1 INFO-VAX	Sat, 16 Aug 2003	Volume 2003 : Issue 452       Contents: Re: 306GB drives!  Re: 306GB drives!  Re: 306GB drives!  Re: 306GB drives!   Re: Compaq BASIC feature or bug?  Re: Compaq BASIC feature or bug?  Re: Compaq BASIC feature or bug?  Re: Compaq BASIC feature or bug? Re: DEC 3000 M500 CXT error ? # decdocument: figures in HTML output  Re: DECnet problem Re: DECnet problem$ Re: ES40's going cheap COV Special !( Re: Init disk for maximal container file( Re: Init disk for maximal container file( Re: Init disk for maximal container file% Re: itrc - H.P. I.T. Resource Center. % Re: itrc - H.P. I.T. Resource Center. % Re: itrc - H.P. I.T. Resource Center. % Re: itrc - H.P. I.T. Resource Center.  Re: more HP layoffs  Re: more HP layoffs  Re: Precompiled zip for VMS 6.2  Re: Precompiled zip for VMS 6.2   Re: Querying UAF from MS Windows  Re: Querying UAF from MS Windows  Re: Querying UAF from MS WindowsP Re: running simh VAX simulator with some other microcode than default? default?d* Re: using mailbox for shared memory access* Re: using mailbox for shared memory accessF Re: Who needs shadowing when you have smart I/O controllers?  (was Re:P Re: Who needs shadowing when you have smart I/O controllers?  (was Re:306GB driv xml web services question   F ----------------------------------------------------------------------    Date: 14 Aug 2003 23:33:12 -0500+ From: young_r@encompasserve.org (Rob Young)  Subject: Re: 306GB drives!3 Message-ID: <Wea6H8AO1F0F@eisner.encompasserve.org>   c In article <5uYZa.78482$7O4.1900368@twister.rdc-kc.rr.com>, "Mike Naime" <mnaime@kc.rr.com> writes: J > According to our sales rep, we should see 306GB EMA style disks shipping > sometime in November.  > F > So, for raw space, thats 84x306 = 25704GB.  25.7 Terrabytes of data.J > So, If I populate a T-5 configuration with those (4 HSG's with 84 drivesG > each), I can get  102.8TB of data in 5 feet of data center row space.  > L > That's the same amount of space that I have in my existing 36 HSG cabinets4 > populated with 18, 36, 72, and a few 144 GB disks. > A > I just hope that the 8.7 VCS cards supports that size of drive.  >    	Why or how would you use them?   6 	Would you put anything mission critical on a 5 member9 	RAID-5 (for example) and serve that LUN up to VMS or NT?   8 	The reason I ask ... is an old favorite subject of mine
 	has come up:   e http://groups.google.com/groups?selm=Lsvhk3%24YxMND%40eisner.encompasserve.org&oe=UTF-8&output=gplain    ---   + From: Rob Young (young_r@encompasserve.org)   Subject: Re: disk benchmarking   Newsgroups: comp.os.vms  Date: 2001-05-29 10:23:36 PST     F  What happens when that RAID5 of 4 - 500 GByte drives takes 2 days to C  restore (5 days?), etc. etc.  That probably won't happen, so where $  is this headed/used (large drives)?   ---   B 	The point is with a 5 member RAID5 and HSG (using those 300 GByteC 	drives) - you have a 1.2 TByte LUN.  When/if the RAID5 array blew  > 	out ( didn't rebuild properly) - you have a 1.2 TByte LUN to 	 	restore.   ? 	So you wouldn't use RAID5.  Okay, a 360 GByte mirror set blows  	out.  You get the picture.   > 	EVA and others get around this by partitioning - whether true> 	virtualization at the hardware level like EVA does - or other# 	clunkier methods of competitors.    	 / 	I wouldn't stick 360 GByte drives in an HSG80, > 	unless of course you could take 5 , 10, 20, 30 hour downtimes> 	for restoration - worst case of course (pick a worst case forC 	you - does it ever get less than 5 hours blowing out and restoring = 	a modest 360 GByte mirrorset - 20 MByte/sec restore speed?).    				Rob    ------------------------------  % Date: Fri, 15 Aug 2003 12:47:56 -0400 8 From: Jim Agnew - VCU/MCV Neurosurgery <jpagnew@vcu.edu> Subject: Re: 306GB drives!' Message-ID: <3F3D0EBC.25708824@vcu.edu>    jlsue wrote: > F > On Fri, 15 Aug 2003 09:43:17 -0400, Jim Agnew - VCU/MCV Neurosurgery > <jpagnew@vcu.edu> wrote: >  > >Rob Young wrote:  > >> >  > >  > >>; > >>         I wouldn't stick 360 GByte drives in an HSG80, J > >>         unless of course you could take 5 , 10, 20, 30 hour downtimesJ > >>         for restoration - worst case of course (pick a worst case forO > >>         you - does it ever get less than 5 hours blowing out and restoring I > >>         a modest 360 GByte mirrorset - 20 MByte/sec restore speed?).  > >>( > >>                                 Rob > > G > >I know a college that blew out it's mail servers, and took 4 days to 
 > >restore...  > >I second that point...  > >  > I > Funny though.  I used to say that about 9GB drives when they first came M > out... I didn't want to have to be the person to restore those.  Of course, ) > that was in the DLT-3 time-frame, iirc.   D Actually, I guess there *is* a need for largeish drives, witness theG large DB's going around now...  But we're really, really need to shadow F those... just switching over to a new array when the primary goes down6 would be a LOT less time than restoreing the primary..  ? and if you're really paranoide, a 3rd array would make sense...  --  F "4,000 years ago I made a mistake."  Elrond Half-Elven, in "Fellowship of the Ring"  F "I try not to be right any more than necessary". -- Larry Wall, author of the Perl Language   ------------------------------  # Date: Fri, 15 Aug 2003 15:35:13 GMT & From: jlsue <jefflsxxxz@sbcglobal.net> Subject: Re: 306GB drives!8 Message-ID: <2dvpjv81b6a7ule4klfc39d2hkehb771ui@4ax.com>  D On Fri, 15 Aug 2003 09:43:17 -0400, Jim Agnew - VCU/MCV Neurosurgery <jpagnew@vcu.edu> wrote:   >Rob Young wrote:  >>     >  >>  9 >>         I wouldn't stick 360 GByte drives in an HSG80, H >>         unless of course you could take 5 , 10, 20, 30 hour downtimesH >>         for restoration - worst case of course (pick a worst case forM >>         you - does it ever get less than 5 hours blowing out and restoring G >>         a modest 360 GByte mirrorset - 20 MByte/sec restore speed?).  >>  & >>                                 Rob > E >I know a college that blew out it's mail servers, and took 4 days to  >restore...  >I second that point...  >   G Funny though.  I used to say that about 9GB drives when they first came K out... I didn't want to have to be the person to restore those.  Of course, ' that was in the DLT-3 time-frame, iirc.    ------------------------------  + Date: Sat, 16 Aug 2003 16:36:07 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney)  Subject: Re: 306GB drives!( Message-ID: <bhlmhn$cpv$1@pcls4.std.com>  - young_r@encompasserve.org (Rob Young) writes:   L >> Personally, I think that shadowing is one of the biggest performance hits' >> that you can take on a VMS system.     ? >	Not really.  A large OEM of storage would try to convince you ; >	of such a farce.  But when pressed for any data at all to & >	back up their FUD - they slink away.  E Yes.  Shadowing is slower for writes (it proceeds at the rate of the  F slowest drive with writes in parallel) and is usually faster for readsG since it shares the read load among the sources.  Things can go to hell H pretty quickly if a full merge is pending or underway.  You have to feed it memory (NPP) though.  --   -Mike    ------------------------------  % Date: Fri, 15 Aug 2003 09:42:31 -0700 0 From: Randy Park <rjpark@mindspring.nospaam.com>) Subject: Re: Compaq BASIC feature or bug? 8 Message-ID: <f53qjvou1r78s4qb8u9liel2p97e8unqqn@4ax.com>  0 On Wed, 13 Aug 2003 18:48:26 -0400, "Neil Rieck" <n.rieck@sympatico.ca> wrote:    >Compaq BASIC feature or bug?  >  > N >While renovating (indenting and remarking) a colleague's program I discoveredI >a missing "END IF" statement that gets by the complier with no errors or K >warnings. While I applaud the compiler authors for their wonderful work, I O >would rather get a warning on this one. (the program's author was surprised by  >this too!)  > J >p.s. since I don't have a software support contract with HP/Compaq I hope& >someone else will report this to them >  >Neil Rieck   C As Ed said, this is expected behavior.  There is an implicit END IF B when an ELSE, a line number, or the end of program is encountered.C This feature can make it difficult to track down missing END IFs if & proper indented formatting isn't done.   ------------------------------  % Date: Fri, 15 Aug 2003 13:31:54 -0400 ( From: David Froble <davef@tsoft-inc.com>) Subject: Re: Compaq BASIC feature or bug? , Message-ID: <3F3D190A.7060306@tsoft-inc.com>   Randy Park wrote:   2 > On Wed, 13 Aug 2003 18:48:26 -0400, "Neil Rieck" > <n.rieck@sympatico.ca> wrote:  >  >  >>Compaq BASIC feature or bug? >> >>O >>While renovating (indenting and remarking) a colleague's program I discovered J >>a missing "END IF" statement that gets by the complier with no errors orL >>warnings. While I applaud the compiler authors for their wonderful work, IP >>would rather get a warning on this one. (the program's author was surprised by >>this too!) >>K >>p.s. since I don't have a software support contract with HP/Compaq I hope ' >>someone else will report this to them  >> >>Neil Rieck >> > E > As Ed said, this is expected behavior.  There is an implicit END IF D > when an ELSE, a line number, or the end of program is encountered.E > This feature can make it difficult to track down missing END IFs if ( > proper indented formatting isn't done. >    Randy,  O Can you be a bit more explicit on the 'ELSE'?  I'm of course aware of the line  N number forcing a new statement, that's why we had so many line numbers before S the advent of the 'EDN IF'.  But I've never been aware of an 'ELSE' doing anything.    Dave   --  4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com6 T-Soft, Inc.  170 Grimplin Road  Vanderbilt, PA  15486   ------------------------------  # Date: Fri, 15 Aug 2003 19:05:48 GMT - From: "Ed Vogel" <edward.vogel_nospam@hp.com> ) Subject: Re: Compaq BASIC feature or bug? 1 Message-ID: <gaa%a.2466$P86.721@news.cpqcorp.net>   5 "David Froble" <davef@tsoft-inc.com> wrote in message & news:3F3D190A.7060306@tsoft-inc.com...  L || Can you be a bit more explicit on the 'ELSE'?  I'm of course aware of the lineH | number forcing a new statement, that's why we had so many line numbers beforeK | the advent of the 'EDN IF'.  But I've never been aware of an 'ELSE' doing 	 anything.   K     I mis-spoke when I said an ELSE implied and END IF.  I should have said *     it implies the end of the THEN clause.  I    So, while both END IF and ELSE (along with a line number and an end of  program)F     terminate the THEN clause of an IF, an END IF terminates the whole	 statement /     while ELSE only terminates the THEN clause.       Sorry...        Ed   ------------------------------    Date: 16 Aug 2003 09:14:07 -0700' From: n.rieck@sympatico.ca (Neil Rieck) ) Subject: Re: Compaq BASIC feature or bug? = Message-ID: <a5396d5d.0308160814.704cbd3c@posting.google.com>   f "Ed Vogel" <edward.vogel_nospam@hp.com> wrote in message news:<oIL_a.2279$sY4.846@news.cpqcorp.net>...8 > This is a feature.  To quote from the documentation of> >     the "IF" statement (From a 1990 version of the VAX BASIC >     Reference Manual): > A >     "All statements between the THEN keyword and the next ELSE, = >      line numer, or END IF are part of the THEN clause...."  > H >     So...if you have an ELSE (or a line number), the END IF is assumed >     by the compiler. > I >     This feature is to support older (I mean "real" old) BASIC programs G >      that were written prior to the addition of END IF as part of the  >      (VAX) BASIC language. >  >     Ed Vogel > 9 >     HP C Engineering (who spent a lot of time on BASIC)  >  > ? Then perhaps a new feature could be added to the complier. When D compiling with switch "/warn=all", it would be cool to see a messageC (warning or informational) indicating that an "END IF" was expected 
 but not seen.   > I'd like to see this for "ELSE" performing an assumed "END IF"D function as demonstrated in my original example as well as the "line% number" case we're all familiar with.   C If you've got to make a case for this with HP management, tell them 1 you did the same thing going from VAX-C to DEC-C.   F If they're worried about backward compatiblity, you could create a newD complier switch ("/informational=all" ?) which would default to off.  
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.! http://www3.sympatico.ca/n.rieck/   F p.s. my web server is currently down because my Bell Canada has pulled the plug to conserve power   ------------------------------  + Date: Fri, 15 Aug 2003 13:57:19 +0000 (UTC) 6 From: "Jeroen M.W. van Dijk" <jmw.vandijk@quicknet.nl>& Subject: Re: DEC 3000 M500 CXT error ?@ Message-ID: <Xns93D8A2217ED3Ejmwvandijkquicknetnl@213.73.255.20>  < >>Firing up Ye Olde DEC3000-500 (Flamingo), the self test is >>fine except this error:  >>CXT     ?? 000 0064   H > I don't know this specific error.  I have the model 400 service manualC > at hand, but unfortunately the built-in graphics is the one thing E > that's missing in all but the 500 and 500X models.  So my manual is  > silent about your error. > I The 300 has also a built-in grahpics card, but I have no manuals of that   model.  K My DEC 3000 300 comes without any errors through the tests and I cant get  . it to test the graphical card from bootlevel.   F EK-FLAMI-SV    DEC 3000-500 SVC INFO                 $130.00          F EK-FLMNG-DK    DEC 3000-500 WK SYSTEM KIT             $70.00          > EK-FLMNG-TM    DEC 3000-500(S) TECH MANUAL            $85.00      F EK-PELCN-OG    DEC 3300 AXP HW REF GUIDE              $25.00          F EK-PELCN-SV    DEC 3000-300/300L SVC GDE              $45.00          F EK-PELHW-IC    DEC 3300 AXP SYS INSTALL               $25.00          F EK-PELSV-DK    DEC 3000-300/300L SERVICE KIT         $130.00            0 The answer is somewhere in one of those manuals.  1 The -SVs are the most likely to give the answer.    ------------------------------  # Date: Fri, 15 Aug 2003 21:38:46 GMT ' From: Steve Thompson <smt@vgersoft.com> , Subject: decdocument: figures in HTML outputJ Message-ID: <Pine.LNX.4.44.0308151736010.11169-100000@honker.vgersoft.com>  H Does anyone know what the first argument to the <FIGURE_FILE> tag shouldD be when embedding images (.gif) in HTML output? I have tried HTML orI ONLINE, neither of which are legal, and my documentation is silent on the G subject. Or perhaps I have to conditionalize the figure preparation and  embed HTML tags in the output?  0 I'm using DECdocument V3.3 on an Alpha, VMS 7.1.   TIA, -steve   ------------------------------  % Date: Sat, 16 Aug 2003 15:58:01 +0200 $ From: Michael Unger <unger@decus.de> Subject: Re: DECnet problem 8 Message-ID: <bhlefp$n77m$1@ID-152801.news.uni-berlin.de>  / On 16-Aug-2003 13:43, Dennis Grevenstein wrote:   ? > I've got two VMS boxes: an Alphastation 200 running 7.3 and a ; > MicroVAX 3100/96 running 7.2. Both of them work prefectly < > standalone, but if they are connected to the same Ethernet? > network they both freeze the moment they have DECnet running. : > They cluster as long as one of them does not run DECnet.> > It's something I have never seen before. They have different- > DECnet addresses. What could be wrong here?    SCS SYSGEN parameters?   Michael    --  ; Real names enhance the probability of getting real answers. @ Please do *not* send "Security Patch Notifications" or "SecurityA Updates"; this system isn't running a Micro$oft operating system. = And don't annoy me <mailto:postmaster@[127.0.0.1]> please ;-)    ------------------------------  % Date: Sun, 17 Aug 2003 03:22:46 +1000 : From: "David McKenzie" <david.mckenzie@paradigm-shift.biz> Subject: Re: DECnet problem C Message-ID: <3f3e6859$0$95044$c30e37c6@lon-reader.news.telstra.net>   + when you say cluster, do you mean "cluster"   H If there is a conflict of scsnode or scssystemid they should not cluster either.   I are the phase IV addresses different. This is derived from the the decnet  address by area*1024 + address  6 Are you running Phase V. If so is the node name unique   more information please?   David    Ex VMS and now a lawyer     ; "Dennis Grevenstein" <dennis@pcde.inka.de> wrote in message & news:bhl5ce$9k1$1@aton.pcde.inka.de... > Hi,  > ? > I've got two VMS boxes: an Alphastation 200 running 7.3 and a ; > MicroVAX 3100/96 running 7.2. Both of them work prefectly < > standalone, but if they are connected to the same Ethernet? > network they both freeze the moment they have DECnet running. : > They cluster as long as one of them does not run DECnet.> > It's something I have never seen before. They have different- > DECnet addresses. What could be wrong here?  >  > mfg  > Dennis >  > --  J > "I remarked to Dennis that easily half the code I was writing in Multics was J > error recovery code. He said, "We left all that stuff out. If there's an error,G > we have this routine called panic, and when it is called, the machine  crashes,2 > and you holler down the hall, 'Hey, reboot it.'"@ >        Tom van Vleck and Dennis Ritchie about Multics <-> UNIX relationship   ------------------------------  % Date: Fri, 15 Aug 2003 07:46:35 -0400 & From: "Island" <dbturner@islandco.com>- Subject: Re: ES40's going cheap COV Special ! / Message-ID: <vjpi4ndk56vm8e@news.supernews.com>    FYI People -  E If you are running an Alphaserver 4100 - this is the perfect upgrade!     4 "David @ Island" <david-@-hpaq.net> wrote in message( news:vjo4v4hjl027e@news.supernews.com...3 > Alphaserver ES40 Model 2's with 1GB Compaq memory + > Come with Keyboard, Mouse SCSI Controller + > 6 Disk Cage and 18GB Disk 10/100 Ethernet  >  > % > Single CPU 500Mhz No license $7,995 % > Single CPU 667Mhz No license $8,995 & > Single CPU 833Mhz No license $12,995 > # > Options available at www.hpaq.net  >  >  > --   > Island Computers US Corp > 2700 Gregory St, Ste 180 > Savannah GA 31404  > Tel; (00) 1 912 447 6622 > Fax; (00) 1 912 201 0402 > dbturner@hpaq.net  > www.hpaq.net > / > Please read our warranty terms and conditions  > before placing your order # > http://www.hpaq.net/warranty.html # > Support: http://www.hpaq.net:8004 & > Support  FTP: ftp://www.islandco.com! > (User name & Password required)  >  >    ------------------------------  % Date: Thu, 14 Aug 2003 22:00:42 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> 1 Subject: Re: Init disk for maximal container file ' Message-ID: <3F3C4CDA.161499E6@fsi.net>    Rob Brown wrote: > ( > On Thu, 14 Aug 2003, Rok Vidmar wrote: > H > >   I am trying to put 4110372 block container file on a 4110480 block > > RZ28M. Is it doable? > >   My > > @ > > $ INITIALIZE DKA100/CLUSTER=16382/DIRECTORIES=16/HEADERS=16-B > > /HOMEBLOCKS=CONTIGUOUS/INDEX=BEGINNING/MAXIMUM_FILES=1/SYSTEM- > > /NOVERIFIED MX2SET > > B > > has a lot to be desired (127433 blocks to be exact) because of > > INDEXF.SYS and BADBLK.SYS  > D > I don't see anything wrong with that.  /HEADERS=16/MAXIMUM_FILES=1@ > should minimize the size of INDEXF.SYS.  /CLUSTER=16382 should" > minimize the size of BITMAP.SYS.  A Well, other posters have noted that the outrageous clustersize is ) contributing heavily to your frustration.   H 4110480 blocks is approx. 1.96GBF. RZ28, eh? ;-) I have six of 'em on my Alpha machine here.   + DJAS01::DDACHTERA$ dir dka600:[000000]*.sys    Directory DKA600:[000000]   ? BACKUP.SYS;1               0/0        28-OCT-2000 16:09:31.93   ? BADBLK.SYS;1               0/0        28-OCT-2000 16:09:31.93   ? BADLOG.SYS;1               0/0        28-OCT-2000 16:09:31.93   ? BITMAP.SYS;1             252/252      28-OCT-2000 16:09:31.93   ? CONTIN.SYS;1               0/0        28-OCT-2000 16:09:31.93   ? CORIMG.SYS;1               0/0        28-OCT-2000 16:09:31.93   ? INDEXF.SYS;1           61792/61792    28-OCT-2000 16:09:31.93   ? SECURITY.SYS;1             1/8        28-OCT-2000 16:09:31.93   ? VOLSET.SYS;1               0/0        28-OCT-2000 16:09:31.93     % Total of 9 files, 62045/62052 blocks.   E The lowest clustersize you can manage (3) will result in a BITMAP.SYS G file of 252 blocks. The trick would be finding the best balance between H /CLUSTERSIZE and resulting the allocations of BITMAP.SYS, INDEXF.SYS and
 SECURITY.SYS.   H Dunno if you'll find a way to eat up only your 180 blocks of "headroom",H but you may come close. The best you can probably manage is likely to be8 a bit less than the target for your container file size.  ! Playing with DVD images, are you?   G If so, and you can build a two-member volume-set, that might solve your > problem, with the caveat that I've never actually tried to useD volume-sets with LD containers, if that is indeed what you're up to.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Fri, 15 Aug 2003 12:46:20 +0100 + From: John Laird <john@laird-towers.org.uk> 1 Subject: Re: Init disk for maximal container file 8 Message-ID: <pdhpjvsdc9n9n97fa4agbqp61r6tiu7uit@4ax.com>  I On Fri, 15 Aug 2003 12:52:09 +0200, Rok Vidmar <Rok.Vidmar@NUK.Uni-Lj.Si>  wrote:  G > And to answer David's question: no, I am not playing with DVD images, F >instead I am trying to build 3 member shadow set from AFAIK identicalG >RZ28M (they all report 2988*86*16 geometry). Two of them are connected F >to VAX 4000-106A and report 4110372 blocks, the last one is connectedH >ty VAXstation 3100-M76/SPX and reports 4110480 blocks. Sigh, 2988*86*16 >equals to 4111488.   K As this is a shadow set, presumably you could afford not to have a real VMS J file system wrapped around the container file ?  Some virtual disk driversH allow you to use the physical disk as a logical disk, so just choose the; size you need.  (VDDRIVER will do this, LDDRIVER will not.)   L [I had been about to suggest that you could not shadow different disk types,D and would have to create logical disks on all of them, but there was5 evidence from you last November that it would work !]    --   John   ------------------------------  % Date: Fri, 15 Aug 2003 13:49:58 -0400 , From: "warren sander" <warren.sander@hp.com>1 Subject: Re: Init disk for maximal container file , Message-ID: <3f3d2362$1@usenet01.boi.hp.com>  F you could try a low-level format to see if the missing blocks are 'bad blocks'   I >   And to answer David's question: no, I am not playing with DVD images, G > instead I am trying to build 3 member shadow set from AFAIK identical H > RZ28M (they all report 2988*86*16 geometry). Two of them are connectedG > to VAX 4000-106A and report 4110372 blocks, the last one is connected I > ty VAXstation 3100-M76/SPX and reports 4110480 blocks. Sigh, 2988*86*16  > equals to 4111488. >    ------------------------------  % Date: Thu, 14 Aug 2003 21:39:20 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> . Subject: Re: itrc - H.P. I.T. Resource Center.' Message-ID: <3F3C47D8.7DE05830@fsi.net>    Mike Rechtman wrote: >  > Chris Moore wrote: > > M > > One other problem is that patches for older versions such as 5.5-2H4, 7.1 O > > etc. have been omitted. (VAX only shows 6.2, 7.2 and 7.3, Alpha is similar)  > > M > > I realize that not much is going on there, but some of those versions are A > > still listed by HP (sorry...'hp') as "prior version support".  > > P > > Looks like I better download everything to my own repository, just in case I< > > encounter yet another "obsolete" system to be supported. > >  > J > PLEASE find the 'feedback' page, and leave a message about this - I did,	 > and was E > contacted back, but I think many voices in unison would make a much  > better
 > impression.   @ I posted feedback and also posted it to two fora on OpenVMS.org.   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  # Date: Fri, 15 Aug 2003 12:02:03 GMT 2 From: "George Pagliarulo" <georgepag@adelphia.net>. Subject: Re: itrc - H.P. I.T. Resource Center.1 Message-ID: <%Y3%a.2433$xX5.357@news.cpqcorp.net>   ! Couple of things on the new site.   G First, FTP access is NOT being removed. Customers will still be able to   access kits via ftp.itrc.hp.com.  L Access to older kits is not being removed. On the new site customers will beI able to access all the kits they could access on the old site. Older kits F will be in an archive area. The archive area won't have the search andJ download capabilities of the active side but instead will look pretty muchK like what is available today on the old site - a list of kits segregated by I architecture and version. The archive kits are still being moved and that L has been slow. They should all be there very soon though.    The decision toL not put all kits up on the active side of the site was pretty much one of 10I lbs of work and a 5 lb bag. Because of differences in the way the new and J old sites handle kits, every VMS kit that got moved to the active side hadJ to be repackaged. It just didn't make sense to do this for the older kits.  L The search engine has some problems. That's being worked. Requests have alsoB been made to allow users to do searches by date and by kit rating.  6 George Pagliarulo - Hewlett-Packard Company, Nashua NH  J george.pagliarulo@takethisout.hp.com (remove "takethisout." when replying)  9 "Jim Strehlow" <JimStrehlow@data911.com> wrote in message 7 news:4b6ec350.0308110920.727cf342@posting.google.com... @ > I have mixed feelings about the new H.P. I.T. Resource Center.G > Microsoft has had Windows Update to help administrators find patches. > > There needed to be something to make things easier for us to > administer OpenVMS patches. = > But H.P. is going to take away their FTP site in September. D > I do not mind finding out about patches via some new tool at ITRC;G > but I prefer using FTP on OpenVMS directly binary copying files to my  > AlphaServer. > E > H.P. will be forcing me to have to either "double transfer" from my D > P.C. to OpenVMS, or I will have to go to a DECWindows terminal (inH > another room away from my regular workspace), launch Mozilla (or other > browser) etc.  > E > Since H.P. has been e-mailing me patch notifications, that has been 0 > good enough for an "update notification tool." > ' > What is YOUR opinion of the new ITRC?  > ? > Jim Strehlow, OpenVMS Systems Administrator, Alameda, CA, USA    ------------------------------  % Date: Fri, 15 Aug 2003 13:57:35 -0400 , From: "warren sander" <warren.sander@hp.com>. Subject: Re: itrc - H.P. I.T. Resource Center.* Message-ID: <3f3d2529@usenet01.boi.hp.com>  J if you want feedback to get to anyone in HP who you don't already know (ie Hoff etc) then% posting to OpenVMS.org isn't the way.   H Sending feedback via the web sites 'feedback' is one way. At least if it* isn't already it is becoming a closed loopJ process with metrics for folks needing to answer the customers. And 'thank( you and go ... yourself' isn't a reponse that mgmt looks kindly upon.  I Not everything can be done. HP does understand 'history' much better than ) Compaq ever did. But even inside HP there F is history and HISTORY. The HP customer base doesn't use as old-crufty5 software at the same time they use new-shiny software C so there is some education. backwards compatiblity is becoming more < important but come on 5.5-2 is old. And even for y2k you had# to have a contract for the patches.   E We are trying to make sure nothing gets deleted and if it does we are < pushing to get it 'undeleted' and/or taking ownership of it.  G so give the ITRC folks some slack (not a lot but some). They are coming 9 around and with the new downloads of layered products and G new ways to get at patches we are trying to get the support side of the K house in order. It will become another order of complexity when the Itanium 3 stuff comes online. Yet another directory tree etc.     < "David J. Dachtera" <djesys.nospam@fsi.net> wrote in message! news:3F3C47D8.7DE05830@fsi.net...  > Mike Rechtman wrote: > >  > > Chris Moore wrote: > > > K > > > One other problem is that patches for older versions such as 5.5-2H4,  7.1 H > > > etc. have been omitted. (VAX only shows 6.2, 7.2 and 7.3, Alpha is similar) > > > K > > > I realize that not much is going on there, but some of those versions  are C > > > still listed by HP (sorry...'hp') as "prior version support".  > > > K > > > Looks like I better download everything to my own repository, just in  case I> > > > encounter yet another "obsolete" system to be supported. > > >  > > L > > PLEASE find the 'feedback' page, and leave a message about this - I did, > > and was G > > contacted back, but I think many voices in unison would make a much 
 > > better > > impression.  > B > I posted feedback and also posted it to two fora on OpenVMS.org. >  > --   > David J. Dachtera  > dba DJE Systems  > http://www.djesys.com/ > * > Unofficial Affordable OpenVMS Home Page:! > http://www.djesys.com/vms/soho/    ------------------------------  % Date: Fri, 15 Aug 2003 17:06:53 -0400 , From: "warren sander" <warren.sander@hp.com>. Subject: Re: itrc - H.P. I.T. Resource Center., Message-ID: <3f3d5188$1@usenet01.boi.hp.com>  A maybe not a contract but you had to contact someone to get them..   7 from the y2k site (I not only have history I live it...    VAXFTY2K0n_U2055 (Not publicly available. Contact your Compaq support  representative for purchase.)   1 if any of you care the y2k site is still alive at I http://h71000.www7.hp.com/openvms/products/year-2000 just in case someone  needs it sometime..   : "Larry Kilgallen" <Kilgallen@SpamCop.net> wrote in message- news:t8EcQ8Hz$2m5@eisner.encompasserve.org... < > In article <3f3d2529@usenet01.boi.hp.com>, "warren sander" <warren.sander@hp.com> writes: > H > > Not everything can be done. HP does understand 'history' much better than- > > Compaq ever did. But even inside HP there J > > is history and HISTORY. The HP customer base doesn't use as old-crufty9 > > software at the same time they use new-shiny software G > > so there is some education. backwards compatiblity is becoming more @ > > important but come on 5.5-2 is old. And even for y2k you had' > > to have a contract for the patches.  > I > My recollection is that Y2K patches for 5.5-2 did not require a support * > contract, even though normal patches do.   ------------------------------  % Date: Thu, 14 Aug 2003 18:48:45 -0500 , From: "Dave Gudewicz" <k9jdk@NOSPAMarrl.net> Subject: Re: more HP layoffs/ Message-ID: <vjsp32lcjfhpca@corp.supernews.com>   H There were 3 or 4 who got it this go around.  Technology and management. Could have been others.    Dave...   4 "Neil Rieck" <n.rieck@sympatico.ca> wrote in message3 news:oWI_a.6506$kp4.869027@news20.bellglobal.com... 9 > "Dave Gudewicz" <k9jdk@NOSPAMarrl.net> wrote in message + > news:vjlq14s6b8og70@corp.supernews.com... I > > Got a call just yesterday from a long term DEC--->CPQ--->HP employee.  NoI > > longer.  He and several others in that category were given their pink  slips.! > > This in the Chicagoland area.  > > H > > Guess that's one way to please Wall Street.  But that's several less highlyE > > qualified people to rely on for support, etc.  Guess that doesn't  matter.  > >  > > Dave...  > >  > [snip] > >  > > I > Yep. Short term gain followed by long term pain. BTW, was your buddy in  > technology or management?  >  > Neil Rieck > Kitchener/Waterloo/Cambridge,  > Ontario, Canada.# > http://www3.sympatico.ca/n.rieck/  >  >    ------------------------------  % Date: Sat, 16 Aug 2003 12:08:48 -0500 , From: "Dave Gudewicz" <k9jdk@NOSPAMarrl.net> Subject: Re: more HP layoffs/ Message-ID: <vjsp8tn9s31132@corp.supernews.com>    Nope.  None were.    Dave...   - "CJT" <cheljuba@prodigy.net> wrote in message $ news:3F3BA700.1060203@prodigy.net... > Dave Gudewicz wrote: > I > > Got a call just yesterday from a long term DEC--->CPQ--->HP employee.  NoI > > longer.  He and several others in that category were given their pink  slips.! > > This in the Chicagoland area.  > % > Was he getting close to retirement?  >  > > H > > Guess that's one way to please Wall Street.  But that's several less highlyE > > qualified people to rely on for support, etc.  Guess that doesn't  matter.  > >  > > Dave...  > > 8 > > "Neil Rieck" <n.rieck@sympatico.ca> wrote in message6 > > news:8ae_a.3867$VG.139059@news20.bellglobal.com... > > 6 > >>I just read the following article at the inquirer:- > >>http://www.theinquirer.net/?article=10996p> > >>If true, it looks like HP is still in slash and burn mode. > >>E > >>BTW, I heard a recent rumour about Fall layoffs in the group thatg supportsC > >>TCP/IP Services for OpenVMS. (something about HP getting rid ofc > >C > > contractors; > >dH > >>and this work being spun off to sub-contractors by Compaq before the > >n > > merger). > >  > >>Has anyone else heard this?. > >> > >> > >>Neil Rieck! > >>Kitchener/Waterloo/Cambridge,e > >>Ontario, Canada.% > >>http://www3.sympatico.ca/n.rieck/s< > >>http://www3.sympatico.ca/n.rieck/links/cool_openvms.html > >> > >> > >> > >M > >  > >  >c   ------------------------------  % Date: Sat, 16 Aug 2003 12:10:56 +0200F+ From: "Hans Vlems" <hvlems.nieuw@zonnet.nl>n( Subject: Re: Precompiled zip for VMS 6.28 Message-ID: <bhkvvj$iep4$1@ID-143435.news.uni-berlin.de>  7 "Leif Holmgren" <nobody@nowhere.com> schreef in berichtd$ news:3F3DCD61.2060807@nowhere.com... > Hello! >v@ > After 13 years as a VMS user of which the first 7 where rather> > intensivem the company I work for has now shut down the lastI > applications on one of our VAX-machines. The software was used during a-I > part of our business that no longer exists, so at least I am pleased tovG > say that the VAX was not replaced by any Windows or Solaris machines.@ >nH > Now I have some old private data on this vax that I would want to takeI > with me. Now this is rather much and I want to zip, tar or whatever thel, > data into a few files before ftp-ing them. >iI > Problem: I can't find any compiles zip, tar or such software for VMS onGG > the internet. I could settle for a .OLD that I linked myself. An evenhI > better solution would be vms-compatible backup and library programs fore
 > windows. >rF > Any suggestions, or information on where I can find anything useful. >i > /Leif  >iK You might want to try and take the VAX home, think of it as "on-line backup  storage" :-)J Did you try to run the precompiled utilities?  I tried to run ZIP compiled- for VMS 7.2 on VMS 6.1 and that did not work.o   Hans   ------------------------------  % Date: Sat, 16 Aug 2003 12:33:13 +0200a+ From: "Hans Vlems" <hvlems.nieuw@zonnet.nl>i( Subject: Re: Precompiled zip for VMS 6.28 Message-ID: <bhl19b$jlbf$1@ID-143435.news.uni-berlin.de>  7 "Leif Holmgren" <nobody@nowhere.com> schreef in berichtr$ news:3F3DCD61.2060807@nowhere.com... > Hello! >g@ > After 13 years as a VMS user of which the first 7 where rather> > intensivem the company I work for has now shut down the lastI > applications on one of our VAX-machines. The software was used during a I > part of our business that no longer exists, so at least I am pleased tomG > say that the VAX was not replaced by any Windows or Solaris machines.  >.H > Now I have some old private data on this vax that I would want to takeI > with me. Now this is rather much and I want to zip, tar or whatever the3, > data into a few files before ftp-ing them. >cI > Problem: I can't find any compiles zip, tar or such software for VMS onvG > the internet. I could settle for a .OLD that I linked myself. An eveneI > better solution would be vms-compatible backup and library programs for0
 > windows. >cF > Any suggestions, or information on where I can find anything useful. >t > /Leifc >s   Leif,   L I just found ZIP.EXE and UNZIP.EXE on a Freeware V3.0 CD and they run on VMSJ 6.1 so will work for you. Send me an email and I'll return the .EXE files.( Remove the .nieuw in my email address...   Hans   ------------------------------  % Date: Thu, 14 Aug 2003 22:10:13 -0500g1 From: "David J. Dachtera" <djesys.nospam@fsi.net>D) Subject: Re: Querying UAF from MS Windowst' Message-ID: <3F3C4F15.58DF8750@fsi.net>4   Brian Conklin wrote: >  > Hello,, >    I am a novice when it comes to OpenVMS.G >    I have been searching for days for a Howto on querying the OpenVMSvE > Alpha 7.3 UAF from a Windows host that will be able to display user2F > names, password expiration date, and whether the account is disabled > or not on an ASP page.G >    I have a fairly large network that includes WinNT/2K servers, UnixtG > servers, Linux servers, and this one OpenVMS server. We currently cans@ > display this user information from all but the OpenVMS server.4 >    Thank you for any help you may be able to give.  H Other posters have mentioned Management Station, but that's a WhineBloze) app., not what you seem to be asking for.r  H The trick might be to get the data your want from a DCL proc. on the VMSG machine via REXEC or RSHELL, massage the output into some suitable HTML F and return it to the browser via the web server on a UN*X, W/NT or W2K box.   -- g David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------    Date: 15 Aug 2003 05:00:09 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) ) Subject: Re: Querying UAF from MS WindowsE3 Message-ID: <+TCKTTG7cuf5@eisner.encompasserve.org>t  [ In article <3F3C4F15.58DF8750@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:- > Brian Conklin wrote: >>  	 >> Hello,1- >>    I am a novice when it comes to OpenVMS.xH >>    I have been searching for days for a Howto on querying the OpenVMSF >> Alpha 7.3 UAF from a Windows host that will be able to display userG >> names, password expiration date, and whether the account is disablede >> or not on an ASP page. H >>    I have a fairly large network that includes WinNT/2K servers, UnixH >> servers, Linux servers, and this one OpenVMS server. We currently canA >> display this user information from all but the OpenVMS server.u5 >>    Thank you for any help you may be able to give.e > J > Other posters have mentioned Management Station, but that's a WhineBloze+ > app., not what you seem to be asking for.R > J > The trick might be to get the data your want from a DCL proc. on the VMSI > machine via REXEC or RSHELL, massage the output into some suitable HTMLtH > and return it to the browser via the web server on a UN*X, W/NT or W2K > box.  B Doing such a thing on your own is fraught with Security issues (ofA course VMS Management Station raises some those concerns as well,01 but one hopes DEC designed protections in there).p   ------------------------------    Date: 15 Aug 2003 13:31:04 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)p) Subject: Re: Querying UAF from MS Windowso3 Message-ID: <S0C1Z3jw9H3k@eisner.encompasserve.org>e  U In article <ZQPaGlF27BVM@elias.decus.ch>, p_sture@elias.decus.ch (Paul Sture) writes: e > In article <+TCKTTG7cuf5@eisner.encompasserve.org>, Kilgallen@SpamCop.net (Larry Kilgallen) writes:k^ >> In article <3F3C4F15.58DF8750@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes: >>> Brian Conklin wrote: >>>> i >>>> Hello,e/ >>>>    I am a novice when it comes to OpenVMS.eJ >>>>    I have been searching for days for a Howto on querying the OpenVMSH >>>> Alpha 7.3 UAF from a Windows host that will be able to display userI >>>> names, password expiration date, and whether the account is disabledu >>>> or not on an ASP page. J >>>>    I have a fairly large network that includes WinNT/2K servers, UnixJ >>>> servers, Linux servers, and this one OpenVMS server. We currently canC >>>> display this user information from all but the OpenVMS server.w7 >>>>    Thank you for any help you may be able to give.r >>> L >>> Other posters have mentioned Management Station, but that's a WhineBloze- >>> app., not what you seem to be asking for.h >>> L >>> The trick might be to get the data your want from a DCL proc. on the VMSK >>> machine via REXEC or RSHELL, massage the output into some suitable HTMLsJ >>> and return it to the browser via the web server on a UN*X, W/NT or W2K >>> box. >> PE >> Doing such a thing on your own is fraught with Security issues (of D >> course VMS Management Station raises some those concerns as well,4 >> but one hopes DEC designed protections in there). > H > From my memories of VMS Management Station there was no way to extractF > data to another file, or print the results except by taking a screen > snapshot (i.e. an image).u  F CMKRNL beats all, but my comment was triggered more by the question ofC what protocols are on the wire carrying what sensitive information.f   ------------------------------  % Date: Fri, 15 Aug 2003 10:25:21 +0200N( From: Andreas Davour <ante@update.uu.se>Y Subject: Re: running simh VAX simulator with some other microcode than default? default?dk0 Message-ID: <cs9d6f7tjim.fsf@tempo.update.uu.se>  , "Stanley F. Quayle" <stan@stanq.com> writes:  0 > On 14 Aug 2003 at 14:14, Andreas Davour wrote:H >> I've been toying with VMS on the simh VAX simulator and was wanderingI >> if the ka655.bin is the microcode for the processor? Then, wouldn't it.I >> be possible to run the simulator as a more competent VAX if you have a> >> fast host CPU?e >tB > The file is the boot ROM, not microcode. Changing the code will F > change the type of VAX reported, but won't change how SIMH emulates  > the VAX instruction set. >nC > CHARON-VAX can emulate several different VAXen by this mechanism.g  A OK, then I see. Thanks for telling! I know about CHARON too... :)    /andreas   ------------------------------  % Date: Thu, 14 Aug 2003 22:15:51 -0500b. From: Lyndon Bartels <lbartels@pressenter.com>3 Subject: Re: using mailbox for shared memory access'. Message-ID: <3F3C0A17.45D8A17E@pressenter.com>  " Thank-you all for the responses...  4 It looks like mailboxes aren't going to work for me.  C There's a possibility of more then two processes accessing the same @ mailbox, so just having two (one for each direction) won't work.  G Every process should have equal read and write possibilities. This is auA shared resource. Like a bulletin board that everyone can see, and  everyone write on.   So...e    # I'm onto mapped sections. $crmpsc()8  H I've started reading, went out to dsnlink and pulled down some examples.  D As with every new thing I try, I write a simple stub program, to try/ things out. Faster to compile, easier to debug.f  & I'll include the latest attempt below.  : I'm currently getting an invalid flag error: SS$_IVSECFLG.  B I've never done work with this, so I'm having problems getting the correct combinations of flags.  H I want to create a shared memory section that multiple processes (in the) same group) can read and write to equaly.t   Any help is appreciated.  D The file is included below... This is almost straight from a dsnlink example.       Thanks in advance,   Lyndon     --  G My opinions are mine and mine alone. They seldom align with those of my 	 employer.-    H The only good thing about putting the cart before the horse is you don't have to look at the horse's butt.      H /* test1.c                                                            */H /*                                                                    */H /* Used to test use of $crmpsc() system call.                         */I /*    crmpsc() creates and maps a section of address space. Either to */  G /*    a file, or to a memory section.                                */wG /* I want to try to map a memory section that multiple processes can */eG /* read and write to.                                                */lG /*                                                                   */s  H #include <descrip.h>            /* For $DESCRIPTOR() macro            */ #include <iodef.h>H #include <psldef.h>             /* For PSL$_ flags                    */H #include <secdef.h>             /* For the SEC$M_ flags               */H #include <ssdef.h>              /* For the SS$_ flags                 */H #include <starlet.h>            /* For the usage of sys$crmpsc()      */H #include <stdio.h>              /* For standard i/o routines          */% #include <stdlib.h>             /* */sH #include <string.h>             /* For string functions               */H #include <syidef.h>             /* For SYI$_ flags                    */% #include <lib$routines.h>       /* */r   #define pagelet_size 512;l  < #define check(stat) if ((stat & SS$_NORMAL) != 1) exit(stat)   struct {     int starting;      int ending;  } g_sns_common;w   typedef struct {     unsigned short      buflen;d"     unsigned short      item_code;      void                *buffer;      unsigned int        *retadr;	 } item_t;      main() {    int             ms_flags;u int             pagcount;d unsigned int    pagesize;r  H int             return_status;        /* return status from sys$ calls*/F char            section_name[31] = "fred";    /* global_section name*/ item_t          *itmlst;  * section_name[strlen(section_name)] = '\0';  H unsigned long   ms_inadr[2];            /* map section input address  */3                                         /* range */?  H unsigned long   ms_retadr[2];           /* map section return address */3                                         /* range */   , $DESCRIPTOR(section_name_desc,section_name);" itmlst = calloc(2,sizeof(item_t));  ( itmlst[0].buflen = sizeof(unsigned int);% itmlst[0].item_code = SYI$_PAGE_SIZE;s  itmlst[0].buffer    = &pagesize; itmlst[0].retadr    = NULL;0   return_status = sys$getsyiw(0,                         0,                         0,                         itmlst,S                         0,0,0);  check(return_status);   4 ms_inadr[0] = (unsigned long)&g_sns_common.starting;; ms_inadr[1] = (unsigned long) ((char *)&g_sns_common.endingLD                 + sizeof(int) + pagesize) / pagesize * pagesize - 1;4 pagcount = (ms_inadr[1] - ms_inadr[0])/pagelet_size;   ms_flags = SEC$M_GBL;   ; return_status = sys$crmpsc(ms_inadr,            /* inadr */ <                         &ms_retadr,             /* retadr */<                         PSL$C_USER,             /* acmode */;                         ms_flags,               /* flags */o<                         &section_name_desc,     /* gsdnam */;                         0,                      /* ident */r<                         0,                      /* relpag */:                         0,                      /* chan */<                         pagcount,               /* pagcnt */9                         0,                      /* vbn */n:                         0,                      /* prot */9                         0);                     /* pfc */. check(return_status);o }t   ------------------------------  % Date: Fri, 15 Aug 2003 13:54:59 +1000 : From: "David McKenzie" <david.mckenzie@paradigm-shift.biz>3 Subject: Re: using mailbox for shared memory accessuC Message-ID: <3f3c5ac5$0$95048$c30e37c6@lon-reader.news.telstra.net>:  J What about InterCluster Communications? ICC. IIRC not only would this workC on a single node, but would work on a cluster without modifciation.o      ; "Lyndon Bartels" <lbartels@pressenter.com> wrote in message/( news:3F3C0A17.45D8A17E@pressenter.com...$ > Thank-you all for the responses... >e6 > It looks like mailboxes aren't going to work for me. >nE > There's a possibility of more then two processes accessing the samesB > mailbox, so just having two (one for each direction) won't work. >dI > Every process should have equal read and write possibilities. This is amC > shared resource. Like a bulletin board that everyone can see, andt > everyone write on. >e > So...  >d >e% > I'm onto mapped sections. $crmpsc()y >pJ > I've started reading, went out to dsnlink and pulled down some examples. >oF > As with every new thing I try, I write a simple stub program, to try1 > things out. Faster to compile, easier to debug.  >r( > I'll include the latest attempt below. >-< > I'm currently getting an invalid flag error: SS$_IVSECFLG. >$D > I've never done work with this, so I'm having problems getting the  > correct combinations of flags. > J > I want to create a shared memory section that multiple processes (in the+ > same group) can read and write to equaly.e >e > Any help is appreciated. > F > The file is included below... This is almost straight from a dsnlink
 > example. >e >f >a > Thanks in advance, >  > Lyndon >w >l > --  I > My opinions are mine and mine alone. They seldom align with those of myc > employer.p >. >aJ > The only good thing about putting the cart before the horse is you don't > have > to look at the horse's butt. >o >o >hJ > /* test1.c                                                            */J > /*                                                                    */J > /* Used to test use of $crmpsc() system call.                         */J > /*    crmpsc() creates and maps a section of address space. Either to */I > /*    a file, or to a memory section.                                */hI > /* I want to try to map a memory section that multiple processes can */dI > /* read and write to.                                                */hI > /*                                                                   */n >oJ > #include <descrip.h>            /* For $DESCRIPTOR() macro            */ > #include <iodef.h>J > #include <psldef.h>             /* For PSL$_ flags                    */J > #include <secdef.h>             /* For the SEC$M_ flags               */J > #include <ssdef.h>              /* For the SS$_ flags                 */J > #include <starlet.h>            /* For the usage of sys$crmpsc()      */J > #include <stdio.h>              /* For standard i/o routines          */' > #include <stdlib.h>             /* */.J > #include <string.h>             /* For string functions               */J > #include <syidef.h>             /* For SYI$_ flags                    */' > #include <lib$routines.h>       /* */$ >9 > #define pagelet_size 512;c >e> > #define check(stat) if ((stat & SS$_NORMAL) != 1) exit(stat) >i
 > struct { >     int starting;  >     int ending;u > } g_sns_common;0 >e > typedef struct {! >     unsigned short      buflen;1$ >     unsigned short      item_code;" >     void                *buffer;" >     unsigned int        *retadr; > } item_t;i >n >  > main() > {g >u > int             ms_flags;e > int             pagcount;h > unsigned int    pagesize;u >rJ > int             return_status;        /* return status from sys$ calls*/H > char            section_name[31] = "fred";    /* global_section name*/ > item_t          *itmlst; >e, > section_name[strlen(section_name)] = '\0'; >mJ > unsigned long   ms_inadr[2];            /* map section input address  */5 >                                         /* range */  > J > unsigned long   ms_retadr[2];           /* map section return address */5 >                                         /* range */E >A. > $DESCRIPTOR(section_name_desc,section_name);$ > itmlst = calloc(2,sizeof(item_t)); >M* > itmlst[0].buflen = sizeof(unsigned int);' > itmlst[0].item_code = SYI$_PAGE_SIZE;7" > itmlst[0].buffer    = &pagesize; > itmlst[0].retadr    = NULL;  >   > return_status = sys$getsyiw(0, >                         0, >                         0,! >                         itmlst,h! >                         0,0,0);. > check(return_status);r >a6 > ms_inadr[0] = (unsigned long)&g_sns_common.starting;= > ms_inadr[1] = (unsigned long) ((char *)&g_sns_common.ending.F >                 + sizeof(int) + pagesize) / pagesize * pagesize - 1;6 > pagcount = (ms_inadr[1] - ms_inadr[0])/pagelet_size; >6 > ms_flags = SEC$M_GBL;S > = > return_status = sys$crmpsc(ms_inadr,            /* inadr */1> >                         &ms_retadr,             /* retadr */> >                         PSL$C_USER,             /* acmode */= >                         ms_flags,               /* flags */N> >                         &section_name_desc,     /* gsdnam */= >                         0,                      /* ident */N> >                         0,                      /* relpag */< >                         0,                      /* chan */> >                         pagcount,               /* pagcnt */; >                         0,                      /* vbn */c< >                         0,                      /* prot */; >                         0);                     /* pfc */  > check(return_status);d > }l   ------------------------------   Date: 16 Aug 03 16:07:41 +0200) From: p_sture@elias.decus.ch (Paul Sture) O Subject: Re: Who needs shadowing when you have smart I/O controllers?  (was Re: ) Message-ID: <qBQNueoUQl4A@elias.decus.ch>   d In article <Pine.LNX.4.56.0308160118130.14612@jaipur>, Ryan Moore <rmoore@rmoore.dyndns.org> writes:( > On Sat, 16 Aug 2003, Mike Naime wrote:E >> It all depends on what your management is willing to risk/pay for.oM >> If I could convince my management to double my HSG count, I would not needi >> to have raidsets.  :-)t >>B >> In 3 years, we have had less than 1% failure rate in our disks.5 >> Auto-sparing has made a drive failure a non-event.- >>G >> Anyone out there actually have an HSG80 raidset that did not rebuildn  >> correctly on a drive failure? > G > Yep.  We had a RAID 0+1 set.  Total of 4 spindles.  One of the drivespL > failed.  The HSG twiddled its thumbs and did nothing.  Turned out it was aJ > bug in the HSG firmware when you have redundant controllers.  Apparently3 > each thought the other was going to do something.  >  > The fix was:, > Set the mirrorset policy to not autospare.8 > Manually add the spare drive to the reduced mirrorset.H > Set the mirrorset policy back to autospare (although this won't do any? > good until we reboot the controller with the fixed firmware).* > E > Of course it took several hours of reading manuals, pulling out ourfH > hair, and a support call to figure out why our controller wasn't doingK > what it was configured to do.  Not to mention worrying that another driveaG > didn't decide to die.  It's my experience that drive failures tend toV > happen in bursts.h > L >> I have lost an entire channel twice in the BA-370 models.  Not yet in the	 >> EMA's.r > K > We once had a scenario where both of our redundant I/O controllers failed G > simultaneously.  It was an HSJ40 or HSJ50 pair.  Don't quite remember I > since it was a couple of years back.  Host-based volume shadowing savedr > our butts.  K I once had something similar with a pair of redundant HSJs. Through a copy/1I paste error on a DECterm window I managed to hit an undocumented HSJ CLI kM command which was accepted. It not only screwed that controller, but confusedgK the other one too. We had to get field service in on that one to physicallyF4 remove the offending controller and reinitialize it.  ' Shadowing certainly saved us that time.X   > I > And given the glitch we've just experienced with autospare not working, K > I sure wouldn't feel comfortable putting all my bags in the basket of the8  > I/O controller not messing up. > D > What happens if a drive fails such that it locks a whole SCSI bus?H > This isn't theoretical, it could happen.  Do you have enough spares inK > one cabient so that if a whole bus fails, you're still covered?  Hardwarei' > failure can cause all kinds of havoc.e > F > What if the power to one of your racks goes nuts and surges for someL > reason and somehow manages to cook a lot of components?  Your rack is onlyJ > as good as the power coming to it.  We've had weird power events before,8 > but never lost a rack.  But nothing would surprise me. > L > You can call host-based shadowing a performance hit.  And if you are doingH > all writes, sure, it may be a performance hit.  But we are doing aboutL > 50-50 reads and writes.  The reads speed up, the writes slow down.  HardlyI > matters much since the write-back cache is on.  Besides, with the HSGs,oK > we have enough performance margin even if a shadow merge kicks in.  Don'te" > you have any performance margin? >    Good points there.  L > And... when I/O controllers decide to do stupid things and/or go belly-up,J > you don't lose access to your data.  It's your data and your time... how > much is it worth to you? >    ------------------------------  # Date: Sat, 16 Aug 2003 16:14:09 GMTr% From: "Mike Naime" <mnaime@kc.rr.com>-Y Subject: Re: Who needs shadowing when you have smart I/O controllers?  (was Re:306GB drivF; Message-ID: <lLs%a.93704$o27.2145512@twister.rdc-kc.rr.com>t  6 Ryan Moore <rmoore@rmoore.dyndns.org> wrote in message0 news:Pine.LNX.4.56.0308160118130.14612@jaipur...( > On Sat, 16 Aug 2003, Mike Naime wrote:F > > It all depends on what your management is willing to risk/pay for.I > > If I could convince my management to double my HSG count, I would nott need > > to have raidsets.  :-) > >uC > > In 3 years, we have had less than 1% failure rate in our disks.r6 > > Auto-sparing has made a drive failure a non-event. > >iH > > Anyone out there actually have an HSG80 raidset that did not rebuild! > > correctly on a drive failure?u >lG > Yep.  We had a RAID 0+1 set.  Total of 4 spindles.  One of the drivesfL > failed.  The HSG twiddled its thumbs and did nothing.  Turned out it was aJ > bug in the HSG firmware when you have redundant controllers.  Apparently3 > each thought the other was going to do something.e >d  L couple of questions here.  What Firmware (VCF) level was that?  How was your; controller configured?  Multibus failover?  Mirrored Cache? 5 4 members...  Gee your management really is paranoid.o  K This is not what I was asking.  I was asking about Raidset re-construction.  not mirrorset.   ADD RAID  vs ADD MIRROR.   > The fix was:, > Set the mirrorset policy to not autospare.8 > Manually add the spare drive to the reduced mirrorset.H > Set the mirrorset policy back to autospare (although this won't do any? > good until we reboot the controller with the fixed firmware).g > E > Of course it took several hours of reading manuals, pulling out our.H > hair, and a support call to figure out why our controller wasn't doingK > what it was configured to do.  Not to mention worrying that another drive G > didn't decide to die.  It's my experience that drive failures tend tow > happen in bursts.h >   L Yes  I have seen that too.  One week we had about 6 72GB drives fail.  TheseK where all purchased in the same PO for 8 HSG's.  (2 T-5 Configurations of 4a controllers each.)  I > > I have lost an entire channel twice in the BA-370 models.  Not yet ino theh
 > > EMA's. > K > We once had a scenario where both of our redundant I/O controllers failed G > simultaneously.  It was an HSJ40 or HSJ50 pair.  Don't quite rememberaI > since it was a couple of years back.  Host-based volume shadowing saved7 > our butts.  L I had a drive failure on a HSJ freeze the entire cluster.  It was a non-prodL cluster using CI. The first drive I had seen fail out of the HSJ's.  Since I= was used to the HSG's I was quite surprised and dissapointed.c   >nI > And given the glitch we've just experienced with autospare not working,tK > I sure wouldn't feel comfortable putting all my bags in the basket of thee  > I/O controller not messing up. >sD > What happens if a drive fails such that it locks a whole SCSI bus?H > This isn't theoretical, it could happen.  Do you have enough spares inK > one cabient so that if a whole bus fails, you're still covered?  Hardware ' > failure can cause all kinds of havoc.  >   J Management is not going to that level of paranoia.  One spare per channel.G When the BUS failed, it froze the HSG system.   It took re-booting bothiI controllers to fix it.  We had 2 production clusters on that HSG.  One ofhJ them the operators/system managers had hit Control-P on the consoles priorK to calling me.  THey called me when the system would not boot.  Not when ittI was first frozen and they where supposed to call me.  It was down until IsI got the controllers re-started.   The other cluster continued after I gott the HSG's re-booted.  F > What if the power to one of your racks goes nuts and surges for someL > reason and somehow manages to cook a lot of components?  Your rack is onlyJ > as good as the power coming to it.  We've had weird power events before,8 > but never lost a rack.  But nothing would surprise me. >.C Redundant power feeds from 2 different power grids, Multiple PDU's, I conditioners/batteries... Etc. Plus the generator.  It would take a whole.I lot for something like that to get through the system.  I would loose thes1 servers or the network before I lost the storage.s    L > You can call host-based shadowing a performance hit.  And if you are doingH > all writes, sure, it may be a performance hit.  But we are doing aboutL > 50-50 reads and writes.  The reads speed up, the writes slow down.  HardlyI > matters much since the write-back cache is on.  Besides, with the HSGs, K > we have enough performance margin even if a shadow merge kicks in.  Don't " > you have any performance margin? >e  I The SAN is not our problem.  The SAN switches and storage controllers arefF not taxed by any means.  During backups, if all traffic to one HSG wasJ entering one HSG port, we would see maybe 50%-60% controller usage.    TheD bottleneck is in the IO subsystem of the VMS boxes. If you are usingK shadowing for both  backups and redundancy, then you are making the OS send J that data Three times. Either backup route has you read/write the data. inK one form or another.  Are you pegging your VMS box during backups?  and for F how long?   I bet that the problem is not the SAN storage for you too.  H Where the shadowing hit is seen the most is when your server fails.  OneI site that I am aware of has clustered GS-160's using shadowed drives in 3 E HSG cabinets.  From their point of view, when a system crashes, their7I systems are unusable until after the shadow merges are finished.  This is.G there opinion based on user responce time.  This is what I am basing myEH opinion on since I do not have that hit when a clustered system crashes.  L > And... when I/O controllers decide to do stupid things and/or go belly-up,J > you don't lose access to your data.  It's your data and your time... how > much is it worth to you?  K NO, How much is it worth to Management!  How much will they buy off on.  IffF I had the design that I really wanted, I would have at least twice the" storage equipment that we do have.  E Overall, we see more system failures (OS/hardware) than we do storage 	 failures.k     Mike   ------------------------------  % Date: Fri, 15 Aug 2003 15:30:16 -0500 + From: Chuck Aaron <caaron@ceris.purdue.edu> " Subject: xml web services question/ Message-ID: <3F3D42D8.9080705@ceris.purdue.edu>    Group,  H Is anyone using xml web services with OpenVMS and if so can you tell me 2 what product you're using and how you're using it?   Thanks,e Chuck Aaronh VMS Systems Mgr. Purdue Univ.   ------------------------------   End of INFO-VAX 2003.452 ************************k" <n.rieck@sympatico.ca> wrote in message3 news:oWI_a.6506$kp4.869027@news20.bellglobal.com... 9 > "Dave Gudewicz" <k9jdk@NOSPAMarrl.nee; use PORT (198,151,12,104,8,119) <<< SIZE xvnc-tip2.txt >>> 213 3362 <<< RETR xvnc-tip2.txt\ >>> 150 ASCII retrieve of /disk$misc/decus/vmslt02b/vu/xvnc-tip2.txt (3362 bytes) started.: >>> 226 Transfer completed.  2867 (8) bytes transferred.
 <<< PASV@ >>> 227 Entering passive mode; use PORT (198,151,12,104,8,123)
 <<< LIST >>> 150 List started.> >>> 226 Transfer completed.V
 <<< PASV@ >>> 227 Entering passive mode; use PORT (198,151,12,104,8,125)
 <<< LIST >>> 150 List started.P >>> 226 Transfer completed.<
 <<< PASV@ >>> 227 Entering passive mode; use PORT (198,151,12,104,8,126) <<< LIST *fl*s >>> 150 List started.x >>> 226 Transfer completed.e <<< TYPE I >>> 200 Type I ok. <<< STRU O VMS >>> 200 Stru O VMS ok.
 <<< PASV@ >>> 227 Entering passive mode; use PORT (198,151,12,104,8,127) <<< SIZE vmsflash401.zip >>> 213 3229940t <<< RETR vmsflash401.zip_ >>> 150 VMS retrieve of /disk$misc/decus/vmslt02b/vu/vmsflash401.zip (3229940 bytes) started.>= >>> 226 Transfer completed.  3229500 (8) bytes transferred.i <<< TYPE A >>> 200 Type A ok. <<< STRU F >>> 200 Stru F ok.
 <<< PASV@ >>> 227 Entering passive mode; use PORT (198,151,12,104,8,130) <<< LIST *bz*V >>> 150 List started.o >>> 226 Transfer completed.i	 <<< PWDm> >>> 257 "/disk$misc/decus/vmslt02b/vu" is current directory. <<< CWD ..1 >>> 250 Connected to /disk$misc/decus/vmslt02b.e
 <<< PASV@ >>> 227 Entering passive mode; use PORT (198,151,1