we get engineering to see /acknowledge these and in turn fix themj > M >    is there a special e-mail address to report bugs to, or do we just crossn
 > our fingers  >    and hope ?  >   1 	why don't you simply call compaq vms support ands report the problem.  a   > -Fredt   -- w> norman lastovica / oracle rdb engineering / usa / 610.696.4685   ------------------------------  # Date: Sun, 17 Dec 2000 15:08:52 GMTi3 From: "MrSignor" <MrSignor@nospam_bellatlantic.net>  Subject: Re: UCX Bug7 Message-ID: <8G4%5.2138$TC3.690005@typhoon2.ba-dsg.net>g  2 > why don't you simply call compaq vms support and > report the problem. 4 being a hoppyist ... I don't have a support contract   ------------------------------  % Date: Sun, 17 Dec 2000 11:13:06 -0500s2 From: rdeininger@mindspring.com (Robert Deininger) Subject: Re: UCX BugL Message-ID: <rdeininger-1712001113060001@user-2iveb1t.dialup.mindspring.com>  l In article <8G4%5.2138$TC3.690005@typhoon2.ba-dsg.net>, "MrSignor" <MrSignor@nospam_bellatlantic.net> wrote:  4 > > why don't you simply call compaq vms support and > > report the problem.r6 > being a hoppyist ... I don't have a support contract     What happens if a non-contracted person calls the support center to report a bug?  Suppose he just wants to report it, and doesn't expect to get a fix?  Will they accept the call?    I think Hoff has said in the past that *security-related* bugs should be reported this way, since Compaq is interested in fixing theses ASAP.a  2 Does anyone have any experience along these lines?   -- n Robert Deininger rdeininger@mindspring.comp   ------------------------------  # Date: Sun, 17 Dec 2000 11:36:44 GMTs( From: Terry Kennedy <terry@gate.tmk.com>6 Subject: Re: VMS tools for Linux: EDT, TPU, TECO, LAT?' Message-ID: <G5pMx8.6nn@spcuna.spc.edu>   1 Jordan Henderson <jordan@lisa.gemair.com> writes:c) > In article <91h8dh$kok$1@joe.rice.edu>,t, > Jerry Leslie <leslie@clio.rice.edu> wrote: >>Charles Sandmann's ED editor:c >>1 >>   http://clio.rice.edu/EDstuff/ED_Overview.txt  >> >l5 > There's also SEDT that was recently discussed here.u >t1 >    http://www.ultranet.com/~anker/sedt/sedt.htm  > 5 > I haven't tried it, but it appears that it has been.9 > successfully ported to a number of UNIX and DOS/Windows-
 > systems.  I   Perhaps I have unusual expectations, but I tried TPU's EDT mode when itlI came out, along with the EMACS EDT mode, and SEDT (when it was bundled w/@H Pathworks) and I found them all woefully inadequate at EDT emulation, in& particular the 2 non-fullscreen modes.  H   I evaluated EDT+ from BBC and I gave them a 3-page, single-spaced memoF listing the incomptibilities with real EDT that I discovered. After a F brief period of not being able to believe this, they agreed that they G were, in fact, incompatibilities. They fixed almost all of them for me,cF leaving only a few complex ones for a future release (this was _years_, ago, so they may have all been fixed by now)  4         Terry Kennedy             http://www.tmk.com5         terry@tmk.com             Jersey City, NJ USAI   ------------------------------   Date: 17 Dec 2000 07:04:44 GMT* From: helbig@astro.rug.nl (Phillip Helbig)/ Subject: Re: when is SYS$NODE normally defined?a. Message-ID: <91hoic$sr6$1@info.service.rug.nl>  A In article <3A3C3A95.44AF1E8E@earthlink.net>, "David J. Dachtera" & <djesys.nospam@earthlink.net> writes:   B > You could do this near the end of your SYSTARTUP_VMS processing: > - > $ IF F$TRNLNM( "SYS$NODE" ) .EQS. "" THEN -t7 > $ DEFINE/SYSTEM/EXEC SYS$NODE 'F$GETSYI( "SCSNODE" )'d  F Yes, that is what I do.  However, I was thinking it was a kludge.  :-)   ------------------------------  % Date: Sun, 17 Dec 2000 04:27:42 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca>d/ Subject: Re: when is SYS$NODE normally defined? + Message-ID: <3A3C86FD.5F005C3@videotron.ca>,   Someone  wrote: / > > $ IF F$TRNLNM( "SYS$NODE" ) .EQS. "" THEN -c9 > > $ DEFINE/SYSTEM/EXEC SYS$NODE 'F$GETSYI( "SCSNODE" )'a     Very basic DCL question here...n  I I was under the impression that when you use the "-" to indicate that the N command continues on the next line, that the next line should NOT begin with a- $.  However, the above example seems to work.t  J What is the official policy and royal etiquette on whether continued lines  should begin or not with a "$" ?   ------------------------------  % Date: Sun, 17 Dec 2000 11:05:28 +0100   From: Paul Sture <paul@sture.ch>/ Subject: Re: when is SYS$NODE normally defined?s+ Message-ID: <VA.000001e6.0f5108a7@sture.ch>f  ; In article <3A3C86FD.5F005C3@videotron.ca>, JF Mezei wrote:i/ > From: JF Mezei <jfmezei.spamnot@videotron.ca>a > Newsgroups: comp.os.vms-1 > Subject: Re: when is SYS$NODE normally defined?3' > Date: Sun, 17 Dec 2000 04:27:42 -0500i >  > Someone  wrote:R1 > > > $ IF F$TRNLNM( "SYS$NODE" ) .EQS. "" THEN -n; > > > $ DEFINE/SYSTEM/EXEC SYS$NODE 'F$GETSYI( "SCSNODE" )'c >  > ! > Very basic DCL question here...  > K > I was under the impression that when you use the "-" to indicate that the.P > command continues on the next line, that the next line should NOT begin with a/ > $.  However, the above example seems to work.s > L > What is the official policy and royal etiquette on whether continued lines" > should begin or not with a "$" ? >aP THEN is a special case, because a $ is allowed before the command which follows.   From the help on IF:   Format  "   $ IF expression THEN [$] command   ___u
 Paul Sture Switzerland    ------------------------------   Date: 17 Dec 2000 12:58:30 GMT* From: helbig@astro.rug.nl (Phillip Helbig)/ Subject: Re: when is SYS$NODE normally defined?t. Message-ID: <91id9m$583$1@info.service.rug.nl>  4 In article <3A3C86FD.5F005C3@videotron.ca>, JF Mezei' <jfmezei.spamnot@videotron.ca> writes:     > Someone  wrote: 1 > > > $ IF F$TRNLNM( "SYS$NODE" ) .EQS. "" THEN --; > > > $ DEFINE/SYSTEM/EXEC SYS$NODE 'F$GETSYI( "SCSNODE" )'S >  > ! > Very basic DCL question here...b  H Note that the above code is not "production".  There might be spaces in F SCSNODE you would want to collapse, perhaps some additional single or  double quotes are needed etc.a  K > I was under the impression that when you use the "-" to indicate that thehP > command continues on the next line, that the next line should NOT begin with a/ > $.  However, the above example seems to work.i > L > What is the official policy and royal etiquette on whether continued lines" > should begin or not with a "$" ?  F While we're at it, what is the official policy and royal etiquette on A whether non-continued lines should begin or not with a "$", both  # interactively and in command files?7  D I used to have the prompt set to the machine name.  Since I started I using DECwindows, I found it nice to have it be "$", since I can cut and V paste a whole command line.E  E For those who do not know it, the book mentioned in the DCL-generated0* quote of the day is, in fact, a real book.  O -------------------------------------------------------------------------------dO ...he wondered why Digital decided to use the dollar sign as the default prompt L character.  It looked like the same dollar sign that glowed from the Digital salesman's eyes....F  O                                   ---Bruce Ellis, THE HITCHHIKER'S GUIDE TO VMS    -- iM Phillip Helbig                       Email .............. helbig@astro.rug.nllM Kapteyn Instituut                    Email ................. helbig@man.ac.ukeM Rijksuniversiteit Groningen          Tel. ................... +31 50 363 6647RM Postbus 800                          Fax .................... +31 50 363 6100cM NL-9700 AV Groningen                 Web ... http://www.astro.rug.nl/~helbig/t  5 My opinions are not necessarily those of my employer.    ------------------------------  % Date: Sun, 17 Dec 2000 15:15:13 +0100e  From: Paul Sture <paul@sture.ch>/ Subject: Re: when is SYS$NODE normally defined?S+ Message-ID: <VA.000001e8.1035acdd@sture.ch>o  D In article <91id9m$583$1@info.service.rug.nl>, Phillip Helbig wrote:, > From: helbig@astro.rug.nl (Phillip Helbig) > Newsgroups: comp.os.vmse1 > Subject: Re: when is SYS$NODE normally defined?s  > Date: 17 Dec 2000 12:58:30 GMT > 6 > In article <3A3C86FD.5F005C3@videotron.ca>, JF Mezei) > <jfmezei.spamnot@videotron.ca> writes: e >  > > Someone  wrote: 3 > > > > $ IF F$TRNLNM( "SYS$NODE" ) .EQS. "" THEN -A= > > > > $ DEFINE/SYSTEM/EXEC SYS$NODE 'F$GETSYI( "SCSNODE" )'F > >  > > # > > Very basic DCL question here...R > J > Note that the above code is not "production".  There might be spaces in H > SCSNODE you would want to collapse, perhaps some additional single or  > double quotes are needed etc.  > M > > I was under the impression that when you use the "-" to indicate that theoR > > command continues on the next line, that the next line should NOT begin with a1 > > $.  However, the above example seems to work., > > N > > What is the official policy and royal etiquette on whether continued lines$ > > should begin or not with a "$" ? > H > While we're at it, what is the official policy and royal etiquette on C > whether non-continued lines should begin or not with a "$", both  % > interactively and in command files?i >r/ Consider the following, in a command procedure:   
    $ dir -
    /date -    /size=all    DG which works as expected, listing all the files in the current directorya   Now,  
    $ dir -
    /date -
    $/size=all   0 This produces %DIRECT-W-NOFILES, no files found.  R Of course, after creating a file called $.TXT, it dutifully lists that. IOW, when L starting a continued line with a dollar, it gets interpreted as part of the D command. Adding a dollar to the start of the second line results in $ %DCL-W-MAXPARM, too many parameters.  P Issuing the same sequence of commands interactively, and recalling that command M demonstrates the command actually passed to DCL. It's fairly clear from this .N behaviour that a dollar should NOT be used to start continuation lines except P where it's part of the command, as in the example which started this discussion.  F > I used to have the prompt set to the machine name.  Since I started K > using DECwindows, I found it nice to have it be "$", since I can cut and m > paste a whole command line.e > R Or a bunch of command lines. That's fine at home, but at work I _really_ need the R prompt to tell me which of many machines I am working on. The downside of that is P I also need to switch to a wider screen or set the prompt back to the dollar to  edit command lines which wrap. ___w
 Paul Sture Switzerlandn   ------------------------------    Date: 17 Dec 2000 11:31:08 -05002 From: malmberg@eisner.decus.org (John E. Malmberg)/ Subject: Re: when is SYS$NODE normally defined? + Message-ID: <J39hAP$VDXGS@eisner.decus.org>c  - In article <3A3C3A95.44AF1E8E@earthlink.net>, 9 "David J. Dachtera" <djesys.nospam@earthlink.net> writes:o > Phillip Helbig wrote:i >>J >> I notice that, with machines with no DECnet installed, I have to define >> SYS$NODE by hand. > [snip] >.B > You could do this near the end of your SYSTARTUP_VMS processing: > - > $ IF F$TRNLNM( "SYS$NODE" ) .EQS. "" THEN -a7 > $ DEFINE/SYSTEM/EXEC SYS$NODE 'F$GETSYI( "SCSNODE" )'   @ If you must declare SYS$NODE manually, please use the following:  ! $if f$trnlnm("SYS$NODE") .eqs. ""l $then.I $   define/system/exec sys$node "''f$edit(f$getsyi("SCSNODE"),"TRIM")'::"  $endif  F f$getsyi(SCSNODE) returns trailing spaces, and SYS$NODE is expected to( have two colons following the node name.  I I would recommend locating the references for SYS$NODE and contacting the K software vendors involved to get the software to use the "SCSNODE" instead.o  J In the past I have only seen this in locally written site specific command@ files that were moved from a system with DECNET to one with out.   -Johnm wb8tyw@qsl.network represnting only my opinion.   ------------------------------   End of INFO-VAX 2000.703 ************************