1 INFO-VAX	Thu, 15 Jun 2006	Volume 2006 : Issue 330       Contents: DS10 big memory  Re: DS10 big memory  Re: DS10 big memory 
 ES40 problems  Re: ES40 problems  Re: ES40 problems  Re: ES40 problems  Re: ES40 problems  Re: ES40 problems  Re: ES40 problems  Re: ES40 problems  Re: FS: MIcroVAX 3100s (5) How To Pick Winning Stocks Re: HSG80 Serial connection  Re: Multi-homed DHCP server  Re: Multi-homed DHCP server # Re: Open VMS OS recovery on Itanium > Re: Slowin' machine while zipping or deleting big binary files> Re: Slowin' machine while zipping or deleting big binary files> Re: Slowin' machine while zipping or deleting big binary files> Re: Slowin' machine while zipping or deleting big binary files Re: Tape drive opinions  Re: Tape drive opinions  Re: Tape drive opinions  Re: Tape drive opinions   F ----------------------------------------------------------------------    Date: 14 Jun 2006 14:31:26 -0700! From: "Rod" <rregier@dymaxion.ca>  Subject: DS10 big memoryC Message-ID: <1150320686.601917.324130@c74g2000cwc.googlegroups.com>   . The Quickspecs for the DS10 say 2G max memory.  G Is that enforced by the hardware, or would it be possible to add 2 - 2G  memory options?   F Looks to me like the 2G DS15 memory option would work if the CPU would permit
 that size.   Tks    ------------------------------  # Date: Wed, 14 Jun 2006 23:42:21 GMT & From: "Hal Kuff" <halkuff@verizon.net> Subject: Re: DS10 big memory, Message-ID: <x91kg.42683$%m5.29669@trnddc04>  I Several DS10 models, some take 1gb and sme max at 2gb... need your model   number  - "Rod" <rregier@dymaxion.ca> wrote in message  = news:1150320686.601917.324130@c74g2000cwc.googlegroups.com... 0 > The Quickspecs for the DS10 say 2G max memory. > I > Is that enforced by the hardware, or would it be possible to add 2 - 2G  > memory options?  > H > Looks to me like the 2G DS15 memory option would work if the CPU would > permit > that size. >  > Tks  >    ------------------------------  % Date: Wed, 14 Jun 2006 20:30:40 -0400 , From: "Richard Tomkins" <tomkinsr@istop.com> Subject: Re: DS10 big memory8 Message-ID: <44909ccc$0$3686$88260bb3@free.teranews.com>  J If the system says 2GB then that's it. If you put in more memory, it mightL appear to work but as the address generation will wrap around, you'll end up crashing a lot.        --  = Posted via a free Usenet account from http://www.teranews.com    ------------------------------  % Date: Wed, 14 Jun 2006 11:03:20 -0700 # From: "Tom Linden" <tom@kednos.com>  Subject: ES40 problems) Message-ID: <op.ta5drua5zgicya@hyrrokkin>   / Just acquired a couple of ES40s and after a few F hiccups (stuck halt button, reseating boards) upgraded the firmware to& latest.  Now on one when I INIT  I get  , *** Error - incompatible DIMM 3 on MMB 0 ***  , *** Error - incompatible DIMM 4 on MMB 0 ***  , *** Error - incompatible DIMM 3 on MMB 1 ***  , *** Error - incompatible DIMM 4 on MMB 1 ***   yet 
 P00>>>sho mem 4    Array       Size       Base Address    Intlv Mode3 ---------  ----------  ----------------  ---------- 1      0       2048Mb     0000000000000000    1-Way   9 So it appears that the problem is Interleave is not 4-way   
 Any thoughts?   > Second problem, on both systems there is a DE500 PCI installedC and it appears that the onboard ethernet is not recognized.  Should ! I just assume that it has failed?   @ (These, BTW, ran Tru64 and were part of a 150 server farm all of? which were all replaced with Linux when HP dropped Tru64.  They 4 said there was no compelling reason to go to HP-UX.)   Tom    ------------------------------  % Date: Wed, 14 Jun 2006 14:27:20 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net>  Subject: Re: ES40 problems: Message-ID: <JsWdnU3uzI6OyA3ZnZ2dnUVZ_qmdnZ2d@comcast.com>   Tom Linden wrote:   1 > Just acquired a couple of ES40s and after a few H > hiccups (stuck halt button, reseating boards) upgraded the firmware to( > latest.  Now on one when I INIT  I get > . > *** Error - incompatible DIMM 3 on MMB 0 *** > . > *** Error - incompatible DIMM 4 on MMB 0 *** > . > *** Error - incompatible DIMM 3 on MMB 1 *** > . > *** Error - incompatible DIMM 4 on MMB 1 *** >  > yet  > P00>>>sho mem 5 >   Array       Size       Base Address    Intlv Mode 5 > ---------  ----------  ----------------  ---------- 2 >     0       2048Mb     0000000000000000    1-Way > ; > So it appears that the problem is Interleave is not 4-way  >  > Any thoughts?  > @ > Second problem, on both systems there is a DE500 PCI installedE > and it appears that the onboard ethernet is not recognized.  Should # > I just assume that it has failed?  > B > (These, BTW, ran Tru64 and were part of a 150 server farm all ofA > which were all replaced with Linux when HP dropped Tru64.  They 6 > said there was no compelling reason to go to HP-UX.) >  > Tom   F My memory is DIMM (sorry, I couldn't resist) but ISTR that there were G four memory boards in those boxes and you had to put matching DIMMs in  I corresponding sockets on each board; e.g. you put a 256MB DIMM in socket  I 0 on each board and you had 1024MB.  Add a second 256MB DIMM in socket 1  E of each board and you had 2048MB.  It's been about six years since I  I installed memory in one of those beasts!   Be very sure that you install  4 the same make and model DIMMs in each set of socket.  F I had two ethernet ports on both of my ES40s.  One pair connected the H two back to back with a crossover cable.  The other pair were connected C to a Cisco switch in the normal fashion.  Never had a problem with  : either set.  What do you see if you remove the two DE500s?   ------------------------------  % Date: Wed, 14 Jun 2006 11:47:19 -0700 # From: "Tom Linden" <tom@kednos.com>  Subject: Re: ES40 problems) Message-ID: <op.ta5fs5e1zgicya@hyrrokkin>   9 On Wed, 14 Jun 2006 11:27:20 -0700, Richard B. Gilbert  =    <rgilbert88@comcast.net> wrote:    > Tom Linden wrote:  > 2 >> Just acquired a couple of ES40s and after a fewI >> hiccups (stuck halt button, reseating boards) upgraded the firmware t=  o ) >> latest.  Now on one when I INIT  I get 0 >>  *** Error - incompatible DIMM 3 on MMB 0 ***0 >>  *** Error - incompatible DIMM 4 on MMB 0 ***0 >>  *** Error - incompatible DIMM 3 on MMB 1 ***0 >>  *** Error - incompatible DIMM 4 on MMB 1 *** >>  yet  >> P00>>>sho mem6 >>   Array       Size       Base Address    Intlv Mode6 >> ---------  ----------  ----------------  ----------3 >>     0       2048Mb     0000000000000000    1-Way = >>  So it appears that the problem is Interleave is not 4-way  >>  Any thoughts? B >>  Second problem, on both systems there is a DE500 PCI installedF >> and it appears that the onboard ethernet is not recognized.  Should$ >> I just assume that it has failed?D >>  (These, BTW, ran Tru64 and were part of a 150 server farm all ofB >> which were all replaced with Linux when HP dropped Tru64.  They7 >> said there was no compelling reason to go to HP-UX.)  >>  Tom  > I > My memory is DIMM (sorry, I couldn't resist) but ISTR that there were =   =  I > four memory boards in those boxes and you had to put matching DIMMs in=    =   I > corresponding sockets on each board; e.g. you put a 256MB DIMM in sock=  et  =   I > 0 on each board and you had 1024MB.  Add a second 256MB DIMM in socket=   1  =   I > of each board and you had 2048MB.  It's been about six years since I  =   I > installed memory in one of those beasts!   Be very sure that you insta=  ll  =   6 > the same make and model DIMMs in each set of socket.  I One system has 4 memory boards, each with 4 DIMMs,  occuping the 4 slots=   4 available and >>> sho mem reveals  4 way interleaved  @ The other system has only two memory boards in 2nd and 4th slots  I Now checking the the boards they are all labeled 54-25582-02, but the DI=  MMs  are different.   > I > I had two ethernet ports on both of my ES40s.  One pair connected the =   =  I > two back to back with a crossover cable.  The other pair were connecte=  d  =  G > to a Cisco switch in the normal fashion.  Never had a problem with  =   < > either set.  What do you see if you remove the two DE500s?  I I guess I was unclear.  There is an ethernet port as part of the basic  =    chassis G on the upper left corner, when viewed from behind.  In addition each  =   
 system hasI a DE500 PCI board installed, and a >>> sho config  on either system only=    =    reveals 	 that one.    ------------------------------  % Date: Wed, 14 Jun 2006 11:50:47 -0700 # From: "Tom Linden" <tom@kednos.com>  Subject: Re: ES40 problems) Message-ID: <op.ta5fyxr2zgicya@hyrrokkin>   F On Wed, 14 Jun 2006 11:47:19 -0700, Tom Linden <tom@kednos.com> wrote:  ; > On Wed, 14 Jun 2006 11:27:20 -0700, Richard B. Gilbert  =   ! > <rgilbert88@comcast.net> wrote:  >  >> Tom Linden wrote: >>3 >>> Just acquired a couple of ES40s and after a few I >>> hiccups (stuck halt button, reseating boards) upgraded the firmware =  to* >>> latest.  Now on one when I INIT  I get1 >>>  *** Error - incompatible DIMM 3 on MMB 0 *** 1 >>>  *** Error - incompatible DIMM 4 on MMB 0 *** 1 >>>  *** Error - incompatible DIMM 3 on MMB 1 *** 1 >>>  *** Error - incompatible DIMM 4 on MMB 1 ***  >>>  yet >>> P00>>>sho mem 7 >>>   Array       Size       Base Address    Intlv Mode 7 >>> ---------  ----------  ----------------  ---------- 4 >>>     0       2048Mb     0000000000000000    1-Way> >>>  So it appears that the problem is Interleave is not 4-way >>>  Any thoughts?C >>>  Second problem, on both systems there is a DE500 PCI installed G >>> and it appears that the onboard ethernet is not recognized.  Should % >>> I just assume that it has failed? E >>>  (These, BTW, ran Tru64 and were part of a 150 server farm all of C >>> which were all replaced with Linux when HP dropped Tru64.  They 8 >>> said there was no compelling reason to go to HP-UX.) >>>  Tom >>I >> My memory is DIMM (sorry, I couldn't resist) but ISTR that there were=    =   I >> four memory boards in those boxes and you had to put matching DIMMs i=  n  =  G >> corresponding sockets on each board; e.g. you put a 256MB DIMM in  =   I >> socket 0 on each board and you had 1024MB.  Add a second 256MB DIMM i=  n  =  I >> socket 1 of each board and you had 2048MB.  It's been about six years=    =   I >> since I installed memory in one of those beasts!   Be very sure that =   =  C >> you install the same make and model DIMMs in each set of socket.  > I > One system has 4 memory boards, each with 4 DIMMs,  occuping the 4 slo=  ts6 > available and >>> sho mem reveals  4 way interleaved > B > The other system has only two memory boards in 2nd and 4th slots > I > Now checking the the boards they are all labeled 54-25582-02, but the =   =   > DIMMs  > are different. >  >>I >> I had two ethernet ports on both of my ES40s.  One pair connected the=    =   I >> two back to back with a crossover cable.  The other pair were connect=  ed  =   H >> to a Cisco switch in the normal fashion.  Never had a problem with  =  = >> either set.  What do you see if you remove the two DE500s?  > I > I guess I was unclear.  There is an ethernet port as part of the basic=    =   	 > chassis I > on the upper left corner, when viewed from behind.  In addition each  =    > system hasI > a DE500 PCI board installed, and a >>> sho config  on either system on=  ly  =   	 > reveals  > that one.   2 Whoops!  bad eyesight, these are MMJ not ethernet.   ------------------------------  % Date: Wed, 14 Jun 2006 15:08:12 -0400   From: "DBT" <dbturner@icusc.com> Subject: Re: ES40 problems0 Message-ID: <1290nm7jru6b0a8@news.supernews.com>   Tom   > ON the memory/cpu cover panel there is an installation diagramK Follow it to the T and I bet your memory problems (with the system that is) 
 will go away. 4 Oh, and ES40 systems DO NOT have a built in ethernet   DT   --     David B Turner Island Computers US Corp 2700 Gregory St, Suite 180 Savannah GA 31404  Tel: 912 447 6622 X201 Cell: 912 447 6622 X251  Fax: 912 201 0402  Email: dbturner@icusc.com  Web: http://www.islandco.com% ===================================== < All orders are subject to the following terms and conditions. of sale. These should be read before ordering.% http://www.islandco.com/warranty.html   . "Tom Linden" <tom@kednos.com> wrote in message# news:op.ta5drua5zgicya@hyrrokkin... 1 > Just acquired a couple of ES40s and after a few H > hiccups (stuck halt button, reseating boards) upgraded the firmware to( > latest.  Now on one when I INIT  I get > . > *** Error - incompatible DIMM 3 on MMB 0 *** > . > *** Error - incompatible DIMM 4 on MMB 0 *** > . > *** Error - incompatible DIMM 3 on MMB 1 *** > . > *** Error - incompatible DIMM 4 on MMB 1 *** >  > yet  > P00>>>sho mem 6 >    Array       Size       Base Address    Intlv Mode5 > ---------  ----------  ----------------  ---------- 3 >      0       2048Mb     0000000000000000    1-Way  > ; > So it appears that the problem is Interleave is not 4-way  >  > Any thoughts?  > @ > Second problem, on both systems there is a DE500 PCI installedE > and it appears that the onboard ethernet is not recognized.  Should # > I just assume that it has failed?  > B > (These, BTW, ran Tru64 and were part of a 150 server farm all ofA > which were all replaced with Linux when HP dropped Tru64.  They 6 > said there was no compelling reason to go to HP-UX.) >  > Tom    ------------------------------  % Date: Wed, 14 Jun 2006 12:57:11 -0700 # From: "Tom Linden" <tom@kednos.com>  Subject: Re: ES40 problems) Message-ID: <op.ta5i1lx5zgicya@hyrrokkin>   C On Wed, 14 Jun 2006 12:08:12 -0700, DBT <dbturner@icusc.com> wrote:    > Tom  > @ > ON the memory/cpu cover panel there is an installation diagramI > Follow it to the T and I bet your memory problems (with the system tha=  t  =   > is)  > will go away. 6 > Oh, and ES40 systems DO NOT have a built in ethernet >  > DT > A Thanks, David.  There are only two memory boards in MMB0 and MMB1 B each with 4 DIMMs, so I guess what is happenning is that I am only5 using the lower two DIMMs (set 0) on the two MMBs for   ' Memory Testing and Configuration Status 4    Array       Size       Base Address    Intlv Mode3 ---------  ----------  ----------------  ---------- 1      0       2048Mb     0000000000000000    1-Way          2048 MB of System Memory  D i.e. the MMBs are 2GB each but I am only able to see the iGB on each board in set 0  F So by adding two more boards of this type I should end up with 8GB and 4-way interleave.   : Would you like to swap these two boards for 4 1 GB boards?   ------------------------------  % Date: Wed, 14 Jun 2006 16:36:57 -0400 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net>  Subject: Re: ES40 problems: Message-ID: <SMWdncfFo6737g3ZnZ2dnUVZ_sOdnZ2d@comcast.com>   Tom Linden wrote:   : > On Wed, 14 Jun 2006 11:27:20 -0700, Richard B. Gilbert  ! > <rgilbert88@comcast.net> wrote:  >  >> Tom Linden wrote: >>3 >>> Just acquired a couple of ES40s and after a few J >>> hiccups (stuck halt button, reseating boards) upgraded the firmware to* >>> latest.  Now on one when I INIT  I get1 >>>  *** Error - incompatible DIMM 3 on MMB 0 *** 1 >>>  *** Error - incompatible DIMM 4 on MMB 0 *** 1 >>>  *** Error - incompatible DIMM 3 on MMB 1 *** 1 >>>  *** Error - incompatible DIMM 4 on MMB 1 ***  >>>  yet >>> P00>>>sho mem 7 >>>   Array       Size       Base Address    Intlv Mode 7 >>> ---------  ----------  ----------------  ---------- 4 >>>     0       2048Mb     0000000000000000    1-Way> >>>  So it appears that the problem is Interleave is not 4-way >>>  Any thoughts?C >>>  Second problem, on both systems there is a DE500 PCI installed G >>> and it appears that the onboard ethernet is not recognized.  Should % >>> I just assume that it has failed? E >>>  (These, BTW, ran Tru64 and were part of a 150 server farm all of C >>> which were all replaced with Linux when HP dropped Tru64.  They 8 >>> said there was no compelling reason to go to HP-UX.) >>>  Tom >> >>J >> My memory is DIMM (sorry, I couldn't resist) but ISTR that there were  G >> four memory boards in those boxes and you had to put matching DIMMs  I >> in  corresponding sockets on each board; e.g. you put a 256MB DIMM in  H >> socket  0 on each board and you had 1024MB.  Add a second 256MB DIMM G >> in socket 1  of each board and you had 2048MB.  It's been about six  J >> years since I  installed memory in one of those beasts!   Be very sure I >> that you install  the same make and model DIMMs in each set of socket.  >  > J > One system has 4 memory boards, each with 4 DIMMs,  occuping the 4 slots6 > available and >>> sho mem reveals  4 way interleaved > B > The other system has only two memory boards in 2nd and 4th slots > M > Now checking the the boards they are all labeled 54-25582-02, but the DIMMs  > are different. >  The DIMMS must be the same!  >>J >> I had two ethernet ports on both of my ES40s.  One pair connected the  A >> two back to back with a crossover cable.  The other pair were  D >> connected  to a Cisco switch in the normal fashion.  Never had a K >> problem with  either set.  What do you see if you remove the two DE500s?  >  > J > I guess I was unclear.  There is an ethernet port as part of the basic  	 > chassis H > on the upper left corner, when viewed from behind.  In addition each   > system hasF > a DE500 PCI board installed, and a >>> sho config  on either system  > only  reveals  > that one.   H I suppose that I was less than clear as well.  If you remove the DE500, % do you then see an on-board Ethernet?   G It has been nearly two years since I last saw the ES40s I once had.  I  I do not recall an on-board ethernet.   My recollection is two DE500-AA in  	 each box.    ------------------------------  % Date: Wed, 14 Jun 2006 17:30:03 -0400 . From: "Carl Friedberg" <frida.fried@gmail.com> Subject: Re: ES40 problemsI Message-ID: <890539d90606141430p632f43a7p4464efde54f081a2@mail.gmail.com>   @ I love ES40's and ES45's, but the design of the memory subsystem leaves a lot to be desired.   F The starting point is either the very clear drawing on the backside ofB the big top cover, or the manual (it can be downloaded in pdf form8 from the HP website, and you should get it and read it).    Here's my experience and advice:  D (1) If you have 4 processors, you probably want to have 4-way memory" interleaving for best performance.  D (2) To get that, you need 16 DIMMs, and I think they probably shouldE be identical (4 sets of 4). A good choice would be 4 sets of 1GB each A (256mb DIMMs). Use Compaq branded memory. I've found that current C pricing has Compaq branded memory in the same ballpark as 3rd party 	 products.   E (3) >>> SHOW FRU || MORE is your friend here. The second column is an D error counter, and will show you accumulated error counts. Those can> be reset with a CLEAR_ALL (can;t find this documented, from my: memory...) followed by an INIT (possibly, by power off/on.  B (4) It can take a LONG time to get through all of the memory tests with 8 or 16 GB.   (5) Here's the kicker A Those memory riser cards (with the 8 DIMM sockets) have a serious C flaw. I have found that it is hard, to nearly impossible, to insert F DIMMs and have them behave correctly the first time around. My rule ofA thumb is: Once I pull out a riser board (one of MMB2, MMB0, MMB1, D MMB3), I don't touch another board until everything on that board is1 recognized. This may require the following steps:    o  power off system    o  pull out all power cords    o  open top cover ,   o  lift up (carefully) a single riser cardC   o  remove all the DIMMs in that card, and reseat them in the same F place. It helps to write down (or record, if you've got a console portF running into a DECserver or PC) the results of show FRU to be sure you? are keeping these DIMMs in the same place. I think only genuine F Digital/Compaq memory displays the serial numbers properly, but again,3 I only put compaq branded memory in these machines. E   o as you are doing the reseating, visually inspect each DIMM socket D for possible damage. Depending on age, a can of compressed air might be helpful. ?   o  inspect the socket on the system board, and clean as above 9   o re-insert the memory riser card into the system board    o  close the top   o  put the cover back on   o  plug in the power cordsE   o power on, and watch the series of displays in the LCD, which will 9 tell you quite specific errors during the memory testing.   C The above takes 15 minutes after you've done it many times. But, my A experience has been that if the DIMMs are not defective, you will 5 eventually get all of them to pass the initial tests.   > IGNORE all errors about other MMBS, except for the one you areD currently working on. They might just go away on their own, when theF current MMB has everything recognized. It sounds crazy, but as I said,4 there's some design issues around these riser cards.  / Then, you can move on to the next memory board.   D Following the above procedure, I have had good luck with 4 different@ ES40 and ES45 models; I've only had 2 bad DIMMs in 6 years. I'veB installed a lot of different memory (mostly watching field service techs, but some of it myself).   Carl Friedberg  . On 6/14/06, Tom Linden <tom@kednos.com> wrote:1 > Just acquired a couple of ES40s and after a few H > hiccups (stuck halt button, reseating boards) upgraded the firmware to( > latest.  Now on one when I INIT  I get > . > *** Error - incompatible DIMM 3 on MMB 0 *** > . > *** Error - incompatible DIMM 4 on MMB 0 *** > . > *** Error - incompatible DIMM 3 on MMB 1 *** > . > *** Error - incompatible DIMM 4 on MMB 1 *** >  > yet  > P00>>>sho mem 6 >    Array       Size       Base Address    Intlv Mode5 > ---------  ----------  ----------------  ---------- 3 >      0       2048Mb     0000000000000000    1-Way  > ; > So it appears that the problem is Interleave is not 4-way  >  > Any thoughts?  > @ > Second problem, on both systems there is a DE500 PCI installedE > and it appears that the onboard ethernet is not recognized.  Should # > I just assume that it has failed?  > B > (These, BTW, ran Tru64 and were part of a 150 server farm all ofA > which were all replaced with Linux when HP dropped Tru64.  They 6 > said there was no compelling reason to go to HP-UX.) >  > Tom  >     . On 6/14/06, Tom Linden <tom@kednos.com> wrote:1 > Just acquired a couple of ES40s and after a few H > hiccups (stuck halt button, reseating boards) upgraded the firmware to( > latest.  Now on one when I INIT  I get > . > *** Error - incompatible DIMM 3 on MMB 0 *** > . > *** Error - incompatible DIMM 4 on MMB 0 *** > . > *** Error - incompatible DIMM 3 on MMB 1 *** > . > *** Error - incompatible DIMM 4 on MMB 1 *** >  > yet  > P00>>>sho mem 6 >    Array       Size       Base Address    Intlv Mode5 > ---------  ----------  ----------------  ---------- 3 >      0       2048Mb     0000000000000000    1-Way  > ; > So it appears that the problem is Interleave is not 4-way  >  > Any thoughts?  > @ > Second problem, on both systems there is a DE500 PCI installedE > and it appears that the onboard ethernet is not recognized.  Should # > I just assume that it has failed?  > B > (These, BTW, ran Tru64 and were part of a 150 server farm all ofA > which were all replaced with Linux when HP dropped Tru64.  They 6 > said there was no compelling reason to go to HP-UX.) >  > Tom  >    ------------------------------  % Date: Wed, 14 Jun 2006 21:10:40 -0500 6 From: "David J. Dachtera" <djesys.no@spam.comcast.net># Subject: Re: FS: MIcroVAX 3100s (5) 0 Message-ID: <4490C1A0.5B9F6F53@spam.comcast.net>   "David J. Dachtera" wrote: >  > Folks, > I > Now that bootcamp is history for this year, I have time to get to these  > machines.  > H > Before I list them on eBay, I want to give the group a chance at them. > D > I'm only selling these for the cost of shipping which is beyond my. > control. These are HEAVY, even without disk. > 4 > They all have only the on-board memory - no SIMMs. >  > They all have no disk or CD. > & > They all boot from CD and pass POST. > H > I will consider international shipping, but understand that the delays6 > may be significant while I gain experience doing so. > C > I recommend e-mailing me at djesys at comcast dot net where I get  > relatively little spam.  > H > These will be posted for sale probably next weekend. The postings will > be Buy-It-Now only.    By the way, folks...  E I haven't forgotten these. The wife wants stuff done around the house H and that's taking priority. Funny how spending money takes priority over building business...   --   David J Dachtera dba DJE Systems  http://www.djesys.com/  & Unofficial OpenVMS Marketing Home Page! http://www.djesys.com/vms/market/   ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/   ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/   ------------------------------  # Date: Thu, 15 Jun 2006 03:45:40 GMT  From: info@stockinvesting.com # Subject: How To Pick Winning Stocks 5 Message-ID: <EJ4kg.40703$hp.20928@read2.cgocable.net>   6 To pick winning stocks, you need the tools to separate8 the companies with superior sales and profit growth from4 the ones who are just making a lot of noise. Besides3 choosing the right companies to invest in, you must 8 learn how to read the market's trend so that your timing
 is right.   8 And picking the next big winner based on the news is not9 as likely as picking the next big loser. The news media's 5 time-frame is usually very short, sometimes just that  particular day!    full article here:5 http://www.swinginvesting.com/winning-stock-pick.html    ------------------------------  % Date: Wed, 14 Jun 2006 21:18:21 -0500 6 From: "David J. Dachtera" <djesys.no@spam.comcast.net>$ Subject: Re: HSG80 Serial connection0 Message-ID: <4490C36D.EB0244CC@spam.comcast.net>  ! "johnhreinhardt@yahoo.com" wrote:  > H > Does anybody know the pinout or a source of a cable to connect a HSG80H > serial console to a 8-pin RJ45 DECserver 900TM port?  The HSG80 uses aE > regular 6-pin RJ12 connector rather than the "usual" DEC MMJ style. H > Unless I've missed something in the FAQ and while googling I don't see? > any DECconnect type cable which has a regular RJ12 connector.   A HSG80 has the same pinout as an MMJ jack. For the flat DECconnect E cables, you could just snip off the MMJ plug and crimp an RJ12 in its = place (facing the same direction so it's still "null modem").    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  & Unofficial OpenVMS Marketing Home Page! http://www.djesys.com/vms/market/   ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/   ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/   ------------------------------  % Date: Wed, 14 Jun 2006 19:50:56 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> $ Subject: Re: Multi-homed DHCP server, Message-ID: <4490A0D3.E4E18727@teksavvy.com>   "Michael T. Davis" wrote: N >         A few weeks ago, we added an interface in TCPIP to provide access to2 > a secondary subnet which resides on our network.  B I am on VAX with older version (but since Join no longer exists, I4 suspect no much has been done with the DHCP server).    E If you look at the TCPIP$DHCP_RUN.LOG , you'll see that when it lists H its configuration, it has and option to ignore a list of interfaces. NotA sure if this is accessible from the DHCP$GUI interface, but it is B probably configurable if you go into the actual text config files.  F Note that a DHCP CLIENT sends a broadcast from a bogus IP address. Any? DHCP server on that ethernet segment can reply to it (also as a N broadcast because at that point, the client still doesn't have an IP address).  C So, from the DHCP Server point of view, it can discriminate between  clients based on:  	-their ethernet address (MAC)7 	-some host name their desire (included in the request) ; 	-or which physical interface the broadcast is coming from.   F In your case, since you have multiple IP subnets on the same ethernet,H the DHCP server has no logical way to know if a request is coming from a+ PC supposed to be on one subnet or another.     A If you do a NETSTAT, do you see port 67 being listened on on each  interface ?   E (On my system, it is strange, it listens on the cluster alias, on the B main interface as well as *.67 and the TCPIP service is defined as; listening to 0.0.0.0 (which results in the NETSTAT *.67 and + automatically listens to any IP interface).    ------------------------------  % Date: Thu, 15 Jun 2006 00:08:12 -0400 - From: "John E. Malmberg" <wb8tyw@qsl.network> $ Subject: Re: Multi-homed DHCP server; Message-ID: <oK6dnab3tK6sQA3ZnZ2dnUVZ_oKdnZ2d@adelphia.com>    Followups set to comp.os.vms.    Michael T. Davis wrote:  > 	We're running...  > 5 >   HP TCP/IP Services for OpenVMS Alpha Version V5.5 / >   on an AlphaServer DS15 running OpenVMS V8.2  > G > 	A few weeks ago, we added an interface in TCPIP to provide access to O > a secondary subnet which resides on our network.  (We have both a /22 network L > [primary] and a /24 network [secondary] on the same wire [single collisionI > domain].)  The TCPIP$DHCP_RUN.LOG file started reporting problems (i.e. M > "network not administered by server"), but DHCP-served clients still seemed I > to be maintaining connectivity.  Here's basically how the system looked 3 > (obfuscated/trimmed TCPIP SHOW INTERFACE output):   * DHCP does not care about collision domain.  . DHCP only cares if DHCP server can be reached.  E By default, routers will block DHCP traffic, but that can be changed.   D All DHCP servers that a client can see must work together.  This is H needed to prevent more than one server giving the same I.P. address out  to different DHCP clients.  @ This is easily done with out clustering and even with different C platforms by having each DHCP server have it's own unique range of   addresses that it gives out.  F If a client needs a specific I.P. address, then all DHCP servers must I know to either give that I.P. address or ignore the request.  Some types  G of DHCP servers do not know how to ignore that type of request so that  
 is harder.  I The simple solution is to have all DHCP servers know about the fixed MAC   to DHCP assignments.  H A DHCP server may be configured to pay attention to the subnet that the H DHCP request came in on.  I have not worked with that configuration, so " I do not know the details of that.  * > Interface   IP_Addr         Network mask > + >  IE1        192.168.80.3    255.255.255.0 ' >  LO0        127.0.0.1       255.0.0.0 + >  WE0        192.168.72.3    255.255.252.0  > G > 	Last night, we were forced to restart the system and after that, our O > DHCP server was no longer honoring requests.  I removed the ...80.3 interface O > and restarted the DHCP server, and now all seems to be working as far as DHCP  > is concerned.  >  >  Questions > G > 	Is there a configuration for HP's DHCP server that will work in this O > environment?  That is, given the server is connected to multiple subnets that M > share the same collision domain, can the DHCP server handle this and if so, O > how?  Strictly speaking, we only need to serve DHCP clients in the ...72.0/22 O > subnet, and we only need a "presence" in the ...80.0/24 subnet on the server. K > On the other hand, if we could also handle DHCP clients in the ...80.0/24 P > subnet, that would be ideal.  If such a configuration isn't possible with HP'sJ > OpenVMS tools, is there something that would work?  (We have a number of% > Windows 2003 servers, for example.)   F What you have is a "router on a stick" configuration, or two networks 0 that have only some systems that can reach both.  D In general, no matter what platform you have, DHCP serving multiple G subnets on the same wire is not something that I would ever want to do.   G It only takes one configuration glitch to start a network shutdown for  ? the DHCP clients, and it may take a while for it to be noticed.   E That is because until a DHCP client is rebooted, it will continue to  G always go to the specific DHCP server that leased it it's I.P. address  , unless that DHCP server becomes unavailable.  I I have not worked enough with HP's DHCP server to know it's capabilities   or limitations in this area.  
 >  Caveats > C > 	Note that the IP addresses used here are not the real values and N > allocation is strictly managed by another entity.  The simple solution wouldL > be to move to a single /21 subnet, but address space to support this isn't, > available.  NATting is also not an option.  H If the routing tables are set up correctly on all systems or there is a I router on a stick, it should not matter if a DHCP client gets an address  * from the 80.0/24 pool or the 72.0/22 pool.  E And if that is the case, then all it is a matter of is that any DHCP  D server that can receive a client broadcast can service that request.  C And the simplest is to allocate each DHCP server a fragment of the  G available address pool.  Of course that requires you to have much more  & I.P. addresses available than clients.  D Effectively for running DHCP servers that do not use a common lease E database or only give out to fixed I.P. address, the total number of  E I.P. addresses you need to be able to offer is the maximum number of  1 clients multiplied by the number of DHCP servers.   D If you can get NAT to be an option, properly applied, it can really C simplify your network management, because then you can put all the  : non-server devices on an reserved RFC non-routable subnet.  G And the NAT only has to be at the firewall to the public Internet.  It  H is not needed to internally use RFC non-routable subnets.  All you have C to do is change your internal routing tables to treat them as valid 2 internal addresses to route with in your intranet.  E If you already have a NAT gateway at the firewall, then it is likely  F that you could migrate all the DHCP allocated systems to something in G the 10.1/16 range without anyone noticing the change, unless they were  K running a server on using an external dynamic DNS provider like dyndns.org.   H It also foils a very popular spammer trick of hosting web sites and DNS & servers zombied systems in DHCP pools.   -John  wb8tyw@qsl.network Personal Opinion Only    ------------------------------  % Date: Wed, 14 Jun 2006 15:36:09 -0500 ' From: "Earl Lakia" <elakia@hotmail.com> , Subject: Re: Open VMS OS recovery on Itanium: Message-ID: <IaGdnfbrubKg7g3ZnZ2dnUVZ_r6dnZ2d@comcast.com>  ' Thanks to John Egoff, problem resolved.   = That though things are working using the EFI, when VMS boots, D the correct serial device must be connected.  I had the console portD connected into the MP serial console port, but had the COM1 selectedB as the output device.  So my Hang was that the output was going to> never never land and since the VGA port window doesn't come toE life until later when VMS boots, nothing was showing up there either.   . Actually, both of my disks boot again as well.   I had to select the following : * Acpi(HWP0002,700)/Pci(1|1)/Uart(9600 N81)/VenMsg(Vt100+)  
 instead of  / * Acpi(PNP0501,0)/Uart(9600 N81)/VenMsg(PcAnsi)   C The problem I had was I had also done a shell command DEFAULT which J sets it back to factory defaults and I hadn't noticed the slight change inA the settings.  I even got my old DAT4 SCSI tape drive from an old ? NAS box connected to the external SCSI port.  Came up as MKB600  (SCSI Id was 6).  
 Life is good.       2 "Earl Lakia" <elakia@hotmail.com> wrote in message4 news:N5ydnfF_IdQouRLZnZ2dneKdnZydnZ2d@comcast.com.... > During one of the HP DSPP sponsored events I > 1 > received an RX2600 server.  This system doesn't  > ) > have any real way of backing itself up.  >  >  > ? > Anyway after an 8.1 Field Test to 8.2, I find that neither of  > 0 > my disks will boot.  Additionally, I was never > < > given any VMS distribution (there weren't enough copies at > 7 > the event).  I was able to get a VMS 8.2 distribution  > 3 > from my customer hoping the DVD would allow me to  > # > restore to a virgin distribution.  >  >  > 1 > I tried to boot it with the EFI and it seems to  > : > hit the DVD for sometime, but then hangs.  Additionally, > < > I am able to boot the Linux kit and the offline diagnostic > 3 > utilities CDs.  Looking at my documentation under  > 3 > "Restoring the OS", only HPUX, Windows, and Linux  > 4 > are mentioned.  What is the procedure for OpenVMS? >  >  > 4 > I also ran the offline diagnostics and all of them > 8 > passed.  Not to say there isn't still something wrong, > 7 > but things seem to be functioning.  Don't know if the  > 5 > diagnostics really test interrupts, etc.  But I can  > 6 > place data to and from the EFI partition on the SCSI > 	 > drives.  >  >  > 6 > Any help out there?  Should the VMS 8.2 distribution > 6 > DVD boot?  I have seen the documentation on updating > : > from a zip download file (this is what I did originally) > : > but not for an OpenVMS system recovery from a DVD or CD. >  >    ------------------------------  % Date: Wed, 14 Jun 2006 20:49:17 +0200 / From: Paul Sture <paul.sture.nospam@hispeed.ch> G Subject: Re: Slowin' machine while zipping or deleting big binary files ; Message-ID: <276c3$44905a2e$50db5015$26116@news.hispeed.ch>    xxrmanza wrote:  > Hi,  > I > I have a XP1000 with OpenVMS V7.2-1 and when I zip or delete a big file H > (120000 blocks) it slows so much that becomes almost useless for otherH > users to work in. I have tried to do it with a default user (no privs,@ > no  big quotas) and it is the same. Can you provide me a hint? >  > Thanx in advance and cheers  >    What version of ZIP?   ------------------------------  # Date: Wed, 14 Jun 2006 19:27:33 GMT ' From: jls <jeffls-nospam@sbcglobal.net> G Subject: Re: Slowin' machine while zipping or deleting big binary files 8 Message-ID: <4fo092dmspsi4ve2r95g1ljlatpj9deo8b@4ax.com>  > On 14 Jun 2006 05:20:15 -0700, "xxrmanza" <rrmj@gmv.es> wrote:   >Hi, > H >I have a XP1000 with OpenVMS V7.2-1 and when I zip or delete a big fileG >(120000 blocks) it slows so much that becomes almost useless for other G >users to work in. I have tried to do it with a default user (no privs, ? >no  big quotas) and it is the same. Can you provide me a hint?  >  >Thanx in advance and cheers   Could be multiple problems:   ? CPU - you could be using more of the CPU for a long time.  On a D single-cpu server this can be very noticeable by others trying to doE things.  This can be helped along if you set your process priority to F at least 1 below the normal priority.  In fact, I'd setup low priority0 batch queue(s) to submit these sorts of jobs to.  C DISK - you could be causing an I/O bottleneck on your disk(s).  CPU F priority adjustment won't help you in this case.  You'd need to chooseF a disk that other users aren't needing to use.  Or submit to batch andF have it execute during a time of day/night when other users are not on the system.   D MEMORY - you could be utilizing so much memory that there isn't muchF available for other users, you could be thrashing with page faults too? much if your working set is too small, you could be causing the ? modified page writer to work harder (and interrupt other users' 	 work)...    E There are just too many possibilities of what could be the problem to > expect a valid answer.  Learning how to manage performance and troubleshooting is imperative.   ------------------------------    Date: 14 Jun 2006 14:43:19 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) G Subject: Re: Slowin' machine while zipping or deleting big binary files 3 Message-ID: <kJfdS6jdG9cr@eisner.encompasserve.org>   c In article <1150287615.010227.35250@u72g2000cwu.googlegroups.com>, "xxrmanza" <rrmj@gmv.es> writes:  > Hi,  > I > I have a XP1000 with OpenVMS V7.2-1 and when I zip or delete a big file H > (120000 blocks) it slows so much that becomes almost useless for otherH > users to work in. I have tried to do it with a default user (no privs,@ > no  big quotas) and it is the same. Can you provide me a hint?      set process/priority=1    ------------------------------    Date: 14 Jun 2006 14:44:26 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) G Subject: Re: Slowin' machine while zipping or deleting big binary files 3 Message-ID: <XgPqV3G+FNeY@eisner.encompasserve.org>   c In article <1150287615.010227.35250@u72g2000cwu.googlegroups.com>, "xxrmanza" <rrmj@gmv.es> writes:  > Hi,  > I > I have a XP1000 with OpenVMS V7.2-1 and when I zip or delete a big file H > (120000 blocks) it slows so much that becomes almost useless for otherH > users to work in. I have tried to do it with a default user (no privs,@ > no  big quotas) and it is the same. Can you provide me a hint?      set volume/nohighwater  ?  B    Normally I would not expect highwater marking to affect delete.   ------------------------------    Date: 14 Jun 2006 20:55:44 -02006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)  Subject: Re: Tape drive opinions, Message-ID: <449077d0$1@news.langstoeger.at>  U In article <e6pe1p$5br$1@naig.caltech.edu>, David Mathog <mathog@caltech.edu> writes: A >Short story:  I'm looking for opinions on the current tape drive B >options for backups in the 400Gb -> 2Tb range.  The new LTO-3 andA >SDLTII cartridges hold 800Gb and 600Gb compressed, respectively. F >A single drive of either would probably satisfy our backup needs for G >another 2, maybe 3 years.  By that time the tape capacity would likely A >have doubled again and we would buy another drive.  Or there are B >tape libraries available for both formats and those would satisfy= >the capacity needs longer, albeit at the cost of multivolume H >backups.  What do you folks suggest?  Please comment upon reliability, E >cost, and/or performance.  Any other tape format we should consider?   J We decided for LTO3 (MSL6000 Series) with MRU and VMS (previously SLS too)   Have you already seen C 	http://www.openvms.digital.com/openvms/storage/smstape_matrix.html    --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  % Date: Wed, 14 Jun 2006 15:09:45 -0400   From: "DBT" <dbturner@icusc.com>  Subject: Re: Tape drive opinions0 Message-ID: <1290nous6ihc6ca@news.supernews.com>  K We just shipped 4 x Quantum LTO3 400/800GB  rackmounted to a customer. (VMS  Customer too!)I They are very very quick and were no problem installation and backup-wise    DT   --     David B Turner Island Computers US Corp 2700 Gregory St, Suite 180 Savannah GA 31404  Tel: 912 447 6622 X201 Cell: 912 447 6622 X251  Fax: 912 201 0402  Email: dbturner@icusc.com  Web: http://www.islandco.com% ===================================== < All orders are subject to the following terms and conditions. of sale. These should be read before ordering.% http://www.islandco.com/warranty.html   4 "David Mathog" <mathog@caltech.edu> wrote in message% news:e6pe1p$5br$1@naig.caltech.edu... B > Short story:  I'm looking for opinions on the current tape driveC > options for backups in the 400Gb -> 2Tb range.  The new LTO-3 and B > SDLTII cartridges hold 800Gb and 600Gb compressed, respectively.F > A single drive of either would probably satisfy our backup needs forH > another 2, maybe 3 years.  By that time the tape capacity would likelyB > have doubled again and we would buy another drive.  Or there areC > tape libraries available for both formats and those would satisfy > > the capacity needs longer, albeit at the cost of multivolumeH > backups.  What do you folks suggest?  Please comment upon reliability,F > cost, and/or performance.  Any other tape format we should consider? > 
 > Long story: D > We've just outgrown the single SDLT320 drive that has been used toC > backup a Solaris system I manage.  This format/drive, and another C > we have on a different machine, have been ultra-reliable, with no G > drive failures and only one bad (from day 1) cartridge.  No cartridge C > that worked the first time has failed subsequently.  This Solaris < > machine is used primarily to run an Oracle database but weA > don't have very stringent uptime requirements.  We also want to ? > be really sure that we can get back on line again following a : > restore.  So backups have consisted of shutting down the> > database, ufsdump of the various partitions and volumes, and> > then start everything up again.  We rotate the tapes through: > an offsite storage location.  The last adjustment of theE > Oracle storage put that data alone over 380Gb.  It took 3 years for B > the database to grow to that size, and us to outgrow the SDLT320B > drive.  I'd rather not do manual multi-cartridge backups on thisB > system, mostly because I don't want to have to come in at 4:00AM= > to feed cartridges into the drive, or to shut down during a  > workday to do backups. > 	 > Thanks,  >  > David Mathog4 > readme backwards sortof: ude DOT hcetlac AT gohtam   ------------------------------  % Date: Wed, 14 Jun 2006 21:50:23 +0100 1 From: Chris Newport <crn.news@nospam.netunix.com>   Subject: Re: Tape drive opinions1 Message-ID: <4490768F.1030204@nospam.netunix.com>    David Mathog wrote: B > Short story:  I'm looking for opinions on the current tape driveC > options for backups in the 400Gb -> 2Tb range.  The new LTO-3 and B > SDLTII cartridges hold 800Gb and 600Gb compressed, respectively.   [big snippage];    I'd rather not do manual multi-cartridge backups on this B > system, mostly because I don't want to have to come in at 4:00AM= > to feed cartridges into the drive, or to shut down during a  > workday to do backups.  D If you can afford it the best way to do backups is to have a backup B server with adequate disk storage. The Production servers back up G overnight to the backup server. Gigabit ether and fsnap are good hints  H here. You now have a fast restore method without fiddling with tapes if H something breaks. The backup server can now be backed up to tape during  working hours.  F A neat trick is to also set up the backup server as an install server G with some extra post-install scripts to restore the entire system from  H scratch in the event of serious failure. This is VERY usefull in a data H centre situation with many servers and a small number of spare machines  reserved for failure recovery.   ------------------------------  % Date: Wed, 14 Jun 2006 19:12:18 -0700 1 From: "Daniel Lang" <invalid@invalid.caltech.edu>   Subject: Re: Tape drive opinions+ Message-ID: <e6qfm3$elj$1@naig.caltech.edu>   5 "David Mathog" <mathog@caltech.edu> wrote in message  % news:e6pe1p$5br$1@naig.caltech.edu... B > Short story:  I'm looking for opinions on the current tape driveC > options for backups in the 400Gb -> 2Tb range.  The new LTO-3 and B > SDLTII cartridges hold 800Gb and 600Gb compressed, respectively.G > A single drive of either would probably satisfy our backup needs for  H > another 2, maybe 3 years.  By that time the tape capacity would likelyB > have doubled again and we would buy another drive.  Or there areC > tape libraries available for both formats and those would satisfy > > the capacity needs longer, albeit at the cost of multivolumeI > backups.  What do you folks suggest?  Please comment upon reliability,  F > cost, and/or performance.  Any other tape format we should consider?   Hello David,  C We have been using a pair of SDLT600 tape drives for nearly 2 years @ without any problems except for 2 cartridges that were initially< defective (we have not had any cartridges go bad over time).? You may want to go to the next generation and get a DLT-S4 tape 3 drive (800 GB native, 1.6 GB compressed) for $4000:   3 http://www.pcnation.com/web/details.asp?item=J40768    Daniel Lang    ------------------------------   End of INFO-VAX 2006.330 ************************                                                                        ------------------------------------------------------------------------ DIRUTL    Version:      10-MAY-1993@    Description:  Change default directories, maintaining a stack4    Author:       Eric M. LaFranchi <eml@emlsoft.com>    Architecture: VAX    # of parts:   -    Language:     MACRO-32sP -------------------------------------------------------------------------------- DISM32    Version:      z-    Description:  DISM32 VAX .EXE disassembler     Author:       Andy Pavlin    Architecture: VAX 