1 INFO-VAX	Thu, 26 Oct 2006	Volume 2006 : Issue 589       Contents: Alpha DS25 and Itanium Re: Backup restoration- Banking security Was:PARSEC: SPAMMERS (not !) 1 Re: Banking security Was:PARSEC: SPAMMERS (not !) 1 Re: Banking security Was:PARSEC: SPAMMERS (not !) & Binary data in multiple adjacent files* Re: Binary data in multiple adjacent files* Re: Binary data in multiple adjacent files* Re: Binary data in multiple adjacent files* Re: Binary data in multiple adjacent files Re: DS10L surgery report freevms 0.2.15 released  Re: freevms 0.2.15 released ! Re: Itanium systems and DECCAMPUS ! Re: Itanium systems and DECCAMPUS ! Re: Itanium systems and DECCAMPUS ! Re: Itanium systems and DECCAMPUS ! Re: Itanium systems and DECCAMPUS ! Re: Itanium systems and DECCAMPUS ! Re: Itanium systems and DECCAMPUS  Re: LAN failover Re: LAN failoverI LSE Language Definitions for Fortran 90 or Fortran 95 Free-Format Source? M Re: LSE Language Definitions for Fortran 90 or Fortran 95 Free-Format Source?  Re: NEW Email address  Perl / VMS Question ) Re: Powerstorm vs Elsa Gloria video card? ) Re: Powerstorm vs Elsa Gloria video card? ) Re: Powerstorm vs Elsa Gloria video card? ) Re: Powerstorm vs Elsa Gloria video card?  Spit Brook location  Re: Spit Brook location  Re: Spit Brook location / Re: Terminal Server Manager and Decnet Phase IV / Re: Terminal Server Manager and Decnet Phase IV / Re: Terminal Server Manager and Decnet Phase IV  TSZ07 Magnetic Tape Drive  Re: TSZ07 Magnetic Tape Drive  Re: TSZ07 Magnetic Tape Drive O Re: Two node ES80 cluster to EVA5000, RWSCS and high CR_WAITS on VMS$VAXcluster O Re: Two node ES80 cluster to EVA5000, RWSCS and high CR_WAITS on VMS$VAXcluster O Re: Two node ES80 cluster to EVA5000, RWSCS and high CR_WAITS on VMS$VAXcluster P Re: Two node ES80 cluster to EVA5000, RWSCS and high CR_WAITS on VMS$VAXcluster   F ----------------------------------------------------------------------  % Date: Thu, 26 Oct 2006 10:09:11 -0400 + From: Chuck Aaron <caaron@ceris.purdue.edu>  Subject: Alpha DS25 and Itanium 5 Message-ID: <ehqfi7$9e7$1@mailhub227.itcs.purdue.edu>   G Has anyone replaced their DS25 with an Itanium yet running Openvms? If  ! so, would you mind to share which C Itanium server you decided on and let me know how it's working out?    Thanks,  Chuck    ------------------------------    Date: 25 Oct 2006 22:55:01 -0700! From: "alok" <alok.net@gmail.com>  Subject: Re: Backup restoration C Message-ID: <1161842101.622564.252490@i42g2000cwa.googlegroups.com>    Thanks a lot guys.  
 Ade wrote:. > "alok" <alok.net@gmail.com> wrote in message? > news:1161691861.910247.220730@b28g2000cwb.googlegroups.com...  > > HiK > > I have three disks on my VAX system, one of which holds Ingres, another J > > is some s/ws, and the third the system. If a new disk needs to replaceK > > one of the current disks,  do all images need to be replaced on all the F > > 3 disks or only on the replaced ones?  is there any s/w dependency > > issue here?  > >  >  > Hi,  > N > Providing you have a CURRENT (and preferably standalone) IMAGE backup of theB > disk in question, you should be able to simply do $ backup/imageF > tape:saveset.sav/save disk: (possibly one or two more qualifiers ??) >  > Ade    ------------------------------  % Date: Thu, 26 Oct 2006 15:25:46 +0200 / From: Paul Sture <paul.sture.nospam@hispeed.ch> 6 Subject: Banking security Was:PARSEC: SPAMMERS (not !)J Message-ID: <paul.sture.nospam-EF719B.15254626102006@mac.sture.homeip.net>  9 In article <JuGdnSvb0q3ICaLYnZ2dnUVZ_oudnZ2d@libcom.com>, )  Dave Froble <davef@tsoft-inc.com> wrote:    > Paul Sture wrote:  > I > > Yep. Drifting slightly from the main topic, this brings us to to the  J > > topic of identity theft. That happened to me nearly 20 years ago, and J > > VAXMAN's recent post here about similar brought back painful memories. > > M > > (Report a substantial loss of money to your bank, get asked to report it  L > > to the police, then find yourself in an interview room being accused of G > > doing it with the aid of an accomplice. The bank's handling of the  M > > matter made me seriously wonder if it was an "inside job" that they were   > > trying to cover up.) > I > This one's easy to see.  The banking system is fatally flawed.  Anyone  K > with your bank routing number and checking account number can initiate a  K > transfer.  But the same information is on every check you write and send  H > to others.  If the bank can get you to not file a claim against them, , > then they don't have to replace the funds.  I I don't know enough about the US banking system to comment, but the last  C time I popped a letter into my local post box, there was a warning  E telling you not to drop any bill payments in there, and instead take  G them to the main post office. Apparently that scam is where fraudsters  I substitute their own paying in slip, and you know nothing about it until  1 a payment reminder comes through the post. Nasty.   H And some people wonder why I am reluctant to use a credit card over the F net. Yes, with a reputable company, I might be persuaded that a given B transaction may be secure, but what happens months later if those ? details are still lying around on a system which happens to be   vulnerable to outside attack?    --  
 Paul Sture   ------------------------------    Date: 26 Oct 2006 09:19:07 -0700, From: "Stanley F. Quayle" <quayle@pobox.com>: Subject: Re: Banking security Was:PARSEC: SPAMMERS (not !)C Message-ID: <1161879547.595960.179800@m73g2000cwd.googlegroups.com>    Paul Sture wrote: I > And some people wonder why I am reluctant to use a credit card over the G > net. Yes, with a reputable company, I might be persuaded that a given C > transaction may be secure, but what happens months later if those @ > details are still lying around on a system which happens to be > vulnerable to outside attack?   C Here in the US, personal liability for credit card fraud is zero by 0 law.  So I use my credit card freely on the Net.  C There's no such law covering debit cards, so I'm much more cautious 
 with them.   ------------------------------  + Date: Thu, 26 Oct 2006 17:36:36 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply): Subject: Re: Banking security Was:PARSEC: SPAMMERS (not !)$ Message-ID: <ehqrn4$mp8$2@online.de>  
 In articleD <paul.sture.nospam-EF719B.15254626102006@mac.sture.homeip.net>, Paul- Sture <paul.sture.nospam@hispeed.ch> writes:    ; > In article <JuGdnSvb0q3ICaLYnZ2dnUVZ_oudnZ2d@libcom.com>, + >  Dave Froble <davef@tsoft-inc.com> wrote:  >  > > Paul Sture wrote:  > > K > > > Yep. Drifting slightly from the main topic, this brings us to to the  L > > > topic of identity theft. That happened to me nearly 20 years ago, and L > > > VAXMAN's recent post here about similar brought back painful memories. > > > O > > > (Report a substantial loss of money to your bank, get asked to report it  N > > > to the police, then find yourself in an interview room being accused of I > > > doing it with the aid of an accomplice. The bank's handling of the  O > > > matter made me seriously wonder if it was an "inside job" that they were   > > > trying to cover up.) > > K > > This one's easy to see.  The banking system is fatally flawed.  Anyone  M > > with your bank routing number and checking account number can initiate a  M > > transfer.  But the same information is on every check you write and send  J > > to others.  If the bank can get you to not file a claim against them, . > > then they don't have to replace the funds.  E This is certainly not the case in Europe.  Anyone who sends me money  G needs this information.  It is NOT enough for him to get money from my   account.  J > And some people wonder why I am reluctant to use a credit card over the H > net. Yes, with a reputable company, I might be persuaded that a given D > transaction may be secure, but what happens months later if those A > details are still lying around on a system which happens to be   > vulnerable to outside attack?   H Yes, but even if you use it otherwise, the details can be on a computer 
 somewhere.  F Actually, this is not a very big problem.  Most credit-card purchases E online require only the data readable on the card.  This is known to  I many people, in particular anyone whom you have used your credit card to  G pay.  (Some, but not all, transactions require the extra number on the  H back.  This might be unknown to some, but not all, people whom you have G paid with your card.)  However, unless they have a physical signature,  H the credit-card company is NOT required to honour the transaction.  So, H when you get your bill, before you pay it, check all the transactions.  I If there is one which you didn't initiate, as the credit-card company to  B cancel it.  By law, they can only refuse if they can prove (via a I signature) that you did initiate the transaction (the burden of proof is  	 on them).   B Yes, credit-card fraud does happen, but only because not everyone  notices the damage in time.    ------------------------------  % Date: Thu, 26 Oct 2006 02:49:30 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> / Subject: Binary data in multiple adjacent files , Message-ID: <45405A54.5AF48084@teksavvy.com>  D I have a program that was setup to provide elevation data for all ofH australia new zealand from 2 large datasets that covered that territory.C The application kpet the 2 files opened and did relative reads into G those binary files to obtain the required data (and cached those record H so that if the next request was for a nearby cell, it would save an IO).  D Now however, updated data comes in much smaller files, each coveringG only 1 lat by 1 long. Each file has 1201 * 1201 values, each value is D 2 bytes. (basically elevation in metres for a 92 square metre area). Each file is roughly 2.8 megs.    G (This is degraded information due to the US government feeling it could E be used by terrorists, but it is still better than what was available B before. The SRTM data is precise to about 1 square metre, and someG sections have been released to 30 square metre accuracy, but most is at 
 92 metre).    C Switching to this would mean that my application would need to open E hundreds of files to cover australia for instance, so this introduces D scalability issues. And since I need to review the code, I figured I& should perhaps consider other methods.  D For my current needs, I need to cover from 45 to 47 north, and from0 -73 to -76, which gives me th following files:     N46W076  N46W075  N46W074  N45W076  N45W075  N46W074   E N46W074 covers 46N 74W at the lower left corner to 47N 73W at the  upper right corner.     H Looking at $CRMPSC, it appears I need to do a RMS $OPEN on each file and/ provide the channel associated with each file.    ' So this would eat into the FILLM quota.   A Now, if i just do the $CRMPSC to map a whole file to some virtual F address in my process space, would this consume 2,8 meg of working setC right away ? Or does VMS just allocate virual memory that is marked F invalid, and only when I try to access a few bytes at a location wouldH VMS load a single page from the actual file that contain those 2 bytes I asked for ?     H The nature of my application means that as I follow a route, I read dataG progressing in one direction, and the odds of having to go back are low F (but not nil). So, blocks containing elevation data read early on are P unlikely to be needed again once I have moved on to points outside those blocks.  G Does VMS provide a means where I can specify that I want at max 2 pages G from a global section to mapped to my working set at ay point in time ? G (this way, when I request a shortword located in a different block, the E system would automatically unmap the oldest block mapped to that file 6 and use that memory to map the new block to the file).      G The other thing I am thinking about is having say 5 files opened at any G point in time, and whevere I access a file, I update some counter based F on the progress in my processing. So when I need to open a new file, IH would then close the file which as the lowest value in the counter (file! that been iddle for the longest).   F So, for every point , I would need to create a file name, and check inG the list of currently opened file if that file has already been opened. - But this would allow me to scale to any size.   3 Also, if I run this on 8.3 (Alpha):  (written in C)   F Say I do an fseek to the 824th byte in the file and then read 2 bytes.> Then, I do an fseek to the 1020th byte and also read 2 bytes.    H Is the underlying IO system smart enough to know that both are accessingH data from the same physical block and use the cache system ? Or is the C: file IO so screwed up that it would bypass this facility ?   ------------------------------    Date: 26 Oct 2006 10:21:27 +0200/ From: huber@NIRWANA-mppmu.mpg.de (Joseph Huber) 3 Subject: Re: Binary data in multiple adjacent files + Message-ID: <XNcTfaOT7weF@vms.mppmu.mpg.de>   \ In article <45405A54.5AF48084@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes:  F > For my current needs, I need to cover from 45 to 47 north, and from2 > -73 to -76, which gives me th following files: >  >  > N46W076  N46W075  N46W074  > N45W076  N45W075  N46W074  > G > N46W074 covers 46N 74W at the lower left corner to 47N 73W at the  > upper right corner.  >  > J > Looking at $CRMPSC, it appears I need to do a RMS $OPEN on each file and1 > provide the channel associated with each file.   > ) > So this would eat into the FILLM quota.  > C > Now, if i just do the $CRMPSC to map a whole file to some virtual H > address in my process space, would this consume 2,8 meg of working setE > right away ? Or does VMS just allocate virual memory that is marked H > invalid, and only when I try to access a few bytes at a location wouldJ > VMS load a single page from the actual file that contain those 2 bytes I
 > asked for ?   I I have the impression You confuse working set with virtual address space: M the mapped blocks will become part of the working set when needed, i.e. when  H it is accessed; and only the accessed pages will be loaded into physical memory. O So I would not worry about the working set, be prepared for the virtual address M space of the process being large enough (although 6*2.8 MB is not really big, > or ?), and have Wsextent as large as possible to avoid paging.   > J > The nature of my application means that as I follow a route, I read dataI > progressing in one direction, and the odds of having to go back are low H > (but not nil). So, blocks containing elevation data read early on are R > unlikely to be needed again once I have moved on to points outside those blocks. > I > Does VMS provide a means where I can specify that I want at max 2 pages I > from a global section to mapped to my working set at ay point in time ? I > (this way, when I request a shortword located in a different block, the G > system would automatically unmap the oldest block mapped to that file 8 > and use that memory to map the new block to the file).  K No , You tell in the $CRMPSC call, how much it maps, and the access pattern F defines which pages are loaded into physical memory (the working set).  L But Your program could explicitly select the page-range mapped by specifyingO the starting vbn and number of blocks to map: so instead of creating 2.8 MB per & file, map only the few blocks needed.       I > The other thing I am thinking about is having say 5 files opened at any I > point in time, and whevere I access a file, I update some counter based H > on the progress in my processing. So when I need to open a new file, IJ > would then close the file which as the lowest value in the counter (file# > that been iddle for the longest).   O But be aware, when opening and closing files, the virtual address space created L with $CRMPSC (assuming an EXPREG mapping) is only released if it is the last one created.L So keep in mind the virtual address space will grow with the number of files opened, and not shrink back.   5 > Also, if I run this on 8.3 (Alpha):  (written in C) H > Say I do an fseek to the 824th byte in the file and then read 2 bytes.@ > Then, I do an fseek to the 1020th byte and also read 2 bytes. J > Is the underlying IO system smart enough to know that both are accessingJ > data from the same physical block and use the cache system ? Or is the C< > file IO so screwed up that it would bypass this facility ?  O Caching happens on the XQP level, not in RMS or the C I/O software. So the disk  block is eventually cached. J The "cache" in C I/O is its block buffer, and depends on the parameters inM the open call or the RMS defaults (see SET/SHOW RMS_DEFAULTS). So in the case J of a single block, it will be "cached" if no read for a different block is between the 2 fseeks/reads.     --  @    Joseph Huber , Muenchen,Germany:  http://www.huber-joseph.de/   ------------------------------  % Date: Thu, 26 Oct 2006 08:36:18 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> 3 Subject: Re: Binary data in multiple adjacent files : Message-ID: <zuedncySfdbQNt3YnZ2dnUVZ_tKdnZ2d@comcast.com>   JF Mezei wrote:   F > I have a program that was setup to provide elevation data for all ofJ > australia new zealand from 2 large datasets that covered that territory.E > The application kpet the 2 files opened and did relative reads into I > those binary files to obtain the required data (and cached those record J > so that if the next request was for a nearby cell, it would save an IO). > F > Now however, updated data comes in much smaller files, each coveringI > only 1 lat by 1 long. Each file has 1201 * 1201 values, each value is F > 2 bytes. (basically elevation in metres for a 92 square metre area).! > Each file is roughly 2.8 megs.   > I > (This is degraded information due to the US government feeling it could G > be used by terrorists, but it is still better than what was available D > before. The SRTM data is precise to about 1 square metre, and someI > sections have been released to 30 square metre accuracy, but most is at  > 92 metre). >  > E > Switching to this would mean that my application would need to open G > hundreds of files to cover australia for instance, so this introduces F > scalability issues. And since I need to review the code, I figured I( > should perhaps consider other methods. > F > For my current needs, I need to cover from 45 to 47 north, and from2 > -73 to -76, which gives me th following files: >  >  > N46W076  N46W075  N46W074  > N45W076  N45W075  N46W074  > G > N46W074 covers 46N 74W at the lower left corner to 47N 73W at the  > upper right corner.  >  > J > Looking at $CRMPSC, it appears I need to do a RMS $OPEN on each file and1 > provide the channel associated with each file.   > ) > So this would eat into the FILLM quota.  > C > Now, if i just do the $CRMPSC to map a whole file to some virtual H > address in my process space, would this consume 2,8 meg of working setE > right away ? Or does VMS just allocate virual memory that is marked H > invalid, and only when I try to access a few bytes at a location wouldJ > VMS load a single page from the actual file that contain those 2 bytes I
 > asked for ?  >   I Virtual memory is just that!  Things only need to occupy physical memory  G when you try to read or write them.  Anything you are not using can be  I pushed out to make room for something else.  So it won't necessarily eat  E your working set which is physical memory.  There are also limits on  G virtual memory (PAGEFILEQUO) and a real constraint.  The later is that  G performance will suck if if each memory reference is to a new page and  # you don't have a large working set.  > J > The nature of my application means that as I follow a route, I read dataI > progressing in one direction, and the odds of having to go back are low H > (but not nil). So, blocks containing elevation data read early on are R > unlikely to be needed again once I have moved on to points outside those blocks. > I > Does VMS provide a means where I can specify that I want at max 2 pages I > from a global section to mapped to my working set at ay point in time ? I > (this way, when I request a shortword located in a different block, the G > system would automatically unmap the oldest block mapped to that file 8 > and use that memory to map the new block to the file). > A Not that I know of.  Virtual memory is managed globally for your  E process.  When you need a new page (you requested a page not in your  B working set) an old page gets pushed out to make room.  The least 7 recently used page will normally be the one pushed out.  >  <snip>   ------------------------------    Date: 26 Oct 2006 10:22:57 -0500 From: briggs@encompasserve.org3 Subject: Re: Binary data in multiple adjacent files 3 Message-ID: <sycPycHTbIIW@eisner.encompasserve.org>   \ In article <45405A54.5AF48084@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes:F > I have a program that was setup to provide elevation data for all ofJ > australia new zealand from 2 large datasets that covered that territory.E > The application kpet the 2 files opened and did relative reads into I > those binary files to obtain the required data (and cached those record J > so that if the next request was for a nearby cell, it would save an IO). > F > Now however, updated data comes in much smaller files, each coveringI > only 1 lat by 1 long. Each file has 1201 * 1201 values, each value is F > 2 bytes. (basically elevation in metres for a 92 square metre area).! > Each file is roughly 2.8 megs.    , So a raster line is 1201*2 = 2402 bytes long  E > Switching to this would mean that my application would need to open G > hundreds of files to cover australia for instance, so this introduces F > scalability issues. And since I need to review the code, I figured I( > should perhaps consider other methods. > F > For my current needs, I need to cover from 45 to 47 north, and from2 > -73 to -76, which gives me th following files: >  >  > N46W076  N46W075  N46W074  > N45W076  N45W075  N46W074  > G > N46W074 covers 46N 74W at the lower left corner to 47N 73W at the  > upper right corner.  >  > J > Looking at $CRMPSC, it appears I need to do a RMS $OPEN on each file and1 > provide the channel associated with each file.   > ) > So this would eat into the FILLM quota.    Yes.  C > Now, if i just do the $CRMPSC to map a whole file to some virtual H > address in my process space, would this consume 2,8 meg of working set > right away ?  @ Mapping it doesn't eat any working set except, possibly, for theB pages devoted to the page table entries that were updated when the> section was mapped and which were, therefore, faulted into theC working set.  (Caveat:  my internals knowledge is VAX-era.  I don't D know whether Alpha process PTE's count against process working set).  8 > Or does VMS just allocate virual memory that is markedH > invalid, and only when I try to access a few bytes at a location wouldJ > VMS load a single page from the actual file that contain those 2 bytes I
 > asked for ?   9 It's just mapped.  It's not faulted into physical memory.   D Yes, the two bytes come into your working set when you try to accessF them, and encounter a hard page fault and VMS reads them into physical memory.   G You also get a whole "page fault cluster" faulted in.  Check the SYSGEN 9 "PFCDEFAULT" parameter or the "pfc" parameter to $CRMPSC.   @ This is roughly analogous to RMS multi-block count.  You read inG multiple pages at one go because contiguous transfers are traditionally 3 cheap and random seeks are traditionally expensive.   J > The nature of my application means that as I follow a route, I read dataI > progressing in one direction, and the odds of having to go back are low H > (but not nil). So, blocks containing elevation data read early on are R > unlikely to be needed again once I have moved on to points outside those blocks.  @ Ok.  So caching one block is a win.  But retaining old blocks inE the cache is near useless.  And read-ahead or read-behind [sic] might 
 be useful.  I > Does VMS provide a means where I can specify that I want at max 2 pages I > from a global section to mapped to my working set at ay point in time ? I > (this way, when I request a shortword located in a different block, the G > system would automatically unmap the oldest block mapped to that file 8 > and use that memory to map the new block to the file).   No.  Not that I know of.  I > The other thing I am thinking about is having say 5 files opened at any I > point in time, and whevere I access a file, I update some counter based H > on the progress in my processing. So when I need to open a new file, IJ > would then close the file which as the lowest value in the counter (file# > that been iddle for the longest).   E Sounds reasonable.  Though I'd probably go for 4 files as a max since I the worst case is when your path hits a 4 way corner between intersecting H regions and makes tiny circles around the intersection point.  Two filesC would be a reasonable alternative, handling the case of a path that @ weaves across a boundary repeatedly but thrashing in the case of> a path that winds around a border between three or more zones.  H > So, for every point , I would need to create a file name, and check inI > the list of currently opened file if that file has already been opened. / > But this would allow me to scale to any size.    Yes.  Sounds like a plan.   5 > Also, if I run this on 8.3 (Alpha):  (written in C)  > H > Say I do an fseek to the 824th byte in the file and then read 2 bytes.@ > Then, I do an fseek to the 1020th byte and also read 2 bytes.   H If you do an fseek to the 824th byte then the next seek should logicallyD be to the 826'th, 822'nd, 3226th, 3228th, 3230th or somewhere in the next file back.   @ [Assumes that your application visits each 92 meter grid section between consecutive waypoints]  B If you read in an 8192 byte page then you're likely to have two orB three complete raster lines in memory already.  That gives you theG grid sections immediately north, south, northeast, northwest, southeast > and southwest of your current location along with all the grid7 sections to your east and west to the edge of the file.   E If you read in a 512 byte block then the only other relevant data you E might have on hand are the grid sections to the east and west of your  current position.   J 1.  Do you want to read in additional pages so that you have ten or twenty raster lines in memory?   @ 2.  If you read in "clusters" to make this happen, are you going9 to be happy with read-ahead, reading the target and a few C subsequent blocks (to the south and east?) or are you going to want 1 some read-behind as well (to the north and west?)   ? One obvious implementation technique would be to ditch VMS page @ fault handling and roll your own with RMS reads or $QIO, reading@ in chunks of (let's say) 100 blocks at a time and always reading' chunks aligned on 100 block boundaries.   F 100 blocks = 51200 bytes ~= 24 raster lines.  Now if you're traversingF a file from north to south you can get away with 50 disk reads instead of 1201.  G And if you're sticking to one grid line all the way east to west across / the file, that's one disk read instead of four.   F West to east traversal along a line of latitude is the only case where6 RMS, C or VMS page fault handling can hope to compete.  = By aligning the reads on 100 block boundaries you tend to get F effective read-ahead whether your path runs north to south or south toA north.  (You don't want to read 100 blocks starting at block 2000 ? then 100 blocks starting at block 1999 then 100 blocks starting  at block 1998...)   J > Is the underlying IO system smart enough to know that both are accessingJ > data from the same physical block and use the cache system ? Or is the C< > file IO so screwed up that it would bypass this facility ?  I C file IO makes use of RMS buffer cache.  My concern is that C/RMS random G I/O is going to tend to fail to make use of the benefits available from > larger block sizes.  And the page fault cluster feature of VMSC memory management is going to fail to optimize access patterns that   proceed in reverse memory order.                                    ------------------------------  % Date: Fri, 27 Oct 2006 00:21:37 +0800  From: prep@prep.synonet.com 3 Subject: Re: Binary data in multiple adjacent files 0 Message-ID: <87ac3jjbe6.fsf@k9.prep.synonet.com>  / JF Mezei <jfmezei.spamnot@teksavvy.com> writes:   F > Now however, updated data comes in much smaller files, each coveringF > only 1 lat by 1 long. Each file has 1201 * 1201 values, each valueB > is 2 bytes. (basically elevation in metres for a 92 square metre( > area).  Each file is roughly 2.8 megs.  B You can get all of oz on 2 CDs with ease. Two niggles, first is itB is bigendian INT16. Second, is that the boundary row and colum are duplicated in each file.  F I intend to read in ALL the files and build a single, or a few, files.H That would be a lot simpler, ESP as I intend to do quite a bit of number crunching over it.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Thu, 26 Oct 2006 14:35:40 +0200 / From: Paul Sture <paul.sture.nospam@hispeed.ch> ! Subject: Re: DS10L surgery report J Message-ID: <paul.sture.nospam-C51AB1.14354026102006@mac.sture.homeip.net>  A In article <1161786520.531813.4650@h48g2000cwc.googlegroups.com>, *  "Rich Jordan" <jordan@ccs4vms.com> wrote:   > Paul Sture wrote: 4 > > In article <12js67l1fhfk8af@news.supernews.com>,B > >  "Island Computers, D B Turner" <dbturner@islandco.com> wrote: > > N > > > Buying a $199 Airconditioner and putting it and everything computer in a > > > closet would work best != > > > Vent it through a  hole in a cheap $28 Home Depot door.  > > > Everything good... > > >  > > 5 > > This was my solution a couple of hot summers ago:  > > 3 > > http://sture.homeip.net/images/hobbyist_vax.jpg  > > L > > Yes, there's a roof above, and the Ibook was indoors when rain came, but > > the VAX lived outside. > >  > > -- > > Paul Sture > H > Glad to  know I don't in fact have the last working (and in use) VR299 > monitor in the world ;)   H Unfortunately I don't have much space in my present abode, so I gave it 7 away. I now regret doing that; it was a fine system :-(    --  
 Paul Sture   ------------------------------    Date: 26 Oct 2006 02:53:21 -0700! From: "mariuz" <mapopa@gmail.com>   Subject: freevms 0.2.15 releasedB Message-ID: <1161856401.773151.21200@f16g2000cwb.googlegroups.com>   Changes:D A minimal and basic SMGRTL was started. New Bliss version, 0.210, is
 available.E The increase in version number is mainly due to the move to the 4.1.1  backend.E System services for capabilities and affinity were added, but are yet 	 untested. ' Exe$nampid now does privilege checking. - Bliss bug fixes are noted in the Bliss HOWTO. . System service dclast did not set access mode.B In i386, PAL was missing user mode in PSL when going to user mode.   Howto use it with qemuD Download the disk images from the site  links  (under the USE link )  5 http://www.systella.fr/~bertrand/FreeVMS/indexGB.html 6 wget http://www.systella.fr/~bertrand/FreeVMS/a.img.gz6 wget http://www.systella.fr/~bertrand/FreeVMS/c.img.gz   gunzip a.img.gz  gunzip c.img.gz   1 qemu -fda a.img -hda c.img -boot a -monitor stdio    Select the default grub option  / You can read more in the USE link (on the site)    ------------------------------  % Date: Thu, 26 Oct 2006 06:07:49 -0700 * From: "Tom Linden" <tom@kednos-remove.com>$ Subject: Re: freevms 0.2.15 released) Message-ID: <op.th05fbuotte90l@hyrrokkin>   D On Thu, 26 Oct 2006 02:53:21 -0700, mariuz <mapopa@gmail.com> wrote:  
 > Changes:F > A minimal and basic SMGRTL was started. New Bliss version, 0.210, is > available.G > The increase in version number is mainly due to the move to the 4.1.1 
 > backend.G > System services for capabilities and affinity were added, but are yet  > untested. ) > Exe$nampid now does privilege checking. / > Bliss bug fixes are noted in the Bliss HOWTO. 0 > System service dclast did not set access mode.D > In i386, PAL was missing user mode in PSL when going to user mode. >  > Howto use it with qemuF > Download the disk images from the site  links  (under the USE link ) > 7 > http://www.systella.fr/~bertrand/FreeVMS/indexGB.html 8 > wget http://www.systella.fr/~bertrand/FreeVMS/a.img.gz8 > wget http://www.systella.fr/~bertrand/FreeVMS/c.img.gz >  > gunzip a.img.gz  > gunzip c.img.gz  > 3 > qemu -fda a.img -hda c.img -boot a -monitor stdio  >   > Select the default grub option > 1 > You can read more in the USE link (on the site)  >   $ What is the implementation language?     --  E Using Opera's revolutionary e-mail client: http://www.opera.com/mail/    ------------------------------   Date: 26 Oct 2006 13:47:26 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)* Subject: Re: Itanium systems and DECCAMPUS+ Message-ID: <4qbsjeFm0u6qU1@individual.net>   , In article <ehqdkn$o1i$1@south.jnrs.ja.net>,! 	david20@alpha2.mdx.ac.uk writes: H > I have just been informed that there is no equivalent to the DECCAMPUSP > agreement which provides VMS layered product software licenses for educational > institutions for Itanium. N > Hence it looks like VMS on Itanium is likely to be rather more expensive for- > Educational institutions than VMS on Alpha.  > : > Maybe the last nail in the coffin for VMS in education ? >  >   K I think most schools dropped academic use of VMS (and thus DECCAMPUS) quite J some time ago so this is merely responding to a program they no longer seeI as needed.  Of course, the right solution would have been to find out why 5 no one was using it and try to win those people back.   J We are still, administratively, a VMS shop.  But, while there is still oneJ VMS machine with accounts for all the students on it I would guess none ofL them ever logs into it (or knows how, or even knows it exists!)  Some of theM diehard faculty may, but since it no longer supports outside email, it really J has no functional purpose any more.  I would imagine the last academic VMSH machines to see actual student use here were mine and they went down forH the last time at the end of last semester.  I still think keeping VMS inJ academia was a good idea, but it was not a battle I could fight by myself.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------    Date: 26 Oct 2006 07:00:27 -0700 From: bob@instantwhip.com * Subject: Re: Itanium systems and DECCAMPUSC Message-ID: <1161871227.915978.305050@m73g2000cwd.googlegroups.com>   7 you do not need to fight it ... one day their will be a 2 major break in into university data that will make them rethink that position ...   ------------------------------    Date: 26 Oct 2006 08:14:51 -0700 From: etmsreec@yahoo.co.uk* Subject: Re: Itanium systems and DECCAMPUSC Message-ID: <1161875691.461812.185880@f16g2000cwb.googlegroups.com>   G There was discussion of this at the VMS TUD in London a week or so ago. ?  The UK offering was, traditionally, different to that in other 9 countries so it is presently being discussed and debated.    I've mailed David directly too.    Steve    bob@instantwhip.com wrote:9 > you do not need to fight it ... one day their will be a 4 > major break in into university data that will make  > them rethink that position ...   ------------------------------    Date: 26 Oct 2006 08:24:47 -0700) From: "DaveG" <david.gudewicz@abbott.com> * Subject: Re: Itanium systems and DECCAMPUSC Message-ID: <1161876287.649598.114240@h48g2000cwc.googlegroups.com>    etmsreec@yahoo.co.uk wrote: I > There was discussion of this at the VMS TUD in London a week or so ago. A >  The UK offering was, traditionally, different to that in other ; > countries so it is presently being discussed and debated.  > ! > I've mailed David directly too.  >   . >From the HPTF held in Houston, TX last month:   CIP - Campus Investment Program C free or steep discounts for OpenVMS for teaching and research only.  (Integrity only)  > All campus software programs found at: www.hp.com/go/education   Hope this helps.   Dave...    ------------------------------    Date: 26 Oct 2006 08:43:51 -0700) From: "DaveG" <david.gudewicz@abbott.com> * Subject: Re: Itanium systems and DECCAMPUSB Message-ID: <1161877431.306143.142600@e3g2000cwe.googlegroups.com>   DaveG wrote: > etmsreec@yahoo.co.uk wrote: K > > There was discussion of this at the VMS TUD in London a week or so ago. C > >  The UK offering was, traditionally, different to that in other = > > countries so it is presently being discussed and debated.  > > # > > I've mailed David directly too.  > >  > 0 > >From the HPTF held in Houston, TX last month: > ! > CIP - Campus Investment Program E > free or steep discounts for OpenVMS for teaching and research only.  > (Integrity only) > @ > All campus software programs found at: www.hp.com/go/education >  > Hope this helps. > 	 > Dave...   D Found this.  More details about edu prrgrams for OpenVMS and others:  ? http://h10038.www1.hp.com/hed/prog_hedsoftware.asp?agencyid=136      Dave...    ------------------------------  + Date: Thu, 26 Oct 2006 16:09:29 +0000 (UTC)  From: david20@alpha2.mdx.ac.uk* Subject: Re: Itanium systems and DECCAMPUS, Message-ID: <ehqmjp$qk3$2@south.jnrs.ja.net>  o In article <1161876287.649598.114240@h48g2000cwc.googlegroups.com>, "DaveG" <david.gudewicz@abbott.com> writes:  >  >etmsreec@yahoo.co.uk wrote:J >> There was discussion of this at the VMS TUD in London a week or so ago.B >>  The UK offering was, traditionally, different to that in other< >> countries so it is presently being discussed and debated. >>" >> I've mailed David directly too. >> > / >>From the HPTF held in Houston, TX last month:  >   >CIP - Campus Investment ProgramD >free or steep discounts for OpenVMS for teaching and research only. >(Integrity only)  >   * Is this purely for teaching and research ?9 The UK DECCAMPUS program covered other uses in Education. H The systems I am looking at replacing are both used as central mailhubs L - though one (a VMS cluster) is also still used for some teaching of Oracle  etc   
 David Webb Security team leader CCSS Middlesex University    ? >All campus software programs found at: www.hp.com/go/education  >  >Hope this helps.  >  >Dave... >    ------------------------------  + Date: Thu, 26 Oct 2006 17:37:31 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)* Subject: Re: Itanium systems and DECCAMPUS$ Message-ID: <ehqror$mp8$3@online.de>  E In article <ehqdkn$o1i$1@south.jnrs.ja.net>, david20@alpha2.mdx.ac.uk  writes:   H > I have just been informed that there is no equivalent to the DECCAMPUSP > agreement which provides VMS layered product software licenses for educational > institutions for Itanium. N > Hence it looks like VMS on Itanium is likely to be rather more expensive for- > Educational institutions than VMS on Alpha.  > : > Maybe the last nail in the coffin for VMS in education ?  G Are you aware of any educational institution making significant use of  
 DECcampus?   ------------------------------    Date: 25 Oct 2006 23:15:05 -0700/ From: "Volker Halle" <volker_halle@hotmail.com>  Subject: Re: LAN failover C Message-ID: <1161843305.028640.269950@h48g2000cwc.googlegroups.com>   D The LLB0 MAC address does not change during failover, but the switchD needs to find out, that this MAC address is now connected to anotherD port on the switch. This needs at least ONE packet to be transmittedC via LLB0: and the newly activated LAN interface to the switch port.   E OpenVMS LAN engineering is planning to address this issue in a future  release.   Volker.    ------------------------------  % Date: Thu, 26 Oct 2006 03:11:09 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>  Subject: Re: LAN failover , Message-ID: <45405F66.DD205150@teksavvy.com>   Volker Halle wrote:  > F > The LLB0 MAC address does not change during failover, but the switchF > needs to find out, that this MAC address is now connected to another > port on the switch.     6 Since they are connected to 2 switches, consider this:  F SWITCH-1 knows that ethernet address X is connected to its port 8. AllM other switches might know that ethernet address X is accessible via SWITCH-1.   F So now, that node cuts off the connection to port 8 on SWITCH-1 and is9 using port 6 on SWITCH-2 for the same ethernet address X.   F Until things change within the switches, the switches will continue toA send packets destined for X to Switch-1 and switch one will still ) deliver packets destined to X via port 8.   H If your switches are configured with VLAN capabilities, you need to makeC sure that SWITCH-2's port 6 is configured to be in the same VLAN as  SWITCH-1 port 8.    F And I would call CISCO or check their manuals to find out exactly whatE happens in your environment when you essentially take a wire out of a H port on one switch and plug it into another switch. You need to find outB what sort of packets cause all switches to update their tables and2 forward packets to the new port on the new switch.   ------------------------------    Date: 26 Oct 2006 08:55:01 -0700( From: "Craig Dedo" <Dedo.Craig@mayo.edu>R Subject: LSE Language Definitions for Fortran 90 or Fortran 95 Free-Format Source?B Message-ID: <1161878101.795067.189900@e3g2000cwe.googlegroups.com>  	 Everyone: ?     Does anyone have a Language Sensitive Editor (LSE) language E definition available for use with Fortran 90, Fortran 95, and Fortran G 2003 free-format source code?  I read through the reference manuals for E LSE 5.0 in DECset 12.7.  The Fortran language definition shipped with E that version only supports the old-style fixed-format source (F77 and E earlier) and the non-standard tab-format source.  It does not support F the free-format source that has been part of ISO/ANSI standard Fortran since 1991.   B     Has anyone created an additional language definition file thatC supports free-format source for Fortran 90, fortran 95, and Fortran % 2003?  If so, where can I get a copy?   E     Recommendation to HP OpenVMS Engineering:  You may wish to update 0 LSE to recognize free-format source for Fortran.     Craig T. Dedo, VMS Consultant $ Mayo Central Lab for Clinical Trials& Superior Drive Support Center 2-110-26 3050 Superior Drive NW Rochester, MN   55905-1700 Voice Phone:  (507) 538-4636 Fax Phone:    (507) 284-0615 Mobile Phone:  (414) 412-5869  E-mail:  <Dedo.Craig@mayo.edu>   ------------------------------  % Date: Thu, 26 Oct 2006 08:54:07 -0700 * From: "Tom Linden" <tom@kednos-remove.com>V Subject: Re: LSE Language Definitions for Fortran 90 or Fortran 95 Free-Format Source?) Message-ID: <op.th1c4hxptte90l@hyrrokkin>   K On Thu, 26 Oct 2006 08:55:01 -0700, Craig Dedo <Dedo.Craig@mayo.edu> wrote:    > Everyone: A >     Does anyone have a Language Sensitive Editor (LSE) language G > definition available for use with Fortran 90, Fortran 95, and Fortran I > 2003 free-format source code?  I read through the reference manuals for G > LSE 5.0 in DECset 12.7.  The Fortran language definition shipped with G > that version only supports the old-style fixed-format source (F77 and G > earlier) and the non-standard tab-format source.  It does not support H > the free-format source that has been part of ISO/ANSI standard Fortran
 > since 1991.  > D >     Has anyone created an additional language definition file thatE > supports free-format source for Fortran 90, fortran 95, and Fortran ' > 2003?  If so, where can I get a copy?   J There are Lisp packages for emacs, of course, you would need to have emacs
 on OpenVMS > G >     Recommendation to HP OpenVMS Engineering:  You may wish to update 2 > LSE to recognize free-format source for Fortran. >  >  > Craig T. Dedo, VMS Consultant & > Mayo Central Lab for Clinical Trials( > Superior Drive Support Center 2-110-26 > 3050 Superior Drive NW > Rochester, MN   55905-1700 > Voice Phone:  (507) 538-4636 > Fax Phone:    (507) 284-0615 > Mobile Phone:  (414) 412-5869   > E-mail:  <Dedo.Craig@mayo.edu> >        --  E Using Opera's revolutionary e-mail client: http://www.opera.com/mail/    ------------------------------  + Date: Thu, 26 Oct 2006 17:31:35 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) Subject: Re: NEW Email address$ Message-ID: <ehqrdn$mp8$1@online.de>  C In article <1161867279.742731.173510@k70g2000cwa.googlegroups.com>,  etmsreec@yahoo.co.uk writes:    I > I thought that non-deliverable mail would either get sent to postmaster   > (whoever that was defined to)   H Should be.  However, the postmaster then has to separate the wheat from  the chaff there.  ( > or would be bounced back with the SMTP > spam management?  I Yes.  Two problems here.  One is that so many undeliverable messages are  D bouncing that YOUR ISP thinks that YOU are spamming, since his SMTP H relay server is getting so many emails from you.  The other is that the G From: addresses are probably dummy addresses, so the mail then bounces   back to your postmaster.   ------------------------------  % Date: Thu, 26 Oct 2006 07:17:48 -0400 + From: Chuck Aaron <caaron@ceris.purdue.edu>  Subject: Perl / VMS Question5 Message-ID: <ehq5gr$48a$1@mailhub227.itcs.purdue.edu>   H I currently have the following directories under my perl_root and under F the cpan directory. What are the commands please to install cpan so I 5 can use XML::SIMPLE on VMS 7.3-2 running perl 5.6.0.?    Thank you in advance.    Chuck    SYSTEM> dird cpan*   Directory PERL_ROOT:[LIB]   5 CPAN.DIR;1                 1  21-MAR-2001 09:44:59.27 5 CPAN.PM;1                295   2-MAR-2000 17:52:52.00    Total of 2 files, 296 blocks.  SYSTEM>    SYSTEM> dird [.cpan...]    Directory PERL_ROOT:[LIB.CPAN]  5 FIRSTTIME.PM;1            31   4-FEB-2000 08:18:44.00 5 NOX.PM;1                   2   4-FEB-2000 08:18:44.00    Total of 2 files, 33 blocks. SYSTEM>    ------------------------------  % Date: Thu, 26 Oct 2006 07:51:55 +0200 ( From: "Rudolf Wingert" <win@fom.fgan.de>2 Subject: Re: Powerstorm vs Elsa Gloria video card?7 Message-ID: <004801c6f8c2$d7d62f80$994614ac@domina.fom>    Hello,  I If this Powerstorm ist a 4D20, then it will be a better card. You can use K (other then Elsa Gloria) Pseudo- and True-color simultaneous. With the Elsa 3 you have to come to a decision during DECW$STARTUP.    Best regards R. Wingert    ------------------------------  % Date: Thu, 26 Oct 2006 07:14:13 -0400 / From: "William Webb" <william.w.webb@gmail.com> 2 Subject: Re: Powerstorm vs Elsa Gloria video card?I Message-ID: <8660a3a10610260414k2f85dc92p845895f71b8e06d4@mail.gmail.com>   4 On 10/26/06, Rudolf Wingert <win@fom.fgan.de> wrote: > Hello, > K > If this Powerstorm ist a 4D20, then it will be a better card. You can use M > (other then Elsa Gloria) Pseudo- and True-color simultaneous. With the Elsa 5 > you have to come to a decision during DECW$STARTUP.  >  > Best regards R. Wingert  >  >   = Has anyone ever gotten other cards besides the three that are 1 "officially supported" to run on a DS10 or DS10L?    Thanks,    WWWebb --     Ajilon Consulting  Site resident at Quest Diagnostics = first.x.last@f$edit(contents of previous line,"COMPRESS").com    ------------------------------  % Date: Thu, 26 Oct 2006 09:59:26 -0400 8 From: Stephen Hoffman <Hoff@HoffmanLabs-RemoveThis-.Org>2 Subject: Re: Powerstorm vs Elsa Gloria video card?) Message-ID: <ehqf01$2kr9$1@pyrite.mv.net>    Rudolf Wingert wrote:  > Hello, > K > If this Powerstorm ist a 4D20, then it will be a better card. You can use M > (other then Elsa Gloria) Pseudo- and True-color simultaneous. With the Elsa 5 > you have to come to a decision during DECW$STARTUP.   C    The PowerStorm 3D30 is the PBXGB-AA, the PowerStorm 4D20 is the   PBXGB-CA.  The two are related.   I    As for whether or not any particular graphics controller works -- and  I if you're asking the sorts of unsupported "does it work?" questions that  C tend to appear here -- try it.   (Follow the usual anti-static and  A grounding and working-around-electrical equipment rules to avoid  ? potential hardware or human damage or death, of course.)   The  F PowerStorm 3D30 and the PowerStorm 4D20 are fast cards, but mayor may E not synch with newer LCD displays.  I've connected a PowerStorm 3D30  G with a TFT2025 series, and it generated artifact; the card I was using  ( seemingly wasn't very good at that task.   ------------------------------  % Date: Fri, 27 Oct 2006 00:08:46 +0800  From: prep@prep.synonet.com 2 Subject: Re: Powerstorm vs Elsa Gloria video card?0 Message-ID: <87iri7jbzl.fsf@k9.prep.synonet.com>  : Stephen Hoffman <Hoff@HoffmanLabs-RemoveThis-.Org> writes:  I >    The PowerStorm 3D30 is one of the fastest 2D cards around.  (Still!)   ? >    I haven't tried any of these cards on the DIGITAL Personal @ > Workstation series, however.  (The ATI Radeon 7500 driver does+ > require an EV6 or later system, however.)   # I have a 40 in a 600au, works fine.    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Fri, 27 Oct 2006 00:24:26 +0800  From: prep@prep.synonet.com  Subject: Spit Brook location0 Message-ID: <8764e7jb9h.fsf@k9.prep.synonet.com>  F The SE buildings are NW of the Spit Brook Rd/Everet interchange?  WhatE is the Sheraton called now days? I seem to remeber a name change from 
 somewhere.   Or a Lat/Long ;)   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------    Date: 26 Oct 2006 09:58:49 -0700- From: "Steve Lionel" <steve.lionel@intel.com>   Subject: Re: Spit Brook locationC Message-ID: <1161881929.418386.186030@b28g2000cwb.googlegroups.com>    prep@prep.synonet.com wrote:H > The SE buildings are NW of the Spit Brook Rd/Everet interchange?  WhatG > is the Sheraton called now days? I seem to remeber a name change from  > somewhere. >  > Or a Lat/Long ;)  @ The Sheraton is still a Sheraton - called "Sheraton Nashua" now.   http://maps.google.com/maps?f=q&hl=en&q=110+spit+brook+rd,+03062&ie=UTF8&z=16&ll=42.71284,-71.457009&spn=0.010059,0.022831&t=k&om=1   , VMS is in building 3, which is the leftmost.   Steve    ------------------------------  + Date: Thu, 26 Oct 2006 17:29:11 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney)   Subject: Re: Spit Brook location( Message-ID: <ehqr97$nct$2@pcls4.std.com>   prep@prep.synonet.com writes:   B >The SE buildings are NW of the Spit Brook Rd/Everet interchange?   C Yes.  If you're looking for it with an aerial image browser, the ZK J buildings are 3 connected rectangles sort of in a line, sort of behind the. hotel.  VMS lives in the left/westernmost one.   ------------------------------  % Date: Thu, 26 Oct 2006 07:49:12 +0100 ' From: "Roger Fraser" <rfraser3@csc.com> 8 Subject: Re: Terminal Server Manager and Decnet Phase IV/ Message-ID: <454057e0_1@glkas0286.greenlnk.net>   2 "Hal Kuff" <HalKuff@Verizon.Net> wrote in message & news:MgT%g.81223$073.75255@trnddc01...H > Does TSM work with Phase IV..? Seems to reference NCL... What version  > would that be..?* You can get the latest (V2.1) for free at 2 http://h71000.www7.hp.com/freeware/freeware60/tsm/" It works with both Phase IV and V.H There are 3 ECOs here 07, 08 & 09. Our site is running ECO9 on VAX V7.3 I Phase IV. We recently decommissioned a VAX V7.3 Phase V system which was   running ECO3 successfully. HTH     Rog     ------------------------------    Date: 26 Oct 2006 00:16:45 -07001 From: "Bart.Zorn@gmail.com" <Bart.Zorn@gmail.com> 8 Subject: Re: Terminal Server Manager and Decnet Phase IVC Message-ID: <1161847005.516793.102010@h48g2000cwc.googlegroups.com>   E I am using TSM V2.1-06 on OpenVMS V8.2 with DECnet-Plus, no problems.   D Like Bob said, TSM predates DECnet Phase V, so I assume it will workF correctly with Phase IV. I do not remember if I ever tried it, though!  	 Bart Zorn    Bob Gezelter wrote:  > Hal Kuff wrote: O > > Does TSM work with Phase IV..? Seems to reference NCL... What version would  > > that be..? >  > Hal, > > > I concur with Richard. Terminal Server Manager (for managingF > DECservers, correct) predates Phase V DECnet by a VERY large margin. > & > - Bob Gezelter, http://www.rlgsc.com   ------------------------------  % Date: Thu, 26 Oct 2006 07:12:26 -0400 / From: "William Webb" <william.w.webb@gmail.com> 8 Subject: Re: Terminal Server Manager and Decnet Phase IVI Message-ID: <8660a3a10610260412p389a81e4s34b30aa648c00882@mail.gmail.com>   O On 26 Oct 2006 00:16:45 -0700, Bart.Zorn@gmail.com <Bart.Zorn@gmail.com> wrote: G > I am using TSM V2.1-06 on OpenVMS V8.2 with DECnet-Plus, no problems.  > F > Like Bob said, TSM predates DECnet Phase V, so I assume it will workH > correctly with Phase IV. I do not remember if I ever tried it, though! >  > Bart Zorn  >  > Bob Gezelter wrote:  > > Hal Kuff wrote: Q > > > Does TSM work with Phase IV..? Seems to reference NCL... What version would  > > > that be..? > >  > > Hal, > > @ > > I concur with Richard. Terminal Server Manager (for managingH > > DECservers, correct) predates Phase V DECnet by a VERY large margin. > > ( > > - Bob Gezelter, http://www.rlgsc.com >  > / I can say unequivocally that it runs with both.   F We used it with Phase IV at USPS, and with Phase V at GlaxoSmithKline.  B There was a product that had most of the functionality of TSM thatD worked over IP called Open DECserver Manager, but I don't know if it) was ever actually released into the wild.    WWWebb --   Ajilon Consulting  Site resident at Quest Diagnostics = first.x.last@f$edit(contents of previous line,"COMPRESS").com    ------------------------------    Date: 26 Oct 2006 04:14:25 -0700 From: apogeusistemas@gmail.com" Subject: TSZ07 Magnetic Tape DriveC Message-ID: <1161861265.701613.198390@b28g2000cwb.googlegroups.com>    Hi all:   . I have an old TSZ07 Digital tape drive in use.2 Is there any way to get more data from this device% than the "sh mag mk" or "sh dev mk" ?  .    Regards.   ------------------------------  + Date: Thu, 26 Oct 2006 07:24:20 -0500 (CDT) * From: sms@antinode.org (Steven M. Schweda)& Subject: Re: TSZ07 Magnetic Tape Drive2 Message-ID: <06102607242046_20224BE8@antinode.org>   From: apogeusistemas@gmail.com  0 > I have an old TSZ07 Digital tape drive in use.4 > Is there any way to get more data from this device' > than the "sh mag mk" or "sh dev mk" ?       Use a longer tape?       "SYS$ETC:SCSI_INFO.EXE"?       What kind of data?   H ------------------------------------------------------------------------  3    Steven M. Schweda               sms@antinode-org 4    382 South Warwick Street        (+1) 651-699-9818    Saint Paul  MN  55105-2547    ------------------------------    Date: 26 Oct 2006 07:23:57 -0700 From: apogeusistemas@gmail.com& Subject: Re: TSZ07 Magnetic Tape DriveB Message-ID: <1161872637.000636.67740@m73g2000cwd.googlegroups.com>   Steven M. Schweda wrote:  > From: apogeusistemas@gmail.com > 2 > > I have an old TSZ07 Digital tape drive in use.6 > > Is there any way to get more data from this device) > > than the "sh mag mk" or "sh dev mk" ?  >  >    Use a longer tape?  >  >    "SYS$ETC:SCSI_INFO.EXE"?  >  >    What kind of data?  > J > ------------------------------------------------------------------------ > 5 >    Steven M. Schweda               sms@antinode-org 6 >    382 South Warwick Street        (+1) 651-699-9818 >    Saint Paul  MN  55105-2547   > I'd like get from the Vax system all TSZ07 display's messages.   ------------------------------    Date: 26 Oct 2006 04:23:40 -0700/ From: "Volker Halle" <volker_halle@hotmail.com> X Subject: Re: Two node ES80 cluster to EVA5000, RWSCS and high CR_WAITS on VMS$VAXclusterC Message-ID: <1161861819.997072.189050@h48g2000cwc.googlegroups.com>    Phil,   F consider checking with MONI DLOCK, if there is lots of non-local (i.e.$ incoming/outgoing) locking activity.  
 $ ANAL/SYS SDA> LCK SHOW ACTIVE  B will show your most active resource trees and where they are being	 mastered.   ; Also check for any lock remastering going on: $ MONI RLOCK     Volker.    ------------------------------  % Date: Thu, 26 Oct 2006 09:26:18 -0400  From: "Jilly" <jilly@hp.com>X Subject: Re: Two node ES80 cluster to EVA5000, RWSCS and high CR_WAITS on VMS$VAXcluster* Message-ID: <4540b77c@usenet01.boi.hp.com>  ; "Volker Halle" <volker_halle@hotmail.com> wrote in message  = news:1161861819.997072.189050@h48g2000cwc.googlegroups.com...  > Phil,  > H > consider checking with MONI DLOCK, if there is lots of non-local (i.e.& > incoming/outgoing) locking activity. >  > $ ANAL/SYS > SDA> LCK SHOW ACTIVE > D > will show your most active resource trees and where they are being > mastered.  > < > Also check for any lock remastering going on: $ MONI RLOCK > 	 > Volker.  >   I Also increase CLUSTER_CREDITS to the max if not already done.  Make sure  K your cluster interconnect path is not saturated (I didn't see what was the  K cluster interconnect).  For the node with the most incoming locks see what  G cpu is controlling the cluster interconnect and try moving the cluster  I interconnect to another cpu with less Int Stack time (SET DEVICE/PREFER).    Jilly    ------------------------------    Date: 26 Oct 2006 10:49:00 -0700) From: "quick6_98" <quick6_98@hotmail.com> X Subject: Re: Two node ES80 cluster to EVA5000, RWSCS and high CR_WAITS on VMS$VAXclusterC Message-ID: <1161884940.188553.102140@f16g2000cwb.googlegroups.com>    Volker Halle wrote:  > Phil,  > H > consider checking with MONI DLOCK, if there is lots of non-local (i.e.& > incoming/outgoing) locking activity. >  > $ ANAL/SYS > SDA> LCK SHOW ACTIVE > D > will show your most active resource trees and where they are being > mastered.  > < > Also check for any lock remastering going on: $ MONI RLOCK > 	 > Volker.   0   A MONITOR DLOCK on NODE A shows the following:       New ENQ Rate@                        (Local)       79.00     177.76      69.66 735.66@                     (Incoming)        0.33       2.17       0.00 9.33@                     (Outgoing)       39.00      41.23      18.00 93.00      Converted ENQ RateA                         (Local)       41.66      54.28      22.00  189.66@                     (Incoming)        1.33       1.85       0.00 8.33@                     (Outgoing)        4.00      12.40       3.00 22.00      DEQ RateB                          (Local)       70.00     180.49      67.66 741.66@                     (Incoming)        0.33       2.18       0.00 8.66@                     (Outgoing)       21.00      46.77       7.00 117.00     Blocking AST Rate B                          (Local)        0.00       0.58       0.00  3.66 @                     (Incoming)        0.00       1.38       0.00 7.00@                     (Outgoing)        0.33       0.77       0.00 2.66D     Dir Functn Rate (Incoming)23.33      11.50       0.00      33.33@                     (Outgoing)       18.00      68.11      14.00 297.33C     Deadlock Message Rate     0.00       0.00       0.00       0.00   0   A MONITOR DLOCK on NODE B shows the following:       New ENQ RateA                         (Local)      658.33     208.57       0.00  658.33@                     (Incoming)       48.00      39.26       6.66 74.33 @                     (Outgoing)        5.33       2.32       0.00 7.00     Converted ENQ RateB                          (Local)       60.33      19.30       4.33 77.33 @                     (Incoming)       19.33      11.85       4.33 21.33 @                     (Outgoing)        3.66       1.57       0.00 4.66     DEQ RateA                         (Local)      658.00     208.71       0.00  658.00@                     (Incoming)       22.66      35.17       2.00 77.66 @                     (Outgoing)        5.33       2.26       0.00 7.00     Blocking AST Rate B                          (Local)        5.00       1.98       0.33  5.00 @                     (Incoming)        1.66       0.70       0.00 2.00@                     (Outgoing)        5.33       1.67       0.00 5.33E     Dir Functn Rate (Incoming) 36.00      79.05       9.33     466.66 @                     (Outgoing)       48.00      18.12       0.00 48.00 C     Deadlock Message Rate     0.00       0.00       0.00       0.00   D   A SDA> LCK SHOW ACTIVE shows the system disk and the cluser common queue disk as most active.  +   A MONITOR RLOCK shows absolutely nothing.    phil   ------------------------------  % Date: Fri, 27 Oct 2006 00:15:19 +0800  From: prep@prep.synonet.com Y Subject: Re: Two node ES80 cluster to EVA5000, RWSCS and high CR_WAITS on VMS$VAXcluster  0 Message-ID: <87ejsvjboo.fsf@k9.prep.synonet.com>  + "quick6_98" <quick6_98@hotmail.com> writes:   F > We have recently started seeing processes in RWSCS and high CR_WAITSC > (860) on the cluster.  The first clue came from AMDS complaining.  > The    ...   A > There are no complaints from AMDS as far as lock contention but E > users are complaing of "stalls".  The problem seemed to start after B > installing VMS_UPDATE_0800 and a reboot of both nodes.  Any help > would be appreciated.   F It is an SCS credit problem. What is the cluster interconect? Have you  altered and SYSGEN SCS settings?   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------   End of INFO-VAX 2006.589 ************************                                                                                                                                                                                                                                                                                                                                                                                                                                                                "=";            CA    A  H   h  H   =  2 !                 I       {;  
  H CA H   A @   @   =  1                 G { H   {     CA @   A  @   =  /                   	I      CA @   A  H   H   =  +                 ,{ C  L$  H   f  CA H   A  X      =  +                  @ "  H  A f  ;                      CA X   A  @   =  2 !                    	I   ,"  CA @   A  @   h@   =  +                   H      ;     CA @   A  @   @   =  )                     	I        CA @   A  @   @   =  )                        ;       CA @   A  0   0   =  '                ICA 0   A  0     =  %                 ICA 0   A        X  =  %                  H       ,'By record number'='@ r
  H  C                  ';  CA    F  (   H| { =  %          CF (   A  @     X=  %                   	H       CA @   A  p   p      =  "                       RELATIVE m      H   S H       H   ;                  CA p   A  X   =  "                 G{ H   { TIVE m      H   S H      CA X   A  H   8H   =  !                          	m        CA H   A  X      =  !                 < H  +   +                str;   CA X   A  @   X=  %                 H{        TIVECA @   A  0   h0   =                   ICA 0   A  8   8   =                   H |     CA 8   As @   @   =                   