1 INFO-VAX	Sun, 31 Jul 2005	Volume 2005 : Issue 424       Contents:9 Best way to install Hobbyist license file when upgrading? = Re: Best way to install Hobbyist license file when upgrading? = Re: Best way to install Hobbyist license file when upgrading? = Re: Best way to install Hobbyist license file when upgrading? ! Re: Collapsing a bound volume set = Re: Heads Up - Prize for the Hobby contest -AlphaServer DS10L - Re: How does sys$pal_swasten work on Itanium. 2 Re: improve file system performance related to BEA2 RE: improve file system performance related to BEAB Re: New Beta Google Groups format (was Re: comp.os.vms archives ?)B Re: New Beta Google Groups format (was Re: comp.os.vms archives ?) Re: OT: Re: LaTeX ---> PDF! Re: Question about the astronauts ! Re: Question about the astronauts ! Re: Question about the astronauts ! Re: Question about the astronauts P Re: Seeking Testing Volunteers W2K MTS/DTC to VMS DECdtm Distributed 2PC Transac Re: What exactly is PGFLQUOTA  Re: What exactly is PGFLQUOTA  Re: What exactly is PGFLQUOTA  Re: What exactly is PGFLQUOTA = Re: [OT] Rounding v Truncation, was: Re: Platform Support vs.   F ----------------------------------------------------------------------  % Date: Sun, 31 Jul 2005 20:38:55 +1000 # From: "Gremlin" <not-here@all.mate> B Subject: Best way to install Hobbyist license file when upgrading?- Message-ID: <42ecaa3d@duster.adelaide.on.net>    Hi All  J When I get my annual update of the license, what would you suggest as the L best way to install the licences?  In the past I have just run the .com and I all is fine, but should I uninstall/delete the previous licenses to keep  2 things clean?  Does it make any difference at all?  L Also, how far in advance can you get the license upgrade?  Mine expires mid K August, but when I tried last week, I received a file with the same expiry  K dates....so, can you get "next" year's expiry dates a week in advance, two   weeks?........   Thanks     ------------------------------    Date: 31 Jul 2005 09:17:02 -0700; From: "johnhreinhardt@yahoo.com" <johnhreinhardt@yahoo.com> F Subject: Re: Best way to install Hobbyist license file when upgrading?C Message-ID: <1122826622.642208.248840@g43g2000cwa.googlegroups.com>    Gremlin wrote: > Hi All > K > When I get my annual update of the license, what would you suggest as the M > best way to install the licences?  In the past I have just run the .com and J > all is fine, but should I uninstall/delete the previous licenses to keep4 > things clean?  Does it make any difference at all? > M > Also, how far in advance can you get the license upgrade?  Mine expires mid L > August, but when I tried last week, I received a file with the same expiryL > dates....so, can you get "next" year's expiry dates a week in advance, two > weeks?........ >  > Thanks  E IIRC it's about 5 days or so in advance that it will generate the new B date.  Once you have loaded and enabled the new license entries itE doesn't matter if the old ones are still there, although it certainly C is a lot easier to look at your current ones if you remove the old. E Either way works fine - just don't remove the old before enabling the  new ones. ;)   ------------------------------  % Date: Sun, 31 Jul 2005 13:06:18 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> F Subject: Re: Best way to install Hobbyist license file when upgrading?, Message-ID: <42ED050A.FA07DD24@teksavvy.com>   Gremlin wrote:K > When I get my annual update of the license, what would you suggest as the $ > best way to install the licences?     M What I do is with 3 windows: one is the editor with the new licence commands. + One is a licence/list/full, and one is DCL.   M I go though each existing licence that is hobbyist (my system has some "real" N licences too), delete it, and then look for it in the "new licence" file, copyF that text and paste it in the dcl window to register and then load it.  N This way, I only register the licenses that I need, and delete the old ones so  they don't clutter the database.   ------------------------------  % Date: Sun, 31 Jul 2005 12:13:17 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net>F Subject: Re: Best way to install Hobbyist license file when upgrading?+ Message-ID: <42ED06AD.F82F7EC6@comcast.net>    Gremlin wrote: >  > Hi All > K > When I get my annual update of the license, what would you suggest as the M > best way to install the licences?  In the past I have just run the .com and J > all is fine, but should I uninstall/delete the previous licenses to keep4 > things clean?  Does it make any difference at all?  E Personal preference, IMO. I just start with a fresh LDB and load only 3 the PAKs I intend to use, then I do a LICENSE LOAD:   1 $ RENAME LMF$LICENSE *.yyyy	! "yyyy" for the year  $ LICENSE CREATE' $! Execute my PAK extract at this point  $ LICENSE LOAD   ...does it for me.  M > Also, how far in advance can you get the license upgrade?  Mine expires mid L > August, but when I tried last week, I received a file with the same expiryL > dates....so, can you get "next" year's expiry dates a week in advance, two > weeks?........  5 Not sure about that. Never had that specific problem.    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  % Date: Sun, 31 Jul 2005 11:53:55 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net>* Subject: Re: Collapsing a bound volume set+ Message-ID: <42ED0223.E27CAD2B@comcast.net>    prep@prep.synonet.com wrote: > ( > "Ed Wilts" <ewilts@ewilts.org> writes: > H > > I'm far from being an expert in bound volume sets so I hope what I'mK > > trying to do is possible...  I've got a 4-member bound volume set on an I > > HSJ-50 that I'd like to migrate to a new single volume on an EVA.  Is G > > it possible to migrate all of the files to a single member and then D > > eliminate the volume set by mirroring the remaining volume to anI > > EVA-based disk?  I'm trying to avoid using VMSbackup for this because K > > the directory hierarchy consists of lots and lots of little files and I K > > don't want to take the volume offline for that long unless I don't have 
 > > a choice.  > F > The help for MOUNT /BIND has the answer, /IMAGE backup, then restoreF > to a volume /NOINIT. INITing it with a `big enough' header count etcH > of course. Tweeking the XQP for it would help too, if you are creating > lots of files.  G Remember to use INIT /MAXIMUM_FILES to pre-extend INDEXF.SYS as well as F /HEADERS or you'll be in deep stuff faster than a spinlock timeout can cause a bug-check/crash.  H > You  end up banging your head against the direcories being on the root/ > volume, and no way to move them off intact...    Yeah - real bummer...    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  % Date: Sun, 31 Jul 2005 11:51:12 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net>F Subject: Re: Heads Up - Prize for the Hobby contest -AlphaServer DS10L* Message-ID: <42ED0180.4D31059@comcast.net>  " susan_skonetski@hotmail.com wrote: > A > Just re-enter JF, only the latest entry will be used.  Skip the C > centrifuges and just find a way to keep trolls out of newsgroups.   G How 'bout if centrifuge the trolls, use what we can and flush the rest?    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  % Date: Sun, 31 Jul 2005 11:56:02 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net>6 Subject: Re: How does sys$pal_swasten work on Itanium.+ Message-ID: <42ED02A1.AEEF1863@comcast.net>    Alan Greig wrote:  >  > Main, Kerry wrote: >  > > @ > > Clare Grant article on porting to OpenVMS Integrity servers:L > > http://h71000.www7.hp.com/openvms/journal/index.html?jumpid=go/openvms/j > > ournal#porting > I > "One of the goals of the project was to make OpenVMS more portable. Who J > knows, maybe someone will be doing this again!" - Clair Grant, June 2005 > OpenVMS Technical Journal. > = > Any truth Clair's working on the PDP-10 port right now? ;-)   E If they were smart, they'd be working on a port to x86-64 so they can 1 survive if Itanic ends up in Davy Jones's locker.    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  % Date: Sun, 31 Jul 2005 14:19:46 +0200 6 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= <arne@vajhoej.dk>; Subject: Re: improve file system performance related to BEA < Message-ID: <42ecc1cc$0$75668$edfadb0f@dread16.news.tele.dk>  & Warning: this going to be a long post.  @ It is practically impossible not to cache at some level, because there are so many levels.    application  Java C  RMS  system  5 For obvious we would expect more effect of caching at  higher levels.  . It is common knowledge in the Java world that:+    - using a big application buffer is good ;    - when using a big application buffer it does not matter ?      whether it is a plain InputStream or a BufferedInputStream 7    - using a small application buffer or single read is 7      not good but acceptable with a BufferedInputStream 7    - using a small application buffer or single read is &      horrible with a plain InputStream  
 Demo program:   
 package test;   # import java.io.BufferedInputStream;  import java.io.FileInputStream;  import java.io.IOException;  import java.io.InputStream;    public class ReadTest { -      public static void main(String[] args) { 1          (new SingleReadUnbufferedTest()).test(); /          (new SingleReadBufferedTest()).test(); 0          (new MultiReadUnbufferedTest()).test();.          (new MultiReadBufferedTest()).test();      } }    abstract class OneTest {E      protected static final String FILE_NAME = "C:\\log4j-1.2.9.jar"; &      private static final int N = 100;      protected int sum;       private String name;       OneTest(String name) {           this.name = name;          sum = 0;       }      void test() {          try {2              long t1 = System.currentTimeMillis();*              for (int i = 0; i < N; i++) {                  dotest();              }2              long t2 = System.currentTimeMillis();I              System.out.println(name + " : result = " + sum + " time = "  
 + (t2 - t1)); "          } catch (IOException e) {!              e.printStackTrace(); 
          }      }/      abstract void dotest() throws IOException;  }   0 class SingleReadUnbufferedTest extends OneTest {!      SingleReadUnbufferedTest() { 5          super("single byte read unbuffered stream");       }'      void dotest() throws IOException { 9          InputStream is = new FileInputStream(FILE_NAME);           int c; '          while ((c = is.read()) >= 0) {               sum += (byte) c; 
          }          is.close();      } }   . class SingleReadBufferedTest extends OneTest {      SingleReadBufferedTest() { 3          super("single byte read buffered stream");       }'      void dotest() throws IOException { ?          BufferedInputStream bis = new BufferedInputStream(new   FileInputStream(FILE_NAME));          int c; (          while ((c = bis.read()) >= 0) {              sum += (byte) c; 
          }          bis.close();       } }   / class MultiReadUnbufferedTest extends OneTest {        MultiReadUnbufferedTest() {4          super("multi byte read unbuffered stream");      }'      void dotest() throws IOException { 9          InputStream is = new FileInputStream(FILE_NAME); $          byte[] b = new byte[10000];          int n; 4          while ((n = is.read(b, 0, b.length)) > 0) {*              for (int i = 0; i < n; i++) {                  sum += b[i];               }
          }          is.close();      } }   - class MultiReadBufferedTest extends OneTest {       MultiReadBufferedTest() {2          super("multi byte read buffered stream");      }'      void dotest() throws IOException { ?          BufferedInputStream bis = new BufferedInputStream(new   FileInputStream(FILE_NAME));$          byte[] b = new byte[10000];          int n; 5          while ((n = bis.read(b, 0, b.length)) > 0) { *              for (int i = 0; i < n; i++) {                  sum += b[i];               }
          }          bis.close();       } }    Results my Windows PC:  D single byte read unbuffered stream : result = 243616700 time = 82140A single byte read buffered stream : result = 243616700 time = 2844 A multi byte read unbuffered stream : result = 243616700 time = 218 ? multi byte read buffered stream : result = 243616700 time = 219    Results my VMS box:   D single byte read unbuffered stream : result = 243616700 time = 81019A single byte read buffered stream : result = 243616700 time = 8056 B multi byte read unbuffered stream : result = 243616700 time = 2074@ multi byte read buffered stream : result = 243616700 time = 2256  ? The results are not identical, but both results fits reasonable  well with expectations.   9 There are no basis for critisizing the VMS port of Java !    So what about your example ?  D I immediatetly focused on the available call. It looked weird to me.7 And if you look at the examples above I did not use it.   
 Demo program:   
 package test;   # import java.io.BufferedInputStream;  import java.io.FileInputStream;  import java.io.IOException;    public class ReadTest2 {-      public static void main(String[] args) { 5          (new NormalSingleReadBufferedTest()).test(); 6          (new CorrectSingleReadBufferedTest()).test();      } }    abstract class OneTest2 { E      protected static final String FILE_NAME = "C:\\log4j-1.2.9.jar"; %      private static final int N = 10;       protected int sum;       private String name;       OneTest2(String name) {          this.name = name;          sum = 0;       }      void test() {          try {2              long t1 = System.currentTimeMillis();*              for (int i = 0; i < N; i++) {                  dotest();              }2              long t2 = System.currentTimeMillis();I              System.out.println(name + " : result = " + sum + " time = "  
 + (t2 - t1)); "          } catch (IOException e) {!              e.printStackTrace(); 
          }      }/      abstract void dotest() throws IOException;  }   5 class NormalSingleReadBufferedTest extends OneTest2 { %      NormalSingleReadBufferedTest() { :          super("normal single byte read buffered stream");      }'      void dotest() throws IOException { ?          BufferedInputStream bis = new BufferedInputStream(new   FileInputStream(FILE_NAME));          int c; (          while ((c = bis.read()) >= 0) {              sum += (byte) c; 
          }          bis.close();      }  }   6 class SpecialSingleReadBufferedTest extends OneTest2 {&      SpecialSingleReadBufferedTest() {;          super("special single byte read buffered stream");       }'      void dotest() throws IOException { ?          BufferedInputStream bis = new BufferedInputStream(new   FileInputStream(FILE_NAME));&          while (bis.available() > 0) {&              sum += (byte) bis.read();
          }          bis.close();       } }    Results my Windows PC:  F normal single byte read buffered stream : result = 24361670 time = 296I special single byte read buffered stream : result = 24361670 time = 16907    Results my VMS box:   F normal single byte read buffered stream : result = 24361670 time = 877K special single byte read buffered stream : result = 24361670 time = 7555923   I We see that it even on Windows is a really bad idea to call available for 
 each read.   On VMS it is a disaster.  C But I do not really consider it a big problem. The available method $ is not intended to be used that way.  A For some reason it seems much more expensive to get the file size B (availeble = file size - current position) on VMS than on Windows.  = Back to the original problem: WebLogic startup is much slower  on VMS than on Windows.   9 It is not a single byte read or available method problem.   8 I am convinced that the biggest factor is that during an< app-server start a lot of files get written (I have not used9 WebLogic, but that is usually the case). Writing files is 9 normally much slower on VMS than on Windows/Unix, because ; VMS has a long history of writing=writing physical to disk, < while other OS's consider the write done as soon as the data> is in a system buffer somewhere. That VMS strategy is good for7 not loosing data and slow for saving stuff that can get " recreated if they should get lost.   Arne   ------------------------------  % Date: Sun, 31 Jul 2005 12:14:06 -0400 ' From: "Main, Kerry" <Kerry.Main@hp.com> ; Subject: RE: improve file system performance related to BEA R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB651759@tayexc19.americas.cpqcorp.net>   =20    > -----Original Message-----1 > From: Arne Vajh=F8j [mailto:arne@vajhoej.dk]=20  > Sent: July 31, 2005 8:20 AM  > To: Info-VAX@Mvb.Saic.Com = > Subject: Re: improve file system performance related to BEA  >=20( > Warning: this going to be a long post. >=20B > It is practically impossible not to cache at some level, because > there are so many levels.  >=20
 > application  > Java > C  > RMS  > system >=207 > For obvious we would expect more effect of caching at  > higher levels. >=200 > It is common knowledge in the Java world that:- >    - using a big application buffer is good = >    - when using a big application buffer it does not matter A >      whether it is a plain InputStream or a BufferedInputStream 9 >    - using a small application buffer or single read is 9 >      not good but acceptable with a BufferedInputStream 9 >    - using a small application buffer or single read is ( >      horrible with a plain InputStream >=20   [major snip ...]   > Results my Windows PC: >=20J > normal single byte read buffered stream : result =3D 24361670 time =3D = 296 C > special single byte read buffered stream : result =3D 24361670=20  > time =3D 16907 >=20 > Results my VMS box:  >=20J > normal single byte read buffered stream : result =3D 24361670 time =3D = 877 C > special single byte read buffered stream : result =3D 24361670=20  > time =3D 7555923 >=20@ > We see that it even on Windows is a really bad idea to call=20 > available for  > each read. >=20 > On VMS it is a disaster. >=20E > But I do not really consider it a big problem. The available method & > is not intended to be used that way. >=20C > For some reason it seems much more expensive to get the file size F > (availeble =3D file size - current position) on VMS than on Windows. >=20? > Back to the original problem: WebLogic startup is much slower  > on VMS than on Windows.  >=20; > It is not a single byte read or available method problem.  >=20: > I am convinced that the biggest factor is that during an> > app-server start a lot of files get written (I have not used; > WebLogic, but that is usually the case). Writing files is ; > normally much slower on VMS than on Windows/Unix, because ? > VMS has a long history of writing=3Dwriting physical to disk, > > while other OS's consider the write done as soon as the data@ > is in a system buffer somewhere. That VMS strategy is good for9 > not loosing data and slow for saving stuff that can get $ > recreated if they should get lost. >=20 > Arne >=20   Arne,   % Excellent post (a definite "keeper").   F Based on what you are outlined about the initial writing of files, I =G wonder if testing BEA start-up (and other Java based files) using VMS = C setup as a UNIX system (write back using dynamic SYSGEN parameter = . RMS_SEQFILE_WBH) might be an interesting test.  
 Reference:F http://h71000.www7.hp.com/doc/731FINAL/6657/6657pro_007.html [scroll =2 down to section 5.11.3 - New write behind feature]  H "An existing user write-behind (RAB$V_WBH setting in RAB$L_ROP) option =H is available for unshared sequential files to request that RMS overlap =J writes to disk. Specifying the write-behind option ensures that at least =G two buffers are allocated by RMS at connect time. When one buffer has = H been filled with records, RMS switches to the second buffer and issues =F an asynchronous QIO to write the contents of the first buffer out to =G disk. RMS does not stall the process, but rather continues to process = D records in the second buffer while the QIO for the first buffer is =C completing. Overlapping I/O can be a big performance win for some =  applications.=20  G OpenVMS Alpha Version 7.3--1 introduces a new system RMS write-behind = G performance option as a dynamic SYSGEN parameter (RMS_SEQFILE_WBH) to = J externally request RMS to implement the write-behind feature as a system =E default (refer to the OpenVMS System Management Utilities Reference = H Manual). If this system option is set, RMS implements the existing RMS =A user write-behind option as a system default (regardless of the = H RAB$L_ROP RAB$V_WBH setting) whenever write access is requested for an =7 unshared sequential file that is opened for image I/O."   J Based on this new dynamic SYSGEN parameter, it should be relatively easy =* to test to see if this makes a difference.   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: Sun, 31 Jul 2005 12:01:00 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net>K Subject: Re: New Beta Google Groups format (was Re: comp.os.vms archives ?) + Message-ID: <42ED03CC.2F5A7242@comcast.net>    "John E. Malmberg" wrote:  >  > Brad Hamilton wrote: > > J > > Do you know of any that allow for authentication to news servers?  I'd& > > love to find one on VMS that does. > 
 > Mozilla.  ( Even the old Netscape V3.03 did as much.   --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  % Date: Sun, 31 Jul 2005 13:44:10 -0400 6 From: Brad Hamilton <brMadAhamPiltSon@coMmcaAstP.neSt>K Subject: Re: New Beta Google Groups format (was Re: comp.os.vms archives ?) 0 Message-ID: <hLWdnXM43vd2kHDfRVn-1A@comcast.com>   David J Dachtera wrote:  > "John E. Malmberg" wrote:  >  >>Brad Hamilton wrote: >>I >>>Do you know of any that allow for authentication to news servers?  I'd % >>>love to find one on VMS that does.  >>
 >>Mozilla. >  > * > Even the old Netscape V3.03 did as much. >   B True bill - of course, I should have specified a _character-cell_ 3 newsreader, but I expect folks to read my mind.	:-)    --   Bradford J. Hamilton "All opinions are my own" * "Lose the MAPS, and replace '-at-' with @"   ------------------------------  + Date: Sun, 31 Jul 2005 06:46:09 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)# Subject: Re: OT: Re: LaTeX ---> PDF $ Message-ID: <dchs3h$qa6$1@online.de>  C In article <k6j7k88o.fsf_-_@ieee.org>, Randy Yates <yates@ieee.org>  writes:   , > rf@cl.cam.ac.uk (Robin Fairbairns) writes:	 > > [...] H > > who does, nowadays, what decus used to do for software distribution? > B > Oh my gosh - "decus" - it's been *decades* since I've heard that > group mentioned.    G The organization still exists.  In some cases, the name has changed to  I reflect the new owners of VMS, but in some cases (such as in Germany) it   hasn't, fortunately.  @ > VMS - what a great OS it was! It's the best OS I've ever used.> > I used to use a VAX cluster during the 80s at GTE Government< > Systems (Mountain View, CA). It was a real joy to work on.  C Not only does VMS still exist, but it is still used in many areas.  D There are the traditional areas: banks, stock exchanges, health careD and, my favourite, the production line at Intel as well as new areasD such as mobile-phone billing systems (I think about 70% of these use VMS, often with Rdb).   G Mistakes both on the part of DEC and within academia led to the decline C of VMS within academia, and many academics assumed that it declined D outside academia as well.  Sure, VMS isn't advertised much, but thenC again it never has been.  The fraction of computers running VMS has G dropped, of course, simply because of all the PCs there are now, though H the absolute number of VMS systems hasn't dropped that much.  VMS hasn'tH made much of a mark for home systems, but if you like VMS, I suggest youH get a used VAX or ALPHA and set it up at home.  Used machines can be had= for free or for a small fee and you can get free licenses for H non-commercial use if you are a DECUS member (which, in some countries, I is free).  I have several VMS machines at home, which among other things  H I use for all internet-related stuff.  It's nice to have a system which > just works and has never been hacked or suffered from a virus.   ------------------------------  # Date: Sun, 31 Jul 2005 07:56:08 GMT   From: Mike M <mikasn@swbell.net>* Subject: Re: Question about the astronauts< Message-ID: <su%Ge.3174$PC7.2249@newssvr29.news.prodigy.net>   Briar Rabbit wrote:  > Harry Potter wrote:  >  >  >>I >> So before any forced circumcision is forced on the male, it should be  G >> done... uh, forced on the female first, right? If there's benefits,  E >> there's benefits and all the females sex organs should be without  4 >> folds of skin for the "cleanest" upkeep possible. >  > q > Guess Rabbit circumcised his dog, as he likes to suck cut dicks..."the foreskin is a dog"?  That said it all... C > Wow, you sure are sore about the fact that the foreskin is a dog.    ------------------------------  % Date: Sun, 31 Jul 2005 11:43:16 +0200 % From: Briar Rabbit <Briar@Rabbit.net> * Subject: Re: Question about the astronauts- Message-ID: <dci6fi$1ta$2@ctb-nnrp2.saix.net>   
 Mike M wrote:    > Briar Rabbit wrote:  >  >> Harry Potter wrote: >> >> >>> J >>> So before any forced circumcision is forced on the male, it should be H >>> done... uh, forced on the female first, right? If there's benefits, F >>> there's benefits and all the females sex organs should be without 5 >>> folds of skin for the "cleanest" upkeep possible.  >> >> >>J >> Guess Rabbit circumcised his dog, as he likes to suck cut dicks..."the + >> foreskin is a dog"?  That said it all... D >> Wow, you sure are sore about the fact that the foreskin is a dog.    > Told you it would get the the skin freaks all in a tizzy.  ;-)   ------------------------------  % Date: Sun, 31 Jul 2005 12:52:59 -0400 6 From: Harry Potter <harrypotter.harrypotter@gmail.com>* Subject: Re: Question about the astronauts7 Message-ID: <Ll7He.387$z91.37237@news20.bellglobal.com>    Briar Rabbit wrote:    > Harry Potter wrote:  >  >  >>I >> So before any forced circumcision is forced on the male, it should be  G >> done... uh, forced on the female first, right? If there's benefits,  E >> there's benefits and all the females sex organs should be without  4 >> folds of skin for the "cleanest" upkeep possible. >  >  > C > Wow, you sure are sore about the fact that the foreskin is a dog.   I C'mon, now; answer the question here. Why not also circumcise the female  G genitals, too (if not FIRST!)? More folds. Even just the clitoral hood  H alone can trap stuff. It happens in Africa and Muslim countries, so why I not widespread wherever male circumcision is preformed... like, the U.S.   for example.   ------------------------------  % Date: Sun, 31 Jul 2005 19:19:07 +0200 % From: Briar Rabbit <Briar@Rabbit.net> * Subject: Re: Question about the astronauts- Message-ID: <dcj169$o55$3@ctb-nnrp2.saix.net>    Harry Potter wrote:    >  >  > Briar Rabbit wrote:  >  >> Harry Potter wrote: >> >> >>> J >>> So before any forced circumcision is forced on the male, it should be H >>> done... uh, forced on the female first, right? If there's benefits, F >>> there's benefits and all the females sex organs should be without 5 >>> folds of skin for the "cleanest" upkeep possible.  >> >> >> >>D >> Wow, you sure are sore about the fact that the foreskin is a dog. >  > K > C'mon, now; answer the question here. Why not also circumcise the female  I > genitals, too (if not FIRST!)? More folds. Even just the clitoral hood  J > alone can trap stuff. It happens in Africa and Muslim countries, so why K > not widespread wherever male circumcision is preformed... like, the U.S.   > for example.    F Show me the evidence that it would achieve the same protective effect?  + PS: don't be such a sore loser, skin freak.    ------------------------------  + Date: Sun, 31 Jul 2005 08:44:28 +0000 (UTC) 3 From: "Richard Maher" <maher_rj@hotspamnotmail.com> Y Subject: Re: Seeking Testing Volunteers W2K MTS/DTC to VMS DECdtm Distributed 2PC Transac ? Message-ID: <dci31a$ff5$1@nwrdmz02.dmz.ncs.ea.ibs-infra.bt.com>    Hi,   K You might recall discussions, some time back, of how I had developed my own @ client/server middleware product called Tier3 and how I was busyF incorporating "hotTIP" a Transaction Internet Protocol (TIP) compliantJ Transaction Manager for VMS. Well, these products are now complete, testedK and even have comprehensive documentation to go with them. What I need from . you is someone to beta-test and evaluate them.  F I won't bore you here with renewed claims of how Tier3 and hotTIP willI breathe new life into VMS as a server platform; I hold these truths to be K self-evident. Just ask your Application Developers, DBAs and/or Windows2000 J System Integrators to glance through the "Tier3 Client/Server Development"I manual (all of 60 pages) and it will be like someone switching a light on " after all these years of darkness!  I So please get in touch if you can at least spend the time to look through C the documentation. If you feel that you may be able to allocate the K resources necessary for a serious evaluation of Tier3, then I will mail you F a copy of the installation kit ASAP (Zipped up on Alpha it's only 1500L blocks).  In all sincerity, for everyone with even the slightest interest inH the longevity of VMS as a development platform, whether at your specificI company or in the world at large, I can promise you that this document is J well worth the read and that you will not be disappointed. (Let me know if6 you'd prefer a PDF or a Word version of the document.)  G Just to let you know, I'm taking the family back to Perth WA next month L after a bit of touring/bomb-dodging, so if you can scope your expressions ofL interest for the next two-weeks that would be just peachy. (Support from the, beaches in the Mauritius will be limited :-)  J Once I'm back in Oz I'll have the facilities, resources and infrastructureG to support the product properly. (Who knows? If enough people get on to H their VMS or Rdb reps and tell them how much they need this product thenL maybe they'll do something sensible like push Tier3 instead of Bridgeworks?)   Cheers Richard Maher  L ---------------------------------------------------------------------------- ----------------------------- C For those who don't remember, here's a brief functional overview of 	 hotTIP: -   E hotTIP allows any DECdtm controlled Resource Manager, such as Rdb, to D participate in a distributed transaction with a cooperating process,F controlled by MTS/DTC on a Windows2000 environment, with the full ACIDJ properties of a true 2PC transaction, regardless of the middleware productK of choice! In case you are unaware, the beauty of TIP's "Two-Pipe" strategy D is it's application-pipe (or middleware) neutrality. Whereas most XAK implementations mandate homogenous Transaction Monitor deployments (such as L Tuxedo everywhere, Encina everywhere, MQSeries everywhere, ACMSxp everywhereK and so on  . . .), hotTIP from TIER3 Software gives you complete freedom to L choose the middleware product(s) that best suite your particular application  and heterogeneous network needs.  F Would  you like to talk to VMS with TIER3 Sockets, COM or DCE/RPC? BEAH MessageQ, IBM MQSeries or HTML? The choice is yours and yours alone. ButJ once you realize that you need to encase your critical transactions withinJ the ACID properties of a true Heterogeneous Two-Phase Commit then you willK come to the conclusion that you need a Transaction Manager that looks a lot H like this. Another drawback of traditional "One-Pipe" strategies is thatF they preclude the run-time determination of transaction  participants.I (Functionality which may be advantageous in a wide-area or Internet based 
 application.)   G "Richard Maher" <maher_rj@hotspamnotmail.com> wrote in message news:...  > Hi,  > E > I am seeking the help of volunteers to test some software that I've H > developed which facilitates distributed two-phase commit transactions,J > encompassing any resource manager (e.g. SQL/Server or Oracle) controlled byB > Microsoft's Distributed Transaction Coordinator in a Windows2000J > environment, with any resource manager under the control of DECdtm (e.g. Rdb 6 > (or Oracle via the XA Veneer)) in a VMS environment. > I > [Yes, at some stage, I hope to sell this software and make money out of  it, L > so unless you have a large philanthropic streak or are simply a techie whoI > likes to stay on top of Windows<->VMS connectivity issues, then you may  wishJ > to look away now. But if you do choose to participate, then rest assuredI > that I have no interest in your personal or company details. (Just your  > work-rate :-)] > C > What differentiates my Transaction Manager software from existing K > Transaction Monitor packages that are already in the marketplace (and why K > you should be interested) is that it is based on the Transaction Internet H > Protocol TIP standard. (RFC 2372) For those of you who don't know, theB > beauty of TIP's "Two-Pipe" strategy is it's application-pipe (or middleware) L > neutrality. Whereas most XA implementations mandate homogenous TransactionD > Monitor deployments (such as Tuxedo everywhere, Encina everywhere, MQSeriesD > everywhere, ACMSxp everywhere and so on  . . .), hotTIP from TIER3 SoftwareJ > gives you complete freedom to choose the middleware product(s) that bestD > suite your particular application and heterogeneous network needs. > H > Would  you like to talk to VMS with TIER3 Sockets, COM or DCE/RPC? BEAJ > MessageQ, IBM MQSeries or HTML? The choice is yours and yours alone. ButL > once you realize that you need to encase your critical transactions withinL > the ACID properties of a true Heterogeneous Two-Phase Commit then you willI > come to the conclusion that you need a Transaction Manager that looks a  lot  > like this. > D > Another drawback of traditional "One-Pipe" strategies is that they precludeI > the run-time determination of transaction  participants. (Functionality J > which may be advantageous in a wide-area or Internet based application.) >   > Anyway, this is what I have: - > I > On the Windows side, you need absolutely *NO* additional software! I'll K > reply to this note with a brief description of the COM+ and DTC functions G > that you would need to invoke in order to successfully push a MTS/DTC H > transaction to VMS. NB: These are standard Windows APIs that are fully > documented on MSDN.  > J > On the VMS side, I have a VMSINSTAL saveset that (all zipped up) is someF > 150KB that I'm happy to e-mail to you along similar lines to the VMSI > hobbyists (non-commercial use) license. I'll reply to this note with an G > Internet Daemon (INETd) example of code that uses my software to cede L > transactional control, over an SQL insert into a Rdb database, to MTS/DTC.J > It's under 500 lines long and contains all of the DCL, 3GL, SQL required toI > produce a working example of a TIP-2PC capable TCP/IP auxiliary server.  ThisL > example will insert a row into the MF_PERSONNEL.Employees table on the VMSJ > side in co-operation with Windows2000 MTS/DTC client that is inserting a row K > into the NORTHWIND.Employee table. Commit them all or roll them all back.  > G > So, in summary, If you'd like to volunteer to put hotTIP through it's  paces ! > then simply reply to this mail.  >  > Regards Richard Maher  >    ------------------------------  # Date: Sun, 31 Jul 2005 13:32:48 GMT * From: "FredK" <fred.nospam@nospam.dec.com>& Subject: Re: What exactly is PGFLQUOTA2 Message-ID: <4q4He.9608$A41.7788@news.cpqcorp.net>  / <dave.baxter@bannerhealth.com> wrote in message < news:1122672794.654461.98530@g49g2000cwa.googlegroups.com...E >       According to the UAF Help, PGFLQUOTA is the number of pages a H > process is allowed to use in the Page File.   Seems eminently sensible > and logical, however;  > C > I have processes which constantly run out of PGFLQUOTA (and these G > processes are starting out with >= 1M pages.     In addition, a "show G > memory/files" command shows little if any pagefile use, i.e. "free" ~ 2 > "size", and monitoring "Page Write I/O Rate = 0) > E > so the question is where does the pgflquota for these processes go. A > Is there some other use for pgflquota which I don't know about.  >    IIRC  A Page file quote doesn't have to do with how much pagefile you are E "actually" using, but how much you "could" consume if the process was E forced to give up its memory.  So the check is being done against how @ much pageable virtual memory you are using, versus how much pageB file you "might" have to consume - not on how much of the pagefile are currently consuming.  D So if you have a 1MB page file quota, and your application grows itsF pageable virtual address space beyond that, you will run out of quota.   ------------------------------    Date: 31 Jul 2005 10:03:29 -0700" From: dave.baxter@bannerhealth.com& Subject: Re: What exactly is PGFLQUOTAC Message-ID: <1122829409.273088.167710@g49g2000cwa.googlegroups.com>    FredK wrote:1 > <dave.baxter@bannerhealth.com> wrote in message > > news:1122672794.654461.98530@g49g2000cwa.googlegroups.com...G > >       According to the UAF Help, PGFLQUOTA is the number of pages a J > > process is allowed to use in the Page File.   Seems eminently sensible > > and logical, however;  > > E > > I have processes which constantly run out of PGFLQUOTA (and these I > > processes are starting out with >= 1M pages.     In addition, a "show I > > memory/files" command shows little if any pagefile use, i.e. "free" ~ 4 > > "size", and monitoring "Page Write I/O Rate = 0) > > G > > so the question is where does the pgflquota for these processes go. C > > Is there some other use for pgflquota which I don't know about.  > >  >  > IIRC > C > Page file quote doesn't have to do with how much pagefile you are G > "actually" using, but how much you "could" consume if the process was G > forced to give up its memory.  So the check is being done against how B > much pageable virtual memory you are using, versus how much pageD > file you "might" have to consume - not on how much of the pagefile > are currently consuming. > F > So if you have a 1MB page file quota, and your application grows itsH > pageable virtual address space beyond that, you will run out of quota.  ? Thanks To all who responded, especially Fred since this was the B information I think I was looking for.     I understand the pointsF everyone else was making, however most were missing the mark, probablyE because I didn't explain properly what I was trying to find out.   My C problem is that these processes were indicating that their pagefile G QUOTA was down to zero, even though they were beginning with very large B (I think) values (>= 1M pages), and even though the system was notD doing any hard paging to or from the page file (there was SOME "ReadD I/O" Page faults which I take to be CODE and/or data being read intoF WSets).   The processes have sufficiently large WS, the Free Page listD is VERY large, and the Modified Page List is sufficiently large thatA the Modified page writer very rarely writes out to the page file.   A What I didn't understand was the connection between PGFLQUOTA and F pageable virtual memory.    In my case, when a process reports that itC has run out of PGFLQUOTA, it really has nothing to do with the Page F file, it is just saying that it has run out of pagable virtual memory.    F However, from what I see here, "Pagable Virtual Memory" allocation forD a process is based on the value of PGFLQUOTA.    So to resolve there@ issue, I keep increasing this quota until the problem goes away.  F Question:   Are there any limiting considerations to how high I can go with this quota??    thanks Dave.   ------------------------------    Date: 31 Jul 2005 10:04:15 -0700" From: dave.baxter@bannerhealth.com& Subject: Re: What exactly is PGFLQUOTAC Message-ID: <1122829455.242616.197440@z14g2000cwz.googlegroups.com>    FredK wrote:1 > <dave.baxter@bannerhealth.com> wrote in message > > news:1122672794.654461.98530@g49g2000cwa.googlegroups.com...G > >       According to the UAF Help, PGFLQUOTA is the number of pages a J > > process is allowed to use in the Page File.   Seems eminently sensible > > and logical, however;  > > E > > I have processes which constantly run out of PGFLQUOTA (and these I > > processes are starting out with >= 1M pages.     In addition, a "show I > > memory/files" command shows little if any pagefile use, i.e. "free" ~ 4 > > "size", and monitoring "Page Write I/O Rate = 0) > > G > > so the question is where does the pgflquota for these processes go. C > > Is there some other use for pgflquota which I don't know about.  > >  >  > IIRC > C > Page file quote doesn't have to do with how much pagefile you are G > "actually" using, but how much you "could" consume if the process was G > forced to give up its memory.  So the check is being done against how B > much pageable virtual memory you are using, versus how much pageD > file you "might" have to consume - not on how much of the pagefile > are currently consuming. > F > So if you have a 1MB page file quota, and your application grows itsH > pageable virtual address space beyond that, you will run out of quota.  ? Thanks To all who responded, especially Fred since this was the B information I think I was looking for.     I understand the pointsF everyone else was making, however most were missing the mark, probablyE because I didn't explain properly what I was trying to find out.   My C problem is that these processes were indicating that their pagefile G QUOTA was down to zero, even though they were beginning with very large B (I think) values (>= 1M pages), and even though the system was notD doing any hard paging to or from the page file (there was SOME "ReadD I/O" Page faults which I take to be CODE and/or data being read intoF WSets).   The processes have sufficiently large WS, the Free Page listD is VERY large, and the Modified Page List is sufficiently large thatA the Modified page writer very rarely writes out to the page file.   A What I didn't understand was the connection between PGFLQUOTA and F pageable virtual memory.    In my case, when a process reports that itC has run out of PGFLQUOTA, it really has nothing to do with the Page F file, it is just saying that it has run out of pagable virtual memory.    F However, from what I see here, "Pagable Virtual Memory" allocation forD a process is based on the value of PGFLQUOTA.    So to resolve there@ issue, I keep increasing this quota until the problem goes away.  F Question:   Are there any limiting considerations to how high I can go with this quota??    thanks Dave.   ------------------------------  % Date: Sun, 31 Jul 2005 13:39:28 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> & Subject: Re: What exactly is PGFLQUOTA, Message-ID: <42ED0CCD.7253B1A3@teksavvy.com>  # dave.baxter@bannerhealth.com wrote: H > However, from what I see here, "Pagable Virtual Memory" allocation for1 > a process is based on the value of PGFLQUOTA.     M I am not sure if this is per process or per job. If it is for a job, remember D that subprocesses will consume from the master process's PGFLQUOTA.   H > Question:   Are there any limiting considerations to how high I can go > with this quota??     # Yes, namely the actual page files.    N As an example, on my all mighty microvax II with 16 megs of memory, I can giveM a process plenty of PGFLQUO, but only if I know that there is enough space in K the actual page file to accept this. (since there is no way that such quota % could reside in memory all the time).   H While this is less obvious on modern machines with plenty of memory, youL should always plan to have a page file(s) that can accomodate most processes$ being paged to file almost entirely.  L If you have a page file which is physically too small, when the process withI large PGFILQUO has to use up the space on disk, and the system can't find K enough contiguous free blocks to write a chunk of memory to it, your system I will grind to a halt with just a small message on opcom. (not even a full  opcom message).   J So you need to look at the physical page files as insurance against giving! processes very large pgfilquotas.    ------------------------------  % Date: Sun, 31 Jul 2005 18:00:08 +1200 $ From: "Lurker" <nowhere@nothing.com>F Subject: Re: [OT] Rounding v Truncation, was: Re: Platform Support vs.3 Message-ID: <7MZGe.4793$PL5.431314@news.xtra.co.nz>   4 "Dave Froble" <davef@tsoft-inc.com> wrote in message* news:11ed7fglvqco402@corp.supernews.com...  C > Say you're $160 late on an installment contract, and your monthly F > payment is $100.  Calculate 160/100 and get 2 months late, and there= > will be some screaming going on.  Not a place for rounding.   9 Hold on a second. If you are late on that $160 you owe it ( right now - no rounding involved at all.  6 However, if you want to take out a loan for those $1602 with a monthly payment of $100, how many months do+ you think you will get to repay it? 1 or 2?    ------------------------------   End of INFO-VAX 2005.424 ************************                                            ij(WR]D,&v*.Q4M#HkH<z/:d<qFq~R职+|2ŨgsH0irj9:.[:ʢ".mO3]}Y8SĩrS+|O4d/=y#
hKVdS!cl4 +|/h_~gzcb*E%J!40I4A;Ew1g>LbNGN<8Ca6M,4ű|Y[D+3,{tҸ":ibԌwok4drXT;kƎQHBEH7vAwCIPF4wfy6jg{{d\)ɡIJP|ڡ 7ͥ XMs(ǛpR
]%/B3s1\~P\L><QF^t]Ѝ
|}|:p+|NM]T4-Fz"O_&w ]j /l($AD1SsF'S=zMUOScsBOp4pFI(d疥d N5`5O*آV%+߁Wo4`=.PFKOuG>	`/U_u
lV]@ >#O' Nk*\;pR3Bn{R=ϔ0/Ef7ȳ~b|,
5A@\w&uK?S.2B;`	/wH =`ЂK⠂$aKѨ5wZpKbwlRF!1GN26??;!wh4_0IR$1aٮYS,$XѮ[\h1F𯖳3?.|hԆ й\i(dFڮ80]6	
ɘ "ZH⬊fua>YHZD
i
;rJyʛҘN+`yK_ǯ[}8sX[/i\
uHM EhQE&0BOzNQUwwň>6vhT;TYó"[ouת1zz}|ToBm+jө[&!i7G5xX=)4z~Z~oy˔.+I(OOe.K/4[̄IڄDLXUCA/a׺V]7__BedW17pѰ$BO`L!s\-IOHs$O̓B!u!>%:@XA{}x?MkMt?&^+U fqMBYG"o *;gMqta HU(8^#(`7ATX*Rtdkh!ąT`(t&p`c`l2sk*@8s\lԺvL:r>t~}<[F