1 INFO-VAX	Wed, 09 Oct 2002	Volume 2002 : Issue 558       Contents:0 Re: "Underqualified tipster backed HP merger"... Re: 'hobbyist' vms0 Re: A warning to use bounds checking in BASIC(?)0 Re: A warning to use bounds checking in BASIC(?)0 Re: A warning to use bounds checking in BASIC(?)0 Re: A warning to use bounds checking in BASIC(?)3 Alphaserver 4100 + 2 FC LP-8000 (freezing/reseting) 7 Re: Alphaserver 4100 + 2 FC LP-8000 (freezing/reseting) 7 Re: Alphaserver 4100 + 2 FC LP-8000 (freezing/reseting) & Booting a VMS cluster node using BOOTP. RE: C - overlaying variables in the same PSECT# Re: C: sizeof question with structs # RE: C: sizeof question with structs ' Re: CDRECORD 1.8.1 privilege(s) needed?  Re: Disable access log on CSWS Re: Disable access log on CSWS Re: Disable access log on CSWS Re: FMS: missing RPC$SHARE% Re: Grant an identifier to an image ? % Re: Grant an identifier to an image ? % Re: Grant an identifier to an image ? % Re: Grant an identifier to an image ? % Re: Grant an identifier to an image ? % Re: Grant an identifier to an image ? % Re: Grant an identifier to an image ? % Re: Grant an identifier to an image ?  Re: Hobbyist kit Contents  Re: Hobbyist kit Contents  legato for Openvms1 Re: logical name weirdness (was: RE: UAF Problem) , Re: MS Unix services for Windows and OpenVMS* Re: OpenVMS 7.3-1 and TCPware 5.6 question Re[2]: DLT8000 do not stream RMS duplicate key question Re: RMS duplicate key question Re: RMS duplicate key question Re: RMS duplicate key question Re: RMS duplicate key question) Scroll mouse question for Fred Kleinsorge  Re: singular and plural of VAX Re: singular and plural of VAX Re: singular and plural of VAX Re: singular and plural of VAXP Re: singular and plural of VAX (was: Re: Another security issue with the UCX POPP Re: singular and plural of VAX (was: Re: Another security issue with the UCX POP  Re: SYS$SYSROOT and SYS$SPECIFIC  Re: SYS$SYSROOT and SYS$SPECIFIC  Re: SYS$SYSROOT and SYS$SPECIFIC  Re: SYS$SYSROOT and SYS$SPECIFIC  Re: SYS$SYSROOT and SYS$SPECIFIC  Re: SYS$SYSROOT and SYS$SPECIFIC  Re: SYS$SYSROOT and SYS$SPECIFIC Re: tcp/ip == UCX / Re: Which UIC to use for a product's username ? / Re: Which UIC to use for a product's username ? " Re: Writing a new VMS filesystem ?  F ----------------------------------------------------------------------   Date: 9 Oct 02 13:43:02 +0200 ) From: p_sture@elias.decus.ch (Paul Sture) 9 Subject: Re: "Underqualified tipster backed HP merger"... ) Message-ID: <WkvLawWVDYHT@elias.decus.ch>   W In article <3DA247E6.3050501@tsoft-inc.com>, David Froble <davef@tsoft-inc.com> writes:  > Jan C. Vorbrggen wrote: > H >>>When I was hired, one of the documents they required was a copy of myF >>>degree. Fortunately I could still find the original paper and got a >>>whole passel of copies made.  >>>  >>  K >> That's not SOP in the US? My, are you believing in the integrity of your  >> future employees. >  > O > Actually, it is not common in the US.  In any case paper can be forged.  The  N > best check is to contact the institution.  Good thing, since I'm not sure I L > could lay hands on my degree.  Not sure if I still deserve it either.  :-) > F FWIW I needed to lay my hands on my degree certificate a few years agoG at short notice, during a stint far from home. The university concerned G was quite happy to provide proof of my degree, even though it wasn't as / impressive looking as the original certificate.    --  
 Paul Sture Switzerland    ------------------------------  # Date: Wed, 09 Oct 2002 06:17:26 GMT ) From: Andrew Balaam <abalaam@yahoo.co.uk>  Subject: Re: 'hobbyist' vms 4 Message-ID: <20021009.6172600.4175751488@imagnu.geo>  * It was 'cripple-ware'. No network adapter.  , The last version only had a console port.=20  I the previous version also had a few terminal lines, but running PPP over=  =20 C one of those connected to the host machine resulted in very poor=20  throughput.   I TS10 in comparison is great - the networking works, and makes the system=  =20  useable.      6 >>>>>>>>>>>>>>>>>> Original Message <<<<<<<<<<<<<<<<<<  I On 09/10/02, 03:19:07, "David J. Dachtera" <djesys.nospam@fsi.net> wrote=  =20  regarding Re: 'hobbyist' vms:      > Bob Koehler wrote: > > C > > In article <anljau02vhc@enews4.newsguy.com>, "Zane H. Healy"=20 $ <healyzh@shell1.aracnet.com> writes: > > > H > > > From what I was told a couple days ago (admitadly not by them),=20
 apparentlyI > > > they've decided to no longer offer a Hobbyist version of their emu=  lator. > > I > >    For a while their hobbyist page had an explanaiton.  Now I can no=   ' > >    longer find their hobbyist page.   E > As it was explained to me, it was partly their fault. They made the I > hobbyist version time restricted, presumably due to the lack of anythi=  ngG > similar to LMF in the host (Windows) environment. Someone figured out I > how to defeat that, posted it on the web, and the rest - as they say -=   ? > is history. Had they just found some creative ways to make it I > cripple-ware, yet still useful to the hobbyist, it might still be arou=  nd > today.   > Hindsight is always 20/20...   > -- > David J. Dachtera  > dba DJE Systems  > http://www.djesys.com/  * > Unofficial Affordable OpenVMS Home Page:! > http://www.djesys.com/vms/soho/    ------------------------------   Date: 9 Oct 02 14:01:29 +0200 ) From: p_sture@elias.decus.ch (Paul Sture) 9 Subject: Re: A warning to use bounds checking in BASIC(?) ) Message-ID: <exYmrGk4hB9t@elias.decus.ch>   g In article <antnq6$9fi$1@slb4.atl.mindspring.net>, "Randy Park" <rjpark@mindspring.nospaam.com> writes:  > : > Gerald Marsh <gerald@cyfer.demon.co.uk> wrote in message4 > news:u4t3qus2ldmnq8rl5ualph58c6odrv2094@4ax.com.../ >> I hope a compiler guru can help on this one.  >>D >> The following is a cut-down version of a program to demonstrate a' >> potential problem with the debugger:  >>! >> $ type big_array_exception.bas  >> !* >> ! Create large integer array and access >> >>8 >>         declare INTEGER Int_Array1(5%),Int_Array2(3%) >> >> >>         N% = 0% >>         FOR I%=1% TO 15% 4 >>                 Int_Array1(I%) = I%*2+123123+I%/3B >>                 PRINT "Element ";I%;" contains ";Int_Array1(I%) >>         NEXT I% >> >> >>         Int_Array2(0%) = 15%  >> >>* >>         PRINT "Array2 =";Int_Array2(0%) >> >> >> >>         PRINT  "Finished..."  >>G >> I know that N% had might as well not been there and the program name I >> is a bit of a misnomer but a compilation without array bounds checking 9 >> I thought would result in an ACCVIO where I% became 6.  >> >> The results are:  >> >>
 >> $ reca bas + >> $ basic big_array_exception/chec=nobound 
 >> $ reca lin  >> $ link big_array_exception  >> $ reca ru >> $ run big_array_exception >> Element  1  contains  123125  >> Element  2  contains  123127  >> Element  3  contains  123130  >> Element  4  contains  123132  >> Element  5  contains  123134  >> Element  6  contains  123137  >> Element  7  contains  123139  >> Element  8  contains  123141  >> Element  9  contains  123144   >> Element  10  contains  123146  >> Element  11  contains  123148  >> Element  12  contains  123151  >> Element  13  contains  123153" >> Element  14  contains  17694728 >> Element  15  contains  65584  >> Array2 = 15 >> Finished... >> >>H >> Up to the non-existent element 13 the calculation works but where the' >> hell did the value for 14 come from?  >>H >> I noted that there was no accvio produced but if I%'s loop goes to 16C >> above the program accvios after the string "Finished" is output.  >>G >> Is this a feature of Alpha architecture and is it a method to ensure  >> we all use /CHECK=BOUNDS? >>I >> I declared the second array to see if it would encroach on the first's E >> boundaries - I suspect that a 64 byte boundary has something to do 
 >> with this.  >>B >> (I've logged a call with CSC regarding strange behaviour of theH >> debugger with this program - SET BREAK/EXCEPTION doesn't! - The stack >> gets corrupted.)  >>* >> Bye for now and keep the flag a-flying! >>
 >> Gerald. >> >> Gerald Marsh  >>2 >> gerald -at- cyfer -dot- demon -dot- co -dot- uk > < > Based upon your output, I would guess that INTEGER default > is LONG (4 bytes). > ? > NoBounds checking means that array bounds will not be checked ? > at run time.  If you exceed the array bound, you get what you : > get.  You may stomp on other data fields, you may access; > allocated but unused memory, or you may attempt to access ? > (read or write) outside your allocated memory.  If the latter = > happens you will get an ACCVIO error.  You might even stomp  > on string headers. > 9 > Int_Array1 has 6 elements (0 to 5) and Int_Array2 has 4 @ > elements (0 to 3).  The compiler probably allocates Int_Array2< > immediatly after Int_Array1 in the same PSECT.  There's no< > gap in allocation between these two arrays.  Other integer= > data may follow in the same PSECT, but aligned on an 8 byte  > boundary.  > = > I have a rather large program that uses no bounds checking, : > but the code that accesses arrays is well controlled and; > doesn't exceed array bounds and on occasion does it's own < > array bounds checking.  I get a 10% reduction in CPU usage  > in some modules by doing this. >   B And on the other side of the coin I once maintained an application= with several hundred .EXEs. The performance hit involved with = bounds checking was not measurable for this application, so I / recompiled the lot with bound checking enabled.   @ The benefit was _huge_ in that I was able to get straight to theA code concerned from a fax of the traceback file and usually get a ? tested fix in the post the same day (back in the days when most  sites didn't have a modem).    --  
 Paul Sture Switzerland    ------------------------------  $ Date: Wed, 9 Oct 2002 11:42:25 -0400; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com> 9 Subject: Re: A warning to use bounds checking in BASIC(?) $ Message-ID: <3da44e9f$1@news.si.com>  , >When run without the debugger, an ACCVIO isE >produced (but no Basic mem managemen violation message!). If stepped 0 >through with the debugger the damn things works  G Very typical.  The debugger instruments the code, adding to the running J image.  This extra could very well mask the out-of-bounds condition.  I've seen it happen.  --A Brian Tillman                   Internet: tillman_brian at si.com A Smiths Aerospace                          tillman at swdev.si.com = 3290 Patterson Ave. SE, MS      Addresses modified to prevent < Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  + Date: Wed, 09 Oct 2002 18:13:09 +0100 (MET) 9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> 9 Subject: Re: A warning to use bounds checking in BASIC(?) ; Message-ID: <01KNGXOR7V8S9QWKOO@sysdev.deutsche-boerse.com>   / > > When run without the debugger, an ACCVIO is H > > produced (but no Basic mem managemen violation message!). If stepped3 > > through with the debugger the damn things works  > I > Very typical.  The debugger instruments the code, adding to the running G > image.  This extra could very well mask the out-of-bounds condition.   > I've seen it happen.    C I suppose this is one reason why /DEBUG=(NOSYMBOLS,TRACEBACK) is a  F sensible default---not too much to weigh you down, but enough to give H SOME hint if something crashes.  I suppose some folks do a /NODEBUG for ' production code delivered to customers.   G While I'm thinking about it, let me float my suggestion again that ALL  G HELP files be rewritten to look like HELP FORTRAN.  (That is, show the  D default of all qualifiers in the "Additional information available" C section, and of course repeat it in the section for each qualifier.    ------------------------------   Date: 9 Oct 2002 11:58:59 -0600  From: briggs@encompasserve.org9 Subject: Re: A warning to use bounds checking in BASIC(?) 3 Message-ID: <2ez7Vz3hko7T@eisner.encompasserve.org>   w In article <01KNGXOR7V8S9QWKOO@sysdev.deutsche-boerse.com>, Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> writes: 0 >> > When run without the debugger, an ACCVIO isI >> > produced (but no Basic mem managemen violation message!). If stepped 4 >> > through with the debugger the damn things works >>  J >> Very typical.  The debugger instruments the code, adding to the runningH >> image.  This extra could very well mask the out-of-bounds condition.  >> I've seen it happen.  > E > I suppose this is one reason why /DEBUG=(NOSYMBOLS,TRACEBACK) is a  H > sensible default---not too much to weigh you down, but enough to give J > SOME hint if something crashes.  I suppose some folks do a /NODEBUG for ) > production code delivered to customers.   = In my experience, the "beneficial" effects of the debugger on ? program execution are typically felt only when the debugger has ! actually been mapped at run time.   A Whether you compile with or without /DEBUG=SYMBOLS is irrelevant. : What matters is whether you actually activate the debuggerH with $ RUN /DEBUG, $ LINK /DEBUG, $ DEBUG or LIB$SIGNAL(%VAL(SS$_DEBUG))  = Yes, having the symbol table floating around in P0 space can  ; change things, but having the debugger mapped (and possibly ; having it write-enable all your code pages) tends to change  things even more.    	John Briggs   ------------------------------  * Date: Wed, 9 Oct 2002 04:49:29 -0700 (PDT). From: Fabio Cardoso <fabiopenvms@yahoo.com.br>< Subject: Alphaserver 4100 + 2 FC LP-8000 (freezing/reseting)@ Message-ID: <20021009114929.58069.qmail@web20208.mail.yahoo.com>  	 Dear Sirs   1 We installed an aditional Fibre Channel (FC) card 9 in our AS-4100, and since then, the server is hanging and > reseting. Any considerations about two FC cards (no multipath): in AS-4100 ? I was thinking in the possibility of this old< server with old bus, dont support two FC cards. We splitted 6 the disks in the EMC for I/O balancing (no multipath).    ; OpenVMS 7.2-1h1  (last patches applied UPDATE 5.0 + FC 5.0)      Regards    FC   =====  ========================== Fbio dos Santos Cardoso OpenVMS System Manager Rio de Janeiro - Brazil  fabiopenvms@yahoo.com.br ==========================  2 __________________________________________________ Do you Yahoo!?2 Faith Hill - Exclusive Performances, Videos & More http://faith.yahoo.com   ------------------------------  % Date: Wed, 09 Oct 2002 14:57:02 +0200  From: Dirk Munk <munk@home.nl>@ Subject: Re: Alphaserver 4100 + 2 FC LP-8000 (freezing/reseting)$ Message-ID: <3DA4279E.40003@home.nl>  J Did you upgrade the firmware of the AS-4100 (and maybe of the FC cards )?.D The 4100 has 2 or 4 pci busses ( I always  confuse the 4000 and the H 4100). Put every FC card in its own PCI bus, and all should work well I  guess.A Are you sure the EMC has all the necesarry patches and settings ? E And please do use multipath, you can still do a bit of I/O balancing  G with that. You will be a lot happier with automatic failover if one of  " the paths fails, I can assure you.   Regards, Dirk   Fabio Cardoso wrote:  
 >Dear Sirs > 2 >We installed an aditional Fibre Channel (FC) card: >in our AS-4100, and since then, the server is hanging and? >reseting. Any considerations about two FC cards (no multipath) ; >in AS-4100 ? I was thinking in the possibility of this old = >server with old bus, dont support two FC cards. We splitted  7 >the disks in the EMC for I/O balancing (no multipath).  >  > < >OpenVMS 7.2-1h1  (last patches applied UPDATE 5.0 + FC 5.0) >  >  >Regards >  >FC  >  >===== >==========================  >Fbio dos Santos Cardoso  >OpenVMS System Manager  >Rio de Janeiro - Brazil >fabiopenvms@yahoo.com.br  >==========================  > 3 >__________________________________________________  >Do you Yahoo!? 3 >Faith Hill - Exclusive Performances, Videos & More  >http://faith.yahoo.com  >    >    ------------------------------  * Date: Wed, 9 Oct 2002 09:28:22 -0700 (PDT). From: Fabio Cardoso <fabiopenvms@yahoo.com.br>@ Subject: Re: Alphaserver 4100 + 2 FC LP-8000 (freezing/reseting)? Message-ID: <20021009162822.5757.qmail@web20208.mail.yahoo.com>    Dirk  A Thew AS-4100 has the lastest versions of firmware and the FC too! G This morning my server freezed two times. At the second time we changed 9 the slot of the new FC to another slot (other bus). Until F now the server didt freeze. I must wait until 15:00-16:00 hrs, because" it is the peak time of server use.  L If the machine dont freeze again, may be the problem is related to the other* bus. I had a machine check related to IOD.   regards  FC  # --- Dirk Munk <munk@home.nl> wrote: L > Did you upgrade the firmware of the AS-4100 (and maybe of the FC cards )?.F > The 4100 has 2 or 4 pci busses ( I always  confuse the 4000 and the J > 4100). Put every FC card in its own PCI bus, and all should work well I  > guess.C > Are you sure the EMC has all the necesarry patches and settings ? G > And please do use multipath, you can still do a bit of I/O balancing  I > with that. You will be a lot happier with automatic failover if one of  $ > the paths fails, I can assure you. > 
 > Regards, > Dirk >  > Fabio Cardoso wrote: >  > >Dear Sirs > > 4 > >We installed an aditional Fibre Channel (FC) card< > >in our AS-4100, and since then, the server is hanging andA > >reseting. Any considerations about two FC cards (no multipath) = > >in AS-4100 ? I was thinking in the possibility of this old ? > >server with old bus, dont support two FC cards. We splitted  9 > >the disks in the EMC for I/O balancing (no multipath).  > >  > > > > >OpenVMS 7.2-1h1  (last patches applied UPDATE 5.0 + FC 5.0) > >  > > 
 > >Regards > >  > >FC  > >  > >===== > >==========================  > >Fbio dos Santos Cardoso  > >OpenVMS System Manager  > >Rio de Janeiro - Brazil > >fabiopenvms@yahoo.com.br  > >==========================  > > 5 > >__________________________________________________  > >Do you Yahoo!? 5 > >Faith Hill - Exclusive Performances, Videos & More  > >http://faith.yahoo.com  > >    > >  >      =====  ========================== Fbio dos Santos Cardoso OpenVMS System Manager Rio de Janeiro - Brazil  fabiopenvms@yahoo.com.br ==========================  2 __________________________________________________ Do you Yahoo!?2 Faith Hill - Exclusive Performances, Videos & More http://faith.yahoo.com   ------------------------------   Date: 9 Oct 2002 07:06:13 -0700 6 From: jaq.vanrooijen@tiscalimail.nl (Hans van Rooijen)/ Subject: Booting a VMS cluster node using BOOTP = Message-ID: <2a3a3daf.0210090606.214fe93e@posting.google.com>   ? I created a OpenVMS clusternode using LANACP instead of DECnet. * The system is booting correctly using MOP.5 Is it also possible to use BOOTP as default protocol.   / On the bootserver is BOOTP already configured.    E EWA0_protocols on the station is set to BOOTP but the machine refuses ? to boot. It seems that the server is not offering a boot to the  station.   ------------------------------  $ Date: Wed, 9 Oct 2002 08:45:06 -0700# From: "Tom Linden" <tom@kednos.com> 7 Subject: RE: C - overlaying variables in the same PSECT 9 Message-ID: <CIEJLCMNHNNDLLOOGNJIKEBCFOAA.tom@kednos.com>   2 As always there are two sides to every coin, but I3 think union is preferrable because it unambiguously 1 documents the overlay; whereas, using pointers is 2 more of dynamic thing with the potential for error6 and is more appropriate for use with templates and the like   >-----Original Message----- ' >From: Kevin Handy [mailto:kth@srv.net] ' >Sent: Monday, October 07, 2002 3:55 PM  >To: Info-VAX@Mvb.Saic.Com8 >Subject: Re: C - overlaying variables in the same PSECT >  >  >Joe wrote: * >> Compaq C V6.5-001 on OpenVMS Alpha V7.3 >>H >> Sent the below to support last Thursday. Haven't heard anything back. >> They must be rather busy... >>E >> Sorry about any wrap on this. Supposing that I want to overlay two I >> variables on the same memory (do the same thing as set up multiple MAP G >> statements in BASIC where the MAP statements have the same "name")   >> how might I do this in C? >>A >> From the docs what I seem to want is some variation of #pragma G >> extern_model. I am trying the following but this doesn't seem end up C >> with the two variables in question starting at the same address:  >>A >> #include "socket.h"     // TPCIP headers among other things...  >> >> struct IOSB { >>    short  status; >>    short  XferSize; >>    void   *ptr; >> };  >> >> typedef struct FTPSRECORD {! >>    int                 RecNum;  >>    struct IOSB         iosb; ) >>    char                DeviceName[16]; B >>    struct sockaddr_in  rhost;   // the standard TCPIP struct..., >>    char                DottedAddress[16];' >>    char                FillSpace[3]; % >>    short               PortNumber;  >> } FTPSRECORD; >> >> #pragma extern_model save/ >> #pragma extern_model strict_refdef "GblSect"   >> nopic,ovr,noexe,wrt,shr,novec >> FTPSRECORD  ftps[64]; >> >>/ >> #pragma extern_model strict_refdef "GblSect"   >> nopic,ovr,noexe,wrt,shr,novec* >> char        GlobalSectionOverlay[8192]; >> #pragma extern_model restore  >> >> void main(void) { >> >> >> } >>F >> According to the linker map I'm ending up with the variables "ftps"G >> and "GlobalSectionOverlay" concatenated in memory. This is confirmed F >> if I run the above under the debugger and look at the locations viaI >> EV/ADDRESS(GlobalSectionOverlay starts roughly where I would expect it 3 >> to start given the size of the variable "ftps").  >>D >> Any pointers in the right direction examples greatly appreciated. > E >Wouldn't a 'union' work? That's the normal way to overlap using 'C', > >and it doesn't require any pragma's or setting up any special >sections to make it work. >  >  >---' >Incoming mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com). A >Version: 6.0.394 / Virus Database: 224 - Release Date: 10/3/2002  >  --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.394 / Virus Database: 224 - Release Date: 10/3/2002   ------------------------------   Date: 9 Oct 2002 16:16:43 GMT ( From: bill@cs.uofs.edu (Bill Gunshannon), Subject: Re: C: sizeof question with structs5 Message-ID: <ao1kpb$i9oi5$3@ID-135708.news.dfncis.de>   9 In article <CIEJLCMNHNNDLLOOGNJIIEPBFNAA.tom@kednos.com>, & 	"Tom Linden" <tom@kednos.com> writes: > ? > (Of course, on some machines, in the past anyway, NULL wasn't  > always zero)     ie. Pr1mes.   ; But I really thought they were the only ones.  Where else??    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  $ Date: Wed, 9 Oct 2002 09:30:11 -0700# From: "Tom Linden" <tom@kednos.com> , Subject: RE: C: sizeof question with structs9 Message-ID: <CIEJLCMNHNNDLLOOGNJIAEBFFOAA.tom@kednos.com>   ? Multics?  The null pointer was intended to be a designation for B a pointer which did NOT point to any storage, using "0" just looks sloppy.    >-----Original Message----- E >From: bill@gw5.cs.uofs.edu [mailto:bill@gw5.cs.uofs.edu]On Behalf Of  >Bill Gunshannon* >Sent: Wednesday, October 09, 2002 9:17 AM >To: Info-VAX@Mvb.Saic.Com- >Subject: Re: C: sizeof question with structs  >  > : >In article <CIEJLCMNHNNDLLOOGNJIIEPBFNAA.tom@kednos.com>,' >	"Tom Linden" <tom@kednos.com> writes:  >>  @ >> (Of course, on some machines, in the past anyway, NULL wasn't >> always zero)    >  >ie. Pr1mes. > < >But I really thought they were the only ones.  Where else?? >  >bill  >  >-- K >Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolves E >bill@cs.scranton.edu     |  and a sheep voting on what's for dinner.  >University of Scranton   | B >Scranton, Pennsylvania   |         #include <std.disclaimer.h>    >  >---' >Incoming mail is certified Virus Free. ; >Checked by AVG anti-virus system (http://www.grisoft.com). A >Version: 6.0.394 / Virus Database: 224 - Release Date: 10/3/2002  >  --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.394 / Virus Database: 224 - Release Date: 10/3/2002   ------------------------------  + Date: Wed, 09 Oct 2002 09:52:07 -0500 (CDT)  From: sms@antinode.org0 Subject: Re: CDRECORD 1.8.1 privilege(s) needed?) Message-ID: <02100909520780@antinode.org>   I From: "Eberhard Heuser-Hofmann" <Eberhard.Heuser-Hofmann@uni-konstanz.de>   $ > cdrecord needs: diag,phy_io,sysprv  F    I tried a few things and decided that DIAGNOSE and PHY_IO are all I
 seem to need.   . > You can install cdrecord with theses priv's.  9    Done.  (After I re-linked with /NOTRACEBACK.)  Thanks.   C    For some mysterious reason, executables linked recently are much F larger than ones linked in 2000 (old: 583, new: 2407, new/notraceback:@ 1689).  Same object files.  Strange.  Stranger still, my old VAXB executable was 1220 blocks, which agrees better with the new Alpha& sizes.  Life is filled with mysteries.  H ------------------------------------------------------------------------  C    Steven M. Schweda               (+1) 651-699-9818  (voice, home) C    382 South Warwick Street        (+1) 763-781-0308  (voice, work) G    Saint Paul  MN  55105-2547      (+1) 763-781-0309  (facsimile, work) 9    sms@antinode.org                sms@provis.com  (work)    ------------------------------   Date: 9 Oct 02 14:54:14 +0200 ) From: p_sture@elias.decus.ch (Paul Sture) ' Subject: Re: Disable access log on CSWS ) Message-ID: <UkPoqIoNrVMv@elias.decus.ch>   b In article <3DA3213D.9080703@vajhoej.dk>, Arne =?ISO-8859-1?Q?Vajh=F8j?= <arne@vajhoej.dk> writes: > Rick Barry wrote:  > O >> There is a serialization bottleneck when the Apache servers are doing access O >> logging that we plan to take a look at. It's not disk I/O latency that's the  >> problem.  > 3 > I am a bit confused. I do not know anything about 9 > Apache. But I can not understand why this serialization 8 > problem should be less for NLA0: than for access.log ? >   * I modified your program to include timers:   program z(input,output);  0 [ external ] procedure lib$init_timer; external;0 [ external ] procedure lib$show_timer; external; var      i,t1,t2 : integer;     context : integer;     status  : integer;     f       : text;      begin      context := 0;      open(f,'z.z',new);     rewrite(f);      t1 := clock;     lib$init_timer;      for i:=1 to 100000 do begin $        writeln(f,'This is a test!');     end;     lib$show_timer;      t2 := clock;
     close(f); (     writeln('Writing to z.z  - ',t2-t1);     open(f,'NLA0:',new);     rewrite(f);      t1 := clock;     lib$init_timer;      for i:=1 to 100000 do begin $        writeln(f,'This is a test!');     end;     lib$show_timer;      t2 := clock;
     close(f); (     writeln('Writing to NLA0 - ',t2-t1); end.  ) And here's the output (sorry if it wraps)    $ r nl_test M  ELAPSED:    0 00:00:03.24  CPU: 0:00:00.49  BUFIO: 49  DIRIO: 158  FAULTS: 3  Writing to z.z  -        490O  ELAPSED:    0 00:00:02.39  CPU: 0:00:02.07  BUFIO: 0  DIRIO: 100000  FAULTS: 0  Writing to NLA0 -       2070 $   C A hefty difference in CPU, and _every_ write to NLA0: results in a   direct I/O.    --  
 Paul Sture Switzerland    ------------------------------  % Date: Wed, 09 Oct 2002 18:24:47 +0200 6 From: Arne =?ISO-8859-1?Q?Vajh=F8j?= <arne@vajhoej.dk>' Subject: Re: Disable access log on CSWS ) Message-ID: <3DA4584F.9000808@vajhoej.dk>    Paul Sture wrote:   , > I modified your program to include timers: >  > program z(input,output); > 2 > [ external ] procedure lib$init_timer; external;2 > [ external ] procedure lib$show_timer; external; > var  >     i,t1,t2 : integer; >     context : integer; >     status  : integer; >     f       : text;  >  >  > begin  >     context := 0;  >     open(f,'z.z',new); >     rewrite(f);  >     t1 := clock; >     lib$init_timer; ! >     for i:=1 to 100000 do begin & >        writeln(f,'This is a test!');
 >     end; >     lib$show_timer;  >     t2 := clock; >     close(f); * >     writeln('Writing to z.z  - ',t2-t1); >     open(f,'NLA0:',new); >     rewrite(f);  >     t1 := clock; >     lib$init_timer; ! >     for i:=1 to 100000 do begin & >        writeln(f,'This is a test!');
 >     end; >     lib$show_timer;  >     t2 := clock; >     close(f); * >     writeln('Writing to NLA0 - ',t2-t1); > end. > + > And here's the output (sorry if it wraps)  > 
 > $ r nl_test O >  ELAPSED:    0 00:00:03.24  CPU: 0:00:00.49  BUFIO: 49  DIRIO: 158  FAULTS: 3  > Writing to z.z  -        490Q >  ELAPSED:    0 00:00:02.39  CPU: 0:00:02.07  BUFIO: 0  DIRIO: 100000  FAULTS: 0  > Writing to NLA0 -       2070 > $  > E > A hefty difference in CPU, and _every_ write to NLA0: results in a  
 > direct I/O.   " Yes. NLA0: is not particular fast.   Arne   ------------------------------  % Date: Wed, 09 Oct 2002 18:30:44 +0200 6 From: Arne =?ISO-8859-1?Q?Vajh=F8j?= <arne@vajhoej.dk>' Subject: Re: Disable access log on CSWS ) Message-ID: <3DA459B4.4070105@vajhoej.dk>    David Jones wrote:  E > In message <3DA3213D.9080703@vajhoej.dk>, <arne@vajhoej.dk> writes: 3 >>I am a bit confused. I do not know anything about 9 >>Apache. But I can not understand why this serialization 8 >>problem should be less for NLA0: than for access.log ? > H > NLA0: is a record-oriented device, so RMS may very well have differentQ > serialization issues (buffering, deferred writes, etc) with it than with a disk  > file.    True.   < But that is part of my confusion. Here is this serialization	 problem ?    It is demonstrated that:.    demo app                           demo app0    Pascal RTL                         Pascal RTL)    RMS              is faster than    RMS (    IO                                 IO+    disk-file                          NLA0:   ! The original suggestion was that: ,    Apache                             Apache+    C RTL                              C RTL )    RMS              is slower than    RMS (    IO                                 IO+    disk-file                          NLA0:    I will assume that: #    * C versus C RTL does not matter A    * Apache itself has not different code for disk-file and NLA0:   5 I suspect that it has to do with the way Apache works 0 with multiple processes, but I am very curious !   Arne   ------------------------------   Date: 9 Oct 02 14:09:18 +0200 ) From: p_sture@elias.decus.ch (Paul Sture) # Subject: Re: FMS: missing RPC$SHARE ) Message-ID: <M00ZYyLXqy8N@elias.decus.ch>   \ In article <3DA2F4AF.E3B3FCEB@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: > John Laird wrote:  >   * >> No, it doesn't.  That's FMS/TRANSLATE.  > N > Oh man, I don't know how I should feel.  Spending all this time fiddling and> > searching for files when I was using the wrong command !!!!! > G > Thanks for the information, even if it thows a big egg on my face :-)   H LOL! For a while I thought you were going to take us all the way throughF the FMS equivalent of Stan Rabinowitz' "A Day in the Life of the Image Activator".    --  
 Paul Sture Switzerland    ------------------------------   Date: 9 Oct 2002 03:25 CDT' From: carl@gerg.tamu.edu (Carl Perkins) . Subject: Re: Grant an identifier to an image ?, Message-ID: <9OCT200203253364@gerg.tamu.edu>  1 JF Mezei <jfmezei.spamnot@videotron.ca> writes...  }Ryan Moore wrote:C }> Oh, and you have to mount the disk with the executable using the 9 }> /SUBSYSTEM qualifier to enable subsystems on the disk.  } O }Oh shucks. That is a showstopper then.  If you make a small utility that would J }benefit from protected subsystems, it may be too much trouble for variousF }system managers to change their systems to support just that utility. } K }Out of curiosity, can the system disk be mounted with /SUBSYSTEM enabled ?   D Do a SHOW DEV/FULL on your system disk and you'll probably find thatC it already is mounted /SUBSYSTEM. (The info is in the Volume Status ! data near the end of the output.)    --- Carl   ------------------------------  % Date: Wed, 09 Oct 2002 09:08:16 +0200 6 From: =?iso-8859-1?Q?Jean=2DFran=E7ois=20PI=C9RONNE?= . Subject: Re: Grant an identifier to an image ?+ Message-ID: <3DA3D5E0.F0B278C7@laposte.net>f   From the documentation:r Using Protected Subsystems, R http://www.pi-net.dyndns.org/docs/openvms0731/731final/6346/6346pro.htm#a654539528  ( winston@SSRL.SLAC.STANFORD.EDU a crit : > ^ > In article <3DA35B3A.BBBB3831@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes:9 > >I know I can INSTALL an image and grant it privileges.f > >nC > >But its it possible to grant an image just a rights identifier ?  > >tO > >For instance, instead of giving an image SYSPRV to access a secured file, itTO > >would be nice to grant it some identifier that would then enable any user to H > >run the image and gain access to only that file. From an audit/systemN > >management point of view, it would make it easier to accept such a utility. > O > I think you just invented subsystem identifiers, which have been around for a Q > couple of releases.  (Unfortunately, I haven't used them, so I can't give you ae > good example.) > 	 > -- Alan    ------------------------------  $ Date: Wed, 9 Oct 2002 18:47:37 +1000: From: forMsytAhm@optusShome.com.aKu (Mark(un-MASK)Forsyth). Subject: Re: Grant an identifier to an image ?8 Message-ID: <slrnaq7r96.366.forMsytAhm@plague.bogus.com>  Y On Wed, 09 Oct 2002 00:56:35 -0400, JF Mezei <jfmezei.spamnot@videotron.ca> gushed forth:u >Ryan Moore wrote:C >> Oh, and you have to mount the disk with the executable using theu9 >> /SUBSYSTEM qualifier to enable subsystems on the disk.e >uO >Oh shucks. That is a showstopper then.  If you make a small utility that would J >benefit from protected subsystems, it may be too much trouble for variousF >system managers to change their systems to support just that utility.  B It's not that bad. You just need to get the system wrangler to putH a "set vol/subsystem <whatever>" in the startups somewhere. No big deal.   > K >Out of curiosity, can the system disk be mounted with /SUBSYSTEM enabled ?O  9 Mine gets set by "set volume/subsystem sys$sysdevice:" in$5 SYS$COMMON:[SYS$STARTUP]VMS$LPBEGIN-050_STARTUP.COM;1F     -- Y     Ooroon	 Mark F...F  $ Another Optus Cable Traffic Monitor.3 http://www.members.optushome.com.au/forsythm/traff/e   ------------------------------  $ Date: Wed, 9 Oct 2002 10:04:55 +0100* From: "Richard Brodie" <R.Brodie@rl.ac.uk>. Subject: Re: Grant an identifier to an image ?, Message-ID: <ao0rfo$1kmq@newton.cc.rl.ac.uk>  a "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message news:3DA3B702.505980AE@videotron.ca...u > Ryan Moore wrote:<D > > Oh, and you have to mount the disk with the executable using the: > > /SUBSYSTEM qualifier to enable subsystems on the disk. >eP > Oh shucks. That is a showstopper then.  If you make a small utility that wouldK > benefit from protected subsystems, it may be too much trouble for various G > system managers to change their systems to support just that utility.t >fL > Out of curiosity, can the system disk be mounted with /SUBSYSTEM enabled ?  @ It gets it by default; you can toggle with SET VOLUME/SUBSYSTEM.   ------------------------------  % Date: Wed, 09 Oct 2002 05:31:37 -0400m- From: JF Mezei <jfmezei.spamnot@videotron.ca>r. Subject: Re: Grant an identifier to an image ?* Message-ID: <3DA3F765.5D9864@videotron.ca>   Carl Perkins wrote: F > Do a SHOW DEV/FULL on your system disk and you'll probably find thatE > it already is mounted /SUBSYSTEM. (The info is in the Volume Statusr# > data near the end of the output.)u  J Yeah, the system disk on a node gets /subsystem. But not the disks I mountN myself. Interestingly, when I issue a MOUNT/cluster of the system disk to makeO it available to other nodes, those other nodes do not see the /subsystem in it.a  0 Someone else mentioned the SET VOLUME/SUBSYSTEM.  F Does the SEL VOLUME/SUBSYSTEM change something on the drive to make itN persistant, or does it apply only to the current mounted status of the drive ?  K i.e. if I do set VOL/SUBSYSTEM $DISK2,  the next time I boot up and issue aiK vanilla MOUNT/SYSTEM/CLUSTER command, will the "/SUBSYSTEM" be activated oniK all nodes or do I really need to add the /"SUBSYSTEM on the MOUNT command ?"  H Is there a reason why /SUBSYSTEM is not default ? Does it degrade system
 performance ?r   ------------------------------   Date: 9 Oct 2002 07:30:36 -0600 - From: koehler@encompasserve.org (Bob Koehler):. Subject: Re: Grant an identifier to an image ?3 Message-ID: <03VtPEDIxaIN@eisner.encompasserve.org>   V In article <9OCT200203253364@gerg.tamu.edu>, carl@gerg.tamu.edu (Carl Perkins) writes: > F > Do a SHOW DEV/FULL on your system disk and you'll probably find thatE > it already is mounted /SUBSYSTEM. (The info is in the Volume Status # > data near the end of the output.)       Not on any ofmine, it isn't.d   ------------------------------   Date: 9 Oct 02 15:18:57 +0200h) From: p_sture@elias.decus.ch (Paul Sture)e. Subject: Re: Grant an identifier to an image ?) Message-ID: <Yt4tjkvCtKpT@elias.decus.ch>h  j In article <Pine.LNX.4.31.0210082116090.5761-100000@jaipur>, Ryan Moore <rmoore@rmoore.dyndns.org> writes:5 > Not quite. You didn't read the manual close enough.s > ? > SET SEC/ACL=(SUBSYSTEM,id=members_subsystem)  member_list.exe  >  > Then on the critical file: > = > set sec /acl=(id=members_subsystem,acc=(whatever)) list.txto > B > Oh, and you have to mount the disk with the executable using the8 > /SUBSYSTEM qualifier to enable subsystems on the disk. >   1 But for the system disk, it's enabled by default.   % There's also SET VOLUME/[NO]SUBSYSTEMc    % > On Tue, 8 Oct 2002, JF Mezei wrote:" >  >> Ryan Moore wrote:M >> > Read the Guide to System security to get the details.  There's a chapter  >> > on protected subsystems.q >> >> Thanks for the pointer. >> >> Here is the relevant info:p >>I >> UAF> ADD/IDENTIFIER MEMBERS_SUBSYSTEM /ATTRIBUTES=(SUBSYSTEM,RESOURCE)n >>T >> $SET SECURITY/ACL=(IDENTIFIER=MEMBERS_SUBSYSTEM,  ACCESS=CONTROL) MEMBER_LIST.EXE >> >># >> So it is simpler than I thought.c >> >  --  
 Paul Sture SwitzerlandR   ------------------------------  $ Date: Wed, 9 Oct 2002 11:57:38 -0400; From: "Brian Tillman" <tillman_brian@notnoone.notnohow.com>2. Subject: Re: Grant an identifier to an image ?$ Message-ID: <3da45231$1@news.si.com>  G >Does the SEL VOLUME/SUBSYSTEM change something on the drive to make itTG >persistant, or does it apply only to the current mounted status of the  drive ?k   The latter.  I just tried it.  --A Brian Tillman                   Internet: tillman_brian at si.comlA Smiths Aerospace                          tillman at swdev.si.coma= 3290 Patterson Ave. SE, MS      Addresses modified to prevent < Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------   Date: 9 Oct 2002 07:46:45 -0600>- From: Kilgallen@SpamCop.net (Larry Kilgallen)c" Subject: Re: Hobbyist kit Contents3 Message-ID: <hlsxZv3JT3Sa@eisner.encompasserve.org>E  c In article <GQW2XEqAGR8l@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes:eI > In article <3DA24A59.E9F7E31@osfn.org>, RCS / RI <rcs@osfn.org> writes:1 >> Stuart Johnson wrote: >>> N >>> Does anyone have any information on what is going to be in the forthcomingK >>> hobbyist kits from Montgar? I am hoping that it will contail the entire G >>> layered products disks, ie. be the full OS distribution and layerede# >>> products. I can dream, can't I?c >> rE >> Has there been any discussion of how the HP merger will affect the  >> Hobbyist program? > D > Certainly I would not expect any effect.  The Hobbyist Program wasF > authorized by the previous head of VMS Development and in my opinionF > the current head of VMS Development is even _more_ inclined to thinkE > it is a good idea.  The previous head of VMS Development is now hisbC > boss.  If upper HP management decided to micro-manage that aspecta. > of VMS it would be totally out of character. > E >> Was there any news about the Hobbyist at the Encompass conference?> > F > I haven't heard any, although I did talk with David Cathey about it.  C In a hallway conversation I spoke to David about this thread and he-@ concurred that there it would be highly unlikely that the New HPA would withdraw support for the hobbyist program.  He also assumedh> that if they were going to do so, they would have told him :-)   ------------------------------  % Date: Wed, 09 Oct 2002 18:08:29 +0200 6 From: Arne =?ISO-8859-1?Q?Vajh=F8j?= <arne@vajhoej.dk>" Subject: Re: Hobbyist kit Contents) Message-ID: <3DA4547D.7010808@vajhoej.dk>    David J. Dachtera wrote:  I > Well, if you can dream up a way to cram a couple gigabytes of data onto - > a 650MB CD-ROM, yes, I guess you can dream.r     I think it is called a DVD !   :-)p   Arne   ------------------------------  # Date: Wed, 09 Oct 2002 13:53:45 GMTC- From: "labadie" <labadie_g.tocardsa@decus.fr>s Subject: legato for Openvms 1 Message-ID: <JxWo9.6$nZ2.321593@news.cpqcorp.net>    Hello    Readingy4 http://www.legato.com/products/networker/openvms.cfm8 I wanted some more information about Legato for Openvms:) is it available for Openvms Alpha 7.2-2 ?b when is it available ?' I naively sent an email on that page at  openvms@legato.com and my mail came back $ The recipient name is not recognized !!   More information appreciated...i   regardsi   Grard   ------------------------------   Date: 9 Oct 02 13:11:02 +0200s) From: p_sture@elias.decus.ch (Paul Sture)c: Subject: Re: logical name weirdness (was: RE: UAF Problem)) Message-ID: <N77g2VhfO+3u@elias.decus.ch>a  _ In article <28f58ade.0210070357.9884818@posting.google.com>, kent.gabrin@home.se (Kega) writes:t~ > Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> wrote in message news:<01KN8BL1Q91E9OF8NK@sysdev.deutsche-boerse.com>... >  > ...cut...- > I >> Again, an interesting question.  IF the idea of separate roots on the lF >> system disk was introduced before search lists were, then it would G >> perhaps make sense.  At least, [SYSn...] would include the stuff in   >> SYS$COMMON. > G > IIRC the [SYSn...] were introduced in VMS V3.0. (I don't think V3 had.F > search lists.) In those days we used to have, at my previous work, aB > separate [SYS1...] tree to be used when we got problems with theG > ordinary [SYS0...]. The SYS1-tree were rather small with a tiny page-eG > & swapfile, no layered product etc. If we screwed up our SYS0-tree in_G > some way we just boot the system into SYS1 and fixed the problem (ford* > instans SYSUAF, BOOT-procedures, etc...)F > (No problem running EDT in line mode on a paper terminal LA120, have > you?)tA > The boot-routines of [SYS1...] had an automatic inquiry of whatyF > password SYSTEM (of the SYSUAF on [SYS1.SYSEXE]) should have. On theE > boot-floppy of the VAX 11/780 I hade made separate boot-routines to % > make it simple to boot the SYS1 up.e > F > Nostalgia: Using FLX to create these alternative to DEFBOO.CMD makes > me feel old. >   < Nostalgia indeed. I recall putting those files onto TU58s in% the correct order to speed things up.s  B > We even had a separate VMS 4 system [sys4...] to use when we didF > laboration tests of features introduced in V4 (VMSCluster, ACLs etc) > during off hours.c >   : I used to put Standalone Backup in [SYSE], and have one on< a data disk too. That was a very real time saver with TU58s.  F I would have used [SYSF], but that was used by the upgrade procedures.  F > In V2 and prior version SYS$SYSTEM was SYS$SYSDEVICE:[SYSEXE]. In V3H > it become [SYSn.SYSEXE] and in V4 it become the well known search listG > (but only when you were running VMSCluster). Correct me if I'm wrong.s >  I believe you have it right.   --  
 Paul Sture Switzerlanda   ------------------------------   Date: 9 Oct 2002 00:49:56 -0700t+ From: usenet_ihc@hotmail.com (Gert de Boom)o5 Subject: Re: MS Unix services for Windows and OpenVMS = Message-ID: <12ce5972.0210082349.48425977@posting.google.com>g  G > Using Microsoft Services for Unix we have installed a NFS server on a2C > Windows 2000 host. Our OpenVMS server connects to this NFS serverK$ > using the NFS client from TCPWare.E > Especially in setting user rights we have some problems. Anyone goti% > any valuable remarks/tips on this ?p >u! > SFU version 2.0, TCPWare V5.5-3l >o, It is working right now. This is what I did:  7 I Checked the NFS data send by the Alpha using EtherealhD (www.ethereal.com) on the W2K server. I saw the correct UID, GID andF even the correct attribute settings arriving, so the problem had to be NFS server on W2K.  C The SFU settings showed nothing weird. We planned to upgrade to 3.0 D anyway, so we deinstalled SFU 2 and installed SFU 3 and after makingB the correct settings (user mapping, etc) it appeared to be working again.  E We still had some trouble setting the correct file access rights, butnC that was solved using Microsoft Knowledge base articles Q262984 and C Q321049 (still don't understand why this isn't in the SFU manual or D even made available via a settings box). After maing some changes inE the registry of the Windows 2000 server and restarting the NFS server/ it worked like we expected.   F A note for other TCPWARE OpenVMS users: the SYSTEM account settings onF OpenVMS are ignored when writing via the TCPWare NFS client. Just likeC the Delete (D)-rights. Only the Owner, Group and World RWE-settingsa@ are transferred and mapped to the corresponding UNIX attributes.   Gert de Boom   ------------------------------  # Date: Wed, 09 Oct 2002 07:29:07 GMTy. From: peter@langstoeger.at (Peter LANGSTOEGER)3 Subject: Re: OpenVMS 7.3-1 and TCPware 5.6 question23 Message-ID: <7VQo9.53159$M52.584009@news.chello.at>   f In article <hZNo9.23661$zU5.971102@news20.bellglobal.com>, "Neil Rieck" <n.rieck@sympatico.ca> writes:! >>[ntp+timezone question snipped]s? >No need to respond to this note. I've discovered the solution.n   And the answer is ?s  I I've now a timezone and NTP config working, but I still wonder if it will : switch timezones at the right time correctly. We'll see...   -- ( Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist= E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  $ Date: Wed, 9 Oct 2002 13:46:19 +05004 From: Valentin Likoum <valentin.likoum@ncc.volga.ru>% Subject: Re[2]: DLT8000 do not stream,4 Message-ID: <8320338297.20021009134619@ncc.volga.ru>  4 On 08.10.2002 Alan Greig <a.greig@virgin.net> wrote:  4 > On Tue, 8 Oct 2002 16:54:13 +0500, Valentin Likoum' > <valentin.likoum@ncc.volga.ru> wrote:t   >>Hello all,  C >>  After reading docs and many advices about tuning backup account04 >>quotas I set up the account with following params:C >>wsextent = wsquota = wsdefault = PQL_MWSEXTENT = WSMAX = 1421312, ; >>fillm = 1000, diolm = 30000, astlm = 31000, biolm = 4000, 5 >>bytlm = 500000, enqlm = 32767, pgflquota = 2000000.s  D > Certainly all these quotas look big enough. However what is sysgenH > param channelcnt? If this isn't (significantly) greater than 1000 then& > backup can never use its fillm quota  D > If you are getting 4.7MB/sec uncompressed and 7.4MB/sec compressesF > sustained over a time period then you are not doing too badly. ThereD > might be scope for a little improvement. My experience is that theB > drives will stop/start from time to time even if continually fedH > although I do think you should be able to get a little better than you > see. [snip]  C   Channelcnt  = 2000 is bigger than fillm (1000), but IMHO it's notVF the issue in any case as I process small (<100) number of big (0.5-5G)B files. Also Dirk suggested cache=tape_data, I tried it though it's$ default for tape device - no effect.D   But it's not the thoughput what worry me the most. I know that theD shoe-shine means frequent start-stops of the tape, but how frequent?? Definitly one start-stop per hour is not the shoe-shine, so ones@ start-stop per 4-8 sec. (my case) is shoe-shine or not? Should I  expect quick death of the drive?     Thank you. -- D
 Best regards, #  Valentin                          u  valentin.likoum@ncc.volga.ru    ------------------------------  % Date: Wed, 09 Oct 2002 02:46:20 -04000- From: JF Mezei <jfmezei.spamnot@videotron.ca>s# Subject: RMS duplicate key question , Message-ID: <3DA3D0B4.F815C9AD@videotron.ca>  M Is there a way to define a file such that duppicate keys are not allowed on a(. secondary key, unless the key is blank/empty ?  K In order words, allow duplicates if blank, but if you fill something in, it  needs to be unique ?   ------------------------------  % Date: Wed, 09 Oct 2002 09:25:38 +0200i From: Dirk Munk <munk@home.nl>' Subject: Re: RMS duplicate key questionw& Message-ID: <3DA3D9F2.5070502@home.nl>   JF Mezei wrote:m  N >Is there a way to define a file such that duppicate keys are not allowed on a/ >secondary key, unless the key is blank/empty ?  > L >In order words, allow duplicates if blank, but if you fill something in, it >needs to be unique ?a >sI In a way, yes. If you define the key with a null-value (for blank), then lH the key will not be added to the index, thus avoiding the duplicate key H problem. However obviously you can't use that index to look for records C with that value in the key, so you will have to search the file in n. another way if you need to find those records.  G Always keep in mind that any key value should be as unique as possible aG if you want fast writes to an indexed file, specially in large files!. iA If the nature of the key is such that many duplicates can not be tI avoided, add data to the key that will make it (more) unique. That extra  H data may not have any function for your application, it is solely there I to make the key as unique as possible. Think of timestamps etc., easy to = add and to make a key unique.b   >  S >o   ------------------------------   Date: 9 Oct 02 15:29:52 +0200 ) From: p_sture@elias.decus.ch (Paul Sture)=' Subject: Re: RMS duplicate key question_) Message-ID: <MLlppKtcJuD8@elias.decus.ch>   \ In article <3DA3D0B4.F815C9AD@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes:O > Is there a way to define a file such that duppicate keys are not allowed on ao0 > secondary key, unless the key is blank/empty ? > M > In order words, allow duplicates if blank, but if you fill something in, it_ > needs to be unique ?  A Not as far as I am aware. Blank duplicates are often the cause ofeA severe performance problems. I try to define a NULL_KEY value for A those, but then of course, you can't do a lookup in the index forn blank keys.)   -- c
 Paul Sture Switzerlandn   ------------------------------   Date: 9 Oct 2002 07:34:18 -0700o4 From: patrick_omalley@hotmail.com (Patrick O'Malley)' Subject: Re: RMS duplicate key questiono= Message-ID: <6fd26891.0210090634.5fb53e42@posting.google.com>   L Dirk Munk <munk@home.nl> wrote in message news:<3DA3D9F2.5070502@home.nl>... > JF Mezei wrote:t > P > >Is there a way to define a file such that duppicate keys are not allowed on a1 > >secondary key, unless the key is blank/empty ?- > >-N > >In order words, allow duplicates if blank, but if you fill something in, it > >needs to be unique ?- > >lK > In a way, yes. If you define the key with a null-value (for blank), then .J > the key will not be added to the index, thus avoiding the duplicate key J > problem. However obviously you can't use that index to look for records E > with that value in the key, so you will have to search the file in e0 > another way if you need to find those records. > I > Always keep in mind that any key value should be as unique as possible hI > if you want fast writes to an indexed file, specially in large files!. cC > If the nature of the key is such that many duplicates can not be  K > avoided, add data to the key that will make it (more) unique. That extra eJ > data may not have any function for your application, it is solely there K > to make the key as unique as possible. Think of timestamps etc., easy to   > add and to make a key unique.  >  > >    > >h     JF,   @ Tried to email you, but the your email address (even without the> spamnot) doesn't work.  Investigate the NULL key attribute forE alternate duplicate keys.  It helps a lot.  In the FDL fil, you'd say  (among other things)   KEY 1e#         DUPLICATES              yesu#         NULL_KEY                yest#         NULL_VALUE              ' '>  C I once did a Decus session on this, and someone in the session went>F home, implemented the principles that I talked about, and cut down hisC processing time from 6 hours to under 10 minutes!!  He posted it on 3 the Internet at one point, but I can't find it now.b  F Let me know if you need more help, and I'll look for the presentation.=  You'd have to remove the "@removethis" in the address below.    Pat & Patrick_OMalley@removethis@hotmail.com   ------------------------------  # Date: Wed, 09 Oct 2002 14:44:18 GMT;9 From: Hein van den Heuvel <hein_netscape@eps.zko.dec.com> ' Subject: Re: RMS duplicate key questione/ Message-ID: <3DA43E74.E5AB0B8A@eps.zko.dec.com>>   JF Mezei wrote:_  O > Is there a way to define a file such that duppicate keys are not allowed on ao0 > secondary key, unless the key is blank/empty ? >tM > In order words, allow duplicates if blank, but if you fill something in, itp > needs to be unique ?  Q Yes, declare DUPLICATES NO (obviously!) and use an all spaces field with NULL KEYrP valeu SPACE to stop RMS from considering that key value for that index.. pace is just an example,Q ZEROES or NULLs or STARS or whatever will work fine also. Teh only caveat is that P you specify only one byte for null where the key will probably be several bytes.   Example:  ) $ convert/fdl=sys$input sys$input tmp.idxn file; org indexed; key 0; seg0_l 4dG key 1; seg0_p 4; seg0_l 1; duplicates no; null_key yes; null_value ' ';m <EXIT> aap  A miesBB nootAA <EXIT>( $ conv/fdl=nl: /key=1 tmp.idx sys$output nootAA miesBB $ conv/merge sys$input tmp.idx teunAi) %CONVERT-I-DUP, duplicate key encountered  teun A vuurB ) %CONVERT-I-DUP, duplicate key encountereds vuurC  <EXIT>( $ conv/fdl=nl: /key=1 tmp.idx sys$output nootAA miesBB vuurC3( $ conv/fdl=nl: /key=0 tmp.idx sys$output aap  A miesBB nootAA teun A vuurC-   ------------------------------  % Date: Wed, 09 Oct 2002 09:32:09 +0200o From: Dirk Munk <munk@home.nl>2 Subject: Scroll mouse question for Fred Kleinsorge& Message-ID: <3DA3DB79.8040004@home.nl>   Fred,L  I Can you give us detailed instructions how to use a USB scroll mouse with  A DECwindows? Since the firmware of my PWS does not know of USB, I  D obviously can't use a USB keyboard (and there is no LK USB keyboard I around yet). But using a scrollmouse with Mozilla to read this newsgroup cG would be great ! I have a supported card installed, and have it almost   working now.   ------------------------------  $ Date: Wed, 9 Oct 2002 10:08:24 +0100* From: "Richard Brodie" <R.Brodie@rl.ac.uk>' Subject: Re: singular and plural of VAXe, Message-ID: <ao0rm9$1d9o@newton.cc.rl.ac.uk>  a "John Laird" <john@yrl.co.uk> wrote in message news:od36quonhpttv29mdro6fv9cuu3ekoot2o@4ax.com...G  I > Shame the plurals of "ax", "tax", "pax", etc don't follow this rule :-) * > (And I've not heard of boxen, coxen...!)  ( http://www.dictionary.com/search?q=boxen   ------------------------------  + Date: Wed, 09 Oct 2002 12:54:53 +0100 (MET)l9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>d' Subject: Re: singular and plural of VAXf; Message-ID: <01KNGMV6JJFG9QWKOO@sysdev.deutsche-boerse.com>u  F From:	IN%"R.Brodie@rl.ac.uk"  "Richard Brodie"  9-OCT-2002 12:47:35.01 To:	IN%"Info-VAX@Mvb.Saic.Com" CC:	$ Subj:	RE: singular and plural of VAX  , Return-path: <Info-VAX-Request@Mvb.Saic.Com>> Received: from CONVERSION-DAEMON by sysdev.deutsche-boerse.comO  (PMDF V5.2-32 #40436) id <01KNGMM5B5G09QX1PI@sysdev.deutsche-boerse.com>; Wed,d   9 Oct 2002 12:47:34 +0100 (MET): Received: from mailp2.deutsche-boerse.de ([172.20.124.50])4  by sysdev.deutsche-boerse.com (PMDF V5.2-32 #40434)D  with ESMTP id <01KNGMM4YDJU9S5ST3@sysdev.deutsche-boerse.com>; Wed,!  09 Oct 2002 12:47:30 +0100 (MET)eL Received: from deimos1.deutsche-boerse.de () by mailhub.exchange.de (V1.3.2)1  id g99AlTt21090; Wed, 09 Oct 2002 12:47:29 +0200aM Received: (from uucp@localhost) by mail2.exchange.de (V1.3) id MAA27303; Wed,G%  09 Oct 2002 12:47:28 +0200 (MET DST)cM Received: from unknown(198.151.12.104) by deimos1.deutsche-boerse.de via smap-5  (V5.0)	id xmab27260; Wed, 09 Oct 2002 12:46:44 +02002% Date: Wed, 09 Oct 2002 10:08:24 +0100g( From: Richard Brodie <R.Brodie@rl.ac.uk>' Subject: Re: singular and plural of VAXi To: Info-VAX@Mvb.Saic.Comr, Reply-to: Richard Brodie <R.Brodie@rl.ac.uk>, Message-id: <ao0rm9$1d9o@newton.cc.rl.ac.uk>6 Organization: Rutherford Appleton Laboratory, Oxon, UK8 X-MIMEOLE: Produced By Microsoft MimeOLE V6.00.2800.1106 Content-type: TEXT/PLAIN
 X-Priority: 3D X-MSMail-priority: Normalt X-Gateway-From: mvb.saic.com	 Lines: 10 H X-Authentication-warning: deimos1.deutsche-boerse.de: uucp set sender to)  <Info-VAX-Request@Mvb.Saic.Com> using -fo= X-Comment: Message Virus scanned by sysdev.deutsche-boerse.del X-Gateway-source-info: USENET2 X-Newsgroups: comp.os.vmse6 X-Newsreader: Microsoft Outlook Express 6.00.2800.1106    a "John Laird" <john@yrl.co.uk> wrote in message news:od36quonhpttv29mdro6fv9cuu3ekoot2o@4ax.com...a  I > Shame the plurals of "ax", "tax", "pax", etc don't follow this rule :-) * > (And I've not heard of boxen, coxen...!)  ( http://www.dictionary.com/search?q=boxen   ------------------------------  + Date: Wed, 09 Oct 2002 13:06:21 +0100 (MET)o9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>:' Subject: Re: singular and plural of VAXh; Message-ID: <01KNGMVHGDPE9QWKOO@sysdev.deutsche-boerse.com>s  , > http://www.dictionary.com/search?q=3Dboxen  E    /bok'sn/ (By analogy with VAXen) A fanciful plural of box often=20vJ    encountered in the phrase "Unix boxen", used to describe commodity Uni= x=20>    hardware. The connotation is that any two Unix boxen are=20    interchangeable.=20    VAXen is a link; let's go there!  F    /vak'sn/ (From "oxen", perhaps influenced by "vixen") The plural=20E    canonically used among hackers for the DEC VAX  computers. "Our=20 3    installation has four PDP-10s and twenty vaxen."   H Yes, VAXen influenced by vixen.  I like that one!  Now let's get Russ=20( Meyer to film that Superbowl ad for VMS!  3 DEC is a link; let's go there!  The first entry is:   +    de=B7ceased   Pronunciation Key  (d-sst)e    adj.   0    No longer living; dead. See Synonyms at dead.   Sic transit gloria mundi!n  E DEC meaning Digital does eventually show up near the end of the list. ; The entry contains a link to Digital Equipment Corporation:        Digital Equipment Corporation   =20 ?    <company> (DEC) A computer manufacturer and software vendor.t   =20 I    Before the killer micro revolution of the late 1980s, hackerdom was=20tJ    closely symbiotic with DEC's pioneering time-sharing machines. The fir= st=20 J    of the group of hacker cultures nucleated around the PDP-1 (see TMRC).= =20gJ    Subsequently, the PDP-6, PDP-10, PDP-20, PDP-11 and VAX were all foci = of=20tI    large and important hackerdoms, and DEC machines long dominated the=20A+    ARPANET and Internet machine population.S   =200J    DEC was the technological leader of the minicomputer era (roughly 1967= =20AJ    to 1987), but its failure to embrace microcomputers and Unix early cos= t=20J    it heavily in profits and prestige after silicon got cheap. However, t= he=20-E    microprocessor design tradition owes a heavy debt to the PDP-11=20-A    instruction set, and every one of the major general-purpose=20nJ    microcomputer operating systems so far (CP/M, MS-DOS, Unix, OS/2) were= =20oJ    either genetically descended from a DEC OS, or incubated on DEC hardwa= re=20eJ    or both. Accordingly, DEC is still regarded with a certain wry affecti= on=20;J    even among many hackers too young to have grown up on DEC machines. Th= e=20$    contrast with IBM is instructive.   =20 6    Quarterly sales $3923M, profits -$1746M (Aug 1994).   =20 @    DEC was taken over by Compaq Computer Corporation in 1998.=20   Sic transit gloria mundi.   A IBM is a link; the entry includes "Inferior But Marketable".  :-|m   ------------------------------    Date: 09 Oct 2002 13:42:50 -0400- From: Rich Alderson <alderson+news@panix.com> ' Subject: Re: singular and plural of VAX . Message-ID: <mddk7krbjad.fsf@panix5.panix.com>  ; Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> writes:   " > VAXen is a link; let's go there! > E >    /vak'sn/ (From "oxen", perhaps influenced by "vixen") The plural D >    canonically used among hackers for the DEC VAX  computers. "Our5 >    installation has four PDP-10s and twenty vaxen."I > G > Yes, VAXen influenced by vixen.  I like that one!  Now let's get Russ0* > Meyer to film that Superbowl ad for VMS!  L Putting on my linguist hat:  The suffix "-en" found in _vixen_ is a feminineK formant, cognate with German "-in" as seen in _Freundin_ "female friend" or:O _Studentin_ "female student", and is unrelated to the homophonous plural suffix0% seen in _oxen_ (cf. German _Ochsen_).=  N However, the near-rhyme undoubtedly influenced the non-linguist hackers at MIT9 and Stanford (the dual original home of the Jargon file).e  G > DEC meaning Digital does eventually show up near the end of the list.r= > The entry contains a link to Digital Equipment Corporation:> > " >    Digital Equipment Corporation >   A >    <company> (DEC) A computer manufacturer and software vendor.? >   H >    Before the killer micro revolution of the late 1980s, hackerdom wasM >    closely symbiotic with DEC's pioneering time-sharing machines. The firstsK >    of the group of hacker cultures nucleated around the PDP-1 (see TMRC).aM >    Subsequently, the PDP-6, PDP-10, PDP-20, PDP-11 and VAX were all foci of ,                                       ^^^^^^H >    large and important hackerdoms, and DEC machines long dominated the- >    ARPANET and Internet machine population.    I There ain't no sech animule.  The difference between a Decsystem-10 and a + DECSYSTEM-20 is software, not architecture.    --  N Rich Alderson                                          alderson+news@panix.comL   "You get what anybody gets.  You get a lifetime."  --Death, of the Endless   ------------------------------   Date: 9 Oct 02 14:16:31 +0200*) From: p_sture@elias.decus.ch (Paul Sture)iY Subject: Re: singular and plural of VAX (was: Re: Another security issue with the UCX POP ) Message-ID: <AjWA76tTOpKc@elias.decus.ch>   ` In article <djv5qu0q2ukrndra2e658h5fnki3d0lodk@4ax.com>, Alan Greig <a.greig@virgin.net> writes:: > On Tue, 08 Oct 2002 17:22:53 +0100 (MET), Phillip Helbig- > <HELBPHI@sysdev.deutsche-boerse.com> wrote:< >  >>> > vac vms 7.2, TCPIP 5.3 >>> ! >>> Ah, the singular form of VAX.i >>>  >>> ;-)i >>H >>By the way, does anyone know where the German-sounding plural "VAXen"  >>came from as used in English?e >  > ox   Pronunciation Key  (ks) > n. pl. oxen (ksn) nF > An adult castrated bull of the genus Bos, especially B. taurus, used > chiefly as a draft animal. y > A bovine mammal. r >   D "VAXen" seems to have been around for a long time. I believe I first? saw it in one of the many DEC related publications around 1983.i  A Google takes me back as far as 1981, but without any explanation:-  D > http://groups.google.ch/groups?selm=anews.Aihuxp.107&output=gplain   --  
 Paul Sture Switzerlandn   ------------------------------  % Date: Wed, 09 Oct 2002 08:06:44 -0400z2 From: Atlant Schmidt <atlantnospam@mindspring.com>Y Subject: Re: singular and plural of VAX (was: Re: Another security issue with the UCX POPS. Message-ID: <3DA41BD4.D441B292@mindspring.com>   "Craig A. Berry" wrote:   ) > One occasionally still hears "brethren"iE > instead of "brothers," generally in a context where the archaism ist. > meant to suggest reverence and/or affection.  7 So you're saying we're "The Brethren of the VAXen?" :-)i   Atlant   ------------------------------   Date: 9 Oct 2002 00:15:53 -0700E% From: Bart.Zorn@xs4all.nl (Bart Zorn)r) Subject: Re: SYS$SYSROOT and SYS$SPECIFICB= Message-ID: <a98cd882.0210082315.18ace15c@posting.google.com>f  | Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> wrote in message news:<01KNFL50NKFY9QWKOO@sysdev.deutsche-boerse.com>...I > > > Also, maybe the form above did *not* work in older versions of VMS,eJ > > > which would explain why SYS$SYSROOT is defined using the translationK > > > of SYS$SPECIFIC instead of SYS$SPECIFIC itself (yes, plus SYS$COMMON,/ > > > of course).r > > E > > Actually, this is right. This would not have worked under VMS 4.0eI > > according to my V4.4 copy of "Guide to File Applications" in which itw* > > says under "New and Changed Features": > > C > > [begin quote] A logical name that is defined with the concealed H > > attribute no longer must translate directly to a physical device butC > > can translate to another logical name as shown in the followingm > > example: > > D > > $ DEFINE MY_ROOT DISK$WORK: [ROOT.]/TRANSLATION=CONCEALED ! [end
 > > quote] > > I > > Therefore, since clusters came out with V4.0, that version of VMS hadeJ > > to use a translation that contains an actual physical device name. And= > > perhaps DEC just never saw any need to modify that since.t > G > On a related topic, one STILL cannot define a concealed device which AI > points to another CONCEALED DEVICE (though, as mentioned above, it can 0G > point to another LOGICAL NAME).  Actually, one "can" DEFINE it (i.e. R> > SHOW LOGICAL will show your definition), but access leads to > ? >    DIRECT-E-OPENIN, error opening test_root:[000000] as input P >    -RMS-F-DEV, error in device name or inappropriate device type for operation > K > In other words, the example above will NOT work if DISK$WORK is actually lJ > a concealed device (e.g. "DISK$WORK" = "DISK$USER:[WORK.]" [concealed]).    B If DISK$USER is defined as a concealed logical which translates to9 only a device name (such as SYS$SYSDEVICE), it DOES work!s  	 Bart Zornt   ------------------------------  % Date: Wed, 09 Oct 2002 03:37:18 -0400Y- From: JF Mezei <jfmezei.spamnot@videotron.ca> ) Subject: Re: SYS$SYSROOT and SYS$SPECIFICr, Message-ID: <3DA3DCA3.5236D3BE@videotron.ca>   Bart Zorn wrote:D > If DISK$USER is defined as a concealed logical which translates to; > only a device name (such as SYS$SYSDEVICE), it DOES work!U  K Doesn't it depend on whether the "terminal" attribute has been set or not ?p   ------------------------------  + Date: Wed, 09 Oct 2002 13:31:52 +0100 (MET) 9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>0) Subject: Re: SYS$SYSROOT and SYS$SPECIFICd; Message-ID: <01KNGO3Y3UOO9QWKOO@sysdev.deutsche-boerse.com>   A > >    DIRECT-E-OPENIN, error opening test_root:[000000] as inputpR > >    -RMS-F-DEV, error in device name or inappropriate device type for operation > > M > > In other words, the example above will NOT work if DISK$WORK is actually  L > > a concealed device (e.g. "DISK$WORK" = "DISK$USER:[WORK.]" [concealed]). > D > If DISK$USER is defined as a concealed logical which translates to; > only a device name (such as SYS$SYSDEVICE), it DOES work!   I Apparently not for me.  Can you provide a working example (best would be iD a .COM file with parameters for the corresponding entries so we can H all just cut and paste and try it out), complete with SHOW LOGICAL/FULL G definitions and, apart from the DEFINE commands (which always "work"), := actual access to the concealed device as well (DIR, SET DEF)?u   ------------------------------   Date: 9 Oct 2002 08:21:34 -0700D. From: spamsink2001@yahoo.com (Alan E. Feldman)) Subject: Re: SYS$SYSROOT and SYS$SPECIFICu= Message-ID: <b096a4ee.0210090721.67e76cc6@posting.google.com>e  j Bart.Zorn@xs4all.nl (Bart Zorn) wrote in message news:<a98cd882.0210082315.18ace15c@posting.google.com>...~ > Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com> wrote in message news:<01KNFL50NKFY9QWKOO@sysdev.deutsche-boerse.com>... [...]tI > > On a related topic, one STILL cannot define a concealed device which lK > > points to another CONCEALED DEVICE (though, as mentioned above, it can eI > > point to another LOGICAL NAME).  Actually, one "can" DEFINE it (i.e. e@ > > SHOW LOGICAL will show your definition), but access leads to > > A > >    DIRECT-E-OPENIN, error opening test_root:[000000] as inputMR > >    -RMS-F-DEV, error in device name or inappropriate device type for operation > > M > > In other words, the example above will NOT work if DISK$WORK is actually ML > > a concealed device (e.g. "DISK$WORK" = "DISK$USER:[WORK.]" [concealed]). >  > D > If DISK$USER is defined as a concealed logical which translates to; > only a device name (such as SYS$SYSDEVICE), it DOES work!a >  > Bart Zorn    Right. You can doe  3 $ DEFINE HIDDEN_DISK DISK$DATA1:[HIDDEN.]/TRAN=CONC." $ DEFINE AB HIDDEN_DISK:/TRAN=CONC $ DEFINE BC AB:/TRAN=CONCa	 $! [etc.]U, $! *** Don't forget the trailing colons! ***   But you can't do  3 $ DEFINE HIDDEN_DISK DISK$DATA1:[HIDDEN.]/TRAN=CONCu0 $ DEFINE WRONG_DISK HIDDEN_DISK:[UTI.]/TRAN=CONC  D That is, you can't add a second-layer [root.]! HIDDEN_DISK will work but WRONG_DISK won't.    Disclaimer: JMHO Alan E. Feldmanu   ------------------------------   Date: 9 Oct 2002 08:24:15 -07004. From: spamsink2001@yahoo.com (Alan E. Feldman)) Subject: Re: SYS$SYSROOT and SYS$SPECIFICR= Message-ID: <b096a4ee.0210090724.4d4772e0@posting.google.com>d  a JF Mezei <jfmezei.spamnot@videotron.ca> wrote in message news:<3DA3DCA3.5236D3BE@videotron.ca>.... > Bart Zorn wrote:F > > If DISK$USER is defined as a concealed logical which translates to= > > only a device name (such as SYS$SYSDEVICE), it DOES work!h > M > Doesn't it depend on whether the "terminal" attribute has been set or not ?   F You can nest concealed logical names as long as you don't add a secondB [root.]. The terminal attribute will always stop further iterativeF translation. IOW, whenever a logical name is translated and the result@ is a terminal equivalence name, then iterative translation stops4 immediately at that point. That is a separate issue.   Disclaimer: JMHO Alan E. FeldmanI   ------------------------------  + Date: Wed, 09 Oct 2002 17:29:07 +0100 (MET)(9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>E) Subject: Re: SYS$SYSROOT and SYS$SPECIFICg; Message-ID: <01KNGWETCA1S9QWKOO@sysdev.deutsche-boerse.com>-  F > > If DISK$USER is defined as a concealed logical which translates to= > > only a device name (such as SYS$SYSDEVICE), it DOES work!t >  > Right. You can doe > 5 > $ DEFINE HIDDEN_DISK DISK$DATA1:[HIDDEN.]/TRAN=CONCO$ > $ DEFINE AB HIDDEN_DISK:/TRAN=CONC > $ DEFINE BC AB:/TRAN=CONCn > $! [etc.] . > $! *** Don't forget the trailing colons! *** >  > But you can't do > 5 > $ DEFINE HIDDEN_DISK DISK$DATA1:[HIDDEN.]/TRAN=CONCn2 > $ DEFINE WRONG_DISK HIDDEN_DISK:[UTI.]/TRAN=CONC   Which is what I meant.  F > That is, you can't add a second-layer [root.]! HIDDEN_DISK will work > but WRONG_DISK won't.s   Exactly.   ------------------------------  + Date: Wed, 09 Oct 2002 17:35:43 +0100 (MET)o9 From: Phillip Helbig <HELBPHI@sysdev.deutsche-boerse.com>c) Subject: Re: SYS$SYSROOT and SYS$SPECIFICm; Message-ID: <01KNGWGPF8VQ9QWKOO@sysdev.deutsche-boerse.com>s  H > You can nest concealed logical names as long as you don't add a second > [root.]. .   Right.  ; > The terminal attribute will always stop further iterativenH > translation. IOW, whenever a logical name is translated and the resultB > is a terminal equivalence name, then iterative translation stops6 > immediately at that point. That is a separate issue.  , This is what TERMINAL means in this context.  D I always thought that it went back to the old days so that when the I final (terminal) definition was reached, resources wouldn't be wasted on bE determining whether the final definition was itself a logical name.  rH Other folks suggested that another reason for the existence of TERMINAL F is to prevent Joe User from defining, say, DSA0 to be some other disk.   ------------------------------  $ Date: Wed, 9 Oct 2002 08:43:05 +0200$ From: "Peter Flunger" <p-i-b@gmx.at> Subject: Re: tcp/ip == UCX0 Message-ID: <ao0j5q$ec8$1@newsreader1.netway.at>  ( "David M Smith" <dsmit115@csc.com> wroteG > Oops! You are, of course, correct Peter. My test was done on my otheroH > system, which is TCP/IP SERVICES V5.1 ECO 3. On my real V5.3 system, I* > got the same behavior you reported, i.e. >h+ > FTP> put CPQ-AXPVMS-OMSVA-V0102--1.PCSI;1i > 200 PORT command successful.! > 150 Opening data connection fort9 > XXX:[XXXXX]CPQ-AXPVMS-OMSVA-V0102--1.PCSI;1 (a.b.c.d,x)gI > XXXXX::XXXXXX 11:22:59 TCPIP$FTP CPU=00:00:01.66 PF=1264 IO=665 MEM=346oG > PUT (IMAGE)           1505280 bytes 00:00:02.02 elapsed (727.36 KB/S):G >       Local: SYS$SYSDEVICE:[DSMIT115]CPQ-AXPVMS-OMSVA-V0102--1.PCSI;1j/ >      Remote: CPQ-AXPVMS-OMSVA-V0102--1.PCSI;15 >. > Sorry for my confusion!i   No problem,r7 and isn't this something we have all been waiting for ?e7 A lot better than doing a TCPIP SHOW DEVICE/PORT=20 andy+ TCPIP SHOW DEVICE BGxxx /FULL or /CONTINOUSD5 to manually compute the throughput of an FTP session.   ) Thank you thHomPaq, for that improvement.l  3 Peter "who is doing a lot of FTP transfers" Flunger    ------------------------------   Date: 9 Oct 2002 07:36:00 -0600e- From: koehler@encompasserve.org (Bob Koehler)s8 Subject: Re: Which UIC to use for a product's username ?3 Message-ID: <lhxVeu4dM0M3@eisner.encompasserve.org>l  j In article <DoNo9.2904$F53.2765002@newssvr28.news.prodigy.com>, "konabear" <maurert@ameritech.net> writes: > N > One additional note.  I would personally using a group number of octal 11 orK > higher.  By default a process with a group number %o10 or less implicitly N > gets SYSPRV priv.  That can cause some confusion as the USERNAME may not mayI > SYSPRV privuledge but acts as if it does.  This can also be a sercurity.4 > hole.  I prefer when SYSPRV is explicitly granted.  C    You can, and I often have, change this via SYSGEN's MAXSYSGROUP.b   ------------------------------   Date: 9 Oct 2002 07:40:51 -0600L- From: Kilgallen@SpamCop.net (Larry Kilgallen) 8 Subject: Re: Which UIC to use for a product's username ?3 Message-ID: <GZLhrGI9mN04@eisner.encompasserve.org>s  c In article <lhxVeu4dM0M3@eisner.encompasserve.org>, koehler@encompasserve.org (Bob Koehler) writes: l > In article <DoNo9.2904$F53.2765002@newssvr28.news.prodigy.com>, "konabear" <maurert@ameritech.net> writes: >> .O >> One additional note.  I would personally using a group number of octal 11 oraL >> higher.  By default a process with a group number %o10 or less implicitlyO >> gets SYSPRV priv.  That can cause some confusion as the USERNAME may not maypJ >> SYSPRV privuledge but acts as if it does.  This can also be a sercurity5 >> hole.  I prefer when SYSPRV is explicitly granted.  > E >    You can, and I often have, change this via SYSGEN's MAXSYSGROUP.   F Therefore, an installation that issues a warning should do so based onI MAXSYSGROUP if that is higher than 10 (IFF the installation is being done.! on the currently running system).t   ------------------------------   Date: 9 Oct 2002 07:28:57 -0600a- From: koehler@encompasserve.org (Bob Koehler)l+ Subject: Re: Writing a new VMS filesystem ? 3 Message-ID: <07wiVpa3+xTA@eisner.encompasserve.org>e  [ In article <3DA35A7E.7452123@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes:e > O > Now, if the engineers build tools to access the files inside that FAT32 file,eP > I wonder if this couldn't be applied to reading diskettes or even disk drives.  G    It actually depends on which FAT they have to implement.  Folks with I    MS knowledge tell me various FAT formats are actually quite different tF    from each other, so the FAT required for the IPF boot may not match6    the one on a floppy or a disk you're interested in.   ------------------------------   End of INFO-VAX 2002.558 ************************