1 INFO-VAX	Wed, 17 Aug 2005	Volume 2005 : Issue 458       Contents: DFG (Defrag) - OpenVMS Re: DFG (Defrag) - OpenVMS Re: DFG (Defrag) - OpenVMS Re: Eisner is down... F Re: How many files can you have in a VMS directory without performanceL How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems? Re: Image restore fails   Re: Investigating ACCVIO problem Reading VMS SAS files on PC?  RE: Reading VMS SAS files on PC?  Re: Reading VMS SAS files on PC?  Re: Replacement for VT terminals  RE: Replacement for VT terminals  RE: Replacement for VT terminals  Re: Replacement for VT terminals  Re: Replacement for VT terminals  Re: Replacement for VT terminals Re: SAMBA for VMS ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' RE: SAMBA for VMS (The saga continues!) # Re: The Educational License Program ! Re: Turning up an Ethernet & tftp   F ----------------------------------------------------------------------  % Date: Wed, 17 Aug 2005 11:00:04 -0500 + From: Chuck Aaron <caaron@ceris.purdue.edu>  Subject: DFG (Defrag) - OpenVMS 5 Message-ID: <ddvmu2$4r8$1@mailhub227.itcs.purdue.edu>   I Can someone tell me what account and what parameter/value needs to be set G to prevent this from occurring when DFG runs to defrag one of my disks?    Thanks,  Chuck     L Begin file evaluation phase:                         17-AUG-2005 00:00:00.72   H     Candidates for defragmentation (based on the defragmentation level):   : %DFG-F-SYSSRVERR, call to system service SYS$EXPREG failed) -SYSTEM-F-EXQUOTA, process quota exceeded  CHUCK>   ------------------------------  + Date: Wed, 17 Aug 2005 16:44:19 +0000 (UTC)  From: david20@alpha2.mdx.ac.uk# Subject: Re: DFG (Defrag) - OpenVMS ) Message-ID: <ddvph3$kc6$1@news.mdx.ac.uk>   c In article <ddvmu2$4r8$1@mailhub227.itcs.purdue.edu>, Chuck Aaron <caaron@ceris.purdue.edu> writes: J >Can someone tell me what account and what parameter/value needs to be setH >to prevent this from occurring when DFG runs to defrag one of my disks? >  >Thanks, >Chuck >  > M >Begin file evaluation phase:                         17-AUG-2005 00:00:00.72  > I >    Candidates for defragmentation (based on the defragmentation level):  > ; >%DFG-F-SYSSRVERR, call to system service SYS$EXPREG failed * >-SYSTEM-F-EXQUOTA, process quota exceeded >CHUCK>   $ Look in sys$startup:dfg$startup.com.! You should find a section like :-    " F $!      The following logicals override the default quotas used by allK $!      defragmentation processes on this system.  The values shown are the N $!      minimim recommended.  In general, the quotas used are conservative andF $!      may be increased without restriction (other than normal systemI $!      restrictions).  DIGITAL does not recommend reducing these values. M $!      Doing so may cause the defragmentation process to fail.  See the user D $!      guide for the POLYCENTER File Optimizer for OpenVMS for moreN $!      information about using these logicals. See Open VMS documentation forM $!      more information about the effects of individual quotas on a process.  $!K $!      To use these logicals, increase the associated value, uncomment the , $!      desired line, and execute this file. $!/ $!  CALL DEFINE_LOGICAL DFG$ASTLM           100 . $!  CALL DEFINE_LOGICAL DFG$BIOLM           181 $!  CALL DEFINE_LOGICAL DFG$BYTLM           61440 . $!  CALL DEFINE_LOGICAL DFG$DIOLM           18/ $!  CALL DEFINE_LOGICAL DFG$ENQLM           500 . $!  CALL DEFINE_LOGICAL DFG$FILLM           50A $!  CALL DEFINE_LOGICAL DFG$JTQUOTA         {uses system default}  .  .  .  "   ! You probably need to increase the    DFG$PGFLQUOTA    and possibly   DFG$WSDEFAULT          DFG$WSEXTENT           DFG$WSQUOTA                 
 David Webb Security team leader CCSS Middlesex University   ------------------------------    Date: 17 Aug 2005 09:39:23 -0700$ From: "Ed Wilts" <ewilts@ewilts.org># Subject: Re: DFG (Defrag) - OpenVMS C Message-ID: <1124296763.427148.254880@o13g2000cwo.googlegroups.com>   
 dfg$pgflquota   	    .../Ed    ------------------------------  # Date: Wed, 17 Aug 2005 10:29:16 GMT " From:   VAXman-  @SendSpamHere.ORG Subject: Re: Eisner is down...0 Message-ID: <00A4867D.7452DB6F@SendSpamHere.ORG>  c In article <8Fs8tpkVzNPb@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes: O >In article <BPmdnb8La_w_YpzeRVn-rw@comcast.com>, Brad Hamilton <brMadAhamPiltS  >h on@coMmcaAstP.neSt> writes: >> Larry Kilgallen wrote: O >>> In article <hpidnZ2dnZ2G4Gu5nZ2dnTBKnN6dnZ2dRVn-zZ2dnZ0@comcast.com>, Brad  6 >h Hamilton <brMadAhamPiltSon@coMmcaAstP.neSt> writes:	 >> <snip> K >>>>I'm going to send mail to Steve now; I've noted the same problem since   >>>>yesterday... >>> L >>> I did better than that -- I telephoned him and it was the first he heard: >>> of the problem.  He seems to have gotten it fixed now. >>>  >>  < >> Thanks, Larry.  I don't have his phone number...      :-) > H >Now that things are working again, you could print out note 343.58 from  >the DECUSERVE_FORUM conference. > I >At one point there was a message (that I cannot find in that conference) H >urging people to copy down the phone number in good times so they would >have it in times of trouble.    Thanks Larry, number noted!  --  K VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM              5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  % Date: Wed, 17 Aug 2005 08:37:00 -0500 ( From: Wayne Sewell <wayne@tachysoft.com>O Subject: Re: How many files can you have in a VMS directory without performance / Message-ID: <00A4868F.4CB2AD70.5@tachysoft.com>   = >From: "Chris Blackburn" <chris.blackburn@uk.thalesgroup.com>  >X-Newsgroups: comp.os.vmsP >Subject: Re: How many files can you have in a VMS directory without performance >         problems? & >Date: Wed, 17 Aug 2005 12:36:42 +0100 >Organization: Thales     K >I have noticed on our VMS boxes that perfomance begins to degrade when the  >number L >of files in a directory exceeds 10,000. This is particularly noticable when
 >one attempts H >to delete files as I suspect the delete works in alphabetical order and >removes theM >first file, tidies up the directory entries (reclaiming space when possible) 	 >and then J >continues to delete the second file etc. The tidying up at the end of the
 >delete seems L >to be what kills performance and it can take several seconds to delete each >file when a large >number are in the directory.  >     M In the original request and in the replies, no mention is made of vms version 8 (yes, the former said 7.x, but I mean *actual* version).  H The reason I bring it up is because there were supposed to be some majorL changes in this area in a fairly recent version of vms, don't remember whichH one.  So the big-directory behavior should change radically between thatJ version and one before, and comments by people on one side of the boundary: would not have much relevance for those on the other side.  J I don't have any direct experience with this at any vms version, because IK simply don't have directories of that size.  If I see one going that way, I N break it up into subdirectories.  And I know how purge/keep=n works and use it every now and then.     ) >"RAQ" <_raq_@yahoo.com> wrote in message > >news:1124275575.684487.287580@z14g2000cwz.googlegroups.com... >> Hi  >> >> Everytime I ask HP about F >> "How many files can you have in a VMS directory without performance
 >> problems?"  >>G >> they send me an article that is about 10-15 years old for VMS 5.2 or D >> even 1.0, that describes that more than 300 files in a catalog is	 >> bad...  >>I >> I have a feeling that <25.000 files works ok (no major complains), but E >> when we reach 100.000 files it gets bad - and with 500.000 it gets  >> really nasty :) >> >>B >> Does anyone know the real numbers? Or how I can calculate them?> >> Its for VMS 7.x + EVA SAN, and is mainly for file creation. >> >> >>	 >> Thanks  >> >  >  >-- P >===============================================================================N >Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxx; >http://www.tachysoft.xxx/www/tachyon.html and wayne.html   L >change .xxx to .com in addresses above, assuming you are not a spambot  :-)P >===============================================================================Q >Jake Blues:"You traded the Caddy for a microphone? ...... Okay, I can buy that."  --  O =============================================================================== M Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxx : http://www.tachysoft.xxx/www/tachyon.html and wayne.html  K change .xxx to .com in addresses above, assuming you are not a spambot  :-) O =============================================================================== P Jake Blues:"You traded the Caddy for a microphone? ...... Okay, I can buy that."O =============================================================================== N Wayne Sewell, Tachyon Software Consulting  (281)812-0738   wayne@tachysoft.com; http://www.tachysoft.com/www/tachyon.html and wayne.html    O =============================================================================== P Jake Blues:"You traded the Caddy for a microphone? ...... Okay, I can buy that."   ------------------------------    Date: 17 Aug 2005 03:46:15 -0700 From: "RAQ" <_raq_@yahoo.com> U Subject: How many files can you have in a VMS directory without performance problems? C Message-ID: <1124275575.684487.287580@z14g2000cwz.googlegroups.com>    Hi   Everytime I ask HP aboutC "How many files can you have in a VMS directory without performance 
 problems?"  D they send me an article that is about 10-15 years old for VMS 5.2 orA even 1.0, that describes that more than 300 files in a catalog is  bad...  F I have a feeling that <25.000 files works ok (no major complains), butB when we reach 100.000 files it gets bad - and with 500.000 it gets really nasty :)     ? Does anyone know the real numbers? Or how I can calculate them? ; Its for VMS 7.x + EVA SAN, and is mainly for file creation.        Thanks   ------------------------------  % Date: Wed, 17 Aug 2005 12:36:42 +0100 < From: "Chris Blackburn" <chris.blackburn@uk.thalesgroup.com>Y Subject: Re: How many files can you have in a VMS directory without performance problems? % Message-ID: <ddv7ge$n24$1@rdel.co.uk>   J I have noticed on our VMS boxes that perfomance begins to degrade when the numberK of files in a directory exceeds 10,000. This is particularly noticable when  one attemptsG to delete files as I suspect the delete works in alphabetical order and  removes the L first file, tidies up the directory entries (reclaiming space when possible) and thenI continues to delete the second file etc. The tidying up at the end of the  delete seemsK to be what kills performance and it can take several seconds to delete each  file when a large  number are in the directory.   Chris.  ( "RAQ" <_raq_@yahoo.com> wrote in message= news:1124275575.684487.287580@z14g2000cwz.googlegroups.com...  > Hi >  > Everytime I ask HP aboutE > "How many files can you have in a VMS directory without performance  > problems?" > F > they send me an article that is about 10-15 years old for VMS 5.2 orC > even 1.0, that describes that more than 300 files in a catalog is  > bad... > H > I have a feeling that <25.000 files works ok (no major complains), butD > when we reach 100.000 files it gets bad - and with 500.000 it gets > really nasty :)  >  > A > Does anyone know the real numbers? Or how I can calculate them? = > Its for VMS 7.x + EVA SAN, and is mainly for file creation.  >  >  >  > Thanks >    ------------------------------  # Date: Wed, 17 Aug 2005 11:48:01 GMT A From: "Colin Butcher" <colin_DOT.butcher_AT@xdelta_DOT.co_DOT.uk> Y Subject: Re: How many files can you have in a VMS directory without performance problems? < Message-ID: <RtFMe.90662$G8.2981@text.news.blueyonder.co.uk>  K You might need to discover DFU (see the Freeware CDs or Freeware download). A Very useful for mass deletions and removing directory trees. Very  informative too.   --     Hope this helps, Colin. ) colin DOT butcher AT xdelta DOT co DOT uk E It's not mine, but I like this definition: Legacy = stuff that works.    ------------------------------    Date: 17 Aug 2005 07:20:27 -0500 From: briggs@encompasserve.orgY Subject: Re: How many files can you have in a VMS directory without performance problems? 3 Message-ID: <zXrucVAZ42g5@eisner.encompasserve.org>   c In article <1124275575.684487.287580@z14g2000cwz.googlegroups.com>, "RAQ" <_raq_@yahoo.com> writes:  > Hi >  > Everytime I ask HP aboutE > "How many files can you have in a VMS directory without performance  > problems?" > F > they send me an article that is about 10-15 years old for VMS 5.2 orC > even 1.0, that describes that more than 300 files in a catalog is  > bad... > H > I have a feeling that <25.000 files works ok (no major complains), butD > when we reach 100.000 files it gets bad - and with 500.000 it gets > really nasty :)  >  > A > Does anyone know the real numbers? Or how I can calculate them? = > Its for VMS 7.x + EVA SAN, and is mainly for file creation.   C As you know from reading the article about VMS 5.2, the real crunch D point is when the directory size exceeds 127 blocks and caching gets thrown out the window.  C The piece that hurts is when you delete the last directory entry in < a directory block and you need to slide the remainder of theD directory down one block to fill in.  Historical implementations didC this by "bubbling" the remainder of the directory down one block at D a time in sequence -- block i is copied from i+1, i+1 from i+2, etc.  C For a 1000 block directory file, that's 2000 disk I/O operations to 2 delete the last file in the first directory block.  E The improvements in directory handling since then have (if I remember D rightly) improved performance by doing multi-block reads and writes.C So now you bubble the directory contents down 127 blocks at a time.   H For a 1000 block directory file, that's 16 disk I/O operations to delete+ the last file in the first directory block.   ' NOTE THAT IT'S STILL AN O(n^2) PROBLEM.   G All that has changed is the constant of proportionality.  Deletions are E about 127 times faster than they were at one time.  But if you double F the directory size, you'll still increase file delete time by a factor of four.  J If you have 500,000 files in a directory, you need to work out a differentO file naming scheme.  By my calculations that's a 50,000 block (or so) directory I file and will require 800 disk I/O's for every tenth file that you delete M at the front or 400 disk I/O's for every tenth file that you create or delete  in the middle.  L (Assumes ten files per directory block -- your mileage may vary depending on file name length).  L One common approach is to put files into a directory tree.  If the names areO constructed right, you might be able to harvest the first two characters in the * file name to be a directory tree position.   e.g.    TY6WDGT.DAT => [.T.Y]TY6WDGT.DAT  W9QRXAG.DAT => [.W.9]W9QRXAG.DAT  ? And now you're down to less than 1000 files per directory.  You F can expand or shrink any given directory in a maximum of 2 disk I/O's.D And you can figure out which directory to use by a simple inspection of the file name.    	John Briggs   ------------------------------    Date: 17 Aug 2005 08:29:39 -0700$ From: "AEF" <spamsink2001@yahoo.com>Y Subject: Re: How many files can you have in a VMS directory without performance problems? C Message-ID: <1124292578.948063.108510@g47g2000cwa.googlegroups.com>   
 RAQ wrote: > Hi >  > Everytime I ask HP aboutE > "How many files can you have in a VMS directory without performance  > problems?" > F > they send me an article that is about 10-15 years old for VMS 5.2 orC > even 1.0, that describes that more than 300 files in a catalog is  > bad... > H > I have a feeling that <25.000 files works ok (no major complains), butD > when we reach 100.000 files it gets bad - and with 500.000 it gets > really nasty :)  >  > A > Does anyone know the real numbers? Or how I can calculate them? = > Its for VMS 7.x + EVA SAN, and is mainly for file creation.  >  >  >  > Thanks  G Well, it depends on a lot of things. What version of VMS, e.g. Also, it C depends on what you want to do and in some cases how your files are C named. Note that you can store fewer large filenames in a directory G block than small filenames. And it is often the size of the .DIR;1 file  that matters (see below).   D There is no hard number limit for the number of files in a directory( that causes anything to abruptly change.   See the VMS FAQ sec. 9.4.   
 Additionally:   B On "older" version of VMS there is a problem in which .DIR;1 filesG larger than 128 blocks (or 127) are not cahced. But I remember one post * saying this only affects wildcard lookups.  G Deleting files with DELETE can be very slow in large directories, but I C never noticed any abrupt change in speed as the .DIR;1 file shrinks E below 128 blocks, so that agrees with the above. Deleting with DELETE D is slow because it goes alphabetically, so a DEL *.*;* command wouldE first delete all the files in the first directory block, then bump up E all the blocks past it by one, and repeat until done. I've never seen F this depend on the 128 block thing. But as it says in the FAQ, the I/OE logic and caching has been improved in newer versions of VMS. See the  FAQ for details.  D If you want to delete all files in a directory, one faster way is toF delete them in reverse order. Some people have command procedures that@ do this (see Jeff's post in this thread). Another trick is to do  3     $ BACKUP/DELETE *.*;* NL:A.B/SAVE/NOCRC/GROUP=0   ? But the fastest is to get the freeware program DFU from the VMS D Freeware CD, unless you want to delete ALL files on a disk, in which case use INIT. (!)  A I do recall seeing a post saying that how you name your files can F affect look-up performance in very large directories. It was somethingG about having the first so-many chars in the filename not being the same " (which screws up some algorithm).    OK?    ------------------------------  % Date: Wed, 17 Aug 2005 11:20:43 -0400 $ From: "Hein" <hein.nomail@hp.nomail>Y Subject: Re: How many files can you have in a VMS directory without performance problems? , Message-ID: <430355ca$1@usenet01.boi.hp.com>  ) "RAQ" <_raq_@yahoo.com> wrote in message  = news:1124275575.684487.287580@z14g2000cwz.googlegroups.com... A > Everytime I ask HP about  "How many files can you have in a VMS * > directory without performance problems?"F > they send me an article that is about 10-15 years old for VMS 5.2 orJ > even 1.0, that describes that more than 300 files in a catalog is bad...  ' We could/should do better at this time.   D > I have a feeling that <25.000 files works ok (no major complains),/ > but when we reach 100.000 files it gets bad - + >  and with 500.000 it gets really nasty :)   # So you found the answer: 20,000 :-)   D Chris B & John B>  The piece that hurts is when you delete the last  directory entry in< a directory block and you need to slide the remainder of the$ directory down one block to fill in.  6 Indeed. One of several other discussion on this topic:K http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=625667   0 John B> So now you bubble the directory contents/ John B> NOTE THAT IT'S STILL AN O(n^2) PROBLEM. #   :W9QRXAG.DAT => [.W.9]W9QRXAG.DAT   B John B> And now you're down to less than 1000 files per directory.  ; Yes on the N-square, but no the < 1000 and double dir goal. I As per RAQs question and implied answer, < 5000 or even  < 10,000 may be   enough. + And I might opt for [.w9] instead of [.w.9]  'It all depends'.   ' Colin B> You might need to discover DFU   ; Absolutely. Best thing snince sliced bread... in ODS space.   C Jeff C> This issue can be shown by my command procedure REV_DEL.COM C Jeff C> (http://support.mti.com/VMS2005cd/supportfiles/rev_del.txt)  : K Norm L> Isn't there a penalty paid for activating the DELETE.EXE image for   every   . Yes. I believe that to be a small extra issue.$ You may want to try my script below.3 (It is from the days before I switched to perl :-). L It just feeds delete a number of files in reverse order over and over again.J You may want to tweak the command line length (there is a test for 200 in  there). L The reverse order will prevent the shuffling when a directory block empties.A The multiple files somewhat reduce the delete image activiations.    fwiw,  Hein.    $if p1.eqs."" then goto help- $directory/out=sys$Login:delete.tmp/col=1 'p1 H $sort /key=(pos:1,siz:-1,desc) sys$Login:delete.tmp sys$Login:delete.tmp5 $open/read/error=clean_up sorted sys$Login:delete.tmp  $init: $ read/end=clean_up sorted f $ l2 = f$len(f) & $ if f$loc(";",f).eq.l2 then goto init $ files = f  $loop: $ read/end=done sorted f $ l1 = f$len(f) & $ if f$loc(";",f).eq.l1 then goto loop $ if l2.lt.200 $  then  $   files = files + "," + f  $   l2 = l2 + l1 + 1 $  else  $   delete 'p2' 'files' 
 $   files = f  $   l2 = l1  $ endif  $ goto loop  $done: $ delete 'p2' 'files'  $ clean_up:  $ close/nolog sorted  $ delete sys$Login:delete.tmp;,; $ exit $help: $ type sys$input   Usage:  K P1 (required) = wild card file spec, with optional DIR selection qualifiers < P2 (optional) = delete command qualifiers (/LOG or /CONFIRM)   $exit    ------------------------------  % Date: Wed, 17 Aug 2005 09:52:51 -0400  From: norm.raphael@metso.comY Subject: Re: How many files can you have in a VMS directory without performance problems? Q Message-ID: <OF92188C2B.57900971-ON85257060.004C1540-85257060.004C3FEF@metso.com>   C Jeff Cameron <roktsci@comcast.net> wrote on 08/17/2005 09:10:04 AM:     > On 8/17/05 3:46 AM, in article> > 1124275575.684487.287580@z14g2000cwz.googlegroups.com, "RAQ" > <_raq_@yahoo.com> wrote: >  > > Hi > >  > > Everytime I ask HP aboutG > > "How many files can you have in a VMS directory without performance  > > problems?" > > H > > they send me an article that is about 10-15 years old for VMS 5.2 orE > > even 1.0, that describes that more than 300 files in a catalog is 
 > > bad... > > J > > I have a feeling that <25.000 files works ok (no major complains), butF > > when we reach 100.000 files it gets bad - and with 500.000 it gets > > really nasty :)  > >  > > C > > Does anyone know the real numbers? Or how I can calculate them? ? > > Its for VMS 7.x + EVA SAN, and is mainly for file creation.  > >  > >  > > 
 > > Thanks > > @ > There are basically three things that can cause directory size performance 	 > issues. $ > 1. SYSGTEN parameter ACP_DIRCACHE. > 2. The speed of your disk.! > 3. The speed of your processor. F > The reason is that VMS keeps the directory entries sorted, and quite often I > the adding and deleting of entries can cause the entire .DIR file to be / > flushed from cache and completely re-written.  > F > As mentioned by Chris Blackburn this is particularly noticeable whenG > deleting files. When you delete files from a directory, it is done in K > alphabetical order, or from the top of the list. Depending on how much of C > the .DIR file is cached (ACP_DIRCACHE) when the cache needs to be  flushed,I > this causes the entire .DIR file to be rewritten to disk. Since entries  wereH > removed from the top, and the directory entries are all shifted up and eachF > block of the directory file has now changed because the entries were shifted  > up.  > = > This issue can be shown by my command procedure REV_DEL.COM = > (http://support.mti.com/VMS2005cd/supportfiles/rev_del.txt)   H Isn't there a penalty paid for activating the DELETE.EXE image for every file, 3 instead of handing it a list, or a wildcarded list?   J (Oh and you spelled deleting as deleteing in one of your comment lines :-| . J > which deletes files in the reverse order, that is from the bottom of the dir I > file. That has the effect of not having to rewrite the entire .DIR file  withE > each cache flush, because the only portion of the DIR file that has  changed H > is the end. So if you are deleting files (even a portion of the files) fromH > the bottom or reverse order will take much less time because the upper4 > blocks will not be changing with each cache flush. > J > If you have a large .DIR file and you delete files in the normal fashionK > with the /LOG qualifier, you will see a hand-full of files being deleted, H > then a pause, then another hand-full, and another pause, over and over again H > with each pause being when the .DIR file is being flushed back to diskF > completely re-written. As the .DIR file gets smaller and smaller the flush J > operation becomes shorter and shorter. If you are using MONITOR/FILE you can J > see your directory cache hit rate toggle back and forth from 100% to lowJ > percentages (again dependent on ACP_DIRCACHE settings). However deleting inI > the reverse order, you will not see these pauses, because the .DIR file  is3 > only being truncated rather than being rewritten.  > I > SO back to your question, what is the number? It depends mainly on what  kindH > of actions you are doing and if the changes are additions or deletions and J > where they are being inserted/removed from the .DIR file. If the changes are J > made near the end of the .DIR file then the consequences are low. If theI > changes are at the top, then when a cache flush occurs more blocks must  beK > rewritten to disk. Given this information you can see that increasing any  ofI > the the three items listed here (in this order of most effective first)  can 0 > help you forgo the cost of directory flushing. >  > 1. ACP_DIRCACHE size.  > 2. Disk speed. > 3. Processor speed.  >  > Jeff Cameron >    ------------------------------  % Date: Wed, 17 Aug 2005 06:10:04 -0700 ( From: Jeff Cameron <roktsci@comcast.net>Y Subject: Re: How many files can you have in a VMS directory without performance problems? 0 Message-ID: <BF28853C.12A2C%roktsci@comcast.net>   On 8/17/05 3:46 AM, in article< 1124275575.684487.287580@z14g2000cwz.googlegroups.com, "RAQ" <_raq_@yahoo.com> wrote:   > Hi >  > Everytime I ask HP aboutE > "How many files can you have in a VMS directory without performance  > problems?" > F > they send me an article that is about 10-15 years old for VMS 5.2 orC > even 1.0, that describes that more than 300 files in a catalog is  > bad... > H > I have a feeling that <25.000 files works ok (no major complains), butD > when we reach 100.000 files it gets bad - and with 500.000 it gets > really nasty :)  >  > A > Does anyone know the real numbers? Or how I can calculate them? = > Its for VMS 7.x + EVA SAN, and is mainly for file creation.  >  >  >  > Thanks > J There are basically three things that can cause directory size performance issues. " 1. SYSGTEN parameter ACP_DIRCACHE. 2. The speed of your disk. 3. The speed of your processor. J The reason is that VMS keeps the directory entries sorted, and quite oftenG the adding and deleting of entries can cause the entire .DIR file to be - flushed from cache and completely re-written.   D As mentioned by Chris Blackburn this is particularly noticeable whenE deleting files. When you delete files from a directory, it is done in I alphabetical order, or from the top of the list. Depending on how much of J the .DIR file is cached (ACP_DIRCACHE) when the cache needs to be flushed,L this causes the entire .DIR file to be rewritten to disk. Since entries wereK removed from the top, and the directory entries are all shifted up and each L block of the directory file has now changed because the entries were shifted up.   ; This issue can be shown by my command procedure REV_DEL.COM ; (http://support.mti.com/VMS2005cd/supportfiles/rev_del.txt) L which deletes files in the reverse order, that is from the bottom of the dirL file. That has the effect of not having to rewrite the entire .DIR file withK each cache flush, because the only portion of the DIR file that has changed K is the end. So if you are deleting files (even a portion of the files) from F the bottom or reverse order will take much less time because the upper2 blocks will not be changing with each cache flush.  H If you have a large .DIR file and you delete files in the normal fashionI with the /LOG qualifier, you will see a hand-full of files being deleted, L then a pause, then another hand-full, and another pause, over and over againF with each pause being when the .DIR file is being flushed back to diskJ completely re-written. As the .DIR file gets smaller and smaller the flushL operation becomes shorter and shorter. If you are using MONITOR/FILE you canH see your directory cache hit rate toggle back and forth from 100% to lowK percentages (again dependent on ACP_DIRCACHE settings). However deleting in J the reverse order, you will not see these pauses, because the .DIR file is1 only being truncated rather than being rewritten.   L SO back to your question, what is the number? It depends mainly on what kindJ of actions you are doing and if the changes are additions or deletions andL where they are being inserted/removed from the .DIR file. If the changes areH made near the end of the .DIR file then the consequences are low. If theJ changes are at the top, then when a cache flush occurs more blocks must beL rewritten to disk. Given this information you can see that increasing any ofK the the three items listed here (in this order of most effective first) can . help you forgo the cost of directory flushing.   1. ACP_DIRCACHE size.  2. Disk speed. 3. Processor speed.    Jeff Cameron   ------------------------------  % Date: Wed, 17 Aug 2005 12:37:31 -0500 / From: Chris Scheers <chris@applied-synergy.com> Y Subject: Re: How many files can you have in a VMS directory without performance problems? 2 Message-ID: <430375DB.9040004@applied-synergy.com>  
 RAQ wrote: > Hi >  > Everytime I ask HP aboutE > "How many files can you have in a VMS directory without performance  > problems?" > F > they send me an article that is about 10-15 years old for VMS 5.2 orC > even 1.0, that describes that more than 300 files in a catalog is  > bad... > H > I have a feeling that <25.000 files works ok (no major complains), butD > when we reach 100.000 files it gets bad - and with 500.000 it gets > really nasty :)  >  > A > Does anyone know the real numbers? Or how I can calculate them? = > Its for VMS 7.x + EVA SAN, and is mainly for file creation.   I As others have mentioned, the performance problems are generally related   to directory size.  D However, there is no direct correlation between number of files and E directory size (other than more files results in bigger directories).   F The directory file format includes some compression.  Long file names G that are unique result in the worst utilization of directory space and  ( quickly result in large directory files.  F Shorter file names allow more file information to be stored per block.  F Multiple versions of the same file name have the best utilization and F quite a few can be in the same directory without performance problems.  I If possible, managing the names of the files in the directory can result  & in a large improvement in performance.   --  G ----------------------------------------------------------------------- $ Chris Scheers, Applied Synergy, Inc.  B Voice: 817-237-3360            Internet: chris@applied-synergy.com    Fax: 817-237-3074   ------------------------------  + Date: Wed, 17 Aug 2005 13:52:38 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney)   Subject: Re: Image restore fails( Message-ID: <ddvff6$6h9$1@pcls4.std.com>  4 David J Dachtera <djesys.nospam@comcast.net> writes:  I >I did it once, long ago. I have neither the resources nor the time to do E >it again just now (I recall this taking most of a week, including an H >abortive attempt to enlist the CSC's assistance.) You might try lookingG >up the ODS-2 documentation and the INITIALIZE documentation. I seem to H >recall (vaguely) one or both of them explaining what files are expectedE >to be where in INDEXF.SYS. I don't recall just now what piece of the G >puzzle indicated that some obscure code expected to find VMS$COMMON by F >FID (15,1,x). Perhaps some older SYSINIT documentation - don't recallH >just now. It might have been that after the basic files are provided byH >($)INITIALIZE, the first available file header ends up being number 15. >Dunno.   D ODS-2 has reserved files with FIDs (1,1,0) through (10,10,0).  TheseF are what you get if you $INITIALIZE a disk and do nothing else.  OtherF FIDs are up for grabs and filled in order, and it just so happens thatG with the usual VMS installation procedure VMS$COMMON.DIR gets (15,1,0). E I see on my system SYS0.DIR has (11,1,0), (12,1,0) is SYSEXE.DIR, but 9 (13,32,0) is unrelated to VMS, it's a Mozilla cache file.   H Old versions of VMS will get you only through (9,9,0) (no SECURITY.SYS).E And Itanium system disks the reserved files go through (11,11,0) with G GPT.SYS.  They can also be created on Alphas with $INIT/GPT (8.2 only).   A I'm sure that with GPT.SYS, Itanium doesn't have (15,1,0) for its I VMS$COMMON.DIR, I'll guess it'll be (16,1,0) if nothing else is different D with the early installation procedure to change the order of things.2 But I don't have an Itanium VMS system to look at.  C The only other thing special about (Alpha) system disks is tbe boot C block, LBN 0, which contains the LBN of the start of the boot image 5 as well as a block count of it.  For VMS this is file A SYS$COMMON:[SYSEXE]APB.EXE.  Nothing special about its FID, on my C system it's (21224,2,0), but VMS has been upgraded several times on C it.  It probably has a low FID on a system disk that has never been ! upgraded since VMS was installed.   E The reserved files are special in one way: $ DELETE of them will give G you "insufficient privilege" even if you have all privileges turned on.    ------------------------------  % Date: Wed, 17 Aug 2005 14:05:36 +0300 7 From: "Guy Peleg" <guy.peleg@remove_this_header@hp.com> ) Subject: Re: Investigating ACCVIO problem , Message-ID: <43031a03$1@usenet01.boi.hp.com>  4 "Dave Froble" <davef@tsoft-inc.com> wrote in message* news:11g4guok0ngg716@corp.supernews.com... > dooleys@snowy.net.au wrote: 9 > > was the program previously installed with privileges?  > > is it since the reboot?  > > Phil > >  > I > Exactly.  It's an access violation, not a quota issue.  It doesn't have - > the privilege to do what it's trying to do.  > ' > Possibly it was installed with privs.  > J > You indicate that it's run from SYSTEM.  Have you changed the authorizedG > and/or default privs for that user account?  Normally, SYSTEM has all  privs. >  > --  6 > David Froble                       Tel: 724-529-04506 > Dave Froble Enterprises, Inc.      Fax: 724-529-0596@ > DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com > 170 Grimplin Road  > Vanderbilt, PA  15486   I This can still be a quota issue....For example if a call to LIB$GET_VM or G SYS$EXPREG to allocate more virtual memory fail , and the return status L is not checked and you just blindly relying that the variable containing the address L of the buffer is valid...and if you initialized it in advance, you may be in a situation 6 of accessing address 0 which will result in an ACCVIO.   ------------------------------  % Date: Wed, 17 Aug 2005 15:33:55 +0200 & From: "Bernhard Fabricius" <bf@dmu.dk>% Subject: Reading VMS SAS files on PC? < Message-ID: <43033cc3$0$84849$edfadb0f@dread11.news.tele.dk>  	 Dear All,   G A long while ago our management decided to ditch SAS on VMS (as did SAS L Institute a little later :-) and our setinits have long expired. Due warningL was given. And extended by a year. Now SAS is GONE from VMS here, though SAS9 on Windows is much used especially against our Rdb bases.   G Sure enough - some chap turns up with 10 MO disks (yikes) with over 400 K *.SASEB$DATA files that he insists on being able to read. I find an ancient L MO drive, I find the right combination of 10 undocumented DIP switches and I+ actually manage to read the stuff onto VMS.   E But how do I read the files with SAS (8.x or 9.x) on Windows PCs? SAS I Institute's on-line help is full of how to EXPORT from VMS. No go. I only L have the .SASEB$DATA files and need to import/convert under Windows. JudgingH from the top of the files, we're talking SAS 6.07 on VAX-VMS 5.5-2, so IF suppose all sorts of floating point represensation problems will arise   Any and all help appreciated!    Cheers   Bernhard bf at dmu dot dk   ------------------------------  % Date: Wed, 17 Aug 2005 10:03:27 -0400 ' From: "Main, Kerry" <Kerry.Main@hp.com> ) Subject: RE: Reading VMS SAS files on PC? R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB6B1ED7@tayexc19.americas.cpqcorp.net>   > -----Original Message-----0 > From: Bernhard Fabricius [mailto:bf@dmu.dk]=20 > Sent: August 17, 2005 9:34 AM  > To: Info-VAX@Mvb.Saic.Com ' > Subject: Reading VMS SAS files on PC?  >=20 > Dear All,  >=20@ > A long while ago our management decided to ditch SAS on VMS=20
 > (as did SAS < > Institute a little later :-) and our setinits have long=20 > expired. Due warning@ > was given. And extended by a year. Now SAS is GONE from VMS=20 > here, though SAS; > on Windows is much used especially against our Rdb bases.  >=20> > Sure enough - some chap turns up with 10 MO disks (yikes)=20 > with over 400 @ > *.SASEB$DATA files that he insists on being able to read. I=20 > find an ancient B > MO drive, I find the right combination of 10 undocumented DIP=20 > switches and I- > actually manage to read the stuff onto VMS.  >=20G > But how do I read the files with SAS (8.x or 9.x) on Windows PCs? SAS @ > Institute's on-line help is full of how to EXPORT from VMS.=20 > No go. I only @ > have the .SASEB$DATA files and need to import/convert under=20 > Windows. JudgingA > from the top of the files, we're talking SAS 6.07 on VAX-VMS=20 
 > 5.5-2, so I H > suppose all sorts of floating point represensation problems will arise >=20 > Any and all help appreciated!  >=20 > Cheers >=20
 > Bernhard > bf at dmu dot dk >=20     Bernard,  5 What makes you think SAS is not available on OpenVMS?   
 Reference:E http://support.sas.com/documentation/installcenter/openvms/index.html  (Alpha) < http://support.sas.com/admin/openvms/vax/admindoc.html (VAX)  H We had recent presentation from SAS on OpenVMS and it was very positive.  F If accessing Rdb files, I would think accessing them directly from theB appropriate RDB system would be better from an overall performanceE perspective than over the network (and a whole lot more reliable than 4 Windows given the virus/trojan/security patches etc)  E The doc's in the links seem to be pretty good in terms of integration E material, but I am not sure if they address your specific question or  not.   Regards   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------  % Date: Wed, 17 Aug 2005 11:04:43 -0400 - From: William Webb <william.w.webb@gmail.com> ) Subject: Re: Reading VMS SAS files on PC? 7 Message-ID: <8660a3a105081708044630a18a@mail.gmail.com>   1 On 8/17/05, Bernhard Fabricius <bf@dmu.dk> wrote:  > Dear All,  >=20I > A long while ago our management decided to ditch SAS on VMS (as did SAS L > Institute a little later :-) and our setinits have long expired. Due warn= ing L > was given. And extended by a year. Now SAS is GONE from VMS here, though = SAS ; > on Windows is much used especially against our Rdb bases.  >=20I > Sure enough - some chap turns up with 10 MO disks (yikes) with over 400 L > *.SASEB$DATA files that he insists on being able to read. I find an ancie= ntL > MO drive, I find the right combination of 10 undocumented DIP switches an= d I - > actually manage to read the stuff onto VMS.  >=20G > But how do I read the files with SAS (8.x or 9.x) on Windows PCs? SAS K > Institute's on-line help is full of how to EXPORT from VMS. No go. I only L > have the .SASEB$DATA files and need to import/convert under Windows. Judg= ing J > from the top of the files, we're talking SAS 6.07 on VAX-VMS 5.5-2, so IH > suppose all sorts of floating point represensation problems will arise >=20 > Any and all help appreciated!  >=20 > Cheers >=20
 > Bernhard > bf at dmu dot dk >=20 >=20 >=20# DISCLAIMER:  I AM NOT A SAS EXPERT.   A I dealt with a similar situation involving several unlabeled ANSI - tapes of SAS data that was dropped in my lap.    Two things went in my favor-  C 1) the files were in export format (even though we were told by SAS . that they weren't, based on the file headers!)  E 2) we still had SAS on the AlphaServer I connected the tape drive to.   & We ended up dumping the tapes to disk.  F I believe they just FTPd them to the UNIX system and managed to import* them into SAS on that system successfully.  C I then cut them several copies of a CD which contained all the data F from the tapes as dumped to a VMS disk, along with the notes on how weF did what we did and the SAS procedure that someone dug up that did the trick.  A I did not keep a copy as customer confidential data was involved.   F I can tell you with a reasonable degree of certainty that if the files? were in export format, you could probably manage to import them  without too many problems.  L However, since they're not in export format, I suspect that a combination o= f=20  6 1) the differences between ODS-2 and FAT32 or NTFS and! 2) the version differences of SAS   * will make this a difficult project indeed.  F There is a utility for Windows that will allow reads of ODS-2 volumes, if that helps.  ' I trust you've talked with SAS support?    WWWebb   --=20 C NOTE: This email address is only used for noncommerical VMS-related  correspondence. C All unsolicited commercial email will be deemed to be a request for 8 services pursuant to the terms and conditions located at# http://bellsouthpwp.net/w/e/webbww/    ------------------------------  # Date: Wed, 17 Aug 2005 10:48:01 GMT " From:   VAXman-  @SendSpamHere.ORG) Subject: Re: Replacement for VT terminals 0 Message-ID: <00A48680.12FFDD30@SendSpamHere.ORG>  g In article <8660a3a105081610532b4a900f@mail.gmail.com>, William Webb <william.w.webb@gmail.com> writes: 8 >On 16 Aug 2005 10:37:55 -0700, tomarsin2015@comcast.net" ><tomarsin2015@comcast.net> wrote: >> HelloI >> I am trying to replace all the VTs connected to the console port on my = >> VAXes. I was thinking of buying a cheap pc and then adding F >> some more serial ports and using Powerterm/Hyperterm. I recall thatH >> a long time ago someone connected the console port to a DECsever. Any  >> thoughts/input would be nice.
 >> phillip >>=20  >>=20  > L >Why would you go for greater complexity and less-than-perfect VT emulation?  I Not only less-than-perfect emulation!  There were many VT "clones" on the J market that I have come across that would cause a system to "break" to theJ console when the terminal was turned off or other features were exercised.J One such "feature" that comes to mind was on a VT clone with multiple con-K nections (like on a VT525) that could be switched using a keyboard function J key much like the VT525.  However, doing so would cause a framing error orK similar noise which, if connected to a console, would appear like a "break" / and drop the connection to the "dead sergeant".   I If using a PeeCee emulation, I'd check out the reaction of the console to E the PeeCee crashing before putting such a "solution" into production.  --  K VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM              5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  % Date: Wed, 17 Aug 2005 08:41:32 -0400 ' From: "Main, Kerry" <Kerry.Main@hp.com> ) Subject: RE: Replacement for VT terminals R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB6B1ECC@tayexc19.americas.cpqcorp.net>   > -----Original Message-----7 > From: Ken Robinson [mailto:sendspamhere@rbnsn.com]=20   > Sent: August 16, 2005 11:51 PM > To: Info-VAX@Mvb.Saic.Com + > Subject: Re: Replacement for VT terminals  >=20- > Wayne Sewell <wayne@tachysoft.com> wrote in 7 > news:00A48635.D81DE825.11@tachysoft.com (in part):=20  >=202 > >>I figured that the DECservers were understood. > >>, > >>I promise to be more explicit next time. > >> > >=20 > >=20A > > Sorry, I'm not familiar with these console consolidators. =20  > Thought itA > > might have some sort of hardware component that brought in=20  > the serial
 > > lines.=20  > >=20G > > The only console consolidator I ever looked at was the original one A > > from the mid-to-late eighties.  Don't remember what it was=20 
 > called, but < > > it *did* have a discrete serial line to every console, aI > > hardware/software combo.  Did it deliberately, using fiber optics.=20 D > > IIRC, electrical isolation was the reason for doing it that way. >=20J > If you're talking about the original one from DEC, it was called VCS.=20J > When I was at Bellcore in the mid '80s we went to see on of the first=20A > prototypes. It was running on a VAX 780. When we implemented=20  > it, we ran=20 F > it on an MicroVAX II running the windowing system before DECwindows. >=20? > VCS turned into PCM (Polycenter Console Manager) which got=20  > sold to CA.=20J > Another company (TDI <http://www.tditx.com/>) came out with a console=20A > manager called ConsoleWorks. I used it when I was at Verizon=20  > Wireless.=20? > The consoles all go into a DECserver and it gets the OPA0:=20  > output from=20? > each console. You can use either a web based interface (it=20  > has it's own=20 9 > webserver) or a command line interface to access the=20  > different consoles.  >=20A > I liked to use any of the console managers, since they allow=20  > you (if you=20= > have the permissions) to do anything you could do from a=20  > local console. >=20 > Ken Robinson > kenrbnsn (at) rbnsn (dot) com  >=20  @ While likely overkill for home use, additional benefits for workG environment of commercial console managers like CA and ConsoleWorks:=20   C - maintain log file of all activity on the console - including that I which happened when the system was down, rebooting error messages etc.=20 D - great for recording console parameter set-ups and things like WWID commands set before booting.A - can handle not only systems, but also disk controllers, network G devices (routers etc) - anything that has a serial interface. Great for  HSC/J/G CI/SAN controllers. H - console log file can be backed up for future reference and/or securityC reasons. Can search file with editor to look for specific activity. H - can setup filters which look for specific keywords in console messagesC and then send system manager alert via email, page etc. e.g. "fatal A bugcheck". These pkgs typically come pre-loaded with many console # messages filter already pre-loaded. E - Both run natively on VMS as well, so you do not need to worry about B the virus and/or security patch of the day issues which would be aG concern for something as critical as a system with direct access to all 	 consoles. C - typically support multi-site config's as well i.e. logging to two F separate systems in different sites. Hence, messages logged on systemsH are not lost in disaster scenario - something that can be critical in DR6 scenario's. Think auditing with Sarbanes Oxley etc ...   Regards   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------  % Date: Wed, 17 Aug 2005 09:00:53 -0400 ' From: "Main, Kerry" <Kerry.Main@hp.com> ) Subject: RE: Replacement for VT terminals R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB6B1ECD@tayexc19.americas.cpqcorp.net>   > -----Original Message----- > From: Main, Kerry=20 > Sent: August 17, 2005 8:42 AM  > To: Info-VAX@Mvb.Saic.Com + > Subject: RE: Replacement for VT terminals  >=20 >=20 > > -----Original Message-----9 > > From: Ken Robinson [mailto:sendspamhere@rbnsn.com]=20 " > > Sent: August 16, 2005 11:51 PM > > To: Info-VAX@Mvb.Saic.Com - > > Subject: Re: Replacement for VT terminals  > >=20/ > > Wayne Sewell <wayne@tachysoft.com> wrote in 9 > > news:00A48635.D81DE825.11@tachysoft.com (in part):=20  > >=204 > > >>I figured that the DECservers were understood. > > >>. > > >>I promise to be more explicit next time. > > >> > > >=20 > > >=20C > > > Sorry, I'm not familiar with these console consolidators. =20  > > Thought itC > > > might have some sort of hardware component that brought in=20  > > the serial > > > lines.=20  > > >=20? > > > The only console consolidator I ever looked at was the=20  > original oneC > > > from the mid-to-late eighties.  Don't remember what it was=20  > > called, but > > > > it *did* have a discrete serial line to every console, a= > > > hardware/software combo.  Did it deliberately, using=20  > fiber optics.=20F > > > IIRC, electrical isolation was the reason for doing it that way. > >=20@ > > If you're talking about the original one from DEC, it was=20 > called VCS.=20B > > When I was at Bellcore in the mid '80s we went to see on of=20 > the first=20C > > prototypes. It was running on a VAX 780. When we implemented=20  > > it, we ran=20 H > > it on an MicroVAX II running the windowing system before DECwindows. > >=20A > > VCS turned into PCM (Polycenter Console Manager) which got=20  > > sold to CA.=20B > > Another company (TDI <http://www.tditx.com/>) came out with=20 > a console=20C > > manager called ConsoleWorks. I used it when I was at Verizon=20  > > Wireless.=20A > > The consoles all go into a DECserver and it gets the OPA0:=20  > > output from=20A > > each console. You can use either a web based interface (it=20  > > has it's own=20 ; > > webserver) or a command line interface to access the=20  > > different consoles.  > >=20C > > I liked to use any of the console managers, since they allow=20  > > you (if you=20? > > have the permissions) to do anything you could do from a=20  > > local console. > >=20 > > Ken Robinson! > > kenrbnsn (at) rbnsn (dot) com  > >=20 >=20@ > While likely overkill for home use, additional benefits for=20@ > work environment of commercial console managers like CA and=20 > ConsoleWorks:=20 >=209 > - maintain log file of all activity on the console -=20 < > including that which happened when the system was down,=20" > rebooting error messages etc.=20? > - great for recording console parameter set-ups and things=20 ( > like WWID commands set before booting.> > - can handle not only systems, but also disk controllers,=20? > network devices (routers etc) - anything that has a serial=20 2 > interface. Great for HSC/J/G CI/SAN controllers.= > - console log file can be backed up for future reference=20 A > and/or security reasons. Can search file with editor to look=20  > for specific activity.< > - can setup filters which look for specific keywords in=20< > console messages and then send system manager alert via=20A > email, page etc. e.g. "fatal bugcheck". These pkgs typically=20 G > come pre-loaded with many console messages filter already pre-loaded. > > - Both run natively on VMS as well, so you do not need to=20B > worry about the virus and/or security patch of the day issues=20< > which would be a concern for something as critical as a=20, > system with direct access to all consoles.A > - typically support multi-site config's as well i.e. logging=20 @ > to two separate systems in different sites. Hence, messages=20: > logged on systems are not lost in disaster scenario -=20; > something that can be critical in DR scenario's. Think=20 & > auditing with Sarbanes Oxley etc ... >=20	 > Regards  >=20 > Kerry Main > Senior Consultant  > HP Services Canada > Voice: 613-592-4660  > Fax: 613-591-4477  > kerryDOTmainAThpDOTcom > (remove the DOT's and AT)=20 >=206 > OpenVMS - the secure, multi-site OS that just works. >=20  < Oops - meant to include some url's for additional reference:  
 ConsoleWorks:  http://www.tditx.com/tds.html A http://www.tditx.com/pdf/pres/cwsolution.pps (diagrams for sample  implementation)   . Unicenter Console Manager for OpenVMS from CA:2 http://www3.ca.com/Solutions/Product.asp?ID=3D1210   Regards   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------    Date: 17 Aug 2005 06:33:16 -0700$ From: "Ed Wilts" <ewilts@ewilts.org>) Subject: Re: Replacement for VT terminals B Message-ID: <1124285596.691563.66590@g47g2000cwa.googlegroups.com>  0 > Unicenter Console Manager for OpenVMS from CA:3 > http://www3.ca.com/Solutions/P roduct.asp?ID=1210   E I've been running this since it was the Vaxcluster Console System and F later became the Polycenter Console Manager before DEC in a display ofF outmost stupidity sold the product set to CA.  CA was happily chargingF maintenance fees without doing a single change to the product.  One of@ the extremely useful features of the PCM product is that you canG generate alerts based on the text messages coming through your console. C  So, say you watch a cluster transition coming through, or a shadow E member dropping out because of disk errors or an HSJ crashing - you'd > want to be notified.  CA, however, never updated their messageE definition files to reflect new releases of the software so the value  degraded over time.   G PCM is still a really nice product and I still run it today to manage a E multi-site cluster.  It was managing my HSJs and Gigaswitches until I F retired them a couple of weeks ago and now it's only got Alphas on it.B I'm thinking of using it now to connect to the back end of the EVAD controllers so the remote guys don't need to call in a field serviceE rep to dump out diagnostic information.  I'll probably connect all my A McData switches too since if they lose their config or need to be / swapped, the console port is all you have left.   C Just remember to put PCM on a non-cluster host since if the cluster B hangs, you really do want to be notified and to have access to the	 consoles.   	    .../Ed    ------------------------------    Date: 17 Aug 2005 06:41:57 -0700$ From: "Ed Wilts" <ewilts@ewilts.org>) Subject: Re: Replacement for VT terminals B Message-ID: <1124286117.760800.56450@o13g2000cwo.googlegroups.com>  G Another commercial product that I've heard lots of good things about is = RoboCentral from Heroix (who also make really good enterprise  monitoring tools).  ? http://www.robomon.com/products/products_detail_robocentral.htm   	    .../Ed   G Disclaimer:  I run Robomon (an older version) but don't run RoboCentral  - we're still a PCM shop.    ------------------------------  % Date: Wed, 17 Aug 2005 09:57:35 -0400 2 From: "Stanley F. Quayle" <squayle@insight.rr.com>) Subject: Re: Replacement for VT terminals / Message-ID: <43030A0F.15934.12D7544E@localhost>   H > There were many VT "clones" on the market that I have come across thatF > would cause a system to "break" to the console when the terminal was/ > turned off or other features were exercised.    $ Funny, my VT220 does the same thing.  D When connected to a terminal server (DECserver), a SET PORT x BREAK 7 DISABLED prevents a break signal from going to the VAX.   ) Unless you power off the terminal server.   @ In my opinion, having the system "break" when a Break signal is B received was not a Good Idea, especially in server usage.  Ctrl-P < (Alpha, some VAX) or Ctrl-B (Itanium) is a much better idea.  
 --Stan Quayle  Quayle Consulting Inc.  
 ----------- Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363 3 8572 North Spring Ct., Pickerington, OH  43147  USA 0 stan-at-stanq-dot-com       http://www.stanq.com) "OpenVMS, when downtime is not an option"    ------------------------------    Date: 17 Aug 2005 09:10:05 -0700# From: "WhoDat?" <whohe@whoever.com>  Subject: Re: SAMBA for VMSC Message-ID: <1124295005.518903.149560@g14g2000cwa.googlegroups.com>    Bob Koehler wrote:e > In article <42FDE0E9.11775.59A6E1E@localhost>, "Stanley F. Quayle" <squayle@insight.rr.com> writes:  > > E > > Shame you can't wait for OpenVMS Engineering to develop their own 9 > > Samba implementation.  Should be ready to go Q2 2006.  > D >    Your a couple decades too late.  And you've got the cart before >    the horse.  > C >    The original author of Samba based it on packet sniffing while  >    running Pathworks.   . You must have missed the news. The horse died.   ------------------------------   Date: 17 Aug 2005 12:20:08 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)0 Subject: Re: SAMBA for VMS (The saga continues!), Message-ID: <3mgoboF16u5blU1@individual.net>  ? Well, I am about out of time for this.  It is not looking good.    After running BUILD.COM I get:  6 %LINK-W-USEUNDEF, undefined symbol SNPRINTF referenced(         in psect $CODE offset %X000006A1M         in module TALLOC file USERS:[HOME.SYSTEM.SAMBA.SOURCE.BIN]SAMBA.OLB;1 7 %LINK-W-USEUNDEF, undefined symbol VSNPRINTF referenced (         in psect $CODE offset %X00000026O         in module SNPRINTF file USERS:[HOME.SYSTEM.SAMBA.SOURCE.BIN]SAMBA.OLB;1 7 %LINK-W-USEUNDEF, undefined symbol VSNPRINTF referenced (         in psect $CODE offset %X00000063O         in module SNPRINTF file USERS:[HOME.SYSTEM.SAMBA.SOURCE.BIN]SAMBA.OLB;1 7 %LINK-W-USEUNDEF, undefined symbol VSNPRINTF referenced (         in psect $CODE offset %X0000009CO         in module SNPRINTF file USERS:[HOME.SYSTEM.SAMBA.SOURCE.BIN]SAMBA.OLB;1 7 %LINK-W-USEUNDEF, undefined symbol VSNPRINTF referenced (         in psect $CODE offset %X000000D1O         in module SNPRINTF file USERS:[HOME.SYSTEM.SAMBA.SOURCE.BIN]SAMBA.OLB;1 6 %LINK-W-USEUNDEF, undefined symbol SNPRINTF referenced(         in psect $CODE offset %X00000344M         in module DOSERR file USERS:[HOME.SYSTEM.SAMBA.SOURCE.BIN]SAMBA.OLB;1   K I can understand the VSNPRINTF references as I can't find that one anywhere L in the sources, but SNPRINTF is in SAMBA.OLB as verified by running "LIBRARY /LIST SAMBA.OLB".   H So, one last time.  Is there anywhere I can find a version of SAMBA thatD will actually build and run on a VAX or am I really wasting my time?   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------   Date: 17 Aug 2005 14:24:39 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)0 Subject: Re: SAMBA for VMS (The saga continues!), Message-ID: <3mgvl7F16f0gvU1@individual.net>  3 In article <fPHMe.10496$ng2.3581@news.cpqcorp.net>, : 	John Malmberg <malmberg@dskwld.zko.dec.compaq.hp> writes: > Bill Gunshannon wrote:B >> Well, I am about out of time for this.  It is not looking good. >>  ! >> After running BUILD.COM I get:  >>  9 >> %LINK-W-USEUNDEF, undefined symbol SNPRINTF referenced + >>         in psect $CODE offset %X000006A1 P >>         in module TALLOC file USERS:[HOME.SYSTEM.SAMBA.SOURCE.BIN]SAMBA.OLB;1 >>  N >> I can understand the VSNPRINTF references as I can't find that one anywhereO >> in the sources, but SNPRINTF is in SAMBA.OLB as verified by running "LIBRARY  >> /LIST SAMBA.OLB". >>  K >> So, one last time.  Is there anywhere I can find a version of SAMBA that G >> will actually build and run on a VAX or am I really wasting my time?  > J > There are people on the SAMBA-VMS mailing list that claim to be running K > this version of SAMBA on VAX.  And the SAMBA-VMS mailing list appears to  H > be the forum that the person maintaining that distribution is reading.+ > They are clearly not reading comp.os.vms.  > D > The missing routines are a result of something being wrong in the J > config.h file and a file being missing from the build that contains the  > replacement routines.   8 OK, I went through config.h and commented out the lines: /* #define HAVE_SNPRINTF     */ ! /* #define HAVE_C99_SNPRINTF   */    and     /* #define HAVE_VSNPRINTF     */" /* #define HAVE_C99_VSNPRINTF   */  I Buld one more time and see if that finally fixes it.  If I make this work H is there anyone who would actually care about the problems I have had to fix?   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  # Date: Wed, 17 Aug 2005 14:27:23 GMT 7 From: John Malmberg <malmberg@dskwld.zko.dec.compaq.hp> 0 Subject: Re: SAMBA for VMS (The saga continues!)3 Message-ID: <fPHMe.10496$ng2.3581@news.cpqcorp.net>    Bill Gunshannon wrote:A > Well, I am about out of time for this.  It is not looking good.  >   > After running BUILD.COM I get: > 8 > %LINK-W-USEUNDEF, undefined symbol SNPRINTF referenced* >         in psect $CODE offset %X000006A1O >         in module TALLOC file USERS:[HOME.SYSTEM.SAMBA.SOURCE.BIN]SAMBA.OLB;1  > M > I can understand the VSNPRINTF references as I can't find that one anywhere N > in the sources, but SNPRINTF is in SAMBA.OLB as verified by running "LIBRARY > /LIST SAMBA.OLB".  > J > So, one last time.  Is there anywhere I can find a version of SAMBA thatF > will actually build and run on a VAX or am I really wasting my time?  H There are people on the SAMBA-VMS mailing list that claim to be running I this version of SAMBA on VAX.  And the SAMBA-VMS mailing list appears to  F be the forum that the person maintaining that distribution is reading.) They are clearly not reading comp.os.vms.   B The missing routines are a result of something being wrong in the H config.h file and a file being missing from the build that contains the  replacement routines.   I This bug is showing up primarily because those replacement routines will  F cause problems with current versions of OpenVMS Alpha and I64 because E they were coded incorrectly to have the exact same names as routines  K defined by the X/Open group that are now in the 64 bit versions of OpenVMS.   G So instead of fixing their symbol names to be unique, and using macros  J for conditional use, the module was just removed from the build procedure.  I The quick fix is to add the missing file to the build procedure, I think  = it is called SNPRINTF.C and should be present in the sources.   I The correct fix is to make sure that the replacement routines are always  H named SAMBA_SNPRINTF and SAMBA_VSNPRINTF in the SNPRINTF.C routine, and ! then in config.h there should be:    #ifdef __CRTL_VER  #if __CRTL_VER < 70312000  #define snprintf samba_snprintf ! #define vsnprintf samba_vsnprintf  #endif #else  #define snprintf samba_snprintf ! #define vsnprintf samba_vsnprintf  #endif  E Then the same build procedure can be used on VAX/ALPHA and that part  - should still work with older versions of VMS.   F The build procedure only needs to add that object file to the library B for older versions of VMS, but once it is coded properly it can be" present with out causing problems.  I This bug in SNPRINTF.C for using the incorrect symbols has been fixed in  C the UNIX version of SAMBA for quite a while because it also causes  " problems on some versions of UNIX.   -John ! malmberg@dskwld.zko.dec.compaq.hp  Personal Opinion Only    ------------------------------   Date: 17 Aug 2005 14:46:44 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)0 Subject: Re: SAMBA for VMS (The saga continues!), Message-ID: <3mh0ujF16f0gvU2@individual.net>  . In article <xVHMe.5$sq6.2485@news.uswest.net>,9 	"Michael D. Ober" <obermd.@.alum.mit.edu.nospam> writes: L > JYC's page talks about this specific issue and there is a download for it.  G Actually, as I mentioned here yesterday, it doesn't.  The "download" is D an Alpha binary which can be included in SAMBA.OLB for those who areF just linking rather than compiling a version.  And, obviously, it onlyC works on Alpha.  The source for SNPRINTF is in the source tree.  As D pointed out by John Malmberg the problem seems to be in CONFIG.H andC hopefully, I have fixed it.  But considering how easy building this A stuff on Unix is, what it takes on VMS is hardly impressive.  And ? we won't even go into having to recompile everything everytime. % Tell me again how much easier VMS is?    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, 17 Aug 2005 08:34:04 -0600 6 From: "Michael D. Ober" <obermd.@.alum.mit.edu.nospam>0 Subject: Re: SAMBA for VMS (The saga continues!). Message-ID: <xVHMe.5$sq6.2485@news.uswest.net>  J JYC's page talks about this specific issue and there is a download for it.  
 Mike Ober.  D "John Malmberg" <malmberg@dskwld.zko.dec.compaq.hp> wrote in message- news:fPHMe.10496$ng2.3581@news.cpqcorp.net...  > Bill Gunshannon wrote:C > > Well, I am about out of time for this.  It is not looking good.  > > " > > After running BUILD.COM I get: > > : > > %LINK-W-USEUNDEF, undefined symbol SNPRINTF referenced, > >         in psect $CODE offset %X000006A1! > >         in module TALLOC file / USERS:[HOME.SYSTEM.SAMBA.SOURCE.BIN]SAMBA.OLB;1  > > F > > I can understand the VSNPRINTF references as I can't find that one anywhereG > > in the sources, but SNPRINTF is in SAMBA.OLB as verified by running  "LIBRARY > > /LIST SAMBA.OLB".  > > L > > So, one last time.  Is there anywhere I can find a version of SAMBA thatH > > will actually build and run on a VAX or am I really wasting my time? > I > There are people on the SAMBA-VMS mailing list that claim to be running J > this version of SAMBA on VAX.  And the SAMBA-VMS mailing list appears toH > be the forum that the person maintaining that distribution is reading.+ > They are clearly not reading comp.os.vms.  > C > The missing routines are a result of something being wrong in the I > config.h file and a file being missing from the build that contains the  > replacement routines.  > J > This bug is showing up primarily because those replacement routines willG > cause problems with current versions of OpenVMS Alpha and I64 because F > they were coded incorrectly to have the exact same names as routinesD > defined by the X/Open group that are now in the 64 bit versions of OpenVMS. > H > So instead of fixing their symbol names to be unique, and using macrosL > for conditional use, the module was just removed from the build procedure. > J > The quick fix is to add the missing file to the build procedure, I think? > it is called SNPRINTF.C and should be present in the sources.  > J > The correct fix is to make sure that the replacement routines are alwaysI > named SAMBA_SNPRINTF and SAMBA_VSNPRINTF in the SNPRINTF.C routine, and # > then in config.h there should be:  >  > #ifdef __CRTL_VER  > #if __CRTL_VER < 70312000 ! > #define snprintf samba_snprintf # > #define vsnprintf samba_vsnprintf  > #endif > #else ! > #define snprintf samba_snprintf # > #define vsnprintf samba_vsnprintf  > #endif > F > Then the same build procedure can be used on VAX/ALPHA and that part/ > should still work with older versions of VMS.  > G > The build procedure only needs to add that object file to the library D > for older versions of VMS, but once it is coded properly it can be$ > present with out causing problems. > J > This bug in SNPRINTF.C for using the incorrect symbols has been fixed inD > the UNIX version of SAMBA for quite a while because it also causes$ > problems on some versions of UNIX. >  > -John # > malmberg@dskwld.zko.dec.compaq.hp  > Personal Opinion Only  >    ------------------------------  + Date: Wed, 17 Aug 2005 09:47:58 -0500 (CDT) * From: sms@antinode.org (Steven M. Schweda)0 Subject: Re: SAMBA for VMS (The saga continues!)2 Message-ID: <05081709475807_20A0026A@antinode.org>  7 From: John Malmberg <malmberg@dskwld.zko.dec.compaq.hp>   K > The correct fix is to make sure that the replacement routines are always  E > named SAMBA_SNPRINTF and SAMBA_VSNPRINTF in the SNPRINTF.C routine,   E    People (maybe just you) keep saying this, but it's not clear to me > what great benefit goofy symbol names like these really offer.   > and # > then in config.h there should be:  >  > #ifdef __CRTL_VER  > #if __CRTL_VER < 70312000 ! > #define snprintf samba_snprintf # > #define vsnprintf samba_vsnprintf  > #endif > #else ! > #define snprintf samba_snprintf # > #define vsnprintf samba_vsnprintf  > #endif  @    If one needs to add this sort of __CRTL_VER conditionality inH config.h, is it really any more painful to add similar conditionality in4 SNPRINTF.C, which would obviate all the goofy names?  8    For example, in my Wget code, (my) config.h_vms says:  0 /* Define if you have the snprintf function.  */2 #if (!defined( __VAX)) && (__CRTL_VER >= 70312000) #define HAVE_SNPRINTF ; #endif /* (!defined( __VAX)) && (__CRTL_VER >= 70312000) */   E And the snprintf() function itself in Wget's snprintf.c (unchanged by / me) is conditional (more or less) according to:    #if !defined(HAVE_SNPRINTF)   D    This compile-'em-if-you-need-'em scheme seems to me to do the job/ just fine without the need for any funny names.   D    Why does the use of the funny names constitute "The correct fix"?  H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode-org     Saint Paul  MN  55105-2547    ------------------------------  # Date: Wed, 17 Aug 2005 16:08:10 GMT 7 From: John Malmberg <malmberg@dskwld.zko.dec.compaq.hp> 0 Subject: Re: SAMBA for VMS (The saga continues!)3 Message-ID: <KhJMe.10506$Yc2.4922@news.cpqcorp.net>    Steven M. Schweda wrote:9 > From: John Malmberg <malmberg@dskwld.zko.dec.compaq.hp>  > F >    This compile-'em-if-you-need-'em scheme seems to me to do the job1 > just fine without the need for any funny names.  > F >    Why does the use of the funny names constitute "The correct fix"?  B Because it is the only way to prevent conflicts with the compiler D optimizer and with shared images that may be loaded on any platform.  I This is the case on some modern UNIX platforms and especially on OpenVMS  I where compiling with optimization on is the norm, and so is using shared  # images instead of object libraries.   E I have not bothered to keep track of which UNIX systems require this  	 behavior.   G It is only if you are using only object libraries and specific linking  I instructions and also disabling the compiler optimizer that you can rely  F on the original UNIX behavior of overlaying operating system supplied @ routines with ones in your program that are in separate modules.    9 The routine names in X/OPEN Single Unix Specification at  H http://www.unix.org/single_unix_specification/ need to be treated as if F they were reserved names as far as global routine names are concerned.  I Or you may not get the result you thought you were getting even on UNIX,  @ even if you are not getting compile or link errors, if you were G intending on having a routine do something different than the standard   routine with the same name.   G Accessing the information about the routine names from X/Open web site  H requires a free registrations and apparently the acceptance of cookies. D   They do sell copies of the specifications for those that want the   information but not the cookies.  H As OpenVMS improves their compliance with the X/Open UNIX specification,H it will cause more problems with routines that are using these names as  global symbols.   H These improvements may come in ECO kits in addition to OpenVMS upgrades.  F So unless you want your source code or object modules to quit working E with each improvement with the VMS CRTL in UNIX compliance, you will  G want to avoid using routines with the same names as in X/OPEN, even if  , they are not currently available on OpenVMS.  F Executable images will continue to work as the symbol conflicts are a  compile/link time issue.  E And if you want your code to work properly on a wide variety of UNIX  < like systems you will also want to follow the same practice.   -John ! malmberg@dskwld.zko.dec.compaq.hp  Personal Opinion Only    ------------------------------  % Date: Wed, 17 Aug 2005 12:04:36 -0400 ' From: "Main, Kerry" <Kerry.Main@hp.com> 0 Subject: RE: SAMBA for VMS (The saga continues!)R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB6B1EE4@tayexc19.americas.cpqcorp.net>   > -----Original Message-----$ > From: bill@triangle.cs.uofs.edu=20A > [mailto:bill@triangle.cs.uofs.edu] On Behalf Of Bill Gunshannon   > Sent: August 17, 2005 10:47 AM > To: Info-VAX@Mvb.Saic.Com 2 > Subject: Re: SAMBA for VMS (The saga continues!) >=200 > In article <xVHMe.5$sq6.2485@news.uswest.net>,; > 	"Michael D. Ober" <obermd.@.alum.mit.edu.nospam> writes: @ > > JYC's page talks about this specific issue and there is a=20 > download for it. >=20> > Actually, as I mentioned here yesterday, it doesn't.  The=20 > "download" is F > an Alpha binary which can be included in SAMBA.OLB for those who areH > just linking rather than compiling a version.  And, obviously, it onlyE > works on Alpha.  The source for SNPRINTF is in the source tree.  As F > pointed out by John Malmberg the problem seems to be in CONFIG.H andE > hopefully, I have fixed it.  But considering how easy building this C > stuff on Unix is, what it takes on VMS is hardly impressive.  And A > we won't even go into having to recompile everything everytime. ' > Tell me again how much easier VMS is?  >=20	 > bill=20  >=20 > --=20     B Wow, do you mean you can build UNIX app's without all of the right! platform source code being there?    That is really impressive.=20    :-)    Regards   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------   Date: 17 Aug 2005 11:14:14 GMT( From: bill@cs.uofs.edu (Bill Gunshannon), Subject: Re: The Educational License Program, Message-ID: <3mgkg6F16evbrU1@individual.net>  , In article <6vFnLqsJmbj2@malvm9.mala.bc.ca>,4 	nothome@spammers.are.scum (Malcolm Dunnett) writes:/ > In article <3mfh2jF16uk96U1@individual.net>,  / >    bill@cs.uofs.edu (Bill Gunshannon) writes:  >  >>> D >>>   Well I'm not Sue ( or Johnny Cash for that matter ) but I can E >>> tell you the new licenses have been available since the beginning B >>> of August or so. You need to re-register your school and get a1 >>> new ID in order to download the new licenses.  >>>  >>  E >> That doesn't make any sense. Why would you have to re-register and @ >> get a new number?  If the number is a throw away, why bother? >>   > & >    Go ahead, shoot the messenger :-)  A Not shooting the messenger.  Just pointing out one more absurdity @ with the edu program.  As has been said before, the harder (and A more obtuse) it is the less and less likely it is to be utilized. > Especially if it falls to a faculty member somewhere to do it.? Not many schools still have fanatic advocates still in their CS  departments.   > J >    It's my experience that I need to re-register each year, after Aug 1,H > in order to get new licenses that are valid for that year. Perhaps I'mJ > not supposed to have to, but I did so and was able to get licenses valid1 > until Aug 31,2006. I suggest you give it a try.   E I'll probably do that later today just so I can keep the machine from 1 dying just when the students are getting started.    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, 17 Aug 2005 09:15:36 -0500 % From: Neil Cherry <njc@wolfgang.uucp> * Subject: Re: Turning up an Ethernet & tftp. Message-ID: <slrndg6hk9.bfc.njc@wolfgang.uucp>  6 On Tue, 16 Aug 2005 08:25:18 -0500, Neil Cherry wrote:= > On Sat, 13 Aug 2005 23:08:26 -0500, David J Dachtera wrote:  >> Alan Greig wrote: >>>  >>> David J Dachtera wrote:  >>> >  >>> >  >>> > Do you get a link light?  / Both interfaces had the link lights up (green).   N >>> > What does the Alpha console think about them? (See the HELP for the SHOW >>> > commands.) > J >>> Maybe he just needs INIT and MOUNT. Difficult to tell the level of VMS% >>> knowledge from the original post.  >>  C >> The OP's .sig included links indicating he was working with UN*X F >> primarily, is likely accustomed to having to do everything manuallyK >> without much assistance, and his comments suggested he was applying such  >> UN*X principles to VMS. > E > First sorry for the delay, yep I come from the Unix world (actually B > from an embedded world ;-) My knowledge of VMS is low but I haveF > managed to keep my MV3100 up and running with DECNet (I did a lot of& > Googling and I asked for help here).  9 I thinks it's an Alpha 610 and I know it's running 7.3-1. C Unfortunately I've forgetten the password and I have to break in (I . have the notes on this, I do it often enough).   --  C Linux Home Automation         Neil Cherry       ncherry@comcast.net ; http://home.comcast.net/~ncherry/               (Text only) 8 http://hcs.sourceforge.net/                     (HCS II): http://linuxha.blogspot.com/                    My HA Blog   ------------------------------   End of INFO-VAX 2005.458 ************************                                                                                                                        ؕM<7<{`G`,̀E
O
HÑmKdllܔ"&OjŪǁ K$"G?5jT_((zc$p¿6+XIi(L&w\Ž'pw)EQ\NAj>{fg
O]yQOc;zޯwK]FysklolՂvsAomooo֣,7ѨV]2u2	ʺǈ+#w R㶻3[}=kw{؛ Ӕ~7x[h[TWT<U^
 XE6]kAv^lK	]eS^}y^tB-%E_tPctxlno6vtsacWXra+,:erR^ԝ
iE
VmFKzʼ'
uU0l=_7mnfOÚ)7-e#xޯ7
zomz.Qa{
rrrrrrr]
?*]CEv.?TomPѿ Gw/?*ǿU[Cſx]swNyߺ[.^A+sӥCOmFKʼwkپվN-Ꝕj<ר	Tz@-Fzj:[P-4uGĝ	#Vݺ;"Sʼw[Zi5h=NX$k5cO=b]JBZhmvUGRd^಼R=~}KdEдi"/Qz@&+U	Q/=o=Z*i4Tzz1Z}x$
G:U^w]cIbqe-u5q}3%D$ Tm{ᥟoteH\'	b<Pp,FMszEկRdk4T?{KM&-#"rGJU:
Wk፮FĸQODdq*pjO[T!+.J=aOZ7%L_e;syhi捪G^7+D3&G*^}z
`h
Tpj:=8Tl}k
ݤmrso8ۛ+Pm>rloPBd>R'Åi<mhO"qmiOۚW-Xz[SQODWɜO7T?af	Voؿ[O4
뛚gSۦfmzZB]SϦzi-OMkbCUoaz7?|jٴXgZK
ִ]LSۦZm۴&m[ڴ2^ϦF_޶i~dvhӺݵugӺݷu;izaӚvG=Z6izkæ֦=6ymz=I]iյVϦVߞ[MkґyM5mz^YSӚNt~ԥS४Gg9=#Dۨu&^צmt}ӓg}6yF)w:EyF;zwg}wg}wg}wg}F93rg៑?#Flٰf柛-
R昛
cn6l(76Wl\qٲfsݢf抛Mn6m>ٴf惛Mn6m>ٴ6TZZ66m^!܍m4mmmsnݤ67l.as
RRoSbR6.6_$^)Ѷf{`e