1 INFO-VAX	Tue, 15 May 2001	Volume 2001 : Issue 267       Contents:  Re: Broken VMS BACKUP on VMS 7.2  Re: Broken VMS BACKUP on VMS 7.2  Re: Broken VMS BACKUP on VMS 7.2  Re: Broken VMS BACKUP on VMS 7.2+ Command to find graphics controller details / Re: Command to find graphics controller details  Finding another processes PC? ! Re: Finding another processes PC? ! Re: Finding another processes PC? ! Re: Finding another processes PC? ! Re: Finding another processes PC? ! Re: Finding another processes PC? ! Re: Finding another processes PC? ! Re: Finding another processes PC? ! Re: Finding another processes PC? ! Re: Finding another processes PC?  FS:  Multia done right How to connect to RDB? Re: How to connect to RDB? Important News Re: Is OpenVMS an UNIX ??? RE: KZPCC under OpenVMS ; Looking for content contributors - OpenVMS on-line training  Re: Mozilla 0.9  Re: Multimedia Doc  netiquette (was Re: SLEEP BUTTON Re: OpenVMS 7.3 and Appletalk? Re: Question re: Java and C++  Re: Setting Up An FTP Account  Re: Setting Up An FTP Account  Re: Setting Up An FTP Account  Re: Setting Up An FTP Account  Re: Setting Up An FTP Account  SLIP on TCPIP servives V5 (VAX) % Re: Some more bits looking for a home % Re: Some more bits looking for a home " Re: The future of VMS - * Humour */ Re: The Internet, Tru64 and other unix variants / Re: The Internet, Tru64 and other unix variants  VAX memory ? Re: VAX memory ? Re: VAX memory ? Re: VAX memory ? RE: VAX memory ? RE: VAX memory ?: RE: XDM logins and Advanced Server external authentication: RE: XDM logins and Advanced Server external authenticationG Re: [EVENT] Last call for =?iso-8859-1?Q?Z=FCrich?= dinner registration   F ----------------------------------------------------------------------  % Date: Mon, 14 May 2001 14:33:38 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)) Subject: Re: Broken VMS BACKUP on VMS 7.2 L Message-ID: <rdeininger-1405011433390001@user-2ivec5g.dialup.mindspring.com>  < In article <3B00172F.74A3B4BE@mmaz.com>, "Barry Treahy, Jr." <treahy@mmaz.com> wrote:  G > Did anyone have any comments or recommended fixes for this particular  > mis-behaviour of BACKUP? > 
 > Regards, >  > Barry   G I'm can't duplicate your symptom, and don't know what to suggest next.   Can you place a service call?    --   Robert Deininger rdeininger@mindspring.com    ------------------------------  % Date: Mon, 14 May 2001 20:26:26 -0000 - From: wspencer@ap.nospam.org (Warren Spencer) ) Subject: Re: Broken VMS BACKUP on VMS 7.2 / Message-ID: <tg0fritlf2veaa@news.supernews.com>   J treahy@mmaz.com (Barry Treahy, Jr.) wrote in <3B00172F.74A3B4BE@mmaz.com>:  F >Did anyone have any comments or recommended fixes for this particular >mis-behaviour of BACKUP?  > 	 >Regards,  >  >Barry$ >-----------------------------------G >I just did two on-line disk to disk backups, with logging off and this  >is what >I've seen:  > D >V4500$ backup/nolog/nofast/noalias/ignore=(noback,inter) $145$dia0: >$145$dia1:/image I >%BACKUP-W-AFNOTSAVED, alias file $145$DIA0:[SYS0]SYSCOMMON.DIR;1 was not  >savedI >%BACKUP-W-AFNOTSAVED, alias file $145$DIA0:[SYSEXE]SYSBOOT.EXE;1 was not  >savedB >V4500$ backup/nolog/nofast/alias/ignore=(noback,inter) $145$dia0: >$145$dia1:/image I >%BACKUP-W-AFNOTSAVED, alias file $145$DIA0:[SYS0]SYSCOMMON.DIR;1 was not  >saved >V4500$  > D >Now I can see why the first failed to boot because the alias to the >SYSBOOT wasH >broke.  What I do not understand is why the second backup still ignored >the> >SYSCOMMON aliases, thereby producing a bad VMS tree, but more >importantly it C >ALSO failed to recreate the SYSBOOT aliases even though it did not  >report any  >errors: > < >V4500$ dir/file $145$dia0:[000000...]sysboot.exe,$145$dia1: > , >Directory $145$DIA0:[SYS0.SYSCOMMON.SYSEXE] >  >SYSBOOT.EXE;1        (113,1,0)  >  >Total of 1 file.  >  >Directory $145$DIA0:[SYSEXE]  >  >SYSBOOT.EXE;1        (113,1,0)  >  >Total of 1 file.  > ( >Directory $145$DIA0:[VMS$COMMON.SYSEXE] >  >SYSBOOT.EXE;1        (113,1,0)  >  >Total of 1 file.  >  >Directory $145$DIA1:[SYSEXE]  >  >SYSBOOT.EXE;1        (113,1,0)  >  >Total of 1 file.  > ' >Grand total of 4 directories, 4 files.  > E >This does appear to be a major flaw in the on-line BACKUP keeping in 
 >mind that3 >stand-alone version does not exhibit this problem.  > $ >Any and all comments appreciated... >  >Barry >  >  >--  > @ >Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIO > B >E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028 >  >   J I could be a million years out-of-date here, but I thought you had to use J the /IMAGE qualifier to get alias-file handling to function correctly.  I I also thought (cob-webs) that standalone back performed /IMAGE backups by   default (or perhaps "only").   ws   --  1 << Marriage is Grand.  Divorce is Fifty Grand. >>    Warren Spencer Senior Software Engineer The Associated Press  ? ** My employer does not necessarily agree with my statements **    ------------------------------  # Date: Mon, 14 May 2001 22:49:43 GMT  From: Dirk Munk <munk@home.nl>) Subject: Re: Broken VMS BACKUP on VMS 7.2 ' Message-ID: <3B006104.B8E98066@home.nl>   < Could this be the infamous syscommon <> vms$common switch ?.  K There was an old version of backup that did a bad restore, and replaced the J vms$common directory by syscommon. Syscommon is just a directory entry for vms$common. B You can see if it the switch has occured by the following command:    show device sys$sysdevice /files  = You should see [vms$common] entries, not [syscommon] entries.   E There is a simple way to repair this, but I would have to look it up.    Regards,   Dirk     "Barry Treahy, Jr." wrote:  G > Did anyone have any comments or recommended fixes for this particular  > mis-behaviour of BACKUP? > 
 > Regards, >  > Barry % > ----------------------------------- H > I just did two on-line disk to disk backups, with logging off and this	 > is what  > I've seen: > E > V4500$ backup/nolog/nofast/noalias/ignore=(noback,inter) $145$dia0:  > $145$dia1:/imageJ > %BACKUP-W-AFNOTSAVED, alias file $145$DIA0:[SYS0]SYSCOMMON.DIR;1 was not > saved J > %BACKUP-W-AFNOTSAVED, alias file $145$DIA0:[SYSEXE]SYSBOOT.EXE;1 was not > saved C > V4500$ backup/nolog/nofast/alias/ignore=(noback,inter) $145$dia0:  > $145$dia1:/imageJ > %BACKUP-W-AFNOTSAVED, alias file $145$DIA0:[SYS0]SYSCOMMON.DIR;1 was not > saved  > V4500$ > E > Now I can see why the first failed to boot because the alias to the 
 > SYSBOOT was I > broke.  What I do not understand is why the second backup still ignored  > the ? > SYSCOMMON aliases, thereby producing a bad VMS tree, but more  > importantly itD > ALSO failed to recreate the SYSBOOT aliases even though it did not > report any	 > errors:  > = > V4500$ dir/file $145$dia0:[000000...]sysboot.exe,$145$dia1:  > - > Directory $145$DIA0:[SYS0.SYSCOMMON.SYSEXE]  >   > SYSBOOT.EXE;1        (113,1,0) >  > Total of 1 file. >  > Directory $145$DIA0:[SYSEXE] >   > SYSBOOT.EXE;1        (113,1,0) >  > Total of 1 file. > ) > Directory $145$DIA0:[VMS$COMMON.SYSEXE]  >   > SYSBOOT.EXE;1        (113,1,0) >  > Total of 1 file. >  > Directory $145$DIA1:[SYSEXE] >   > SYSBOOT.EXE;1        (113,1,0) >  > Total of 1 file. > ( > Grand total of 4 directories, 4 files. > F > This does appear to be a major flaw in the on-line BACKUP keeping in > mind that 4 > stand-alone version does not exhibit this problem. > % > Any and all comments appreciated...  >  > Barry  >  > -- > A > Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIO  > C > E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028    ------------------------------  % Date: Mon, 14 May 2001 16:02:28 -0700 + From: "Barry Treahy, Jr." <treahy@mmaz.com> ) Subject: Re: Broken VMS BACKUP on VMS 7.2 ( Message-ID: <3B006404.F298A4AD@mmaz.com>   Dirk Munk wrote:  > > Could this be the infamous syscommon <> vms$common switch ?.  G Nope, my earlier postings demonstrated that the common directories were  properly aliased.    Regards,   Barry    >  > M > There was an old version of backup that did a bad restore, and replaced the L > vms$common directory by syscommon. Syscommon is just a directory entry for
 > vms$common. D > You can see if it the switch has occured by the following command: > " > show device sys$sysdevice /files > ? > You should see [vms$common] entries, not [syscommon] entries.  > G > There is a simple way to repair this, but I would have to look it up.  > 
 > Regards, >  > Dirk >  > "Barry Treahy, Jr." wrote: > I > > Did anyone have any comments or recommended fixes for this particular  > > mis-behaviour of BACKUP? > >  > > Regards, > > 	 > > Barry ' > > ----------------------------------- J > > I just did two on-line disk to disk backups, with logging off and this > > is what  > > I've seen: > > G > > V4500$ backup/nolog/nofast/noalias/ignore=(noback,inter) $145$dia0:  > > $145$dia1:/imageL > > %BACKUP-W-AFNOTSAVED, alias file $145$DIA0:[SYS0]SYSCOMMON.DIR;1 was not	 > > saved L > > %BACKUP-W-AFNOTSAVED, alias file $145$DIA0:[SYSEXE]SYSBOOT.EXE;1 was not	 > > saved E > > V4500$ backup/nolog/nofast/alias/ignore=(noback,inter) $145$dia0:  > > $145$dia1:/imageL > > %BACKUP-W-AFNOTSAVED, alias file $145$DIA0:[SYS0]SYSCOMMON.DIR;1 was not	 > > saved 
 > > V4500$ > > G > > Now I can see why the first failed to boot because the alias to the  > > SYSBOOT was K > > broke.  What I do not understand is why the second backup still ignored  > > the A > > SYSCOMMON aliases, thereby producing a bad VMS tree, but more  > > importantly itF > > ALSO failed to recreate the SYSBOOT aliases even though it did not > > report any > > errors:  > > ? > > V4500$ dir/file $145$dia0:[000000...]sysboot.exe,$145$dia1:  > > / > > Directory $145$DIA0:[SYS0.SYSCOMMON.SYSEXE]  > > " > > SYSBOOT.EXE;1        (113,1,0) > >  > > Total of 1 file. > >   > > Directory $145$DIA0:[SYSEXE] > > " > > SYSBOOT.EXE;1        (113,1,0) > >  > > Total of 1 file. > > + > > Directory $145$DIA0:[VMS$COMMON.SYSEXE]  > > " > > SYSBOOT.EXE;1        (113,1,0) > >  > > Total of 1 file. > >   > > Directory $145$DIA1:[SYSEXE] > > " > > SYSBOOT.EXE;1        (113,1,0) > >  > > Total of 1 file. > > * > > Grand total of 4 directories, 4 files. > > H > > This does appear to be a major flaw in the on-line BACKUP keeping in
 > > mind that 6 > > stand-alone version does not exhibit this problem. > > ' > > Any and all comments appreciated...  > > 	 > > Barry  > >  > > -- > > C > > Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIO  > > E > > E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028    --  ? Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIO   A E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028    ------------------------------  + Date: Mon, 14 May 2001 23:22:10 +0000 (UTC) ) From: jon@eoin.demon.co.uk (Jon Laughton) 4 Subject: Command to find graphics controller details0 Message-ID: <9dppb2$9o3$2@INDY.eoin.demon.co.uk> Keywords: graphics controller   ? Is there any way to find the model no. of a graphics controller ? from the command line? (OpenVMS on an Alphastation.) Best I can B get is "Digital PV-PCI Graphics Controller" from ">>>show device".     TIA    Jon    ------------------------------  % Date: Mon, 14 May 2001 22:42:50 -0500 C From: "Craig A. Berry" <craig.berry@nospam.SignalTreeSolutions.com> 8 Subject: Re: Command to find graphics controller detailsI Message-ID: <craig.berry-0637E4.22425014052001@newsrump.sjc.telocity.net>   0 In article <9dppb2$9o3$2@INDY.eoin.demon.co.uk>,+  jon@eoin.demon.co.uk (Jon Laughton) wrote:   A > Is there any way to find the model no. of a graphics controller A > from the command line? (OpenVMS on an Alphastation.) Best I can D > get is "Digital PV-PCI Graphics Controller" from ">>>show device". >   
 You might try    $ ANALYZE/SYSTEM SDA> CLUE CONFIG   ------------------------------    Date: 14 May 2001 10:31:12 -0700+ From: Tim Shoppa <shoppa@trailing-edge.com> & Subject: Finding another processes PC?( Message-ID: <9dp4p004l0@drn.newsguy.com>  < Is there a supported interface to finding another process's @ current PC (program counter)?  SHOW PROCESS/CONTINUOUS obviously? does this, but my look through SYS$GETJPI (where I would expect @ this functionality to be) in the documentation doesn't show this? as an option there, so I'm either looking in the wrong place or * SHOW PROC/CONT uses a different interface.  A If there is no supported interface, I'd be very happy to be shown H an example (somewhere in the DECUS tapes or in the Freeware collections)$ of an unsupported way of doing this.  F This is all to track down the bug in a process that runs continuously,E normally getting a few percent of CPU usage as it services requests,  G but it goes into 90+% CPU usage for a few minutes once every few months J and then recovers. (My gut feeling is that it's a buffer overflow in the CN source code.)  I've already written a program that uses SYS$GETJPI to sound anG alert when this happens, but by the time someone gets to the console in N central control to do the SHOW PROC/CONT things have (somehow) recovered.  AnyF other hints on how we might debug such an intermittent looping will be very welcome :-)   Tim.   ------------------------------  % Date: Mon, 14 May 2001 14:46:19 -0400 2 From: rdeininger@mindspring.com (Robert Deininger)* Subject: Re: Finding another processes PC?L Message-ID: <rdeininger-1405011446190001@user-2ivec5g.dialup.mindspring.com>  3 In article <9dp4p004l0@drn.newsguy.com>, Tim Shoppa ! <shoppa@trailing-edge.com> wrote:   > > Is there a supported interface to finding another process's B > current PC (program counter)?  SHOW PROCESS/CONTINUOUS obviouslyA > does this, but my look through SYS$GETJPI (where I would expect B > this functionality to be) in the documentation doesn't show thisA > as an option there, so I'm either looking in the wrong place or , > SHOW PROC/CONT uses a different interface. > C > If there is no supported interface, I'd be very happy to be shown J > an example (somewhere in the DECUS tapes or in the Freeware collections)& > of an unsupported way of doing this. > H > This is all to track down the bug in a process that runs continuously,G > normally getting a few percent of CPU usage as it services requests,  I > but it goes into 90+% CPU usage for a few minutes once every few months L > and then recovers. (My gut feeling is that it's a buffer overflow in the CP > source code.)  I've already written a program that uses SYS$GETJPI to sound anI > alert when this happens, but by the time someone gets to the console in P > central control to do the SHOW PROC/CONT things have (somehow) recovered.  AnyH > other hints on how we might debug such an intermittent looping will be > very welcome :-)  @ Can you set it up to run with a detached debugger?  Sort of likeH interrupting a running program with control-y and then typing DEBUG.  IfC this is a non-interactive process, I don't know for sure how to get : control to attach the debugger.  This is what I might try:  & First, have the debugger manual handy.  B Second, run your program in an environment where the DBG$INPUT andH DBG$OUTPUT logical names point to a terminal.  The terminal can't have aJ process logged in. (I think those are the right logical names, but this is/ from memory and I don't have the manual handy.)   E Third, have your process signal SS$_DEBUG (again, from memory).  That > should bring your program under debugger control.  You can setJ breakpoints, watchpoints, etc, and then let the program run.  If you can't< change the source code, I don't know how to do this offhand.  G You may need to set another logical to put the debugger in multiprocess . mode or something.  Memory is getting vague...    J This seems like a pretty desparate strategy.  Can you beef up the run-time4 checks and diagnostics when you compile the program?   --   Robert Deininger rdeininger@mindspring.com    ------------------------------   Date: 14 MAY 2001 19:44:03 GMT+ From: Dave Greenwood <greenwoodde@ornl.gov> * Subject: Re: Finding another processes PC?2 Message-ID: <14MAY01.19440328@feda01.fed.ornl.gov>  , Tim Shoppa <shoppa@trailing-edge.com> wrote:> > Is there a supported interface to finding another process's B > current PC (program counter)?  SHOW PROCESS/CONTINUOUS obviouslyA > does this, but my look through SYS$GETJPI (where I would expectpB > this functionality to be) in the documentation doesn't show thisA > as an option there, so I'm either looking in the wrong place ori, > SHOW PROC/CONT uses a different interface.  A Assuming you're using a relatively recent version of VMS, look atn SYS$PROCESS_SCAN.    Dave --------------9 Dave Greenwood                Email: Greenwoodde@ORNL.GOVnH Oak Ridge National Lab        %STD-W-DISCLAIMER, I only speak for myself   ------------------------------  % Date: Mon, 14 May 2001 20:01:48 -0000 - From: wspencer@ap.nospam.org (Warren Spencer)i* Subject: Re: Finding another processes PC?/ Message-ID: <tg0edcgmfn4q03@news.supernews.com>   . shoppa@trailing-edge.com (Tim Shoppa) wrote in <9dp4p004l0@drn.newsguy.com>:   = >Is there a supported interface to finding another process's eA >current PC (program counter)?  SHOW PROCESS/CONTINUOUS obviouslyn@ >does this, but my look through SYS$GETJPI (where I would expectA >this functionality to be) in the documentation doesn't show this @ >as an option there, so I'm either looking in the wrong place or+ >SHOW PROC/CONT uses a different interface.  > B >If there is no supported interface, I'd be very happy to be shownI >an example (somewhere in the DECUS tapes or in the Freeware collections)d% >of an unsupported way of doing this.n >EG >This is all to track down the bug in a process that runs continuously,tF >normally getting a few percent of CPU usage as it services requests, H >but it goes into 90+% CPU usage for a few minutes once every few monthsI >and then recovers. (My gut feeling is that it's a buffer overflow in theeH >C source code.)  I've already written a program that uses SYS$GETJPI toF >sound an alert when this happens, but by the time someone gets to the@ >console in central control to do the SHOW PROC/CONT things haveD >(somehow) recovered.  Any other hints on how we might debug such an. >intermittent looping will be very welcome :-) >, >Tim.  >r  I More details would help.  What version of C? Any other languages involved K (like Java)? Version of OpenVMS?  What is the app doing?  Does it access a rL database?  Whose?  Version?  Any other significant products glued into this , app?  Is the box on which it runs clustered?  H Run-time profiling of the executing code might yield some clues, if you & catch the problem when it's happening.   ws   -- T1 << Marriage is Grand.  Divorce is Fifty Grand. >>    Warren Spencer Senior Software Engineer The Associated Press  ? ** My employer does not necessarily agree with my statements **-   ------------------------------  % Date: Mon, 14 May 2001 13:10:14 -0700u+ From: "Barry Treahy, Jr." <treahy@mmaz.com>v* Subject: Re: Finding another processes PC?( Message-ID: <3B003BA6.D944CDA1@mmaz.com>  5 > In article <9dp4p004l0@drn.newsguy.com>, Tim ShoppaD# > <shoppa@trailing-edge.com> wrote:s >d? > > Is there a supported interface to finding another process's;D > > current PC (program counter)?  SHOW PROCESS/CONTINUOUS obviouslyC > > does this, but my look through SYS$GETJPI (where I would expectlD > > this functionality to be) in the documentation doesn't show thisC > > as an option there, so I'm either looking in the wrong place orS. > > SHOW PROC/CONT uses a different interface. > >e >c  P The problem with doing a show process is that the PC that is displayed can oftenF change so quickly that you can miss it, but commonly used addresses orQ 'stall-points' can be cross-referenced back to the exact point in the code if you[S have the link map and the object listings.  If the PC addresses are extremely large N (ie 7fxxxxxx or 80xxxxxx), chances are you are in a sharable library or in theP executive with a system service vector and with the help the the link map, or anN ANA/IMAGE you can determine which library that is and can even determine whick5 routine in that library if you spend a little time...E   BarryS   --  ? Barry Treahy, Jr  *  Midwest Microwave  *  Vice President & CIOo  A E-mail: Treahy@mmaz.com * Phone: 480/314-1320 * FAX: 480/661-7028i   ------------------------------    Date: 14 May 2001 13:27:18 -0700+ From: Tim Shoppa <shoppa@trailing-edge.com>l* Subject: Re: Finding another processes PC?) Message-ID: <9dpf3601i8f@drn.newsguy.com>i  L In article <rdeininger-1405011446190001@user-2ivec5g.dialup.mindspring.com>,! rdeininger@mindspring.com says... A >Can you set it up to run with a detached debugger?  Sort of liketI >interrupting a running program with control-y and then typing DEBUG.  IflD >this is a non-interactive process, I don't know for sure how to get  >control to attach the debugger.  D It runs non-interactive, but we can start it up from the debugger ifI necessary.  Problem is, it's several months between application restarts,M/ we don't have a programmer at the console 24*7,tL and even if we did the users would complain about the fact that the computerJ isn't running their $10Billion peripheral while we're doing the debugging.I If the computer goes down for a couple minutes in rush hour, we're in theoJ newspaper the next day, and we don't like that one bit, so we really don'tN want the process stopping while it waits for someone to come use the debugger.  M My belief is that the problem is in a wild input packet from a remote stationhL (93 of them) or one of the several hundred console CRT's in central control,H possibly as a result of flaky hardware in the console or remote station,M that sends this process into a tight loop.  Clearly having better diagnosticskG in the process would be a good thing, especially if it throws out inputoJ packets filled with garbage rather than trying to work on them.  But rightF now I'm looking for a shortcut so that we can do some diagnosis before the next build.o  K >This seems like a pretty desparate strategy.  Can you beef up the run-time$5 >checks and diagnostics when you compile the program?$  M We will be adding run-time checks for the next build, but that's months away.SI If we can find the high-CPU-loop problem before the next scheduled build,0I that's far superior - thus my hope that I could find the PC or maybe evensF map into the CPU-hogging process's address space and look at a few key locations on the fly.$   Tim.   ------------------------------  % Date: Mon, 14 May 2001 17:02:08 -0400 2 From: norm lastovica <norman.lastovica@oracle.com>* Subject: Re: Finding another processes PC?* Message-ID: <3B0047D0.28122F09@oracle.com>  * You want to get a VMS listings CD and look+ at [LIUTL.LIS]INFO.LIS.  The following code  appears to do most of the work:B  E IF .resflg                        ! If resident then okay to send AST$ THEN   BEGINd=   CH$FILL(0,16,pcpsl_buf);        ! Initialize PC/PSL to zeroA9   EXE$READ_PROCESS(EXE$CVT_EPID_TO_IPID(.proc_l_pid), 16,,?      eacb$k_pc, pcpsl_buf, eacb$k_general_register, ast_count);    END;  , From there, follow the listings to [SYS.LIS]- PROC_READ_WRITE.LIS and read the comments forE) EXE$READ_PROCESS.  Presto - you're there.      Tim Shoppa wrote:  > = > Is there a supported interface to finding another process's,B > current PC (program counter)?  SHOW PROCESS/CONTINUOUS obviouslyA > does this, but my look through SYS$GETJPI (where I would expectCB > this functionality to be) in the documentation doesn't show thisA > as an option there, so I'm either looking in the wrong place ora, > SHOW PROC/CONT uses a different interface. > C > If there is no supported interface, I'd be very happy to be shown J > an example (somewhere in the DECUS tapes or in the Freeware collections)& > of an unsupported way of doing this. > H > This is all to track down the bug in a process that runs continuously,F > normally getting a few percent of CPU usage as it services requests,I > but it goes into 90+% CPU usage for a few minutes once every few monthseL > and then recovers. (My gut feeling is that it's a buffer overflow in the CP > source code.)  I've already written a program that uses SYS$GETJPI to sound anI > alert when this happens, but by the time someone gets to the console inoP > central control to do the SHOW PROC/CONT things have (somehow) recovered.  AnyH > other hints on how we might debug such an intermittent looping will be > very welcome :-) >  > Tim.   -- $> norman lastovica / oracle rdb engineering / usa / 610.696.4685   ------------------------------   Date: 14 MAY 2001 21:38:46 GMT+ From: Dave Greenwood <greenwoodde@ornl.gov>r* Subject: Re: Finding another processes PC?2 Message-ID: <14MAY01.21384694@feda34.fed.ornl.gov>  . , Dave Greenwood <greenwoodde@ornl.gov> wrote:. > Tim Shoppa <shoppa@trailing-edge.com> wrote:@ > > Is there a supported interface to finding another process's D > > current PC (program counter)?  SHOW PROCESS/CONTINUOUS obviouslyC > > does this, but my look through SYS$GETJPI (where I would expectaD > > this functionality to be) in the documentation doesn't show thisC > > as an option there, so I'm either looking in the wrong place or0. > > SHOW PROC/CONT uses a different interface. >  $C > Assuming you're using a relatively recent version of VMS, look at5 > SYS$PROCESS_SCAN.C  , Er - you said "PC" not "PID".  Oops..blush..   >  D > Dave > --------------; > Dave Greenwood                Email: Greenwoodde@ORNL.GOVoJ > Oak Ridge National Lab        %STD-W-DISCLAIMER, I only speak for myself   ------------------------------    Date: 14 May 2001 13:58:31 -0700+ From: Tim Shoppa <shoppa@trailing-edge.com>,* Subject: Re: Finding another processes PC?) Message-ID: <9dpgtn01qak@drn.newsguy.com>o  N In article <tg0edcgmfn4q03@news.supernews.com>, wspencer@ap.nospam.org says... >0/ >shoppa@trailing-edge.com (Tim Shoppa) wrote inr ><9dp4p004l0@drn.newsguy.com>: E > > >>Is there a supported interface to finding another process's B >>current PC (program counter)?  SHOW PROCESS/CONTINUOUS obviouslyA >>does this, but my look through SYS$GETJPI (where I would expect B >>this functionality to be) in the documentation doesn't show thisA >>as an option there, so I'm either looking in the wrong place orS, >>SHOW PROC/CONT uses a different interface. >>C >>If there is no supported interface, I'd be very happy to be showniJ >>an example (somewhere in the DECUS tapes or in the Freeware collections)& >>of an unsupported way of doing this. >>H >>This is all to track down the bug in a process that runs continuously,G >>normally getting a few percent of CPU usage as it services requests,  I >>but it goes into 90+% CPU usage for a few minutes once every few months.J >>and then recovers. (My gut feeling is that it's a buffer overflow in theI >>C source code.)  I've already written a program that uses SYS$GETJPI tohG >>sound an alert when this happens, but by the time someone gets to the A >>console in central control to do the SHOW PROC/CONT things havenE >>(somehow) recovered.  Any other hints on how we might debug such ant/ >>intermittent looping will be very welcome :-)o >> >>Tim. >> >pJ >More details would help.  What version of C? Any other languages involved! >(like Java)? Version of OpenVMS?p  D VMS 5.5-2, VAX C 3.2-044, lots of DECnet and mailbox stuff going on,J and a really decrepit old version of UCX (2.0B) that causes me to fume and: cuss and swear at the guy who installed it just last year.G (For those on comp.os.vms who know me personally, that means something,RB as I normally don't go nonlinear over such things... it's just theF concept of someone installing such a crappy version of UCX so recentlyG makes me start bouncing off the wall!  In any event, UCX apparently haswC nothing to do with the process that's going into a tight CPU loop.)r   > What is the app doing?  O Central control on the second-largest US mass transit system's subway.  Tens ofoK thousands of SCADA-type status and control points.  Several dozen processesDA running at a variety of priorities.  Lots of processing on status0F points coming back to do train tracking.  Hundreds of X-terminal CRT'sE in central control for the line control supervisors, superintendents,>J passenger ops, and maintenance personel, all with live real-time graphical9 displays.  Disk and tape farms to record historical data.s   >  Does it access a  >database?  Whose?  E ODB, if that means anything to you.  A real-time database accessed bys= thirty proceses as a giant (150 Mbyte) shared global section.g  ) >  Is the box on which it runs clustered?n  G No, but there is a lot of DECnet communications with remote nodes (bothw2 Ultrix workstations and Q-bus data concentrators.)  I >Run-time profiling of the executing code might yield some clues, if you  ' >catch the problem when it's happening.u  I I *can* catch it when it's happening and I've even wired an audible alarmrG to go off when the offending process starts chewing up CPU.  But by thefD time we run into central to try debugging it, it has somehow clearedF itself up.  That's why I want to have something automatically find theK PC during high CPU usage, and maybe even map into the CPU-hogging process'stD address space and look at a few key variables.  We've nearly got allH the bugs out of the system by now, but this is the last one, and it only comes up a couple times a year!y   Tim.   ------------------------------    Date: 14 May 2001 14:33:10 -0700+ From: Tim Shoppa <shoppa@trailing-edge.com>n* Subject: Re: Finding another processes PC?) Message-ID: <9dpium022hb@drn.newsguy.com>n  ? In article <14MAY01.19440328@feda01.fed.ornl.gov>, Dave says...t >e- >Tim Shoppa <shoppa@trailing-edge.com> wrote:o? >> Is there a supported interface to finding another process's eC >> current PC (program counter)?  SHOW PROCESS/CONTINUOUS obviously B >> does this, but my look through SYS$GETJPI (where I would expectC >> this functionality to be) in the documentation doesn't show thismB >> as an option there, so I'm either looking in the wrong place or- >> SHOW PROC/CONT uses a different interface.e > B >Assuming you're using a relatively recent version of VMS, look at >SYS$PROCESS_SCAN.  E It's 5.5-2.  And while it does have $PROCESS_SCAN, it looks like thislG is a way to *select* which processes are scanned by a wildcard $GETJPI,aJ not a way to get additional information beyond what's provided by $GETJPI.  C Is there some way to use $PROCESS_SCAN to get a process's PC?  I've9F been staring at the description in the System Service book and I don't see how it'll do the job.t   Tim    ------------------------------  % Date: Tue, 15 May 2001 22:38:21 -0400h From: william.webb@juno.comn Subject: FS:  Multia done rightb= Message-ID: <20010515.223822.-164545.0.william.webb@juno.com>o  C Doing this from home since the place from which I usually email theiH newsgroup from takes a dim view of their email facilities being used for/ personal transactions of a commercial nature...h  H I've gotten my PW433au up and running so my other Alpha box is for sale:  + 133 MHz Alpha-based Multia with 128 MB RAM;t" Installed external SCSI connector;   Upgraded cooling fan as per < http://www.netbsd.org/Ports/alpha/multiafaq.html#overheat --E Installed the 1 inch deep 18 CFM McMaster-Carr fan that they mention.t  H DECpc expansion cabinet with RZ28 (2GB) drive, extra cooling fan mounted inside a and RRD45 CD-ROM.   3 SCSI cable to connect expansion box and terminator.i  I I've done all the gruntwork and have VMS 7.2 running on it under hobbyisto  4 licenses (which, of course, are not transferrable).   H You supply the keyboard, mouse, monitor (and hobbyist licenses) and it's yours.C I'm even willing to install *your* licenses before I ship it out if - you'll email them when we've reached a deal. o   I'm asking $400 or best offer.   WWWebb@ ________________________________________________________________ GET INTERNET ACCESS FROM JUNO!5 Juno offers FREE or PREMIUM Internet access for less!e0 Join Juno today!  For your FREE software, visit:  http://dl.www.juno.com/get/tagj.   ------------------------------  % Date: Tue, 15 May 2001 01:09:30 +020024 From: "Robert Schmoelzer" <robert.schmoelzer@aon.at> Subject: How to connect to RDB?o0 Message-ID: <3b006603$0$34576$62ce1842@SSP1NO55>  
 Hi everybody,o  < I am trying to figure out the best, the worst, the usual, or6 the most modern ways to connect to an RDB database ;-)  8 Reading all kinds of .html and .pdf files on the stealth/ part of the Oracle (RDB) WebSite i came up withw the following:  # - i can use the RDB - ODBC drivers.e/     This seems to be the "Oracle preferred" wayt  % - on RDB 7.1 i am able to use SQL*Netp5     is it really compatible to SQL*Net on 8i, 9i,...?   7 - there are some signs which indicate that it should be <     possible to use JDBC via the drivers for Classic Oracle.1     Can anyone confirm this? Does anybody use it?   = - i can use some third party software like Attunity Connector      (aka ISG Navigator)-    G We want to use JDBC if possible, since most of our software developmentoI is based on JAVA now, and it would give us good flexibility on the clients side (can you say LINUX on Alpha ;-}c  - any suggestions, comments, hints, flames,....t Robert Schmoelzerr Telemed Communications Servicesu   ------------------------------  % Date: Mon, 14 May 2001 19:45:13 -0400t- From: Michael Austin <miaustin@bellsouth.net> # Subject: Re: How to connect to RDB?(- Message-ID: <3B006E09.BCFAA45A@bellsouth.net>)   Robert,u  L It depends on what you want "connected" to Rdb... an application, a website,M and ODBC-enable application such as MSQuery/Excel, interactively... there arenH alot of different methods dependent upon the technology you want to use.  " What version of Rdb, OpenVMS etc..   see comments below as well.o   Robert Schmoelzer wrote:   > Hi everybody,r > > > I am trying to figure out the best, the worst, the usual, or8 > the most modern ways to connect to an RDB database ;-) >b: > Reading all kinds of .html and .pdf files on the stealth1 > part of the Oracle (RDB) WebSite i came up witht > the following: >r% > - i can use the RDB - ODBC drivers.r1 >     This seems to be the "Oracle preferred" wayu >a  H Yes this works.. you need the Oracle ODBC Driver for Rdb (2.10.17 or theJ latest 3.0.2.1 for NT, 2K,98,95).  I would say "preferred" is a little tooD strong of a phrase and it entirely depends on what you are trying to1 accomplish and what platforms you want to target.t    ' > - on RDB 7.1 i am able to use SQL*Netr7 >     is it really compatible to SQL*Net on 8i, 9i,...?h >i  $ Yes.  Works just like 7.3, 8, 8i, 9i( Again depends on what you want connected   >t9 > - there are some signs which indicate that it should ben> >     possible to use JDBC via the drivers for Classic Oracle.3 >     Can anyone confirm this? Does anybody use it?  >p  J There is a JDBC thick/fat client that currently works and looks like there# will be thin client available soon.r   > ? > - i can use some third party software like Attunity Connectorh >     (aka ISG Navigator)- >-  % Yes, many companies use this "driver"    >1I > We want to use JDBC if possible, since most of our software developmentFK > is based on JAVA now, and it would give us good flexibility on the clientn > side! > (can you say LINUX on Alpha ;-}S >p  K Yes I can say Linux on Alpha, however Rdb is OpenVMS only. can you also sayeM very little "real" support... after all it is open source freeware... And you/  want to bet your business on it?   >t. > any suggestions, comments, hints, flames,...   > Robert Schmoelzeri! > Telemed Communications Servicesn  M If you are in need of an Oracle Rdb DBA consultant, you can always contact meS9 privately as I will be available at the end of next week.    Michael Austin Rdb DBA Consultant   ------------------------------   Date: 14 May 2001 22:07:32 GMT From: <sasx888er_666@yahoo.com>e Subject: Important Newsi+ Message-ID: <9dpkv4$2kcm$6961@news.ukr.net>t   Herpes Sufferers,   new guaranteed treatment! http://reflect.to/life   ------------------------------  # Date: Mon, 14 May 2001 23:43:20 GMTs/ From: StevenU@POBoxes.com (Steven P. Underwood)c# Subject: Re: Is OpenVMS an UNIX ???d2 Message-ID: <3b006d60.939772493@news.telocity.com>  A It is currently in the Other section, but still only the 1 entry:o  @ http://serverwatch.internet.com/serversort/ossort-other.html#vms   Steve   D On 14 May 2001 17:23:17 GMT, "Bill Pedersen" <pedersen@ccsscorp.com> wrote:  + >That's because they have changed the site.i > 2 >It USED to be under the "Commercial Unix" list... >2 >Now it is no place at all...h >aM >I had sent them a note to suggest they add it separately and get information91 >about the various web servers that run on VMS...g >1 >--1 >Bill Pedersen >CCSS Corporationp >www.CCSScorp.comg
 >831-336-2708o >a? >"Bill Gunshannon" <bill@triangle.cs.uofs.edu> wrote in message & >news:9dp2lv$93l$3@info.cs.uofs.edu...
 >> In articleyB ><OF7BBFB7CE.D58EFF16-ON03256A49.004A233D@ep-bc.petrobras.com.br>,/ >>  fabio_compaq@ep-bc.petrobras.com.br writes:h >> |> Just click >> |> A >> |> http://serverwatch.internet.com/serversort/ossort-main.htmls >>F >> Did I miss something??  I didn't see VMS, Open or otherwise, listed1 >> at all.  Not that I would expect it to be.....n< >>            After all, it is the stealth operating system. >> >> bill  >> >> --iM >> Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesdG >> bill@cs.scranton.edu     |  and a sheep voting on what's for dinner.d >> University of Scranton   |bA >> Scranton, Pennsylvania   |         #include <std.disclaimer.h>r >m >e   Steven P. Underwood,DNRC Whitinsville,MAa StevenU@POBoxes.com    ------------------------------  % Date: Mon, 14 May 2001 21:31:51 +0100l5 From: "Steeples, Oliver" <Oliver.Steeples@compaq.com>   Subject: RE: KZPCC under OpenVMSN Message-ID: <F498D199EDB12D468CD2C66680D308018B144B@reoexc04.emea.cpqcorp.net>   Hi,b@ 	as you've found the KZPCC isn't supported on vms at all and i'mK unsure of any future support.  As for it working, I doubt it as vms doesn't  have the drivers.r  I You maybe better off with a RA3000 as this is supported and has expansion 
 capabilities.o   Regards, 	Olivera   -----Original Message-----> From: Jeffrey_Hiris@brown.edu [mailto:Jeffrey_Hiris@brown.edu]" Sent: Monday, May 14, 2001 1:36 PM To: Info-VAX@Mvb.Saic.Comh Subject: KZPCC under OpenVMS     Hi all:w  I the KZPCC adapter (PCI-bus Ultra-2 LVD RAID controller) is not listed as sE supported under OpenVMS, at least not yet, which leaves a "hole" for aL high-performance disks in the low-end VMS line.  (The KZPAC Ultra SCSI-2 has  A always left something to be desired in terms of raw performance.),  J I was wondering if any brave soul has tried a KZPCC out under OpenVMS, and if u5 so, whether it worked, at least as a non-boot device.3   thanks   ------------------------------   Date: 14 May 2001 19:28:37 GMT- From: "Bill Pedersen" <pedersen@ccsscorp.com>eD Subject: Looking for content contributors - OpenVMS on-line training0 Message-ID: <3b003217$1@kerberos.linuxpuppy.net>  J Last Fall we started developing an on-line remote learning site.  The siteI is at http://plato.CCSScorp.com/.  The site is primarily OpenVMS Centric.cJ As such, both the courses we have completed and those we are developing atK present are targeted in the OpenVMS arena.  We have started some additionalEJ discussions with additional vendors (in the OpenVMS application space) whoB are also interested in having easily available courses on the web.  K I come to the news group today because I am looking to increase the rate atSI which we develop and deploy courses.  I am looking for content providers.n9 In exchange for the content you provide you will receive:e  <  a percentage of the revenue for the courses you contributed/  a by-line on the course description and courset   a bio-page or link to your site  L To contribute a course, we would need content from you.  It would be best ifK this content were at least laid out in lesson form, although we can massage K it into lesson form if necessary.  We also need questions and answers for a ' test to be associated with each course.   G We will maintain the course, making corrections and additions as either I discovered by us or by you, the contributor.  We will administer the siteoB which includes working to make all courses as similar in style andD presentation as possible, developing examples and providing studentsE with support. We will also maintain editorial control of the site and( the courses.  D The news group has many talented people, many of you might even haveI content available without very much effort.  This gives an opportunity toaL help OpenVMS get some more exposure on the web as well as provide a resourceF for those looking to increase their knowledge in a structured fashion.  4 Please feel free to visit our site and check it out.  I We are also implementing a referral program.  This will reward people who@I refer students to the site who then sign up for a course. This will be ine place later this week.  < If you have any questions or comments please get back to me.  
 Thank you,  
 Bill Pederseno CCSS Corporation www.CCSScorp.com 831-336-2708   --
 Bill Pedersenc CCSS Corporation www.CCSScorp.com 831-336-2708   ------------------------------    Date: 14 May 2001 17:42:09 -0500- From: koehler@encompasserve.org (Bob Koehler)  Subject: Re: Mozilla 0.93 Message-ID: <D7cfrtgq9Iby@eisner.encompasserve.org>o  Y In article <9doaca$34do@newton.cc.rl.ac.uk>, "Richard Brodie" <R.Brodie@rl.ac.uk> writes:  > C > When the integration work was done, some of the fixes were rollednI > up into a maintenance upgrade to the Java libraries. It's this version   > (1.3.0-01) that is required. >   C I assume that's the same as the 1.3.0-1 listed as "New" for OpenVMSh Alpha on Compaq's web page?j  H Which is news since the last time I looked 1.3.0 was listed as pre-beta.  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporationi= NASA GSFC Flight Software       | Federal Sector, Civil GroupdE                                 | please remove ".aspm" when replying)   ------------------------------  % Date: Mon, 14 May 2001 17:44:09 -0400G* From: "Paul A. Jacobi" <nospan@nospam.com> Subject: Re: Multimedia Doca1 Message-ID: <wrYL6.263$fi2.6540@news.cpqcorp.net>m  ) <paul.beaudoin@hsbc.com> wrote in messageW? news:OFCAA4855C.03DBE5C1-ON80256A4C.0054AB0D@systems.uk.hsbc...oF >Main interest - can I play MP3 files? Any pointers to useful doc most
 gratefully > received.t  L See the MPEG3 player available on the OpenVMS Freeware CD.  It seems to work for most MP3 files.   D If you add a PCI sound card to a Wildfire, you have the world's mostK expensive MP3 player!  Yes, I tied it and it works.  It was just a bit hard  to hear over the fan noise.f    @ http://www.openvms.compaq.com/freeware/FREEWARE40/MPEG3PLAY-096/     Paul A. Jacobi Compaq Computer Corporation ! OpenVMS Systems Group, ZKO3-4/U14J 110 Spitbrook Road Nashua, NH 03062-2698  Email: Paul.Jacobi@compaq.com    ------------------------------   Date: 14 May 2001 14:16:12 CDT= From: wayne@tachysoft.xxx.310887.killspam.0162 (Wayne Sewell)e) Subject: netiquette (was Re: SLEEP BUTTONe. Message-ID: <206wRuZkhfAm@tachxxsoftxxconsult>  H In article <3AFF8E3A.AE3CF2E1@home.nl>, Dirk Munk <munk@home.nl> writes: > Wayne Sewell wrote:S > S >> In article <3.0.6.32.20010513083722.00794780@pop.indy.net>, mwe@indy.net writes:dM >> > I own a Compaq Presario 7360. When I press the sleep button, the monitor   O >> Throw the entire system into a dumpster.  Once it departs on its trip to thesF >> landfill to fulfill its true destiny, you will have absolute quiet. >>R >> If and when you get a real computer, as opposed to a nintendo with delusions of$ >> grandeur, come back and visit us. > V > But I seriously doubt if he ever will after receiving a reply like this. So he is inS > the wrong newsgroup, and the most of us have a dislike for M$ Windooz stuff. That-R > doesn't mean we have to be rude to someone who asks very politely if we can helpN > him. I sometimes get the feeling that certain contributers to this newsgroupT > mistakenly think they are in a kind of virtual Jerry Springer show. Let us take an8 > example from VMS itself, and stay elegant and stylish.    J Okay, so my post was obnoxious, but it is equally obnoxious to jump into aN newsgroup with both feet without the slightest clue of the group's topic.  AllM the netiquette guides suggest *reading* a group a little bit before posting. hK How did he get here in the first place?   What is there about "comp.os.vms"5N which suggests any sort of tie with billyboxes?   If he was picking newsgroups% at random, why not ask the readers ofuM rec.music.country.dixie-chicks.i-like-emily-the-best to answer his question? a   I got annoyed.  Sorry.   -- tO ===============================================================================hM Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxxl: http://www.tachysoft.xxx/www/tachyon.html and wayne.html  K change .xxx to .com in addresses above, assuming you are not a spambot  :-)sO ===============================================================================dK Hotel guy (after bed demolition):  That bed goes back to Henry the eighth!!sO    Curly: That's nothin'!  We had a bed go back to Sears and Roebuck the fifth!u   ------------------------------   Date: 14 May 2001 19:49:59 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)' Subject: Re: OpenVMS 7.3 and Appletalk?0, Message-ID: <9dpct7$4md@gap.cco.caltech.edu>   In article <paul.r.anderson-1405011010450001@dhcp-16-21-34-107.eng.lkg.dec.com>, paul.r.anderson@compaq.com (Paul Anderson) writes:8 >dH >In our qualification of DCPS V1.8 and V2.0 with OpenVMS V7.3, we testedF >with AppleTalk and discovered no problems.  The PATHWORKS for OpenVMSI >(Macintosh) product installed and ran on both OpenVMS Alpha and VAX V7.3 D >systems.  The product is unsupported but does seem to work on V7.3.  I It's not at all clear how much longer it will continue to work though, oroJ even if it actually works on all machines (the GS series, for instance, isJ a big question mark in my mind.)  And "work" is a relative term.  AlthoughD I've had no problem with the appletalk transport lately, the exampleI programs didn't work for me at 7.2-1 on a DS10 using the current Compaq Ch* compiler (whatever that was at the time.)   F We had a few go arounds with trying to get this code released into theF public domain (onto the Freeware, for instance) but that seems to haveE ground to a halt inside Compaq, and I'm out of this game so won't be d putting any more time into it.  G Still, it would probably reassure a few people (and less FUD is a good 2E thinkg) to know that the DCPS group (or at least Paul, inside Compaq)bG actually had access to all of the relevant code and was able to build a I working distribution now.  I have a sinking feeling that not only is thisrR transport "unsupported", but also that either the code itself may be lost or is noH longer buildable.  Both would spell big trouble if (make that, when) an # incompatibility does finally arise.H   Regards,   David Mathog mathog@caltech.edu? Manager, sequence analysis facility, biology division, Caltech /   ------------------------------    Date: 14 May 2001 16:56:54 -0500- From: koehler@encompasserve.org (Bob Koehler)m& Subject: Re: Question re: Java and C++3 Message-ID: <QVrzyy5GrrhU@eisner.encompasserve.org>a  j In article <20010514110155.02866.00003636@ng-fo1.aol.com>, ejheller@aol.com.nojunk (Edward Heller) writes: > O > So, here is the posed question: Is anyone aware of any limitations in OpenVMSr8 > relating to the interaction between Java, JNI and C++? >   ; Did you use the name mangling tools that come with the JDK?l  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporation_= NASA GSFC Flight Software       | Federal Sector, Civil GroupEE                                 | please remove ".aspm" when replyings   ------------------------------  # Date: Mon, 14 May 2001 18:49:09 GMT.& From: adroso@home.com (Still Learning)& Subject: Re: Setting Up An FTP Account9 Message-ID: <3b001f10.37855468@news.jamison1.pa.home.com>e   Richard,  E When I copied SYSTEM to NEWUSER in AUTHORIZE, I used /PASS=NEWPASS in2F the COPY command (along with /PWDMINIMUM=6 and /NOPWDLIFETIME) and I'mF able to login interactively with the NEWUSER account.  If it makes anyE difference, I'm using WS_FTP Pro on a 98SE PC to log into my VAX on ao LAN.   ------------------------------   Date: 14 MAY 2001 19:35:12 GMT+ From: Dave Greenwood <greenwoodde@ornl.gov>e& Subject: Re: Setting Up An FTP Account2 Message-ID: <14MAY01.19351297@feda01.fed.ornl.gov>  ' adroso@home.com (Still Learning) wrote:r
 > Richard, >  wG > When I copied SYSTEM to NEWUSER in AUTHORIZE, I used /PASS=NEWPASS insH > the COPY command (along with /PWDMINIMUM=6 and /NOPWDLIFETIME) and I'mH > able to login interactively with the NEWUSER account.  If it makes anyG > difference, I'm using WS_FTP Pro on a 98SE PC to log into my VAX on as > LAN.  % A couple of things you might look at:t  = Is there anything in the accounting logs?  Use a command likey     $ ACC/FULL/USER=newuseru  G You'd want to look for the entries labelled "Final status code" and thenA "Final status text" (they give the same info in different forms).1  7 Does AUTHORIZE show any login restrictions for NEWUSER?b  D Can you try another ftp client?  "Login Incorrect" (which I think isE what you said was the error message) is not a VMS message - unless it D comes from your ftp server.  Another ftp client may give a different( message which might be more informative.   Dave --------------9 Dave Greenwood                Email: Greenwoodde@ORNL.GOVtH Oak Ridge National Lab        %STD-W-DISCLAIMER, I only speak for myself   ------------------------------  % Date: Mon, 14 May 2001 15:47:12 -0400g- From: Michael Austin <miaustin@bellsouth.net>h& Subject: Re: Setting Up An FTP Account, Message-ID: <3B003640.6EBE8F6@bellsouth.net>  < I did not see you mention which IP stack you are using (UCX,* MULTINET...).  is the FTP service enabled?  H Also, when you copy a username in AUTHORIZE you "generally" want to give it a new UIC as well.     Michael Austin  - available now.7 Rdb DBA Consultant  -- I also do VMS System management.2   Still Learning wrote:r  
 > Richard, >tG > When I copied SYSTEM to NEWUSER in AUTHORIZE, I used /PASS=NEWPASS in H > the COPY command (along with /PWDMINIMUM=6 and /NOPWDLIFETIME) and I'mH > able to login interactively with the NEWUSER account.  If it makes anyG > difference, I'm using WS_FTP Pro on a 98SE PC to log into my VAX on aw > LAN.   ------------------------------    Date: 14 May 2001 16:44:40 -05003 From: malmberg@encompasserve.org (John E. Malmberg)s& Subject: Re: Setting Up An FTP Account3 Message-ID: <Ma+PtAKSqF0r@eisner.encompasserve.org>J  9 In article <3b001f10.37855468@news.jamison1.pa.home.com>,e, adroso@home.company (Still Learning) writes:
 > Richard, >oG > When I copied SYSTEM to NEWUSER in AUTHORIZE, I used /PASS=NEWPASS inrH > the COPY command (along with /PWDMINIMUM=6 and /NOPWDLIFETIME) and I'mH > able to login interactively with the NEWUSER account.  If it makes anyG > difference, I'm using WS_FTP Pro on a 98SE PC to log into my VAX on ac > LAN.  J Check to see if any new *.LOG files are showing up in the NEWUSER accounts% default directory with your attempts.l  M Check to make sure that you do not have something in SYLOGIN.COM or LOGIN.COM ( that does not work from a network login.  J Make sure that acounting is enabled for the process level and check to see> what the status of the network login attempts for NEWUSER are.   -John  wb8tyw@qsl.network Personal Opinion Only ! OpenVMS, No space suits required!k   ------------------------------    Date: 14 May 2001 16:51:48 -0500- From: koehler@encompasserve.org (Bob Koehler)s& Subject: Re: Setting Up An FTP Account3 Message-ID: <Zf9EhVd14L2h@eisner.encompasserve.org>t  a In article <3aff941c.2277627@news.jamison1.pa.home.com>, adroso@home.com (Still Learning) writes:-H > I'm new at this, so go easy on me ... I have a user account on my V7.2H > VAX that I want to login to via FTP from a PC on my LAN, but each timeH > it tells me Login Incorrect even though I *KNOW* I'm using the correctE > username and password.  I can get in using my SYSTEM account, and I/G > copied the SYSTEM account to the username I want to use in AUTHORIZE,,G > but it won't accept the new username.  Obvious, I'm missing somethingh > ... what is it?e  H Is WS_FTP telling you "Login Incorrect"?  Can you set it to verbose modeC and see what VMS is telling WS_FTP?  Possibly you have something int> your login.com (or sylogin.com) which needs to be protected by an  #    $if f$mode() .eqs. "INTERACTIVE"s  such as "set terminal" commands.  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences CorporationI= NASA GSFC Flight Software       | Federal Sector, Civil GrouprE                                 | please remove ".aspm" when replying    ------------------------------  % Date: Mon, 14 May 2001 23:51:59 -0400-- From: JF Mezei <jfmezei.spamnot@videotron.ca>d( Subject: SLIP on TCPIP servives V5 (VAX), Message-ID: <3B00A7DD.D68F7486@videotron.ca>  L Since the nice engineers at Compaq have not seen fit to provide the PPP codeN to allow VAXes to keep up with the times, I decided to try a SLIP interface to plug in a PDA to a VMS VAX.o  9 Slip seems very very simple. But alas, it doesn't work...i  & However, i have a couple of questions:  ( The VMS host has the address of 10.0.0.3J The ethernet interface has hosts that are all in the range 0 to 15 for the
 last byte.  & Then I create the SLIP interface with:M TCPIP> SET INTERFACE SL1 /network=255.255.255.240/host=10.0.0.20/SERIAL=TTA2:c  J Question: is the /HOST address the actual address that will be used by the3 PDA, or is that some "middle of the line" address ?   H SHOW INTERFACE shows an increasing number of RECEIVED packets as the PDAD generates data (PINGs for instance), but the SEND number stays at 0.  J If I use TCPTRACE 10.0.0.20 , I will then see the PINGs (or other packets)N sent by the PSION. If I used the basic web browser of the PDA, it does attemptK to connect to port 80 of the node. So it would seem to me that the VMS hostyN does receive properly formatted TCP packets but either does not act on them or5 is unable to reverse the connection back to the PDA. D  N However, the VMS host never seems to respond. I have a small RS232 led plug inC and I see data sent by the PDA, but never any data by the VMS host.    I added a route with:C  M SET ROUTE 10.0.0.20/gateway=10.0.0.3  (.3 being the host on which TTA2 exists  and on which I try stuff).  + But that doesn't seem to make a difference.   N I know this is very vague, but any hints on what I should be looking for would be appreciated.-  H I had originally set the PDA to be in the same subnet as the rest of theL machines, but that didn't work and decided to plane it in a different subnet3 hoping it would help the VMS host route the packet.    ------------------------------  % Date: Mon, 14 May 2001 21:22:58 -0700e* From: "Ian King" <iking@killthewabbit.org>. Subject: Re: Some more bits looking for a home> Message-ID: <m92M6.11362$55.4381768@newsrump.sjc.telocity.net>   "Apples & oranges" alert:  :-)  D 10A @ *240V*.  That's a 20A circuit for us 110V folks, which is whatJ currently powers my 11/34.  The 11/34 (with two RK05Js) does OK, even withL the hot tub (and some incidental lighting) on that circuit; it's only when II power up an RM02 and the hot tub heater kicks in that I blow the breaker..J :-)  I'm currently planning a small rewire project, to recycle a 240v lineH I'm not using (it's not as common in the US, usually only for heavy drawK appliances) to run the 11/34 and accessories on one leg, and the hot tub onrI the other.  ("Recycle" means rewire to two 110 circuits, in case anyone'su wondering.)e  H Also, I'm not the least bit surprised that a newer machine like an 11/70J would probably have less draw; one of the areas of progress in electronicsL over the years has been decreasing power draw.  My DELUA probably draws more than your CPU.  :-)c  K Heating is all well and good if you need it; now if I could just figure outeI some way to use the 11/34's energy draw to keep beer cold, it'd run 24x7.i
 Cheers -- Iana  9 "Paul Repacholi" <prep@prep.synonet.com> wrote in messagen' news:8766f37n3p.fsf@prep.synonet.com...i > Aidan KarleyF <aidan@karley.go-and-spam-me-you-sweaty-toads.freeserve.co.uk> writes: >pE > >      3~4KW though is a worry- that's a full 13 amps, which is the E > > maximum for floor cables in domestic use. Sounds like I really dosD > > need to split the unit into a couple of racks on a separate wall > > socket each. > I > That is a VERY high number. I ran an 11/70 of a single 10A 240v socket. G > Plus 3 ph for the RP06, and 2 RM03s on a second 10A. Without some oldoI > SMD drives, yu would be working VERY hard to pull 1KW in a qbus system.l >e >e > --> > Paul Repacholi                               1 Crescent Rd.,9 > +61 (08) 9257-1001                           Kalamunda.rB >                                              West Australia 60760 > Raw, Cooked or Well-done, it's all half baked.J > Spam-To: uce@ftc.gov,enforcement@sec.gov,sness@fcc.gov,hfurchtg@fcc.gov,$ >   mpowell@fcc.gov,gtristan@fcc.gov   ------------------------------  % Date: Tue, 15 May 2001 00:55:27 -0400g' From: Doug Carman <pdp11@bellsouth.net>l. Subject: Re: Some more bits looking for a home- Message-ID: <3B00B6BF.3F357E78@bellsouth.net>e   Ian King wrote:  > J > Also, I'm not the least bit surprised that a newer machine like an 11/70L > would probably have less draw; one of the areas of progress in electronicsN > over the years has been decreasing power draw.  My DELUA probably draws more > than your CPU.  :-)m > D The 11/70 is actually older and much larger than an 11/34.  I postedG this a while back based on the observed current draw from my 11/70.  My G system no longer has any MASSBUS devices or RH70 controllers installed,aH so it draws much less power than a complete system from mid to late 70'sD would have. Even at that, I don't run it on anything close to a 7x24 schedule these days :)   System configuration:"  9 PDP-11/70 CPU KB11-C w/FP11-C, 0 RH70's, 1 DL11W, 1 RX211D MK11 1024KWf0 BA11 w/1 DZ11, 1 RL11, 1 KLESI, 1 DELUA, 1 UDA50 RX02
 SA72 w/4 RA72W RL02 TU81+8 LA120B  F The entire system is fed with a single 220V 30A circuit which is split9 into two 120V circuits through an 861-D power controller.e  < Configuration           Run current             Peak Current:                         L1      L2              L1      L2< CPU only                5.8A    5.2A            6.2A    5.7A= CPU+MK11                5.8A    8.9A            6.2A    10.5Aa= CPU+MK11+BA11           9.9A    8.9A            14.7A   10.5As= Full load at idle       12.1A   14.1A           17.4A   17.5Ao% Full load with all      12.2A   15.4AM disks spun up, tape- loading, CPU running
 EMKA diag.  E The total current draw would be the sum of the values from L1 and L2.-  H These are observed values from my own system.  Your system will never be2 exactly the same as mine.  Your mileage will vary.   -- : Doug Carmane pdp11@bellsouth.netZ   ------------------------------  % Date: Mon, 14 May 2001 17:12:37 -0400M  From: John Santos <JOHN@egh.com>+ Subject: Re: The future of VMS - * Humour *.5 Message-ID: <1010514170652.8541A-100000@Ives.egh.com>t  $ On Sun, 13 May 2001, JF Mezei wrote:   > John Santos wrote:F > > I'm unfamiliar with the VAX hardware ;-), but many vacuum cleanersD > > have an output port, too.  You can plug the hose into it and use@ > > it to inflate low-pressure stuff, such as air mattresses and > > inflateable boats. > O > My VAX have an output port too (I beleive some call it a printer). It is usedeL > to move paper from a paper tray out to someone who quickly files it in theO > wastebasket. So it can be said that VAXes have an output port that is used toa > fill wastebaskets.  D One of my customers used to use a wastebasket as the output bin on aG high speed printer.  I think it was the fastest LP DEC made at the time B (1200 lpm?), printed continously all day right into a wastebasket.H They were worried about losing stuff and if something bad happened, theyD would pull the paper out of the wastebasket and re-enter it by hand.  F As I recall, they printed several boxes of paper a day.  Periodically,D they would empty the wastebasket back into the original paper boxes,E seal them up, and let them sit in the corner for a week or two beforeh sending them to the recyclers.  E I think after several years of this, they realized the disks (RP04's)s= were more reliable than the printer, we just saved everythingr? on two different disks and backed it up to tape every night and  they saved several forests.t   -- = John Santos= Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Mon, 14 May 2001 18:58:53 +0100a0 From: andrew harrison <andrew.nospam@uk.sun.com>8 Subject: Re: The Internet, Tru64 and other unix variants* Message-ID: <3B001CDD.8BDB839E@uk.sun.com>   Christopher Smith wrote: >  > > -----Original Message-----4 > > From: Christof Brass [mailto:brass@infopuls.com] >  > > andrew harrison wrote: >  > > Christof Brass wrote:i > C > > > > From what I know the WINE people don't put much effort into-G > > > > their project. These are pasttime programmers and honestly they.C > > > > don't use very advanced techniques from my point of view. AkF > > > > friend of mine has reverse engineered substantial parts of theD > > > > M$ COM DLLs and he could easily do the same for every single > > > > service within WNT.s > M > I have had occasion to look at the wine source code.  I must say that while F > it's not perfect by any means, it seems very well done in most partsN > (different parts being done by different people, the quality varies some...) > K > I really think that the problem is not in the reverse engineering, but innM > the re-implementation.  They must come up with work-alike code that, to M$,tJ > looks nothing like M$ code. :) Certainly a very difficult task, since M$1 > takes even other peoples' code for their own ;)r >   : The reverse engineering issue is a problem as reading the 8 WINE release notes showed. The documentation constantly 4 refered to questions about how one or other of MS's  features worked.  9 So much so that Wine itself includes an extensive set of e9 debug facilites to trace system calls, trace messages etcn8 to help along with additional freeware apps like PE and  IDA. n      9 > > > Why not post this to the WINE newsgroup. They mighti8 > > > disagree, they were at one point putting out a new& > > > WINE release roughly every week. > A > > Interesting that you are using this pointless measurement forsC > > quality (remember what I pointed out with reagard to complexityiA > > of the problem?) and quantity of development effort - did youpC > > work at Micro$hit? They build every night a new NT. Very modernr? > > and efficient way to develop something, don't you think so?t > I > I have to agree with you here.  An automated weekly build means nothingg= > about the amount of productive changes made to the product.o >   = Except of course that Christof has no experience of building M; and running WINE while I do and the weekly releases always KB included code changes which were either bug fixes or enhancements.     Regardsr Andrew Harrisong Enterprise IT Architect4   ------------------------------  % Date: Tue, 15 May 2001 01:17:10 +0200e) From: Christof Brass <brass@infopuls.com>s8 Subject: Re: The Internet, Tru64 and other unix variants+ Message-ID: <3B006776.4CF17F0@infopuls.com>n   Fred Kleinsorge wrote: > K > Ignoring potential sources of *useful* software is simple stupidity.  The-K > fact that an application is written to use UNIX library/system interfacesoE > does not mean that it has no value, or no value to an OpenVMS user.  > M > Rewriting much of this software to replace UNIX calls with VMS calls is notrL > only mostly pointless, but counterproductive in that it then takes you outJ > of the mainstream and increases the diffculty of obtaining new versions. > E > Christof Brass wrote in message <3AF9E5AD.476C6E94@infopuls.com>...h > >Jan Vorbrueggen wrote:M > >>0 > >> Christof Brass <brass@infopuls.com> writes: > >>4 > >> > There is no point in having UNIX apps on VMS. > >>N > >> I suggest you first define what a "UNIX app" is - possibly in contrast toA > >> a "VMS app" - and then we might have a basis for discussion.A > >> > >>         Jan > >-> > >Simply an app which uses mostly UNIX system calls and whichB > >doesn't take full advantage of the VMS features like RMS files.3 > >Poorly ported UNIX apps which should run on VMS.o  = It is a value e.g. to have properly used index files. I don'tc? know of any UNIX app ported to VMS doing that. If a UNIX app is ? properly ported to VMS and the maintenance is still done on thed> UNIX side I would propose to automate this replacement process9 where necessary. It could be that the OS interface can beE< isolated in a way the these parts are kept on the respective OSs.   ------------------------------  % Date: Mon, 14 May 2001 16:19:55 -0700y! From: Tom Linden <tom@kednos.com>o Subject: VAX memory ?e9 Message-ID: <CIEJLCMNHNNDLLOOGNJIEEDKCJAA.tom@kednos.com>u  ; I have a 4000/60 and a 4000/90  each have 4 memory modules.A  + When I boot the 60 it claims to have 72 MB.R+ When I boot the 90 it claims to have 64 MB.   J All the modules have 36 chips  each of which is 4 Mb.  the 60 use Goldstar0 and the 90 Hitachi but they are interchangeable.  L Is there perhaps a firmware issue here.  I would guess that the modules mustL be interchangeable.  The designation on the modules is however CA-082 on the 60 and CA-CFK on the 90, which makes me suspicious.  If they aren't interchangeable, I am-J sure Dec had a reason, but I am also certain that I would never understand why?  
 Anybody know?n   ------------------------------  % Date: Mon, 14 May 2001 17:50:46 -06003( From: emanuel stiebler <emu@ecubics.com> Subject: Re: VAX memory ?k+ Message-ID: <3B006F56.73ABB13A@ecubics.com>T   Tom Linden wrote:  > = > I have a 4000/60 and a 4000/90  each have 4 memory modules.  > - > When I boot the 60 it claims to have 72 MB.v- > When I boot the 90 it claims to have 64 MB.V  M > on the 90, which makes me suspicious.  If they aren't interchangeable, I ameL > sure Dec had a reason, but I am also certain that I would never understand > why?   That one was easy ;-)O  7 The 4000/60 has 8 MByte onboard. So everything is fine.e   cheers & have fun    ------------------------------   Date: 14 May 2001 23:51:45 GMT- From: "Bill Pedersen" <pedersen@ccsscorp.com>- Subject: Re: VAX memory ?y. Message-ID: <3b006ff0@kerberos.linuxpuppy.net>  J The VAXstation 4000 Model 90 had a limitation of 64 MB or 16 MB increments
 of memory.  ( The Model 60 allows for 8 MB increments.  L I think this will explain the difference of how it is looking at the memory.   --
 Bill Pedersenh CCSS Corporation www.CCSScorp.com 831-336-2708. "Tom Linden" <tom@kednos.com> wrote in message3 news:CIEJLCMNHNNDLLOOGNJIEEDKCJAA.tom@kednos.com...d= > I have a 4000/60 and a 4000/90  each have 4 memory modules.I >t- > When I boot the 60 it claims to have 72 MB.o- > When I boot the 90 it claims to have 64 MB.  >cL > All the modules have 36 chips  each of which is 4 Mb.  the 60 use Goldstar2 > and the 90 Hitachi but they are interchangeable. >tI > Is there perhaps a firmware issue here.  I would guess that the moduleso mustJ > be interchangeable.  The designation on the modules is however CA-082 on then > 60 and CA-CFJ > on the 90, which makes me suspicious.  If they aren't interchangeable, I amL > sure Dec had a reason, but I am also certain that I would never understand > why? >- > Anybody know?  >a   ------------------------------  % Date: Mon, 14 May 2001 19:57:23 -0400r- From: Michael Austin <miaustin@bellsouth.net>s Subject: Re: VAX memory ?r- Message-ID: <3B0070E3.8895B84E@bellsouth.net>7   From the VMS FAQ:.  4 Why doesn't OpenVMS see the new memory I just added?  J  When adding memory to an OpenVMS system, one should check for an existing definition of the PHYSICALPAGESrB  (OpenVMS VAX) or PHYSICAL_MEMORY (OpenVMS Alpha) parameter in the SYS$SYSTEM:MODPARAMS.DATP  parameter database, use a text editor to reset the value in the file to the new/ correct value as required, and then perform the   following command:I  4        $ @SYS$UPDATE:AUTOGEN GETDATA REBOOT FEEDBACK    J  This AUTOGEN command will reset various system parameters based on recent$ system usage (FEEDBACK), and it willO  reset the value for the PHYSICALPAGES parameter to the new value. It will alson reboot the OpenVMS system.  M  PHYSICALPAGES and PHYSICAL_MEMORY can also be used to deliberately lower thet amount of memory available forL  use by OpenVMS. This ability can be useful in a few specific circumstances,4 such as testing the behaviour of an application in aP  system environment with a particular (lower) amount of system memory available.  I  PHYSICALPAGES and PHYSICAL_MEMORY can be set to -1, to indicate that all-  available memory should be used.     Michael Austin DBA Consultant     Tom Linden wrote:a  = > I have a 4000/60 and a 4000/90  each have 4 memory modules.D >6- > When I boot the 60 it claims to have 72 MB.s- > When I boot the 90 it claims to have 64 MB.  >eL > All the modules have 36 chips  each of which is 4 Mb.  the 60 use Goldstar2 > and the 90 Hitachi but they are interchangeable. >yN > Is there perhaps a firmware issue here.  I would guess that the modules mustN > be interchangeable.  The designation on the modules is however CA-082 on the > 60 and CA-CFM > on the 90, which makes me suspicious.  If they aren't interchangeable, I am L > sure Dec had a reason, but I am also certain that I would never understand > why? >n > Anybody know?d   ------------------------------  % Date: Mon, 14 May 2001 16:52:01 -0700d! From: Tom Linden <tom@kednos.com>I Subject: RE: VAX memory ?g9 Message-ID: <CIEJLCMNHNNDLLOOGNJIGEDLCJAA.tom@kednos.com>G  7 Thanks.  I take that to be a YES to interchangeability?e   > -----Original Message-----1 > From: emanuel stiebler [mailto:emu@ecubics.com]O$ > Sent: Monday, May 14, 2001 4:51 PM > To: Info-VAX@Mvb.Saic.Comc > Subject: Re: VAX memory ?. >  >  > Tom Linden wrote:t > > ? > > I have a 4000/60 and a 4000/90  each have 4 memory modules.e > > / > > When I boot the 60 it claims to have 72 MB.o/ > > When I boot the 90 it claims to have 64 MB.e > : > > on the 90, which makes me suspicious.  If they aren't  > interchangeable, I am D > > sure Dec had a reason, but I am also certain that I would never  > understand > > why? >  > That one was easy ;-)h > 9 > The 4000/60 has 8 MByte onboard. So everything is fine.t >  > cheers & have fune >    ------------------------------  % Date: Mon, 14 May 2001 17:04:05 -0700w! From: Tom Linden <tom@kednos.com>g Subject: RE: VAX memory ? 9 Message-ID: <CIEJLCMNHNNDLLOOGNJIMEDMCJAA.tom@kednos.com>'   Michael,  H Thanks for the reponse.  My problem wasn't VMS related, this is what was being reported from the console.-       > -----Original Message-----6 > From: Michael Austin [mailto:miaustin@bellsouth.net]$ > Sent: Monday, May 14, 2001 4:57 PM > To: Info-VAX@Mvb.Saic.Com- > Subject: Re: VAX memory ?e >u >s > >From the VMS FAQ: >e6 > Why doesn't OpenVMS see the new memory I just added? >cL >  When adding memory to an OpenVMS system, one should check for an existing! > definition of the PHYSICALPAGES D >  (OpenVMS VAX) or PHYSICAL_MEMORY (OpenVMS Alpha) parameter in the > SYS$SYSTEM:MODPARAMS.DATB >  parameter database, use a text editor to reset the value in the > file to the newm1 > correct value as required, and then perform the  >  following command:o >n6 >        $ @SYS$UPDATE:AUTOGEN GETDATA REBOOT FEEDBACK >e >/L >  This AUTOGEN command will reset various system parameters based on recent& > system usage (FEEDBACK), and it will= >  reset the value for the PHYSICALPAGES parameter to the new  > value. It will alsok > reboot the OpenVMS system. >e8 >  PHYSICALPAGES and PHYSICAL_MEMORY can also be used to > deliberately lower the  > amount of memory available for? >  use by OpenVMS. This ability can be useful in a few specificd > circumstances,6 > such as testing the behaviour of an application in a@ >  system environment with a particular (lower) amount of system > memory available.< >pK >  PHYSICALPAGES and PHYSICAL_MEMORY can be set to -1, to indicate that alli" > available memory should be used. >s >- > Michael Austin > DBA Consultant >i >: > Tom Linden wrote:t >g? > > I have a 4000/60 and a 4000/90  each have 4 memory modules.o > >o/ > > When I boot the 60 it claims to have 72 MB.s/ > > When I boot the 90 it claims to have 64 MB.k > >aA > > All the modules have 36 chips  each of which is 4 Mb.  the 60  > use Goldstar4 > > and the 90 Hitachi but they are interchangeable. > > C > > Is there perhaps a firmware issue here.  I would guess that the  > modules mustB > > be interchangeable.  The designation on the modules is however > CA-082 on the  > > 60 and CA-CF9 > > on the 90, which makes me suspicious.  If they aren'tm > interchangeable, I am C > > sure Dec had a reason, but I am also certain that I would neverl > understand > > why? > >  > > Anybody know?g >e   ------------------------------  % Date: Mon, 14 May 2001 18:12:36 -0400i- From: "Mitchell, David R." <mitchell@WPI.EDU> C Subject: RE: XDM logins and Advanced Server external authenticationlH Message-ID: <FE1835D68492D311BF7900508B5BEB0D2CA3F5@petra.admin.wpi.edu>  G Yes.  I'm aware of that.  The question was why doesn't authentication =  workE from an X session when Advanced Servers' external authentication is =h used.  IE would expect all athentication to be handled the same, so if externalbF authentication works from a telnet or rlogin session why wouldn't it = workG from an X session?  Apparently some other mechanism is at work when X =t loginsA are being established.  Strange, but perhaps someone else knows = 	 something 
 about this???M   Davidv   > -----Original Message-----6 > From: Jack Patteeuw [mailto:jjpatteeuw@peoplepc.com]& > Sent: Saturday, May 12, 2001 4:03 PM > To: Info-VAX@Mvb.Saic.ComnE > Subject: Re: XDM logins and Advanced Server external authentication" >=20 >=20G > XDMCP has nothing to do with "authentication", other than providing a18 > screen where you can enter your username and password. >=20G > Having said that, Thanks Compaq, for **FINALLY** providing some basiciD > "X" functionality that has been around for years in every Un*x OS.? > Finally brings VMS up to where Ultrix was over 5 years ago !!k >=20 > Jack Patteeuw  >=20 >=20 > "Mitchell, David R." wrote:2 > >=20A > > We run Advanced Server 7.2 on our OpenVMS 7.2-1 system and=20  > I've recently been@ > > playing around with external authentication.  Most of our=20
 > users don't7= > > directly log into our OpenVMS systems on a daily basis=204 > anymore so they tendA > > to forget their password by the time they do need to login=20i > again.  They dohA > > however login to their NT/2000 accounts on a daily basis. =20- > I'd therefore like= > > to use the "extauth" flag on their OpenVMS accounts to=20  > allow them to useuA > > their NT/2000 usernames and passwords when they do need to=20  > login to our> > > OpenVMS system.  Everything seems to work fine with the=20 > testing I've done soB > > far when logging in via telnet.  But, there are still a few=20 > users (including? > > myself) that use PC-Xware to login.  I've found that any=20n > login attempt fails B > > when using the extauth flag on my account with an X login. =20 > We'd actuallyr? > > been running some old NCDWare software that was handling=20  > our XDM logins until> > > this morning when I upgraded to TCP/IP 5.1 and shutdown=20 > NCD's software so I @ > > could use the Compaq XDM server.  I thought that since we=20 > were using thirdA > > party software for XDM logins it didn't know how to handle=20a > the external< > > authentication.  But, after switching to Compaq's XDM=20 > server I still can'tB > > get this to work.  Anyone have any ideas on how to get this=20 > to work or any? > > information indicating that this simply wont work?  Thanks.a > >=20	 > > David  >=20   ------------------------------  % Date: Mon, 14 May 2001 16:51:20 -0600 % From: Dan O'Reilly <dano@process.com>aC Subject: RE: XDM logins and Advanced Server external authenticationaA Message-ID: <5.0.2.1.2.20010514164759.00ae6030@ntbsod.psccos.com>a  M Generally, with X logins, there is one of two mechanisms employed: XauthorityoB or Xhost.  Virtually all UNIX systems us Xauthority (also known asL "MIT Magic Cookie" for authentication.  This isn't part of XDM, it's part of8 the X specification.  And DECwindows doesn't support it.  0 At 04:12 PM 5/14/2001, Mitchell, David R. wrote:K >Yes.  I'm aware of that.  The question was why doesn't authentication workhM >from an X session when Advanced Servers' external authentication is used.  I F >would expect all athentication to be handled the same, so if externalJ >authentication works from a telnet or rlogin session why wouldn't it workM >from an X session?  Apparently some other mechanism is at work when X loginsOJ >are being established.  Strange, but perhaps someone else knows something >about this???   ------I +-------------------------------+---------------------------------------+ I | Dan O'Reilly                  |                                       |mI | Principal Engineer            |  "Why should I care about posterity?  |sI | Process Software              |   What's posterity ever done for me?" |uI | http://www.process.com        |                    -- Groucho Marx    | I +-------------------------------+---------------------------------------+o   ------------------------------  % Date: Tue, 15 May 2001 01:29:39 +0200b) From: Christof Brass <brass@infopuls.com>tP Subject: Re: [EVENT] Last call for =?iso-8859-1?Q?Z=FCrich?= dinner registration, Message-ID: <3B006A63.42F80F7E@infopuls.com>   Didier Morandi wrote:e > H > On the 18th of May, this Friday, will take place the OpenVMS Technical" > Update Seminar in Compaq Zrich. > E > After this show and totally independant in terms of organization, ar > dinner has been set up.n > J > We are already 8 people. If you will attend this seminar and if you read& > this, you are welcome to the dinner. > ? > Feel free to send mail to Didier.Morandi@ubs.com to register.r	 > Thanks.  >  > D. > --  > Compaq Computer (Schweiz) GmbH# > berlandstrasse 1, 8600 Dbendorfs > GSM: +41 79 7054670a  @ As a complement: I have still some ideas to continue the evening= after the dinner. OTOH it could be that most of the attendeese, are too tired (jet lag, conference, dinner).   ------------------------------   End of INFO-VAX 2001.267 ************************