1 INFO-VAX	Sat, 26 Aug 2000	Volume 2000 : Issue 476       Contents:& August 2000 edition of the OpenVMS FAQ Re: Big AlphaServer Sale Re: carl lydick 7 Re: Compaq Marketing: Agressively making new "inroads"? $ Re: Convert question - DCPS printing Re: DECNet Phase IV Manuals ?  Re: DECNet Phase IV Manuals ? # Re: Disaster Tolerance - Terrorists + Don't global pages count against WS Limits? / Re: Don't global pages count against WS Limits? 0 Earn at Least $5,000/Week. FREE 32 Page Booklet!- File name saved from EVE (TPU) on ODS5 volume   Re: how to connect to NTP server  Re: how to connect to NTP server Re: LEDs & VAXStation 3100 M48 Re: MEMORY LEAK DETECTION ? 6 Re: Minor Nit In OpenVMS ALPHA Version of SHUTDOWN.COM6 Re: Minor Nit In OpenVMS ALPHA Version of SHUTDOWN.COM Pathworks for VMS $ Re: Portable GUIs (VMS+Windows-NT) ?$ Re: Portable GUIs (VMS+Windows-NT) ?* Problems with Accounts " CAPTIVE " in VMS.. Re: Problems with Accounts " CAPTIVE " in VMS. Re: Renaming SYS$NODE_nodename Re: Renaming SYS$NODE_nodename7 Replace licenses for VAXcluster for Education/Research? ; Re: Replace licenses for VAXcluster for Education/Research? ; Re: Replace licenses for VAXcluster for Education/Research? 
 Re: run *.com 
 Re: run *.com 
 RE: run *.com 
 Re: run *.com 
 Re: run *.com 
 Re: run *.com  Re: UCX routing problem 2 Upgrade path from VMS 6.1 for my "New" uVAX 3100 ?6 Re: Upgrade path from VMS 6.1 for my "New" uVAX 3100 ? Re: VMS 7.2 can't see a cd. Re: Working Set Size Limits and Process Quotas. Re: Working Set Size Limits and Process Quotas  F ----------------------------------------------------------------------  # Date: Fri, 25 Aug 2000 23:40:55 GMT 2 From: "Zane H. Healy" <healyzh@shell1.aracnet.com>/ Subject: August 2000 edition of the OpenVMS FAQ 6 Message-ID: <buDp5.475$M62.182077@typhoon.aracnet.com>  H Normally I just access the copy of the OpenVMS FAQ available via http onE eisner, however it's out of date.  Where can one find the August 2000  edition on the web?   ) In other words where does it really live?   	 		Thanks, 
 		    Zane   ------------------------------    Date: 25 Aug 2000 17:03:07 -0500* From: young_r@eisner.decus.org (Rob Young)! Subject: Re: Big AlphaServer Sale + Message-ID: <UTMh51mVALrI@eisner.decus.org>   R In article <8o61pa$sjq$1@pyrite.mv.net>, "Bill Todd" <billtodd@foo.mv.com> writes:   > G > Andrew Harrison SUNUK Consultancy <andrew.nospam@uk.sun.com> wrote in . > message news:39A59AE0.BF0E7CC7@uk.sun.com...   >>C >> Hang on, in order to provide continuous availability to data and A >> applications in an OpenVMS cluster the application has to have C >> been modified to make use of the cluster facilites. This is myth D >> numero uno the one which says that there is never any requirementF >> to code an application to use the services provided by the cluster.  E 	By the way... "modified" is not correct.  Several large applications = 	were/are written with clusters in mind.  They just so happen % 	to run standalone also (DLM is DLM).    > I > I'm afraid you've only succeeded in demolishing a straw man of your own M > devising:  the post to which you responded did not suggest the above 'myth' B > in any way, but simply provided examples of features which, on aK > feature-by-feature basis, one would prefer to have in the OS to having to M > implement oneself or purchase from some third party who may or may not have  > integrated them competently. >  >>F >> Where do you stop, if the application has to be modified to make it< >> re-startable or what ever else you may require to make it? >> highly available in an OpenVMS cluster then you are allready D >> on the slippery slope to applications providing the functionality: >> of a cluster without requiring cluster services at all. > J > Which is similar to the slippery slope to applications providing all theK > functionality of an OS without requiring an OS at all.  Most applications K > opt not to do so, and prefer, all other considerations being more-or-less J > equal, to use an OS that provides the features they need rather than oneF > which forces them to create them or acquire them from third parties. >  	 ; 	Good zingers Bill.  To touch on the last paragraph .. this 5 	is what Oracle's Raw Iron is all about.  Why use the + 	Sun OS when we can do it all via Raw Iron?   8 	If you follow Andrew closely , he most always clings to> 	Oracle OPS because what he states there is true.  With Oracle8 	OPS you can actually sleep at night with all your Unix ; 	boxes effectively redundant (lose a Unix box, OPS is smart < 	to go elsewhere as your storage failsover also).  Likewise,@ 	with failover Unix clusters and true Unix clusters other angles 	here.  : 	Interestingly , as other Unixes get a true DLM and a trueF 	cluster filesystem (whenever it happens) you could write applicationsF 	to take advantage of the OS (like VMS with DLM, ASTs, RMS JournallingB 	you can build a world-class application) and perhaps the argument? 	then becomes more like ours.  Why steer to Oracle?  Because if A 	you are running certain OSes , you may not have much of a choice @ 	for TRUE application high availability without compromises (1).     				Rob     K (1)  Know someone that is doing a failover cluster using a well known Unix. G      8 CPUs in the primary and the failover box.  Question to him:  "So G 	what do you do with the 8 CPU failover box?  Quite a bit of money just D 	laying there waiting to be used.  Answer:  We'll use it for testingB 	and what not.  My tease in reply:  Must be nice to have that kind@ 	of money for a test platform, why not a single CPU?"  Of course@ 	the argument then goes downhill badly.. point is you either useC 	the resources or you make compromises.  Years ago a large customer G 	of a former employer had a duplicate Unix server ($80000 worth) parked ? 	in a closet in case the primary went down.  Things have gotten A 	better on the Unix side... not nearly has funny but still a good  	laugh every now and then.   ------------------------------  % Date: Sat, 26 Aug 2000 00:28:59 +0200 " From: "Hans Vlems" <hvlems@iae.nl> Subject: Re: carl lydick( Message-ID: <8o6rkh$9um$1@news.IAEhv.nl>  B Carl's contribution to this group has been invaluable to this day.! Many of us still use Carl's code. I As for his social skills, his reputation was a lot worse than he actually D deserved. Some problems required a lot more in terms of mind reading than VMS knowledge. C What I remember of Carl (public and private communications) is that G a carefully prepared question got an extremely accurate answer, usually  phrased in pleasant english.H However, the opposite was also true and besides learning a lot from CarlG about VMS internals and C syntax, my vocabulary in that strange foreign ( language called english also "improved".   Four years, Carl, cheers,   
 Hans Vlems  : barbara trumpinski-roberts heeft geschreven in bericht ...F >i don't know from vax.vms but i just thought i would come in here andH >remind you guys that carl 'speaker-to-minerals' lydick died 4 years ago >yesterday.  > ) >so raise a glass and toast his memory...  >  >hugs, >  >kitten  > 9 >/\ /\   'ah, but you don't have to know everything.  you : >{=.=}   just have to know where to find it.' john brunner7 >  ~     kittent@uiuc.edu             _shockwave rider_ 7 >        http://members.tripod.com/~barbarakitten smotu  >    ------------------------------  % Date: Sat, 26 Aug 2000 01:57:50 +0100 % From: Alan Greig <a.greig@virgin.net> @ Subject: Re: Compaq Marketing: Agressively making new "inroads"?* Message-ID: <39A7160E.87B4D9D6@virgin.net>  % SYSMIKE@alpha1.ocis.uncwil.edu wrote:   G > Returning from vacation last week, I saw the following small roadsign 5 > at mile 173 on I-81 South in the Wilkes-Barre area:  >  >         FUTURE VMS SITE  > H > At first I started wondering what future job opportunities might awaitM > with PA DOT (maybe an alpha-controlled toll road?). Then I started thinking M > about the ultimate fate of many refrigerators here in NC (on the roadside). ' > Sure hope that's not what they meant.  >    Vehicle Management System?   --
 Alan Greig   ------------------------------  % Date: Fri, 25 Aug 2000 21:07:08 +0200 > From: "Jean-Franois Marchal" <jean-francois.marchal@x9000.fr>- Subject: Re: Convert question - DCPS printing 2 Message-ID: <8o6g02$qku$1@s1.read.news.oleane.net>   Thanks ! I'm going to test your program. < I've already done it with DCL, interpreting each vfc control> to produce the correct output record, but it is not very fast!  
 Jean-Franois     2 "Bill Hobbs" <hobbsb@my-deja.com> wrote in message# news:8o63gl$qh5$1@nnrp1.deja.com... 4 > In article <8o08kh$9et$1@s1.read.news.oleane.net>,C >   "Jean-Franois Marchal" <jean-francois.marchal@x9000.fr> wrote: J > > Thanks for the info ... I looked at the RMS Utilities Ref Manual too . > > F > > As this file structure is standard, I thought DCPS should print itF > > properly. It should skip 66 lines, to start printing at the top ofD > > the second sheet, but the first line is printed on line 66 (used* > > DATA_TYPE=LIST to check line numbers). > > I > > I will need to convert the file to "standard text file" with variable J > > format and carriage_return carriage control. A simple convert does notF > > process carriage control information. There is no form feed in the	 converted I > > file, neither does it process "skip n lines" specifications. (the "5"  stands0 > > for 0 0 0 1 1 0 1 0 1 that is skip 53 lines) > > J > > Is there any tool to do it automatically ? or should I write some code > > to convert the file ?  > K > I had to work with VFC files some time ago and needed to convert them for G > e-mailing.  As I recall, none of the VMS utilities did what I wanted:  convert H > a VFC file to variable length records while preserving the formatting. > K > What I did was write a C copy program. The C RTL routines handled all the F > format conversions.  Here is the code that I used (I'm typing from a& > printout, so there may be mistakes): >  >  > /*** > prn-vfc-to-cr-var.c  > G > Convert the contents of a print/vfc file to carriage_return/variable.  > ***/ >  > #include <stdio.h> > #include <stdlib.h>  > #include <string.h>  > E > #define RECORD_SIZE 16384  /* largest record that can be handled */  >  > " > int main(int argc, char *argv[]) > {  >   FILE *file_in, *file_out;  >   char record[RECORD_SIZE];  >   int status = EXIT_FAILURE; >  >   if (argc == 3) >   { 2 >     (void *) printf("program: <%s>\n", argv[0]); > # >     file_in = fopen(argv[1], "r", E >       "rat=prn", "rfm=vfc", "fsz=2", "rop=rah", "mbc=16", "mbf=4");  >     if (file_in != NULL) >     { 7 >       (void *) printf("input file: <%s>\n", argv[1]);  > & >       file_out = fopen(argv[2], "w",= >         "rat=cr", "rfm=var", "rop=wbh", "mbc=16", "mbf=4");  >       if (file_out != NULL) 	 >       { : >         (void *) printf("output file: <%s>\n", argv[2]); > G > /* this is where the action is - the C I/O routines are doing all the  formatJ > work */  while (fgets(record_in, RECORD_SIZE, file_in) != NULL)  { (void *) > fputs(record_in, file_out); }  > $ >         (void *) fclose(file_out);  >         status = EXIT_SUCCESS;	 >       }  >       else9 >       { (void *) printf("cannot open output file\n"); }  > ! >       (void *) fclose(file_in);  >     } 
 >     else6 >     { (void *) printf("cannot open input file\n"); } >   }  >   elseC >   { (void *) printf("must specify an input and output file\n"); }  >  >   exit(status);  > }  >  >  > ( > Sent via Deja.com http://www.deja.com/ > Before you buy.    ------------------------------  % Date: Fri, 25 Aug 2000 15:47:16 -0400 % From: "Brian Tillman" <tillman_brian> & Subject: Re: DECNet Phase IV Manuals ?$ Message-ID: <39a6cceb$1@news.si.com>   Hoff writes:  K >  I may be able to pry loose the bookreader format manuals for an archival H >  documentation page, but I can't think of a justification for the HTMLG >  conversion of the Phase IV manuals.  Put another way, do you want me I >  doing the HTML conversion of this manual -- because I can quite likely J >  manage this myself -- but because of a deficit in my copious spare time? >  allowance, something else on my schedule has to "give".  :-)   I Why would ANYONE need to "HTML-ize" Bookreader format files, when WEBBOOK G exists and can display Bookreader documents is a web browser just fine?  --A Brian Tillman                   Internet: tillman_brian at si.com A Smiths Industries, Inc.                   tillman at swdev.si.com = 3290 Patterson Ave. SE, MS      Addresses modified to prevent < Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  # Date: Fri, 25 Aug 2000 22:00:33 GMT ( From: Terry Kennedy <terry@gate.tmk.com>& Subject: Re: DECNet Phase IV Manuals ?' Message-ID: <FzvBsx.Lyo@spcuna.spc.edu>   * Yuri Ermakov <ermak@cbr.ryazan.su> writes:= > http://gatekeeper.dec.com/pub/DEC/DECnet/PhaseIV/index.html   J   However, those are the manuals for *writing* a DECnet Phase IV implemen-F tation, not for *using* the *VMS* implementation that Compaq supplies.  K   Also, those manuals are out-of-date - for example, the MOP manual at that J site is V3.0.0, September 1983, while the one you get if you order it fromI Compaq is V4.0.0, June 1992 (EK-DNA11-FS), and there are a number of sup- J plements to it which have never been formally released, like "MOP Registry 22 March 1993".   - 	Terry Kennedy             http://www.tmk.com 5         terry@tmk.com             Jersey City, NJ USA    ------------------------------  % Date: Fri, 25 Aug 2000 13:09:34 -0700 ! From: Shane.F.Smith@Healthnet.com , Subject: Re: Disaster Tolerance - TerroristsC Message-ID: <OF67E04BEE.F6167271-ON88256946.006E930B@HEALTHNET.COM>   H Not that I am aware of, and I believe most of the VMS sites in that areaK were customers of the company I worked for at the time. My office was about H a mile away, we lost windows (unfortunately, the wrong kind...). Our VMS ystems didn't even hiccup.   Shane           8 Steve.Spires@yellowpages.co.uk on 08/25/2000 08:15:53 AM   To:   Info-VAX@Mvb.Saic.Com  cc:   - Subject:  Re: Disaster Tolerance - Terrorists     = Contact:   Tel: 3063  -  VSSG, 1st Floor, Bridge Street Plaza     J I don't know for sure, but I would hazard a guess at it being when the IRA7 bombed the City of London to break their 'ceasefire'...    Steve Spires        F robert_jm_barron@HOTMAIL.COM (Robert Barron) on 25/08/2000 03:00:38 PM    To:        Info-VAX@Mvb.Saic.Com+ cc:         (bcc: Steve Spires/YellowPages) H From:      robert_jm_barron@HOTMAIL.COM (Robert Barron), 25 August 2000, 3:00            p.m.    Disaster Tolerance - Terrorists          Hi, @ Whiles reading "OpenVMS and NT Integration (for dummies)" I cameF across a line on page 171 where they claim that due to a fiber ClusterB Interconnect, an OpenVMS application survived a terrorist bombing.   are any details available?  
 Thank you, Robet    ------------------------------  # Date: Fri, 25 Aug 2000 20:27:24 GMT * From: Alan E. Feldman <alan48@my-deja.com>4 Subject: Don't global pages count against WS Limits?) Message-ID: <8o6kqj$gbt$1@nnrp1.deja.com>    Hello,  C (If I have misread the below-quoted Wizard's posting, I apologize.)   E In Ask the Wizard no. (1338) Global Pages and Ingres, the Wizard says D that global pages are not counted against the working set.  However,C the DCL procedure WORKING_SET.COM (aka WORKSET.COM) explicitly adds  global pages to "process pages"   / $ globalpages   =       F$GETJPI('pid,"GPGCNT") / $ processpages  =       F$GETJPI('pid,"PPGCNT") 1 $ pagefaults    =       F$GETJPI('pid,"PAGEFLTS") 2 $ pages         =       globalpages + processpages  A and then displays "pages" in the "Pages in WS" column. I found by @ experimentation that the value from WORKSET.COM agrees with SHOWC PROC/CONT Working Set value which agrees with the output of ^T. The  performance manual defines  G Working set count: The actual number of pages the working set requires. D It consists of the process's pages plus any global pages the process uses.   F Are global section pages counted differently than global pages of codeF installed as shared are? How is the counting done for various kinds of; gbl pages? Are global section pages usually or always data?l  E I would have surmised that if a process needs to map to any page thatgC it's not already mapped to, global or otherwise, it would involve aaE page fault and therefore AWSA would count the global page against thee+ process's working set limit. Is that right?n   --, Disclaimer: JMHO or JASQ as the case may be. Alan E. Feldmann alan48  &-)0 dellnet.com     & Sent via Deja.com http://www.deja.com/ Before you buy.m   ------------------------------  % Date: Fri, 25 Aug 2000 19:08:22 -0500 1 From: Andy Goldstein <Goldstein@star.zko.dec.com>O8 Subject: Re: Don't global pages count against WS Limits?0 Message-ID: <39A6C426.372E880B@star.zko.dec.com>   Alan E. Feldman wrote: >  > Hello, > E > (If I have misread the below-quoted Wizard's posting, I apologize.)w > G > In Ask the Wizard no. (1338) Global Pages and Ingres, the Wizard saysM< > that global pages are not counted against the working set. > C Unless this got changed in the V7 memory management rewrite, globalcF pages are counted against the working set. I'd want to verify with ourD memory management folks to be sure, but without counting them in theD working set we would be unable to manage real memory use of pageableH global sections. Some of the oddball stuff, like PFN mapped sections andH resident images may not be counted in the working set because they never page.   H > Are global section pages counted differently than global pages of codeH > installed as shared are? How is the counting done for various kinds of > gbl pages?  F They're all the same. Pages of code that's installed shared constituteB global sections that are not inherently distinguishable from other read-only global sections.  2 > Are global section pages usually or always data? > E Obviously depends on what you're doing, but as I said, read-only code G pages that don't require fixup are also global pages. Fixup is requiredb> when code (or data) pages contain address constants. These areA initialized by the linker to be relative to the base of the image7E section, and must be turned into an absolute address when the sectionBE base address is known. This occurs at image activation time. Sectionsn1 requiring fixup obviously become process-private.<   ------------------------------  + Date: Fri, 25 Aug 2000 12:44:10 -0500 (EST)o From: wedgegr@yahoo.com 9 Subject: Earn at Least $5,000/Week. FREE 32 Page Booklet!H; Message-ID: <iss.12ba.39a6d304.ebe79.1@mclmx.mail.saic.com>n  3 You can earn more than $5,000 a week within 28 dayso0 with our amazing new Internet marketing program.  C InfoShare is a breakthrough Web Marketing concept that cannot fail.a? Find out how to earn more money than you ever dreamed possible.e! Download our FREE 32-page booklete* "Inside Secrets to Wealth on the Web" NOW!  ? Just hit REPLY and type "Send the Booklet" in the subject line!n   ------------------------------   Date: 25 Aug 2000 23:39:32 GMT6 From: DAVISM@er6.eng.ohio-state.edu (Michael T. Davis)6 Subject: File name saved from EVE (TPU) on ODS5 volume: Message-ID: <8o703k$eev$1@charm.magnus.acs.ohio-state.edu>  F 	We're running under OVMS v7.2-1 on an Alpha and the file system(s) inM question is (are) ODS5.  Say I fire up EVE without specifying a file name.  IsL write a couple lines, then hit CTRL/Z. I'm prompted for a file name, which IL supply--in lower-case, since my CAPS LOCK is typically off.  EVE (TPU) savesL the file with the name _exactly_ as I typed it (preserving case).  If I saveJ to a file name which already exists, the case of the existing file name isM maintained on the new version of the file created.  Should this really be the L behavior, even when my process' parse style is set to "TRADITIONAL"?  If so,H why?  ...Any way to get the TPU to behave dependent on my process' parse style?   Thanks,S  Mike. --K              Michael T. Davis              |    Systems Specialist: ChE,MSEiN   E-mail: davism@er6.eng.ohio-state.edu    | Departmental Networking/ComputingJ            -or- DAVISM+@osu.edu            |     The Ohio State UniversityJ http://www.er6.eng.ohio-state.edu/~davism/ |     197 Watts, (614) 292-6928   ------------------------------  # Date: Sat, 26 Aug 2000 01:36:57 GMTw/ From: StevenU@POBoxes.com (Steven P. Underwood)A) Subject: Re: how to connect to NTP serverO1 Message-ID: <39a71dea.88045152@news.telocity.com>t  C Where is the Official FAQ located?  The OpenVMS FAQ link off of thexE OpenVMS web page (wizard area) shows an edit date of 18-AUG-2000, yeteB does not seem to have the time information.  The 2 ftp.digital.comF links from that FAQ both have an FAQ dated 29-NOV-1999.  The decus.orgE site is dated mid-March. The link to the FAQ at mit.edu (part 1) doeseE include the TIME section?  Seems pretty sad that the MIT site is moreS! up to date than the OpenVMS site.n   Say it ain't so....a   Stevea  @ On 25 Aug 2000 17:34:12 GMT, hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) wrote:    >eK >In article <8o5too$j9b$1@nnrp1.deja.com>, tony_barratt@my-deja.com writes:o> >:...how to sysnc the VMS boxen to an external (NTP?) clock... >1G >  Please see the OpenVMS FAQ.  This question and variations get asked uJ >  rather often of late, and I've added an entire section on OpenVMS time C >  and timekeeping with the most recent edition of the OpenVMS FAQ.  >tO > --------------------------- pure personal opinion ---------------------------fM >   Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com  >    Steven P. Underwood,DNRC Whitinsville,MAP StevenU@POBoxes.coma   ------------------------------  # Date: Sat, 26 Aug 2000 01:43:08 GMTs/ From: StevenU@POBoxes.com (Steven P. Underwood)o) Subject: Re: how to connect to NTP server 1 Message-ID: <39a72054.88663041@news.telocity.com>    Other than USENET.  @ On Sat, 26 Aug 2000 01:36:57 GMT, StevenU@POBoxes.com (Steven P. Underwood) wrote:a  D >Where is the Official FAQ located?  The OpenVMS FAQ link off of theF >OpenVMS web page (wizard area) shows an edit date of 18-AUG-2000, yetC >does not seem to have the time information.  The 2 ftp.digital.com>G >links from that FAQ both have an FAQ dated 29-NOV-1999.  The decus.orgaF >site is dated mid-March. The link to the FAQ at mit.edu (part 1) doesF >include the TIME section?  Seems pretty sad that the MIT site is more" >up to date than the OpenVMS site. >n >Say it ain't so.... >r >Steve >CA >On 25 Aug 2000 17:34:12 GMT, hoffman@xdelta.zko.dec.nospam (Hoffr >Hoffman) wrote: >t >>L >>In article <8o5too$j9b$1@nnrp1.deja.com>, tony_barratt@my-deja.com writes:? >>:...how to sysnc the VMS boxen to an external (NTP?) clock...  >>H >>  Please see the OpenVMS FAQ.  This question and variations get asked K >>  rather often of late, and I've added an entire section on OpenVMS time bD >>  and timekeeping with the most recent edition of the OpenVMS FAQ. >>P >> --------------------------- pure personal opinion ---------------------------N >>   Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com >> >o >Steven P. Underwood,DNRCg >Whitinsville,MA >StevenU@POBoxes.com   Steven P. Underwood,DNRC Whitinsville,MA. StevenU@POBoxes.comw   ------------------------------  % Date: Fri, 25 Aug 2000 15:52:08 -0400o% From: "Brian Tillman" <tillman_brian>.' Subject: Re: LEDs & VAXStation 3100 M48-$ Message-ID: <39a6ce0f$1@news.si.com>  
 Ale wrote:  G >When I wrote "Think Free" I was thinking something like "to be free tom thinkk* >different", not only "free = at no cost",@ >probably it should be better if I'd write "Think Different" :-)  L Perhaps if you had used the correct phrase ("Think Freely") there would haveD been no error in interpretation.  Your suggested replacement is alsoE grammatically incorrect.  "Think Differently" is what you should say.  --A Brian Tillman                   Internet: tillman_brian at si.comtA Smiths Industries, Inc.                   tillman at swdev.si.comf= 3290 Patterson Ave. SE, MS      Addresses modified to prevente< Grand Rapids, MI 49512-1991     SPAM.  Replace "at" with "@"8        This opinion doesn't represent that of my company   ------------------------------  % Date: Fri, 25 Aug 2000 20:59:32 +0200j From: Cor Mom <cor@momss.nl>$ Subject: Re: MEMORY LEAK DETECTION ?( Message-ID: <39A6C214.42231756@momss.nl>  C You may try VIP - VMS Information Provider. It monitors Working Sete@ values and a lot of other interesting values. You can find it at http://www.momss.nl/vip.  
 Good luck!   Cor Moms      " Softwaregroep TB/GA Gasunie wrote: >  > Hi there,  > L > I have an application which shortly after running reaches his WSEXTENT and& > has a rapidly decreasing  PAGFILCNT.K > Obviously there is a memory leak in this application (which is written iniN > C). Does anybody know a tool to watch the behaviour of memory consumption of3 > a C application. (i.e detection of a memory leak)  > Any suggestions are welcome. >  > Herman Behllingg   ------------------------------  % Date: Fri, 25 Aug 2000 19:08:25 +0100   From: steven.reece@quintiles.com? Subject: Re: Minor Nit In OpenVMS ALPHA Version of SHUTDOWN.COM"H Message-ID: <OF2152E238.3206A36F-ON80256946.0063570D@qedi.quintiles.com>  6 "Because it thought it needed to try and catch up withI (NT/Solaris/HP-Ughs/AIX) for the most downtime in a week so that it coulde. get more attention from the system managers"??   or  D "Because it wanted the system manager to come and do some work on itH instead of just patting it on the top and saying how reliable it was and+ how little downtime these VMS servers have"o  + but maybe that's too many letters....  :-))o   Hoff commented: @ >>>  Ok, I'll propose that "Standalone" means "Standalone System Operations",H   and quite obviously this has nothing to do with that Standalone BACKUPH   stuff.  (Problem solved. :-) I can pass this proposal to the technical:   writers, unless you can suggest a better alternative.<<<   ------------------------------  % Date: Fri, 25 Aug 2000 19:29:20 -0500** From: Keith Brown <kbrown780@usfamily.net>? Subject: Re: Minor Nit In OpenVMS ALPHA Version of SHUTDOWN.COM*, Message-ID: <39A70F60.B976B517@usfamily.net>   Jerry Leslie wrote:i > G > The OpenVMS ALPHA version of SYS$SYSTEM:SHUTDOWN.COM has "Standalone"d) > as the default reason for the shutdown:g > % >    Reason for shutdown [Standalone]  > G > but OpenVMS ALPHA no longer has a "Standalone" backup, per the outputr > from STABACKIT.COM:  >  >    $ @sys$update:stabackit > L >         Standalone Backup is no longer part of the OpenVMS Alpha operatingL >         system.  Please refer to the "Backing Up and Restoring the  SystemL >         Disk"  appendix  in  the  "OpenVMS  Alpha Upgrade and Installation' >         Manual" for more information.t > 6 > --Jerry Leslie     (my opinions are strictly my own)  = OpenVMS Alpha still supports standalone backup. It just bootsu; from a CD rather than disk or tape.  Just my 2 cents worth.  -- t Keith Browni kbrown780@usfamily.net   ------------------------------  % Date: Fri, 25 Aug 2000 14:38:27 -0400{% From: James Griffin <griffin@vol.com>) Subject: Pathworks for VMSO Message-ID: <374692323616CB7C.F53FE532CA69077B.09B6979AE7C2FA40@lp.airnews.net>    Good afternoon.i  G I have a VAXStation 4000/200, currently running VMS 5.4-2 and PATHWORKS B v4.0.  I've been trying to update to VMS 5.5-2 and have run into a stumbling block wrt PATHWORKS.   Here's the problem:s  B When running SYS$STARTUP:PCFS_STARTUP.COM, I get a message saying:?   "File server for PATHWORKS for VMS will not start.  Check thee PCFS$STARTUP.log"-  E Which I do.  The first line in PCFS$LOG_FILES:PCFS$STARTUP.LOG reads:   G %PCFS$STARTUP-F-LOWWS, Process Quota PQL_WORKING_SET is low, minimum ise 512-  @ I have looked everywhere to find that parameter.  I've found and8 checked other PQL parameters, but not the specified one.  
 Any ideas?   TIA.   Jim Griffini griffin@vol.comt   ------------------------------  % Date: Fri, 25 Aug 2000 18:37:15 -0400d) From: yyyc186.illegaltospam_@flashcom.netn- Subject: Re: Portable GUIs (VMS+Windows-NT) ? 9 Message-ID: <39a6f538$1$lllp186$mr2ice@news.flashcom.com>i  4 In <spqk458487v104@corp.supernews.com>, on 08/25/00 B    at 06:37 PM, "Jim Jennis" <jjennis@discovery.fuentez.com> said:    G Isn't WRQ the symbol for what used to be/is Wind River Systems?  If so,t the library is Zinc.   Roland  1 >"bz" <bernd.zebedin@arcs.ac.at> wrote in message_% >news:399D073A.D05B0FB2@arcs.ac.at...cI >> I am looking for a mature Tool for development of cross-platform GUIs, K >> that can be run without major modifications both under OpenVMS and undera >> Windows-NT.J >> I have tested the VMS-port of Qt, but it seems to be faulty. Has anyoneG >> experience with porting Java-GUI-applications to VMS or with the ISAnF >> Dialog Manager  or with any other multi-platform toolkit supporting >> OpenVMS?. >> >> Grateful for any advice, bz.< >Hi VMS Colleagues,w  B >I faced this problem in a major integration effort with VMS in myA >previous job (and also with our current customers) and found theV> >availability of good mature tools for VMS to be very lacking.I >Nevertheless I did find a very solid solution, and it MAY be of interesttE >to others on the list so I decided to post the solution we have usedo >here.  I >For cross platform VMS application development and integration, we use aa$ >product from WRQ called Verastream.  ( >http://www.wrq.com/products/verastream/  > >It is a very powerful, mature object oriented cross-platform,C >cross-database gui/application development tool kit. Runs on 25-30 I >different OS's (including OVMS and even OS/400) supports 25-30 different G >databases and about a dozen different inter-application communicationsr >protocols.g  E >It generates platform independent code that is REALLY write once runtF >anywhere (unlike Java which is usually "write once test everywhere").  5 >We use it on OVMS, Linux, Solaris, HP-UX, NT, 95/98.c  J >You can generate apps with Motif, Windows, Character based, WEB interfaceG >and have a single source application that runs on any platform withoutlB >recompilation or source code changes. IMHO it is a great product.  F >You can write an application on VMS and run it on Linux, NT, Unix, or >vice-versa.  G >If anyone would like additional information, feel free to write me offw
 >the list.  	 >Regards,    >Jim       --  ; -----------------------------------------------------------tD yyyc186@flashcom.net              To Respond delete ".illegaltospam"6                             MR/2 Internet Cruiser 1.528                             For a Microsoft free univers; -----------------------------------------------------------,   ------------------------------  % Date: Fri, 25 Aug 2000 20:54:07 -0500 7 From: "David J. Dachtera" <djesys.nospam@earthlink.net> - Subject: Re: Portable GUIs (VMS+Windows-NT) ?a- Message-ID: <39A7233F.5790403B@earthlink.net>n  * yyyc186.illegaltospam_@flashcom.net wrote: > 5 > In <spqk458487v104@corp.supernews.com>, on 08/25/00 D >    at 06:37 PM, "Jim Jennis" <jjennis@discovery.fuentez.com> said: > I > Isn't WRQ the symbol for what used to be/is Wind River Systems?  If so,  > the library is Zinc.  3 WRQ is Walker, Richer & Quinn - http://www.wrq.com/   4 I saw no mention of a buyout or merger on that site.   --   David J. Dachterar dba DJE Systemsa http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/c  H This *IS* comp.os.vms. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged."   ------------------------------  % Date: Fri, 25 Aug 2000 18:01:11 -0500 + From: Jorge Valenzuela <jvalen@alefdata.cl>b3 Subject: Problems with Accounts " CAPTIVE " in VMS.g+ Message-ID: <39A6FAB6.86D05057@alefdata.cl>h  * Problems with Accounts " CAPTIVE " in VMS.  @ 1. Now in the menus of the clients are constructed in DCL, whichE received options or data through commando " INQUIRE ", but this it isnH not applicable when the FLAG " Captive " is turned on, thus is that they@ had to modify the menus that used this command, by the command " READ/PROMPT ".  E 2, After fixing the menus, all the accounts that the client used wereaH modified  and the flag " Captive " was turned on. But a problem with allH the programs that use the function " RTL " LIB$SPAWN, which makes a callE or exit to the S.O. and executes a commando that goes like parameter,i= this comand doesnt work with the flag " CAPTIVE " turned on.i  = In order to give solution the following options were tested :n  ,  a) To give a high privilege to the account.3  b) Change the  flag " Captive " by other similars.-*  c) Change the call to function LIB$SPAWN.   Result:0  Option a) and b) was not work,iE   but option c) presents some possibilities to give solution, like byL	 example -s  F   When function LIB$SPAWN is being used to eliminate archives, can beG changed by the function LIB$DELETE_FILE, which executes of similar form G   When function LIB$SPAWN makes a commando " COPY ", can be changed by  function LIB$RENAME_FILE.oA   But other possibilities exist who was not solution them, like:      - To execute a procedure DCL/   - To make a TYPE to archives with format FMS.   $     Any idea... help or suggestion ?  1                                 thanks in advance   A                                                 Jorge V.S.'000825e   --H ////////////////////////////////////////////////////////////////////////   //  ALEF DATA S.A.  //5 //  Fn:(562) 695-6600  //  Jorge Valenzuela Sepulveday2 //  Fx:(562) 697-2037  //  Jefe Area de Ingeniera1 //  Agustinas 1141 piso 7. //  jvalen@alefdata.cl - //  Santiago, Chile  //  alefdata@alefdata.cliG ///////////////////////////////////////////////////////////////////////i   ------------------------------   Date: 25 Aug 2000 23:22:52 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)7 Subject: Re: Problems with Accounts " CAPTIVE " in VMS. 6 Message-ID: <8o6v4c$82h$1@mailint03.im.hou.compaq.com>  Y In article <39A6FAB6.86D05057@alefdata.cl>, Jorge Valenzuela <jvalen@alefdata.cl> writes: + :Problems with Accounts " CAPTIVE " in VMS.s  H   Please take the time to skim through the documentation for CAPTIVE andH   RESTRICTED accounts in the security manual, as this will save you someD   effort.  In the OpenVMS V7.2 documentation, please take a look at:  '     http://www.openvms.compaq.com:8000/pB       72final/6346/6346pro_003.html#prv_creating_a_captive_account     :1. ..." INQUIRE "...T  H   As stated in the manuals and as you have found, use READ.  INQUIRE is K   a powerful and mystical command with strange and wonderous capabilities,  J   and (though both the command and the capabilities are fully documented) 7   is a DCL command that should best be left to wizards.   F :2, After fixing the menus, all the accounts that the client used wereI :modified  and the flag " Captive " was turned on. But a problem with alldI :the programs that use the function " RTL " LIB$SPAWN, which makes a calloF :or exit to the S.O. and executes a commando that goes like parameter,> :this comand doesnt work with the flag " CAPTIVE " turned on.  H   Use the TRUSTED bit flag on lib$spawn, or the SPAWN/TRUSTED qualifier,G   depending on the context -- if you need to use a subprocess, that is.g  J   If you are working primarily in DCL, I added a section (shameless plug) I   to the Writing Real Programs in DCL book that covers security aspects, l@   and particularly around correct use of CAPTIVE and RESTRICTED.  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Fri, 25 Aug 2000 21:05:35 +0200n> From: "Jean-Franois Marchal" <jean-francois.marchal@x9000.fr>' Subject: Re: Renaming SYS$NODE_nodenamea2 Message-ID: <8o6ft5$qhi$1@s1.read.news.oleane.net>  H Changing the SCSNODE name also does have an impact on the Queue Manager.E Is there any solution to avoid creating a new version to keep th Qmany	 running ?a   Jean-Franois Marchal   ? "Hoff Hoffman" <hoffman@xdelta.zko.dec.nospam> wrote in messageg0 news:8o66so$48a$1@mailint03.im.hou.compaq.com... >e; > In article <8o3vu3$elj$1@nnrp1.deja.com>, Alan E. Feldmanv <alan48@my-deja.com> writes: >9J > :In the VMS FAQ, MGMT19., it says that one of the steps needed to rename > :a node is > :o > :<H > :6. Rename the SYS$NODE_oldnodename rightslist identifier to match theA > :new name. (Do not change the binary value of this identifier.)c > :> > :aI > :But, in the OVMS Guide to System Security, Section 4.1.2.1 (Manual forsI > :OVMS V6.2) it says that the identifier SYS$NODE_nodename is created byiC > :the system startup procedure SYS$SYSTEM:STARTUP.COM. I looked indA > :STARTUP.COM and it gets the nodename from F$GETSYI("NODENAME")t > ..4 > :What is the resolution of this apparent conflict? >eC >   Both are quite correct, and certainly not mutually exclusive...  >eH >   As stated, the technique in the FAQ ensures you keep the same binaryJ >   value for the identifier, which is of interest to folks that have thisJ >   identifier scattered around in various ACLs.   You may or may not haveH >   this, but it is simpler and far easier to simply assume that you do. >AG >   The cited documentation provides details of how the identifier getsd  >   created, if it is not found. >VJ >   If you have no references to the identifier in ACLs (or if you wish toI >   locate and alter individual ACEs containing the binary value for this-G >   identifier), then you can delete the identifier and permit it to bexE >   recreated -- the new identifier might arrive with the same binarytH >   value, or it might also obviously get a new binary value assigned... >0I >   Note that if you do not somehow remove (via renaming or via deletion)RL >   the text value of this identifier, the old text and the old binary valueI >   will remain present in the rightslist.  As documented, you will get aO> >   new text identifier that matches your new SCSNODE setting. >FH > :Also, why does it say "DECnet Phase IV area number"? What if you have > :Phase V or DECnet Plus? TIA.2 >,F >   DECnet-Plus routing does not particularly have areas, as that is a' >   concept of DECnet Phase IV routing.  >e, >  --------------------------- pure personal# opinion ---------------------------t1 >    Hoff (Stephen) Hoffman   OpenVMS Engineeringm hoffman#xdelta.zko.dec.com >i   ------------------------------   Date: 25 Aug 2000 23:02:11 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)' Subject: Re: Renaming SYS$NODE_nodenamen6 Message-ID: <8o6ttj$7op$1@mailint03.im.hou.compaq.com>  s In article <8o6ft5$qhi$1@s1.read.news.oleane.net>, "Jean-Franois Marchal" <jean-francois.marchal@x9000.fr> writes:eI :Changing the SCSNODE name also does have an impact on the Queue Manager.oF :Is there any solution to avoid creating a new version to keep th Qman
 :running ?  5   From the OpenVMS FAQ section under discussion here:>  C       "Modify the host node name on the various queues in the queue-@       database.  (each queue has a host name, and it defaults toD       the SCS node name of the queue's host system.  See the command+       INIT/QUEUE/ON=node for information.)"w  E   I usually prefer to specify the /ON qualifier on most (all?) queue -E   commands in startup and elsewhere, rather than defaulting it.  But n9   obviously rebuilding the queue database will work, too.-  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Fri, 25 Aug 2000 15:59:01 -0400f# From: Jim Agnew <agnew@hsc.vcu.edu> @ Subject: Replace licenses for VAXcluster for Education/Research?+ Message-ID: <39A6D005.43B859E1@hsc.vcu.edu>e  D Our university is letting the Campuswide Site Licenses go next year.J I need to come up with licenses or a replacement licenses for our VAXen...   Where do I call?  My biggest problem is that I can't navigate the automated phone systems as i'm hard of hearing, and i want optionn X, get a real person... sigh...     	 Jim Agnewe   ------------------------------   Date: 25 Aug 2000 23:11:23 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)D Subject: Re: Replace licenses for VAXcluster for Education/Research?6 Message-ID: <8o6uer$7op$3@mailint03.im.hou.compaq.com>  Q In article <39A6D005.43B859E1@hsc.vcu.edu>, Jim Agnew <agnew@hsc.vcu.edu> writes: E :Our university is letting the Campuswide Site Licenses go next year./  $   When do your CSLG licenses expire?  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Fri, 25 Aug 2000 20:47:29 -0400i2 From: rdeininger@mindspring.com (Robert Deininger)D Subject: Re: Replace licenses for VAXcluster for Education/Research?L Message-ID: <rdeininger-2508002047290001@user-2ive7eb.dialup.mindspring.com>  [ In article <8o6uer$7op$3@mailint03.im.hou.compaq.com>, hoffman@xdelta.zko.dec.nospam wrote:   S > In article <39A6D005.43B859E1@hsc.vcu.edu>, Jim Agnew <agnew@hsc.vcu.edu> writes:UG > :Our university is letting the Campuswide Site Licenses go next year.n > & >   When do your CSLG licenses expire?  4 Hmm.  This may turn out to be an interesting thread.   --   Robert Deininger rdeininger@mindspring.come   ------------------------------  # Date: Fri, 25 Aug 2000 18:13:08 GMTs From: byatesiii@my-deja.comm Subject: Re: run *.com) Message-ID: <8o6cv0$6hn$1@nnrp1.deja.com>r  6 In article <8o67gf$48a$2@mailint03.im.hou.compaq.com>,&   hoffman@xdelta.zko.dec.nospam wrote: >.= > In article <8o6558$qtp$1@goliat.eik.bme.hu>, FAZEKAS Mihalyn# <michael@goliat.eik.bme.hu> writes:rF > :How can i simply run all .COM file in current (or spec.) directory? > H >   As this is a rather unusual request, what are you really up to here? >cG Ha Ha! Reminds me of when I was training a new junior technical person,-F who wanted to run each .COM in our pos_com directory, just to see what they did!!!k    & Sent via Deja.com http://www.deja.com/ Before you buy.H   ------------------------------  % Date: Fri, 25 Aug 2000 14:19:55 -0400 ' From: David Beatty <dabeat@wnt.sas.com>L Subject: Re: run *.com2 Message-ID: <lrimORZjOTZ3kObK3XHWAVZAmW1c@4ax.com>  E OOPS, I misread your question.  See Hoff's post above -- a loop using , the F$Search lexical would be the way to go.  E On Fri, 25 Aug 2000 13:11:53 -0400, David Beatty <dabeat@wnt.sas.com>e wrote:   >  >@filename -- .COM is assumed. > , >On 25 Aug 2000 15:59:36 GMT, FAZEKAS Mihaly# ><michael@goliat.eik.bme.hu> wrote:" >kE >>How can i simply run all .COM file in current (or spec.) directory?O   ------------------------------  % Date: Fri, 25 Aug 2000 15:56:36 -0300i1 From: "Boyle, Darren" <boyledj@bankofbermuda.com>i Subject: RE: run *.comK Message-ID: <9F664D538536D411BD3200508B6FF01A8AE96F@bdant027.bda.bobda.com>   3 As long as he didn't do it in SYS$MANAGER directoryw   > ----------: > From: 	byatesiii@my-deja.com[SMTP:byatesiii@my-deja.com]( > Sent: 	Friday, August 25, 2000 3:13 PM > To: 	Info-VAX@Mvb.Saic.Com > Subject: 	Re: run *.coms > 8 > In article <8o67gf$48a$2@mailint03.im.hou.compaq.com>,( >   hoffman@xdelta.zko.dec.nospam wrote: > >V? > > In article <8o6558$qtp$1@goliat.eik.bme.hu>, FAZEKAS Mihaly4% > <michael@goliat.eik.bme.hu> writes:5H > > :How can i simply run all .COM file in current (or spec.) directory? > >sJ > >   As this is a rather unusual request, what are you really up to here? > >eI > Ha Ha! Reminds me of when I was training a new junior technical person, H > who wanted to run each .COM in our pos_com directory, just to see what
 > they did!!!- >  > ( > Sent via Deja.com http://www.deja.com/ > Before you buy.  >     F **********************************************************************C This message and any files transmitted with it are confidential and J may be privileged and/or subject to the provisions of privacy legislation.M They are intended solely for the use of the individual or entity to whom they:L are addressed. If the reader of this message is not the intended recipient, B please notify the sender immediately and then delete this message.I You are notified that reliance on, disclosure of, distribution or copying: of this message is prohibited.   Bank of BermudaiF **********************************************************************   ------------------------------    Date: 25 Aug 2000 17:24:45 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)n Subject: Re: run *.com+ Message-ID: <2hRhz$DnbJ1p@eisner.decus.org>t  _ In article <8o6558$qtp$1@goliat.eik.bme.hu>, FAZEKAS Mihaly <michael@goliat.eik.bme.hu> writes:mE > How can i simply run all .COM file in current (or spec.) directory?t  @ I think the first step would be to back all files up to tape :-)   ------------------------------  % Date: Fri, 25 Aug 2000 17:55:45 -0400i+ From: Tim Shoppa <shoppa@trailing-edge.com>i Subject: Re: run *.com0 Message-ID: <39A6B321.EE6F61F@trailing-edge.com>   byatesiii@my-deja.com wrote: > 8 > In article <8o67gf$48a$2@mailint03.im.hou.compaq.com>,( >   hoffman@xdelta.zko.dec.nospam wrote: > >v? > > In article <8o6558$qtp$1@goliat.eik.bme.hu>, FAZEKAS Mihaly % > <michael@goliat.eik.bme.hu> writes:wH > > :How can i simply run all .COM file in current (or spec.) directory? > >eJ > >   As this is a rather unusual request, what are you really up to here? > >rI > Ha Ha! Reminds me of when I was training a new junior technical person,pH > who wanted to run each .COM in our pos_com directory, just to see what
 > they did!!!t  G I once made a file in my home directory called DO_NOT_TYPE_ME.TXT whichsA contained several hundred thousand control-G's.  It was less thano< a half hour later that I caught someone down the hall poking through my files :-).a  B All this proves is that if you want to be sure that something *is*? done on a computer, just put a big sign up saying not to do it.c   Tim.   ------------------------------  % Date: Sat, 26 Aug 2000 02:03:49 +0100i% From: Alan Greig <a.greig@virgin.net>y Subject: Re: run *.com* Message-ID: <39A71775.20CB26FF@virgin.net>   Hoff Hoffman wrote:Y  a > In article <8o6558$qtp$1@goliat.eik.bme.hu>, FAZEKAS Mihaly <michael@goliat.eik.bme.hu> writes:dF > :How can i simply run all .COM file in current (or spec.) directory? >gH >   As this is a rather unusual request, what are you really up to here? >-  H I'm guessing he wants to have them defined as foreign commands. In which& case DEFINE DCL$PATH is the  way to go   >eH >   The literal answer involves a simple DCL command procedure using theI >   f$search lexical to scan the directory, with each *.COM located being-F >   invoked as a command procedure in the sequence they are retrieved. >JE >   An example that simply lists the files with a .COM file extension(E >   follows, and could easily be altered to invoke the located files:y >8 > $ fn = f$search("",1)" > $Next: > $ fn = f$search("*.COM",1) > $ if fn .eqs. "" then exit > $ write sys$output "@''fn'""
 > $ goto Nextv > $u >e% >   To invoke a command file you can:l >r >     @filenamei >     @filename.COM  >     filename >rI >   The latter syntax assumes you have DCL$PATH logical name defined, andaF >   that it includes the current directory in its list of directories. >rH >   Also please indicate your OpenVMS version when posting, as potentialK >   suggestions -- such as the DCL$PATH stuff -- have specific (and usually " >   minimum) version requirements. >iP >  --------------------------- pure personal opinion ---------------------------N >    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   --
 Alan Greig   ------------------------------  % Date: Fri, 25 Aug 2000 19:23:18 -04001, From: Jack Pergal <jpergal@nospam.erols.com>  Subject: Re: UCX routing problem8 Message-ID: <flvdqsca3j3js4rfdgl28erh9ifeh7580p@4ax.com>  3 On Thu, 24 Aug 2000 09:33:55 +0200, Oswald Knoppersi& <Oswald.Knoppers@whitehouse.nl> wrote:   >Jack Pergal wrote:  >> wI >> I'm running ucx 2.0B on vms 5.52. I have routing enabled/supply. On annI >> alpha running osf1 I see the routes advertised by the vax, but I can't H >> ping any nodes on those routes. On the vax I see routes advertised byG >> the alpha and I can ping nodes on those routes. What am I missing ond >> the vax?i > I >I guess you haven't enabled IP forwarding on the vax. To enable this yousI >use the command 'ucx set prot ip/forward'. I am not sure if this commande6 >was the same in the (very old) version you are using. >h >Oswald   C Thanks that was setting I missed. It works with all of the networkssA except one we created using the private network 10.1.1.0 . I alsopF noticed one system runnning vms 6.x had ucx 3.3 installed. I'll try toD install ucx 3.3 on vms 5.52. Ever forward! maybe we'll get up to 4-5
 years behind!   	 			Thanks  			Jackn   ------------------------------  % Date: Fri, 25 Aug 2000 16:27:36 +0100  From: Paul Sture <paul@sture.ch>; Subject: Upgrade path from VMS 6.1 for my "New" uVAX 3100 ?I+ Message-ID: <VA.000000a9.000fdb28@sture.ch>   H After much ado finding the right cable connectors, I have just fired up   an old VAX 3100 (M38 I believe).  G I know I'm just being a bit lazy asking this, but does anyone know off u: the top of their heads what the upgrade path to V7.2-1 is?   E.g. do I have to go via 6.2?o ___g
 Paul Sture Switzerland    ------------------------------   Date: 25 Aug 2000 23:09:29 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)? Subject: Re: Upgrade path from VMS 6.1 for my "New" uVAX 3100 ?/6 Message-ID: <8o6ub9$7op$2@mailint03.im.hou.compaq.com>  N In article <VA.000000a9.000fdb28@sture.ch>, Paul Sture <paul@sture.ch> writes:H :I know I'm just being a bit lazy asking this, but does anyone know off ; :the top of their heads what the upgrade path to V7.2-1 is?/  H   Um, I know I could be called lazy for mentioning this :-), but please 5   check the OpenVMS Frequently Asked Questions (FAQ).c  I   In the August 2000 edition of the OpenVMS FAQ, you will find a section SC   with (freshly-updated) OpenVMS upgrade path information included.c  0   Notes:  1: there is no V7.2-1 for OpenVMS VAX.G           2: The FAQ has details of a "wrinkle" lurking in the upgrade SF              path specifically from OpenVMS VAX V6.1 to later OpenVMS D              VAX releases, and the information you need to avoid it.  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Fri, 25 Aug 2000 22:06:57 GMTe( From: a.koopman@hccnet.nl (Adri Koopman)# Subject: Re: VMS 7.2 can't see a cdd- Message-ID: <39a6ec12.6038823@news.hccnet.nl>p   Hi Charles,   > Reading your problem, it sounds familiar to me. I had the sameC problem, trying to use a Pioneer DR-U12X SCSI cdrom on my mVAX 31002E mod 80. The cdrom is shown on the >>> SHO DEV, but after booting backmE to openVMS 7.2, I wan't able to mount any cdrom. When I exchanged thed* unit with an Digital RRD42, it works fine.F I've called techsupp about this problem and they told me that the scsi< bus in a mVAX is not a 100% standard scsi, but with some DECD modifications. That's why the RRD42 worked, because of the different internal wiring.F So, if you want to use a cdrom on your mVAX, you'll have to look for a  DEC cdrom device like the RRD42.  F Also, if someone else has different experiences with third party cdrom devices, I'm interested too.   Adri Koopman    A On 24 Aug 2000 15:52:49 -0500, ckronenw@earth.execpc.com (CharlesI Kronenwetter) wrote:  D >I have a Vaxstation 3100 mod 30 with which I am attempting to use aJ >Toshiba scsi cdrom. I can see the drive on the bus correctly with the SHOJ >DEV command at bootup. I can run standalone backup (from the VMS HobbyestK >cdrom) without problem. I can copy the saveset information without problemjJ >to one of the other scsi disks.  When, however, I run VMS 7.2 and attemptE >to mount the same cdrom, I get an error indicating that media is nothH >present and the drive will not mount.  Is this a bug with 7.2 or is VMSI >just fussier in its choice of cdroms? Why the difference between VMS 7.2f >and standalone backup?> >  >C Kronenwettere >Milwaukee, Wi >l   ------------------------------  % Date: Fri, 25 Aug 2000 18:49:11 -0500 1 From: Andy Goldstein <Goldstein@star.zko.dec.com> 7 Subject: Re: Working Set Size Limits and Process Quotas / Message-ID: <39A6BFA7.2A86BDA@star.zko.dec.com>    Alan E. Feldman wrote: > B > Why are recommended working set size limits (WSDEFAULT, WSQUOTA,= > WSEXTENT, etc.) and process quotas like DIOLM, FILLM, etc.,eF > often in powers of 2 or sums of powers of 2? Sometimes they are not.D > Is there any disadvantage to not using powers of 2? If so, why are > they sometimes used?  G No particular reason other than your typical programmer's affection fordB powers of two. Some parameters consume resources (i.e., memory) inE powers of two groups. The significant ones are SYSGEN parameters, not B quotas. Working set and page file quotas are expressed in 512 byteH "pagelets" to make the units consistent with the VAX. However, even here7 the granularity is only 16 (8kb page vs 512 byte page).a   ------------------------------   Date: 25 Aug 2000 23:24:52 GMT2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)7 Subject: Re: Working Set Size Limits and Process Quotasi6 Message-ID: <8o6v84$82h$2@mailint03.im.hou.compaq.com>  c In article <39A6BFA7.2A86BDA@star.zko.dec.com>, Andy Goldstein <Goldstein@star.zko.dec.com> writes:d :Alan E. Feldman wrote:n :> nC :> Why are recommended working set size limits (WSDEFAULT, WSQUOTA,e> :> WSEXTENT, etc.) and process quotas like DIOLM, FILLM, etc.,G :> often in powers of 2 or sums of powers of 2? Sometimes they are not.iE :> Is there any disadvantage to not using powers of 2? If so, why areo :> they sometimes used?a :gH :No particular reason other than your typical programmer's affection for :powers of two...l  >   Some things are just a "two-bit affair", in other words. :-)  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   End of INFO-VAX 2000.476 ************************