1 INFO-VAX	Tue, 15 Jul 2003	Volume 2003 : Issue 387       Contents: Re: A rant about HP's website. Re: A rant about HP's website. Re: A rant about HP's website. Re: Can VMS ZIP encrypt? Re: Can VMS ZIP encrypt?& Cheap Memory Options for Alpha Servers* Re: Cheap Memory Options for Alpha Servers" Re: Cleaning up the proxy database# Re: Cluster satellite fails to boot  Re: cxx performance  Re: cxx performance N Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful Performance Re: Getting rid of DECwindows? Re: Getting rid of DECwindows? Re: Getting rid of DECwindows? Re: Getting rid of DECwindows? Re: Getting rid of DECwindows?C Re: How do I determine which processor I am running on (alpha SMP)? : Re: How to close and reopen detached process's Sys$Output? Re: iSCSI anyone?  Re: iSCSI anyone? & Move dumpfile to alternate disk oddity* Re: Move dumpfile to alternate disk oddity" Re: My Alpha has no panties on !!!" Re: My Alpha has no panties on !!!" Re: My Alpha has no panties on !!!% Re: OpenVMS - Remot emailbox (Decnet) % Re: OpenVMS - Remot emailbox (Decnet) % Re: OpenVMS - Remot emailbox (Decnet) % Re: OpenVMS - Remot emailbox (Decnet) % Re: OpenVMS - Remot emailbox (Decnet) L Re: OpenVMS Pearl - OpenVMS V8.0, first release on Itanium, ships   today!!!P Re: OpenVMS Pearl - OpenVMS V8.0, first release on Itanium, ships   today!!!   tP Re: OpenVMS Pearl - OpenVMS V8.0, first release on Itanium, ships   today!!!   t, Re: OpenVMS running under simh, ucx problem.& OT: HP's PeeCee customer statisfaction* Re: OT: HP's PeeCee customer statisfaction* Re: OT: HP's PeeCee customer statisfaction* Re: OT: HP's PeeCee customer statisfaction* Re: OT: HP's PeeCee customer statisfaction* Re: OT: HP's PeeCee customer statisfaction= Re: Q: How to close and reopen detached process's Sys$Output? ' Re: Questions from Pat St. Laurent @ HP ' Re: Questions from Pat St. Laurent @ HP ' Re: Questions from Pat St. Laurent @ HP  Re: RF31 hardware problem  Re: RF31 hardware problem # Re: Serial Connection over Ethernet & Re: VAXstation 4000m90 with no monitor1 What color were the VMS manuals for old versions? 5 Re: What color were the VMS manuals for old versions? 5 Re: What color were the VMS manuals for old versions? 5 Re: What color were the VMS manuals for old versions? 5 Re: What color were the VMS manuals for old versions? 5 Re: What color were the VMS manuals for old versions? 5 Re: What color were the VMS manuals for old versions? 5 Re: What color were the VMS manuals for old versions?   F ----------------------------------------------------------------------  % Date: Mon, 14 Jul 2003 13:03:40 -0500 ( From: brandon@dalsemi.com (John Brandon)' Subject: Re: A rant about HP's website. 1 Message-ID: <03071413034098@dscis6-0.dalsemi.com>   . Doc.Cypher [Use-Author-Address-Header@[127.1]]N > Could someone please ask people to at least check that pages/forms work with > Lynx.   K Doc - you just did "check" and found that it did not work... you should get  paid for that methinks.  ;)          J*o*h*n B*r*a*n*d*o*n  VMS Systems Administrator * firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  % Date: Tue, 15 Jul 2003 08:58:48 +0800 ( From: Peter Sutter <nospam@sopac.com.au>' Subject: Re: A rant about HP's website. = Message-ID: <3f1351d2$0$23588$5a62ac22@freenews.iinet.net.au>    warren sander wrote:  I > Doc.. could you give some exact url's you are  talking about. what page  > has a bad url link on it? L > what page says 'content starts here' and doesn't display anything? this is5 > usually a table problem not a deliberate attempt to  > make the page not show up. > I > The problem with making sure everything works with LYNX just adds about A > 20+ more systems things have to be checked on. Which version of L > LYNX on what platform do you want to make sure things work on? is it ok ifH > it works with the version on NSK but not the one on MPE? how about VMSL > V5.5-2? or OSF1? or non-HP systems.. We do try to make sure things work onI > the most common browsers that our users (of specific areas) use. I make F > sure for instanance that the OpenVMS site works on Netscape 3.03b. II > figure if that works then most will. It is after all 1996 technology so L > everything that's been created since then doesn't much work. thats 7 years > of back support. > K > so please put in the specfiic url's that don't work. I checked www.hp.com K > with lynx v2.8.2 rel.1 on OpenVMS Alpha V7.3 and navigated that page. You ! > don't give any other page url's  > @ > the OpenVMS SSL download does work with this version of lynx..0 > http://h71000.www7.hp.com/openvms/products/sslG > all the way to getting the mail message with the url and the download I > worked.. so what's the problem? if it isn't OpenVMS then don't gripe in I > the OpenVMS newsgroup. IF it is then let me know.. it's not like we are  > hiding or something. > L It's not even html compliant. Try http://validator.w3.org/ and validate the F web page. Its a great service of w3c.org. There are after all certain G standards for html and xhtml and it might be a wise move to adhere tho   these standards.  I Most companies do their websites the quick and dirty with some Micro$oft  G software, their wisdom being that if it runs on IE, then we are ok, we  J don't need the rest, it is us who set the standard. Well, I disagree with  this attitude.   Peter Sutter   Apropos http://www.hp.com   ; http://validator.w3.org/check?uri=http://www.hp.com returns   . This page is not Valid HTML 4.01 Transitional!  I  Below are the results of attempting to parse this document with an SGML   parser.   G Line 51, column 99: there is no attribute "MARGINHEIGHT" (explain...).  M   ...003366" vlink="#660066" marginheight="0" marginwidth="0" leftmargin="0"   topma +                                           ^ G Line 51, column 115: there is no attribute "MARGINWIDTH" (explain...).  K   ...660066" marginheight="0" marginwidth="0" leftmargin="0" topmargin="0"> +                                           ^ F Line 51, column 130: there is no attribute "LEFTMARGIN" (explain...). <   ...eight="0" marginwidth="0" leftmargin="0" topmargin="0">+                                           ^ E Line 51, column 144: there is no attribute "TOPMARGIN" (explain...).  .   ...inwidth="0" leftmargin="0" topmargin="0">+                                           ^ M Line 217, column 58: required attribute "ACTION" not specified (explain...).  =                       <form name="countryForm" method="POST"> =                                                             ^ G Line 217, column 58: document type does not allow element "FORM" here;  $ missing one of "TH", "TD" start-tag =                       <form name="countryForm" method="POST"> =                                                             ^ D Line 218, column 65: document type does not allow element "TD" here  (explain...). :   ...       <td height="30" align="right" valign="middle">:                                                          ^J Line 324, column 22: end tag for "TR" which is not finished (explain...).                      </tr>                          ^ B Line 384, column 334: end tag for element "MAP" which is not open  (explain...). -   ...e-hho&amp;mtxb=B1&amp;mtxl=L1"> </map>'; +                                           ^ C Line 387, column 1189: end tag for element "MAP" which is not open   (explain...). -   ...e-hho&amp;mtxb=B3&amp;mtxl=L2"> </map>'; +                                           ^ B Line 397, column 329: end tag for element "MAP" which is not open  (explain...). -   ...e-smb&amp;mtxb=B1&amp;mtxl=L1"> </map>'; +                                           ^ C Line 400, column 1274: end tag for element "MAP" which is not open   (explain...). -   ...e-smb&amp;mtxb=B3&amp;mtxl=L2"> </map>'; +                                           ^ B Line 411, column 923: end tag for element "MAP" which is not open  (explain...). -   ...e-ent&amp;mtxb=B1&amp;mtxl=L3"> </map>'; +                                           ^ B Line 414, column 915: end tag for element "MAP" which is not open  (explain...). -   ...e-ent&amp;mtxb=B2&amp;mtxl=L3"> </map>'; +                                           ^ G Line 700, column 87: document type does not allow element "FORM" here;  $ missing one of "TH", "TD" start-tag +   ...tp://www.hp.com/search/" method="GET"> +                                           ^ D Line 703, column 59: document type does not allow element "TD" here  (explain...). >               <td align="right" height="33" bgcolor="#E7E7E7">>                                                              ^C Line 708, column 118: there is no attribute "BORDER" (explain...).  J   ...m/img/hpc60_arrow_submit.gif" border="0" alt="Begin your search"></a>+                                           ^    ------------------------------    Date: 14 Jul 2003 20:03:12 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen) ' Subject: Re: A rant about HP's website. 3 Message-ID: <JJibep19J7M4@eisner.encompasserve.org>   h In article <3f1351d2$0$23588$5a62ac22@freenews.iinet.net.au>, Peter Sutter <nospam@sopac.com.au> writes:  N > It's not even html compliant. Try http://validator.w3.org/ and validate the H > web page. Its a great service of w3c.org. There are after all certain I > standards for html and xhtml and it might be a wise move to adhere tho   > these standards.   That is a good point.   K > Most companies do their websites the quick and dirty with some Micro$oft  I > software, their wisdom being that if it runs on IE, then we are ok, we  L > don't need the rest, it is us who set the standard. Well, I disagree with  > this attitude.  F Of course it is possible that some HP Standard requires something thatJ violates the HTML Standard, but certainly not so many as your test showed.   ------------------------------  % Date: Mon, 14 Jul 2003 15:56:53 -0700 , From: Ken Fairfield <My.Full.Name@intel.com>! Subject: Re: Can VMS ZIP encrypt? ) Message-ID: <3F133535.C6940C86@intel.com>    David Gray wrote:  > 3 > Ok thanks for that, got it and it's working fine.  > D > One more thing... I saw once that ZIP could be used with VMS styleF > qualifiers by way of a CLD - anyone know how to get hold of the CLD?' > I cannot find it on the net anywhere.   : 	See if there's a ZIP_CLI.EXE in the distribution.  That's> the one that's been linked with the .CLD.  Define your foreign& command to point to this image, e.g.,    	zip == "$dev:[dir]zip_cli"   ? and you'll find that $ZIP/HELP shows you the command syntax in   DCL-eze.   	-Ken > P.S. I'm sure you could get the actual .CLD file from the full7      ZIP distribution on any one of the freeware sites.  --6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfield " D1C Automation VMS System Support ) kenneth[dt]h[dt]fairfield[ta]intel[dt]com    ------------------------------  % Date: Mon, 14 Jul 2003 20:26:33 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> ! Subject: Re: Can VMS ZIP encrypt? ' Message-ID: <3F135849.60155767@fsi.net>    Ken Fairfield wrote: >  > David Gray wrote:  > > 5 > > Ok thanks for that, got it and it's working fine.  > > F > > One more thing... I saw once that ZIP could be used with VMS styleH > > qualifiers by way of a CLD - anyone know how to get hold of the CLD?) > > I cannot find it on the net anywhere.  > C >         See if there's a ZIP_CLI.EXE in the distribution.  That's @ > the one that's been linked with the .CLD.  Define your foreign' > command to point to this image, e.g.,  > $ >         zip == "$dev:[dir]zip_cli" > @ > and you'll find that $ZIP/HELP shows you the command syntax in
 > DCL-eze. >  >         -Ken@ > P.S. I'm sure you could get the actual .CLD file from the full9 >      ZIP distribution on any one of the freeware sites.   H ZIP and UNZIP as distributed by Hunter Goatley will produce both a plainA and _CLI variant of the .EXE when you "@LINK" after unpacking the  distro.    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Mon, 14 Jul 2003 13:39:38 -0500 ( From: brandon@dalsemi.com (John Brandon)/ Subject: Cheap Memory Options for Alpha Servers 1 Message-ID: <03071413393837@dscis6-0.dalsemi.com>   G Has anyone had any experience with using non-HP memory for use in Alpha , Servers?  Specifically DS20, ES40, or GS160?   What was the result?   Thoughts on why or why not?        J*o*h*n B*r*a*n*d*o*n  VMS Systems Administrator * firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  % Date: Mon, 14 Jul 2003 22:00:37 -0500 , From: "Dave Gudewicz" <k9jdk@NOSPAMarrl.net>3 Subject: Re: Cheap Memory Options for Alpha Servers / Message-ID: <vh6rinjvt65055@corp.supernews.com>   J I've used Dataram memory in a couple of DS20Es.  No problems after severalJ years.  I haven't a clue why vendors charge so much for memory, when there- are but a few foundries for it on the planet.    Dave...   5 "John Brandon" <brandon@dalsemi.com> wrote in message + news:03071413393837@dscis6-0.dalsemi.com... I > Has anyone had any experience with using non-HP memory for use in Alpha . > Servers?  Specifically DS20, ES40, or GS160? >  > What was the result? >  > Thoughts on why or why not?  >  >  >  > J*o*h*n B*r*a*n*d*o*n  > VMS Systems Administrator , > firstname.lastname.spam.me.not@dalsemi.com   ------------------------------  % Date: Mon, 14 Jul 2003 20:30:15 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net> + Subject: Re: Cleaning up the proxy database ' Message-ID: <3F135927.E0720D1C@fsi.net>    Dave Brennan wrote:  >  > OK I figured it out:-  >  > set server secuirity/exit  > del net$proxy.dat;*  > del netproxy.dat;* > set server security/start  > mc authorize create/proxy   / Yeah - bumped my head on that one a while back.    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------    Date: 14 Jul 2003 14:57:38 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) , Subject: Re: Cluster satellite fails to boot3 Message-ID: <YYsEHxA8SILw@eisner.encompasserve.org>   o In article <dd7df538.0307140517.2bf75623@posting.google.com>, dan.williams@btconnect.com (Dan Williams) writes: E > I have set up a cluster satellite, which joins the cluster and then A > does nothing else. It does come with an error about not finding F > net$config.dat but I couldn't find anything about this in the cluter	 > manual.  > D > I've tried deleteing this node and trying different settings no no > avail.E > Is there any reason why it shouldn't boot into a login prompt. I am B > trying to install vms on this other machine, which only has dssi > drives and a tk70.   [...]   7 > %SYSINIT, waiting to form or join a VMScluster system H > %PAB0, Port Error Bit(s) Set - CNF/PMC/PSR  03060022/00000004/00000040 > E > %PAB0, Port is Reinitializing (  49 Retries Left).  Check the Error  > Log. > H > %PAA0, Port Error Bit(s) Set - CNF/PMC/PSR  03060022/00000004/00000040 > E > %PAA0, Port is Reinitializing (  49 Retries Left).  Check the Error  > Log. > * > %CNXMAN,  lost connection to system OZZY@ > %CNXMAN,  sending VAXcluster membership request to system OZZY4 > %CNXMAN,  re-established connection to system OZZY4 > %CNXMAN,  now a VAXcluster member -- system MARVIN  C   Are you using cluster_config.com from the boot server to add the  B   satellite node?  Are you sure you've set up the boot server as aJ   disk server?  Note that cluster_config.com requires interaction at both H   machines during 1st boot of the satellite, but it does know what it's    doing.  H   The errors and the connection loss could be network hardware, (PA is aF   software device and it's not clear why you have two of them) or just4   maybe something's flaky in the satellite's copy ofC   cluster_authorize.dat.  Make sure both the satellite and the boot H   server are using the cluster_authorize.dat in sys$common, the command:  8    $dir sys$sysdevice:[sys*.sysexe]cluster_authorize.dat     should find no files.    ------------------------------  # Date: Mon, 14 Jul 2003 19:54:32 GMT & From: jlsue <jefflsxxxz@sbcglobal.net> Subject: Re: cxx performance8 Message-ID: <dd26hv8bpckvpk67dge4ndvusaohsv6uff@4ax.com>  7 On Fri, 04 Jul 2003 11:22:22 -0500, "David J. Dachtera"  <djesys.nospam@fsi.net> wrote:     > I >Actually, its not that complex. WSDEF harks back to the days when memory C >was expensive and tended to be somewhat dear. Now that VMS is more G >frequently found on memory-rich systems, it makes sense to let WSDEF = D >WSQUOTA in such cases. That way, the working set doesn't have to beG >expanded just to do useful work. You can set it to a value most likely I >to be needed. Then, when an image is activated, all the system has to do 9 >is read it in - no need to expand the working set first.   F I totally agree that WSDEF=WSQUOTA is fine for many, many systems now.J But make sure that everyone understands that it's NOT the working set thatJ expands, necessarily, but the WS List Size.  I.e., as Bruce Ellis was fond
 of saying:  : 	Q:  "How much memory does VMS give your working set when 
 		you login?" $ 	A:  "None!  You've got to earn it."  H What this means is that you WS List Size=WSDEF(=WSQUOTA if you make thisG change).  However, you don't actually have that memory allocated to you   unless you need it (ask for it).   ------------------------------  % Date: Mon, 14 Jul 2003 20:15:20 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>  Subject: Re: cxx performance' Message-ID: <3F1355A8.D4BA5105@fsi.net>    jlsue wrote: > 9 > On Fri, 04 Jul 2003 11:22:22 -0500, "David J. Dachtera"   > <djesys.nospam@fsi.net> wrote: >  > > K > >Actually, its not that complex. WSDEF harks back to the days when memory E > >was expensive and tended to be somewhat dear. Now that VMS is more I > >frequently found on memory-rich systems, it makes sense to let WSDEF = F > >WSQUOTA in such cases. That way, the working set doesn't have to beI > >expanded just to do useful work. You can set it to a value most likely K > >to be needed. Then, when an image is activated, all the system has to do ; > >is read it in - no need to expand the working set first.  > H > I totally agree that WSDEF=WSQUOTA is fine for many, many systems now.L > But make sure that everyone understands that it's NOT the working set thatL > expands, necessarily, but the WS List Size.  I.e., as Bruce Ellis was fond > of saying: > B >         Q:  "How much memory does VMS give your working set when >                 you login?" - >         A:  "None!  You've got to earn it."  > J > What this means is that you WS List Size=WSDEF(=WSQUOTA if you make thisI > change).  However, you don't actually have that memory allocated to you " > unless you need it (ask for it).  H Doesn't take much to "ask for it", of course... SET TERM/INQ, check your* MAIL, most any other image activation, ...   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  # Date: Mon, 14 Jul 2003 22:31:02 GMT & From: jlsue <jefflsxxxz@sbcglobal.net>W Subject: Re: D.H. Brown: EV7 AlphaServers Deliver Enhanced RAS and Powerful Performance 8 Message-ID: <3nb6hvomhnqaknf2dajefn81ko58baqjh4@4ax.com>  H On Fri, 11 Jul 2003 04:21:40 -0400, "Bill Todd" <billtodd@metrocast.net> wrote:   > E >"Fred Kleinsorge" <my-last-name@stardotzko.dec.com> wrote in message , >news:CxjPa.4355$9X2.944@news.cpqcorp.net...   >>F >> In context, the point was that the cost/benefit long term to remain$ >> competetive with Alpha was bleak. > * >Perhaps to the arithmetically-challenged. >   I The only arithmetic that matters is the profit calculation.  Nobody knows F what that would be with EV8 investment:  design, manufacturing, system1 design/implementation, and price/volume of sales.    ------------------------------   Date: 14 Jul 2003 19:18:35 GMT/ From: Chris Scheers <chris@applied-synergy.com> ' Subject: Re: Getting rid of DECwindows? 2 Message-ID: <3F3BE090.B7EFD41@applied-synergy.com>   Christoph Gartmann wrote:  >  > Hello, > G > we have a VAX-cluster here that started life somewhen in 1987. On the M > system-disk there is a lot of DECwindows stuff. But none of the VAX cluster N > members starts or uses DECwindows anymore. Thus, I would like to remove this6 > stuff from the disk. Is it safe to do something like& >  DELETE sysdevice:[sys0...]decw*.*;*L > ? Or is there some deinstallation procedure? It seems to be version 1.2 of
 > DECwindows.  > O > System parameters are set up to supress DECWndows. Anything else to consider?   6 Instead of deleting the files directly, take a look at? SYS$UPDATE:DECW$TAILOR.EXE or SYS$UPDATE:DECW$MOTIF_TAILOR.EXE.   G ----------------------------------------------------------------------- $ Chris Scheers, Applied Synergy, Inc.  C Voice: 817-237-3360            Internet: chris@applied-synergy.com     Fax: 817-237-3074    ------------------------------  % Date: Mon, 14 Jul 2003 15:28:52 -0400 * From: "Stanley F. Quayle" <stan@stanq.com>' Subject: Re: Getting rid of DECwindows? . Message-ID: <3F12CC34.23699.9F9A8C8@localhost>  2 On 14 Jul 2003 at 11:12, Christoph Gartmann wrote:R > I would like to remove this stuff from the disk. Is it safe to do something like& >  DELETE sysdevice:[sys0...]decw*.*;*   Nope.   B There is a procedure called VMS TAILOR which can reduce the space D taken by things like DECwindows.  It's on the VMS boot CD (and it's ' probably on the system somewhere, too).   @ Depending on your VMS version, you might be able to do "PRODUCT  REMOVE DWMOTIF".  8 I recommend backing up the system disk in either case...  F However, rather than "cleaning up" your disk, you should probably get B a bigger disk.  The few hundreds of megs you save isn't worth the  pain...   
 --Stan Quayle  Quayle Consulting Inc.  
 ----------C Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363  Fax: +1 614 868-1671 1 8572 North Spring Ct. NW, Pickerington, OH  43147 = Preferred address:  stan@stanq.com       http://www.stanq.com    ------------------------------  # Date: Mon, 14 Jul 2003 20:01:17 GMT 3 From: hammond@not@peek.ssr.hp.com (Charlie Hammond) ' Subject: Re: Getting rid of DECwindows? / Message-ID: <h_DQa.182$kh.143@news.cpqcorp.net>   3 In article <3F3BE090.B7EFD41@applied-synergy.com>,  1 Chris Scheers <chris@applied-synergy.com> writes:   7 >Instead of deleting the files directly, take a look at @ >SYS$UPDATE:DECW$TAILOR.EXE or SYS$UPDATE:DECW$MOTIF_TAILOR.EXE.  B This does *NOT* get rid of the DECwindows layered product, but it D can be used to get rid of the DECwindows support files that are part of OpenVMS.   D That said, I would not necessarily recommend this.  It is surprisingD how often something in one "unused" part of OpenVMS is actually usedF by some other part of the O/S.  It is generally better and less costlyH to by more and/or larger disks than to remove "unused" parts of OpenVMS.   --  J       Charlie Hammond -- Hewlett-Packard Company -- Ft Lauderdale  FL  USAF           (hammond@not@peek.ssr.hp.com -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------    Date: 14 Jul 2003 14:40:34 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) ' Subject: Re: Getting rid of DECwindows? 3 Message-ID: <n8BBIeO0n8k0@eisner.encompasserve.org>   f In article <beu36u$67q$1@n.ruf.uni-freiburg.de>, gartmann@immunbio.mpg.de (Christoph Gartmann) writes: > Hello, > G > we have a VAX-cluster here that started life somewhen in 1987. On the M > system-disk there is a lot of DECwindows stuff. But none of the VAX cluster N > members starts or uses DECwindows anymore. Thus, I would like to remove this6 > stuff from the disk. Is it safe to do something like& >  DELETE sysdevice:[sys0...]decw*.*;*L > ? Or is there some deinstallation procedure? It seems to be version 1.2 of
 > DECwindows.  > O > System parameters are set up to supress DECWndows. Anything else to consider?   G    Just because your VAXen doesn't use DECwindows doesn't mean you want D    to delete it.  Both server and client parts are in the DECwindowsD    files.  Someone may be bringing up one of those clients remotely.  B    If you just delete all the files you'll probably get errors for8    those which are INSTALLed as known files during boot.   ------------------------------   Date: 14 Jul 2003 20:06:16 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann) ' Subject: Re: Getting rid of DECwindows? 0 Message-ID: <bev2fo$nmg$1@n.ruf.uni-freiburg.de>  d In article <3F3BE090.B7EFD41@applied-synergy.com>, Chris Scheers <chris@applied-synergy.com> writes: >Christoph Gartmann wrote: >>  	 >> Hello,  >>  H >> we have a VAX-cluster here that started life somewhen in 1987. On theN >> system-disk there is a lot of DECwindows stuff. But none of the VAX clusterO >> members starts or uses DECwindows anymore. Thus, I would like to remove this 7 >> stuff from the disk. Is it safe to do something like ' >>  DELETE sysdevice:[sys0...]decw*.*;* M >> ? Or is there some deinstallation procedure? It seems to be version 1.2 of  >> DECwindows. >>  P >> System parameters are set up to supress DECWndows. Anything else to consider? > 7 >Instead of deleting the files directly, take a look at @ >SYS$UPDATE:DECW$TAILOR.EXE or SYS$UPDATE:DECW$MOTIF_TAILOR.EXE.  > Ok, the first one didn't remove much, the latter removed most.   What is still there is:       Directory DISK1:[SYS0.SYSCOMMON]   DECW$DEFAULTS.DIR;1    Total of 1 file.  5 Directory DISK1:[SYS0.SYSCOMMON.DECW$DEFAULTS.SYSTEM]    DECW$SESSION.UID;1     Total of 1 file.  + Directory DISK1:[SYS0.SYSCOMMON.SYS$KEYMAP]    DECW.DIR;1             Total of 1 file.  ' Directory DISK1:[SYS0.SYSCOMMON.SYSCBI]    DECW$CBI.DIR;1         Total of 1 file.  ' Directory DISK1:[SYS0.SYSCOMMON.SYSEXE]   B DECW$DWT_DECNET.EXE;1                   DECW$DWT_FONT_DAEMON.EXE;1( DECW$DWT_STARTXTDRIVER.EXE;1               Total of 3 files.   ( Directory DISK1:[SYS0.SYSCOMMON.SYSFONT]   DECW.DIR;1             Total of 1 file.  9 Directory DISK1:[SYS0.SYSCOMMON.SYSFONT.DECW.USER_100DPI]   ( DECW$FONT_DIRECTORY.DAT;2                  Total of 1 file.  8 Directory DISK1:[SYS0.SYSCOMMON.SYSFONT.DECW.USER_75DPI]  ( DECW$FONT_DIRECTORY.DAT;2                  Total of 1 file.  0 Directory DISK1:[SYS0.SYSCOMMON.SYSHLP.EXAMPLES]  < DECW.DIR;1          EVE$DECWINDOWS.TPU;1                       Total of 2 files.   ' Directory DISK1:[SYS0.SYSCOMMON.SYSLIB]   a DECW$DWT_.ADA;1     DECW$SECURITY_VMS.EXE;1                 DECW$SERVER_EXTENSION_XTRAP_V54.EXE;1 O DECW$TRANSPORT_LAT.EXE;1                DECW$UIL.ENV;1      DECW$X_.ADA;1        LSE$DECWSHR.EXE;1      Total of 7 files.   ' Directory DISK1:[SYS0.SYSCOMMON.SYSMGR]   O DECW$AUTOGEN.COM;2  DECW$CHECK_PARAMS.COM;3                 DECW$LOGICALS.COM;3 < DECW$STARTI18N.COM;2                    DECW$STARTLIBS.COM;3< DECW$STARTUP.COM;3  DECW$STARTXTERMINAL.COM;1                  Total of 7 files.   ' Directory DISK1:[SYS0.SYSCOMMON.SYSMSG]   ` DECW$DEVICE_GF.COM;1                    DECW$DWTERRDB.DAT;1 DECW$PRIVATE_SERVER_SETUP.TEMPLATE;1( DECW$XLIBERRDB.DAT;1                       Total of 4 files.   ( Directory DISK1:[SYS0.SYSCOMMON.SYSTEST]   DECW$IVP.COM;2         Total of 1 file.  ' Directory DISK1:[SYS0.SYSCOMMON.SYSUPD]   D DECW$COMPARE_VERSIONS.COM;2             DECW$COMPILE_ADA_UNITS.COM;1O DECW$GET_IMAGE_VERSION.COM;2            DECW$KITBLD.DAT;1   DECW$KITBLD.IDX;1   ? DECW$MOTIF_KITBLD.DAT;2                 DECW$MOTIF_KITBLD.IDX;2 G DECW$MOTIF_TAILOR.EXE;2                 DECW$MOTIF_TAILOR_CLEANUP.COM;2 U DECW$OBSOLETE.DAT;1 DECW$OBSOLETE.IDX;1 DECW$TAILOR.EXE;1   DECW$TAILOR_ON.TEMPLATE;1  DECW$VERSIONS.COM;2    Total of 14 files.   Directory DISK1:[SYS0.SYSMGR]   ( DECW$DXMLIBSHR.EXE;1                       Total of 1 file.  ( Grand total of 15 directories, 46 files.       Any comments on these?   Regards,    Christoph Gartmann    --  E  Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452   ImmunbiologieA  Postfach 1169                 Internet: gartmann@immunbio.mpg.de   D-79011  Freiburg, Germany 9                http://www.immunbio.mpg.de/home/menue.html    ------------------------------    Date: 14 Jul 2003 12:09:21 -0700$ From: rcbryan@hotmail.com (RC Bryan)L Subject: Re: How do I determine which processor I am running on (alpha SMP)?= Message-ID: <fbcf38dc.0307141109.5c08480a@posting.google.com>   E I kind of figured it was stored in the operating system somewhere but < I was hoping there would be an instruction or PAL routine or
 something.  A Let me get out my OpenVMS internals manual and see about the PCB.    Thanks, 
 /Bob Bryan  b jemby@gtech.com (Jerry) wrote in message news:<8ac07e79.0307111425.5fbda56d@posting.google.com>...k > rcbryan@hotmail.com (RC Bryan) wrote in message news:<fbcf38dc.0307101119.767ac76d@posting.google.com>... I > > I have been reading about the RSCC PALcode instruction that returns a C > > 64 bit count of clocks since some base time.  Since this is per A > > processor, I have to ask is there a fast way to find out what F > > processor I am running on?  The SYS$GETJPI system service can giveH > > this information but where does it get the information from?  I haveH > > been searching my AARM for something that would return the processor3 > > number and have not been able to find anything.  > >  > > Any ideas? > > 
 > > Thanks
 > > /RC Bryan K > That information for the CPU number is kept in the PCB (  PCB$L_CPU_ID ).    ------------------------------  % Date: Mon, 14 Jul 2003 14:15:09 -0700 , From: Ken Fairfield <My.Full.Name@intel.com>C Subject: Re: How to close and reopen detached process's Sys$Output? ) Message-ID: <3F131D5D.75AC0FB1@intel.com>    Richard Maher wrote: >  > Hi,  > 1 > You probably already know this but if not. .  .  > L > Could you change your DCL to run sys$system:loginout.exe as the image withN > an input .COM file that runs your C program and an output .LOG file that VMSN > will flush to disk periodically for you? (I think there's a SYSGEN parameter- > that controls how frequently this is done?)  > L > Then you can just do your standard DISPLAYs, PRINTs, fprint() (or whatever9 > cryptic nonsense C requires) and it's all done for you.   E Thanks for the reply, but you've solved a problem I don't have... :-) H I think VAXman has answered the question I do have, but I haven't tested yet.   	-Ken  --6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfield " D1C Automation VMS System Support ) kenneth[dt]h[dt]fairfield[ta]intel[dt]com    ------------------------------  # Date: Mon, 14 Jul 2003 19:14:17 GMT < From: "John E. Malmberg" <Malmberg@dskwld.zko.dec.compaq.hp> Subject: Re: iSCSI anyone?. Message-ID: <diDQa.179$ch.42@news.cpqcorp.net>   Hoff Hoffman wrote:  > L >   At least one of the locals has an iSCSI client running on OpenVMS Alpha.H >   Donno if he's got a server running.  The code is a port of the Intel >   iSCSI software, IIRC.   I The server serves a container file, and the other is a test program that  < pretends to be a client and verifies the server's operation.  H This is a mostly direct compile and go of the Intel iSCSI software from B SourceForge with some minor changes to fix syntax issues with the E HP/COMPAQ C compiler.  As with the Intel source, the names are still  
 hardcoded.  F I have not tried anything other than using the loopback interface for  access.   9 Building it currently requires OpenVMS 7.3-2 FT or later.    -John ! malmberg@dskwld.zko.dec.compaq.hp  Personal Opinion Only    ------------------------------  # Date: Tue, 15 Jul 2003 04:06:03 GMT ; From: "John Gemignani, Jr." <jon-nope@thiswontworkossc.net>  Subject: Re: iSCSI anyone?9 Message-ID: <L4LQa.906$KZ.410449@news1.news.adelphia.net>   K "Andrew Harrison SUNUK Consultancy" <Andrew_No.Harrison_No@nospamn.sun.com> ; wrote in message news:beuop0$f80$1@new-usenet.uk.sun.com...  > John Gemignani, Jr. wrote:' > > "Andrew Harrison SUNUK Consultancy" ' <Andrew_No.Harrison_No@nospamn.sun.com> ? > > wrote in message news:be4479$lnv$1@new-usenet.uk.sun.com...  > >  > >>Jim Brankin wrote: > >>K > >>>"John Gemignani, Jr." <jon-nope@thiswontworkossc.net> wrote in message  > > = > > news:<6lZMa.26115$Jw6.9846123@news1.news.adelphia.net>...l > >	G > >>>>Is there anyone interested in running software-based iSCSI client  and/orK > >>>>server on OpenVMS?  I have some rather advanced work in this area andp amK > >>>>seeking suggestions and needs on the way to productization.  In fact,e > >O > > when > >a2 > >>>>ready, I could use a few test sites as well. > >>>> > >>>>-John  > >>>  > >>>x
 > >>>Hi John,  > >>> J > >>>   I can see that there is a market for a software based solution butI > >>>I would have thought it was small. Surely HP must eventually producedH > >>>iSCSI products based on one of the host-based adapters. Several areK > >>>mentioned in the FAQ at http://technomagesinc.com/iscsi_faq.html. WhennI > >>>that happens, they will take the serious users of iSCSI and you willu? > >>>be left with the folk who are using it on an ah-hoc basis.r > >>>s > >> > >e > >     Couple of points:eJ > >         The protocol is still emerging, and purchasing a card soon may yield! > > headaches.J > >         For redundency you may have to purchase additional cards (akin to > > redundant NICs) G > >         if a problem should arise.  I also don't believe that thosen cardse > > will be cheap, but if(K > >         you have the dough for a lot of storage, you'll have it for the  > > adapters.  But then, if J > >         you have the dough for your big servers, what about your other > > systems? > > L > >         You are also assuming that the iSCSI targets are not going to be on > > any of yourPE > >         servers.  For those with VAXes yet, you are assuming thatc
 someone is
 > > making anu > >         adapter for them.P > >eL > >         You forget that most of Windows is used on an ad-hoc basis.  VMS is > > just oneL > >         version of the product.  I am confident that my phone will ring. > >  > >uA > >>You really nead a Gigabit card that supports TOE for iSCSI toe> > >>be performant otherwise the stack ends up being to much of > >>a performance drag.h > >r > >b) > >     Is this measured, or theoretical?  > >A >a, > A number of articles cite TOE performance. > . > http://www.iscsistorage.com/iscsidevices.htm, > http://www.trebia.com/pdf/Tech_note_44.pdf  J And that's all that they do.  They cite for the adapters and storage (that
 aren't cheap) H what they will do.  Although, one says that at 245 I/O's per second they have offloadedJ 99% of the load -- at this moment I am running a fair mix on ONE DEVICE of aboutoJ 180 /second at the 65% (of one CPU) with the client and server on the same system.-K A good guess is to lower that number by half to just about 35%, and this is: with oldI SCSI hardware.  The SANs also would implement some more special features,s whiche: I have not yet gotten to, such as the caching on the data.   >h >[D > >         I assume that's theoretical.  Here are the measurements: > > K > >         With an AS255 4/300 reading a disk (COPY) from an XP1000 with a" > > 6/667, I was able toL > >         transfer about 1.2MBYTES/second.  Using the same 6/667 writing a > > disk on an ES40 6/1000K > >         (BACKUP/IMAGE/INIT) I get 2.3 MBYTES/second.  In both cases the- > > systems certainly DOK > >         NOT run out of CPU time.  The EV6's spend only a few percentageR! > > points on the interrupt stacksL > >         (the ES40 is about 2-3%, which REALLY astonished me).  The AS255 has  > > 10Mbit through aL > >         10/100 switch to the XP1000 with 100Mbit.  The XP's 100MBit goes > > through switches and1 > >         eventually through fiber to the ES40.t > >s > F > 1.2-2.3 MB/s however is very slow compared with what you can achieveB > either with directly attached storage or with SAN based storage.  L This starts with 10Mb ethernet.  1.2MB/second really isn't that bad.  I have alsoK been improving the performance in such a way that I can actually get bettero performancerH using the iSCSIdriver against the local SCSI disk than I can against the disk directly.   >sE > based on your results pushing ~60MB/s over Gigabit ethernet using aeB > standard NIC could result a CPU load of between 50 and 75%. ThisA > would be much less acceptable when the alternative using DAS or ) > a SAN would have a much lower overhead.1  L If you take the current (first trial) code base and interpolate the results, then youK get these numbers.  There are a LOT of sites using MSCP right now.  Are youn( saying that they are doing it all wrong?  L The storage (and adapter) businesses are working on iSCSI at the moment.  MyK question is: who is willing to invest BIG BUX in an immature technology?  Ir would J rather play with the technology in software and move into hardware when it hasgH proven itself.  But then, that's just me.  It also lets me play with the
 pieces that I H have now ... VMS systems with disks on them ... client and server on the same0 hardware BEFORE I SPEND THE BUX ON THE BIG IRON.  L You are right with the TOE that you will get the best performance.  iSCSI is anJ emerging technology and, as I said in my last post, at this moment in time it's a leap J of faith and headaches in the wings until they get the kinks out and prove that you can mix and match hardware.e  I Thank you for your post and if there are people interested in testing thei	 software,E please do contact me.    -Johnd   ------------------------------   Date: 14 Jul 2003 19:20:47 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann)./ Subject: Move dumpfile to alternate disk oddityw0 Message-ID: <beuvqf$mi8$1@n.ruf.uni-freiburg.de>   Hello,  N now under OpenVMS 7.3-1 I thought I give the possibility to move dumpfiles offJ the system disk a try. I tried it on three satellite nodes of the cluster. What I did: (   - system parameter DUMPSTYLE set to 119   - added DUMPFILE_DEVICE="mpi7$dka200:" to MODPARAMS.DAT I   - performed AUTOGEN from SAVPARAMS to GENFILES in order to have the newe     dumpfile created   - shut down the node!   - modified dump_dev to "dka200"t   - rebootedL Now the original dumpfile on the system disk is still open, the new dumpfile, cannot been seen with a "SHOW DEVICE/FILES".N Next I did the same with a differenct satellite with a small difference. WhileG the node was down I deleted the original dumpfile from the system disk.iJ Now the node has no dumpfile open at all. I forced a crash with OPCRASH onM this node but found no dump in the new dumpfile. Thus, what I am doing wrong?e   Regards,    Christoph Gartmann(  l -- .E  Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452   ImmunbiologieA  Postfach 1169                 Internet: gartmann@immunbio.mpg.ded  D-79011  Freiburg, Germanyl9                http://www.immunbio.mpg.de/home/menue.html    ------------------------------   Date: 14 Jul 2003 20:08:26 GMT3 From: gartmann@immunbio.mpg.de (Christoph Gartmann) 3 Subject: Re: Move dumpfile to alternate disk oddityo0 Message-ID: <bev2jq$nmg$2@n.ruf.uni-freiburg.de>  f In article <beuvqf$mi8$1@n.ruf.uni-freiburg.de>, gartmann@immunbio.mpg.de (Christoph Gartmann) writes: >iO >now under OpenVMS 7.3-1 I thought I give the possibility to move dumpfiles off K >the system disk a try. I tried it on three satellite nodes of the cluster.  >What I did:) >  - system parameter DUMPSTYLE set to 11.: >  - added DUMPFILE_DEVICE="mpi7$dka200:" to MODPARAMS.DATJ >  - performed AUTOGEN from SAVPARAMS to GENFILES in order to have the new >    dumpfile created  >  - shut down the nodeo" >  - modified dump_dev to "dka200"
 >  - rebooted M >Now the original dumpfile on the system disk is still open, the new dumpfile - >cannot been seen with a "SHOW DEVICE/FILES".aO >Next I did the same with a differenct satellite with a small difference. While H >the node was down I deleted the original dumpfile from the system disk.K >Now the node has no dumpfile open at all. I forced a crash with OPCRASH on N >this node but found no dump in the new dumpfile. Thus, what I am doing wrong?  E I discovered my mistake: DUMPSTYLE should be 13 (bit 2 with value 4).pL Although it is dynamic parameter, a change still keeps the original dumpfile open...l   Regards,    Christoph Gartmannp   -- )E  Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452   ImmunbiologieA  Postfach 1169                 Internet: gartmann@immunbio.mpg.de   D-79011  Freiburg, Germanyf9                http://www.immunbio.mpg.de/home/menue.htmld   ------------------------------  % Date: Mon, 14 Jul 2003 15:44:13 -0400:8 From: Jim Agnew - VCU/MCV Neurosurgery <jpagnew@vcu.edu>+ Subject: Re: My Alpha has no panties on !!!t' Message-ID: <3F13080D.413BD794@vcu.edu>    Bob Koehler wrote: > d > In article <3F12B082.914A14A4@vcu.edu>, Jim Agnew - VCU/MCV Neurosurgery <jpagnew@vcu.edu> writes: > >0" > > I thought that was a girdle... > >y > D >    Can be done, but often a girdle is targetted to the other side.  # i see, no if's ands or buts.... ;-)2   jim0   -- 5F "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: 14 Jul 2003 14:35:23 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler).+ Subject: Re: My Alpha has no panties on !!!a3 Message-ID: <ICa+PV5nBGpy@eisner.encompasserve.org>   b In article <3F12B082.914A14A4@vcu.edu>, Jim Agnew - VCU/MCV Neurosurgery <jpagnew@vcu.edu> writes: >   > I thought that was a girdle... >   B    Can be done, but often a girdle is targetted to the other side.   ------------------------------  % Date: Mon, 14 Jul 2003 20:22:15 -0500t1 From: "David J. Dachtera" <djesys.nospam@fsi.net>a+ Subject: Re: My Alpha has no panties on !!!i' Message-ID: <3F135747.2ECD3534@fsi.net>   ' Jim Agnew - VCU/MCV Neurosurgery wrote:  >  > Bob Koehler wrote: > >uf > > In article <3F12B082.914A14A4@vcu.edu>, Jim Agnew - VCU/MCV Neurosurgery <jpagnew@vcu.edu> writes: > > > $ > > > I thought that was a girdle... > > >S > >eF > >    Can be done, but often a girdle is targetted to the other side. > % > i see, no if's ands or buts.... ;-)'  ! *CHOKE*   *HORK*   *GAG*   *PUKE*g   ...but I love it, anyway!    --   David J. Dachterau dba DJE Systemsf http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Mon, 14 Jul 2003 15:30:51 -0400k* From: "Stanley F. Quayle" <stan@stanq.com>. Subject: Re: OpenVMS - Remot emailbox (Decnet). Message-ID: <3F12CCAB.30211.9FB7770@localhost>  , On 14 Jul 2003 at 17:07, Jean Mertens wrote:I > Using SYS$CREMBX, how to connect to a mailbox on another node (OpenVMS) L > or using regular OPEN Fortran statement in order to send mail box messages > that are quite long ?e  2 Sorry -- mailboxes are not shared between systems.  F You can use DECnet task-to-task communications without too much pain, 1 although I don't have any Fortran examples handy.   
 --Stan Quaylev Quayle Consulting Inc.  
 ----------C Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363  Fax: +1 614 868-167171 8572 North Spring Ct. NW, Pickerington, OH  43147y= Preferred address:  stan@stanq.com       http://www.stanq.com    ------------------------------    Date: 14 Jul 2003 14:41:35 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)h. Subject: Re: OpenVMS - Remot emailbox (Decnet)3 Message-ID: <S5k3Eex0ySxf@eisner.encompasserve.org>a  h In article <3f12c6e7$0$271$ba620e4c@reader0.news.skynet.be>, "Jean Mertens" <jmertens@skynet.be> writes:I > Using SYS$CREMBX, how to connect to a mailbox on another node (OpenVMS)] > (using Fortran preferrably) L > or using regular OPEN Fortran statement in order to send mail box messages > that are quite long ?l  A    Standard VMS mailboxes are not network accessable.  Try DECneta    mailboxes instead.s   ------------------------------  % Date: Mon, 14 Jul 2003 22:30:05 +0200a) From: "Jean Mertens" <jmertens@skynet.be>o. Subject: Re: OpenVMS - Remot emailbox (Decnet)< Message-ID: <3f131279$0$275$ba620e4c@reader1.news.skynet.be>   Hi,f  H What is a Decnet Mailbox as opposed to the one created with SYS$CREMBX ?> Is this the mailbox to return network information defined with LIB$ASN_WTH_MBX ?o  Where to find doc (VMS manual ?)  ( Do you have an example in Fortran or C ?  G "Bob Koehler" <koehler@eisner.nospam.encompasserve.org> a crit dans lee8 message de news:S5k3Eex0ySxf@eisner.encompasserve.org...D > In article <3f12c6e7$0$271$ba620e4c@reader0.news.skynet.be>, "Jean% Mertens" <jmertens@skynet.be> writes:*K > > Using SYS$CREMBX, how to connect to a mailbox on another node (OpenVMS)d > > (using Fortran preferrably)-E > > or using regular OPEN Fortran statement in order to send mail boxi messages > > that are quite long ?R >CC >    Standard VMS mailboxes are not network accessable.  Try DECneto >    mailboxes instead.e >e   ------------------------------   Date: 14 Jul 2003 20:36 CDTl' From: carl@gerg.tamu.edu (Carl Perkins)a. Subject: Re: OpenVMS - Remot emailbox (Decnet)- Message-ID: <14JUL200320364419@gerg.tamu.edu>   - "Jean Mertens" <jmertens@skynet.be> writes...eH }"Bob Koehler" <koehler@eisner.nospam.encompasserve.org> a crit dans le9 }message de news:S5k3Eex0ySxf@eisner.encompasserve.org...eE }> In article <3f12c6e7$0$271$ba620e4c@reader0.news.skynet.be>, "Jean & }Mertens" <jmertens@skynet.be> writes:L }> > Using SYS$CREMBX, how to connect to a mailbox on another node (OpenVMS)  }> > (using Fortran preferrably)F }> > or using regular OPEN Fortran statement in order to send mail box	 }messagesv }> > that are quite long ? }>D }>    Standard VMS mailboxes are not network accessable.  Try DECnet }>    mailboxes instead. }>I }What is a Decnet Mailbox as opposed to the one created with SYS$CREMBX ?i   Jumping in here...  J I'm not sure why it is sometimes called that, as it isn't really a mailboxG as such. In some respects the data reading and writing is very similar.   B SYS$CREMBX creates a "mailbox" which is an MBAx: type device. It'sE essentially a chunk of memory that you can write to and read from via C the mailbox drivers. As such, it is specific to the system it is on B and there is no network access to it (unless you program a network" server process that accesses it).   F A "DECnet mailbox" isn't just a mailbox. In fact it isn't a mailbox inF the MBAx: device sense at all, although there are mailboxes involved -H one is used for the network command data at each end. The network driverG writes various pieces of info to this, such as notification of new link  requests, link failure, etc.  ? }Is this the mailbox to return network information defined withu }LIB$ASN_WTH_MBX ?  G That function does do that - when used to connect to the "_NET:" devicetG (which is the DECnet device template) the mailbox that this produces isg. the network command mailbox I mentioned above.  I This routine really is just a shortcut to do in one step what is normallyj. done in two steps (sys$crembx and sys$assign).  ! }Where to find doc (VMS manual ?)0 } ) }Do you have an example in Fortran or C ?n  J You are probably looking for what is known as non-transparent task-to-taskE DECnet communications. This is analogous to TCP/IP communications butRM perhaps a tad easier. The documentation for this used to be in the NetworkingoE manual (volume 5A of the system management set, if you happen to have H access to the old printed documentation from a couple of versions back).D I checked Google and it now appears to be in chapter 8 of the manual
 located here:o  I h71000.www7.hp.com/doc/73final/documentation/pdf/DECNET_OVMS_NET_MAN.PDF o  L There is an example of the non-transparent sort of thing in the SYS$EXAMPLESI directory. The various DB_*.C files are involved, with the actual networkbI stuff being in DB_SERVER.C and DB_REQUESTER.C (as the server and client).sG These examples are aparently the basis for most programs written to userJ this and are fairly complete in showing you how to do things - as I recallC the client does it's IO sunchronously and the server does it mostlyaC asynchronously, also demonstrating how to use IO completion ASTs tom handle the asynch IO.$  E I don't know what exactly you are trying to do. You might actually befM able to get by with the "transparent" version instead of the "nontransparent" H version. This basically runs something on a remote system on demand. ForG this you can specify a .com file to execute on the remote node just buy2 doing an open on something likeh,  node"username password"::task=my_proceedureH which will run my_proceedure.com from the default login directory of theI username you specified on the node you specified. That can send back datafG by writing to sys$net, or receive data by reading from sys$net. As bothWG are the same, you have to be careful if you want to communicate in bothnF directions (I've never botherd doing so - not that I've ever used this transparent DECnet for much).i  J Here is a brief and not incredibly clear example of the use of transparent DECent communications:   $ set noverify! $ if p1 .eqs. "" then goto serverm0 $ location = "''p1'""''p2' ''p3'""::""task=sq""" $ type 'location $ exit $server:  $ define/user sys$output sys$net $ show process/quota $ exit  B This is both the client and the server (you could separate them if' you wanted, but this way is more tidy).A  E Put it in the default login directory of an account. If you have that 8 directory as your current default directory and do this:    $ @sq.com node username password  E specifying the correct data for the 3 parameters, it will not jump to-D the server section, staying in the client section it will connect toD node and run the same sq.com file in a process owned by the specfiedE username. On the server end it will jump to the server section in theJH code where it will send back the process quotas for the process in whichK it is running. Back on the client end, the data will be read and displayed.yI Not real exciting, but it works. It cheats somewhat by not using explicitMG open and read statments in the client part, instead relying on the TYPE # command do do these things for you.s  E If I had a more clear example, I'd have used it. This is the only onee I could find laying around.   J The "non-transparent" version requires you to do all the setup, monitoringI of the command mailbox for messages, and shutdown manually. Note that allrI the reads and writes are done with sys$qio[w] rather than native languagemF read and write calls. In return for the added complexity, you can haveG a process declare itself to be a DECnet object to which connections cancF be made (so any connection to "TASK=MY_OBJECT_NAME" is handled by it).G This is similar to a process declaring itself to be the handler for all H TCP/IP connections to a specific port number (via the listen() routine).  C I get the impression that I'm not explaining any of this very well.n" Maybe someone else will do better.  I Anyhow, if you need to write to a mailbox on anyother system, you have tooH create a server that runs on that system which will accept data over theG network and write it to the mailbox. This can be done via DECnet or viadE TCP/IP (or UDP/IP if you don't mind loosing the occasional packet) ory/ any other networking software available to you.-  J Note that if what you areally mean by "mailbox" is "the place where e-mailJ is stored" rather than the VMS specific meaning of the term, then this has% probably not been very useful to you.h   --- Carl   ------------------------------  % Date: Tue, 15 Jul 2003 05:33:12 +0200b) From: "Jean Mertens" <jmertens@skynet.be>d. Subject: Re: OpenVMS - Remot emailbox (Decnet)< Message-ID: <3f1375a3$0$286$ba620e4c@reader0.news.skynet.be>  4 What is surprising is that the following is working:  D 1) From NodeA $CREMBX mail box NODEA_MBX (logical name) as permanent8 2) Define proxy on NodeA user A to point to NodeB user B 3) From NodeB in DCL# DCL>OPEN/WRITE tmp NodeB::NODEA_MBX 2 DCL> WRITE tmp "hello there this is remote node B" DCL> CLOSE tmp  J NodeA receives in NODEA_MBX the string : hello there this is remote node B  C The same works with pure Fortran Open and Write if the file name isr1 NodeB::"NODEA_MBX" with the mail box logical nameo between to double quotes.   + It must be an undocumented VMS feature ....t  . Thanks anyway. I'll go and try the tas to task  > "Carl Perkins" <carl@gerg.tamu.edu> a crit dans le message de' news:14JUL200320364419@gerg.tamu.edu...s/ > "Jean Mertens" <jmertens@skynet.be> writes...rJ > }"Bob Koehler" <koehler@eisner.nospam.encompasserve.org> a crit dans le; > }message de news:S5k3Eex0ySxf@eisner.encompasserve.org...vG > }> In article <3f12c6e7$0$271$ba620e4c@reader0.news.skynet.be>, "Jean ( > }Mertens" <jmertens@skynet.be> writes:D > }> > Using SYS$CREMBX, how to connect to a mailbox on another node	 (OpenVMS)d" > }> > (using Fortran preferrably)H > }> > or using regular OPEN Fortran statement in order to send mail box > }messagesE > }> > that are quite long ? > }>F > }>    Standard VMS mailboxes are not network accessable.  Try DECnet > }>    mailboxes instead. > }>K > }What is a Decnet Mailbox as opposed to the one created with SYS$CREMBX ?- >- > Jumping in here... > L > I'm not sure why it is sometimes called that, as it isn't really a mailboxI > as such. In some respects the data reading and writing is very similar.. >AD > SYS$CREMBX creates a "mailbox" which is an MBAx: type device. It'sG > essentially a chunk of memory that you can write to and read from viahE > the mailbox drivers. As such, it is specific to the system it is onaD > and there is no network access to it (unless you program a network# > server process that accesses it).  >tH > A "DECnet mailbox" isn't just a mailbox. In fact it isn't a mailbox inH > the MBAx: device sense at all, although there are mailboxes involved -J > one is used for the network command data at each end. The network driverI > writes various pieces of info to this, such as notification of new linke > requests, link failure, etc. >oA > }Is this the mailbox to return network information defined with- > }LIB$ASN_WTH_MBX ? >,I > That function does do that - when used to connect to the "_NET:" deviceWI > (which is the DECnet device template) the mailbox that this produces is/0 > the network command mailbox I mentioned above. >aK > This routine really is just a shortcut to do in one step what is normallye0 > done in two steps (sys$crembx and sys$assign). > # > }Where to find doc (VMS manual ?)q > }n+ > }Do you have an example in Fortran or C ?s >rL > You are probably looking for what is known as non-transparent task-to-taskG > DECnet communications. This is analogous to TCP/IP communications but D > perhaps a tad easier. The documentation for this used to be in the
 NetworkingG > manual (volume 5A of the system management set, if you happen to havetJ > access to the old printed documentation from a couple of versions back).F > I checked Google and it now appears to be in chapter 8 of the manual > located here:t >lJ > h71000.www7.hp.com/doc/73final/documentation/pdf/DECNET_OVMS_NET_MAN.PDF > A > There is an example of the non-transparent sort of thing in then SYS$EXAMPLESK > directory. The various DB_*.C files are involved, with the actual network K > stuff being in DB_SERVER.C and DB_REQUESTER.C (as the server and client).lI > These examples are aparently the basis for most programs written to use-L > this and are fairly complete in showing you how to do things - as I recallE > the client does it's IO sunchronously and the server does it mostlyoE > asynchronously, also demonstrating how to use IO completion ASTs to  > handle the asynch IO.n >fG > I don't know what exactly you are trying to do. You might actually be > > able to get by with the "transparent" version instead of the "nontransparent"J > version. This basically runs something on a remote system on demand. ForI > this you can specify a .com file to execute on the remote node just buyo! > doing an open on something like.. >  node"username password"::task=my_proceedureJ > which will run my_proceedure.com from the default login directory of theK > username you specified on the node you specified. That can send back datapI > by writing to sys$net, or receive data by reading from sys$net. As bothVI > are the same, you have to be careful if you want to communicate in bothsH > directions (I've never botherd doing so - not that I've ever used this > transparent DECnet for much).m >lL > Here is a brief and not incredibly clear example of the use of transparent > DECent communications: >n > $ set noverify# > $ if p1 .eqs. "" then goto server-2 > $ location = "''p1'""''p2' ''p3'""::""task=sq""" > $ type 'location > $ exit
 > $server:" > $ define/user sys$output sys$net > $ show process/quota > $ exit >-D > This is both the client and the server (you could separate them if) > you wanted, but this way is more tidy).e > G > Put it in the default login directory of an account. If you have thatt: > directory as your current default directory and do this: >i" > $ @sq.com node username password >BG > specifying the correct data for the 3 parameters, it will not jump to F > the server section, staying in the client section it will connect toF > node and run the same sq.com file in a process owned by the specfiedG > username. On the server end it will jump to the server section in the>J > code where it will send back the process quotas for the process in whichB > it is running. Back on the client end, the data will be read and
 displayed.K > Not real exciting, but it works. It cheats somewhat by not using explicitkI > open and read statments in the client part, instead relying on the TYPEl% > command do do these things for you.. >CG > If I had a more clear example, I'd have used it. This is the only oneS > I could find laying around.M > L > The "non-transparent" version requires you to do all the setup, monitoringK > of the command mailbox for messages, and shutdown manually. Note that allSK > the reads and writes are done with sys$qio[w] rather than native languagerH > read and write calls. In return for the added complexity, you can haveI > a process declare itself to be a DECnet object to which connections canoH > be made (so any connection to "TASK=MY_OBJECT_NAME" is handled by it).I > This is similar to a process declaring itself to be the handler for allEJ > TCP/IP connections to a specific port number (via the listen() routine). >.E > I get the impression that I'm not explaining any of this very well.N$ > Maybe someone else will do better. > K > Anyhow, if you need to write to a mailbox on anyother system, you have to J > create a server that runs on that system which will accept data over theI > network and write it to the mailbox. This can be done via DECnet or via_G > TCP/IP (or UDP/IP if you don't mind loosing the occasional packet) orA1 > any other networking software available to you.  >WL > Note that if what you areally mean by "mailbox" is "the place where e-mailL > is stored" rather than the VMS specific meaning of the term, then this has' > probably not been very useful to you.A >1
 > --- Carl   ------------------------------  # Date: Mon, 14 Jul 2003 20:36:17 GMTI9 From: "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> U Subject: Re: OpenVMS Pearl - OpenVMS V8.0, first release on Itanium, ships   today!!!./ Message-ID: <5vEQa.187$0o.185@news.cpqcorp.net>X  5 "Alan Frisbie" <Abuse@NelsonUSA.com> wrote in message & news:3F12F7AD.7030203@NelsonUSA.com... > 9 > > "Alan Frisbie" <Abuse@NelsonUSA.com> wrote in messageK( > > news:3F0B2ED6.10900@NelsonUSA.com... >RG > >> Which brings up the big question: What is the lowest cost IA64 boxI9 > >> that will run (NOTE: I did *NOT* say "support") VMS?  >E > > Fred Kleinsorge wrote:8 >  > The Itanium-1 Intel Reference platform "white box". > @ > Thank you.   Can anyone tell me of a good source for this box? >E  K Used.  They aren't made anymore.  Nor would I expect them to work on VMS in.J the 2004 timeframe, since we will be targeting only Itanium-2 architectureI and forward (so if it breaks... we won't fix it - and the code schedulinge will really suck for I1).n  K If you want to send me a Dell Madison box, I'm sure we'll make sure that itn1 works - even though I doubt we will "support" it.-   ------------------------------  % Date: Mon, 14 Jul 2003 11:34:21 -0700u( From: Alan Frisbie <Abuse@NelsonUSA.com>Y Subject: Re: OpenVMS Pearl - OpenVMS V8.0, first release on Itanium, ships   today!!!   t-, Message-ID: <3F12F7AD.7030203@NelsonUSA.com>  7 > "Alan Frisbie" <Abuse@NelsonUSA.com> wrote in messagem& > news:3F0B2ED6.10900@NelsonUSA.com...  E >> Which brings up the big question: What is the lowest cost IA64 box 7 >> that will run (NOTE: I did *NOT* say "support") VMS?    > Fred Kleinsorge wrote:6  > The Itanium-1 Intel Reference platform "white box".  > Thank you.   Can anyone tell me of a good source for this box?   Thanks,  Alan   ------------------------------  % Date: Mon, 14 Jul 2003 15:48:45 -07003( From: Alan Frisbie <Abuse@NelsonUSA.com>Y Subject: Re: OpenVMS Pearl - OpenVMS V8.0, first release on Itanium, ships   today!!!   tg, Message-ID: <3F13334D.1050707@NelsonUSA.com>   Fred Kleinsorge wrote:7 > "Alan Frisbie" <Abuse@NelsonUSA.com> wrote in messagem( > news:3F12F7AD.7030203@NelsonUSA.com... > 8 >>>"Alan Frisbie" <Abuse@NelsonUSA.com> wrote in message' >>>news:3F0B2ED6.10900@NelsonUSA.com...J >>F >>>>Which brings up the big question: What is the lowest cost IA64 box8 >>>>that will run (NOTE: I did *NOT* say "support") VMS? >>>i >>>Fred Kleinsorge wrote:o >>8 >> > The Itanium-1 Intel Reference platform "white box". >>A >> Thank you.   Can anyone tell me of a good source for this box?i  M > Used.  They aren't made anymore.  Nor would I expect them to work on VMS in-L > the 2004 timeframe, since we will be targeting only Itanium-2 architectureK > and forward (so if it breaks... we won't fix it - and the code scheduling' > will really suck for I1).3 > M > If you want to send me a Dell Madison box, I'm sure we'll make sure that ity3 > works - even though I doubt we will "support" it.   B Given all that, what would you suggest that a developer on a tight budget buy?i   Alan   ------------------------------  % Date: Tue, 15 Jul 2003 00:24:35 +0200p+ From: "Hans Vlems" <hvlems.nieuw@zonnet.nl>N5 Subject: Re: OpenVMS running under simh, ucx problem.?9 Message-ID: <bevaj6$9c5up$1@ID-143435.news.uni-berlin.de>T  7 "Peter Sutter" <nospam@sopac.com.au> schreef in berichtd7 news:3f123bb5$0$23614$5a62ac22@freenews.iinet.net.au...'K > Sorry if this is the wrong newsgroup, but I could not find one that dealss > with simh and OpenVMS. > I > I had a bit of a misfortune in that my microvax 3100 died. Fortunately,rK > there was a VaxStation 4000 VLC with which I could salvage the disks from]. > the 3100. (Created container files with LD). >-L > These container files I tranferred to a Linux box, where I installed simh.3 > Simh accepts the container files created with LD.a >nH > First time, OpenVMS comes up and the only thing I had to change was toH > define some system wide logicals in systartup_vms to map the old disks' > (dkannn) to the emulated ones (duan).r >iJ > Everything works fine, except for one slight problem with ucx. ucx seems toJ > work fine, with one exception: There is one node on the network which isH > unreachable, and that is the linux box the emulator runs on. This goes bothJ > ways, the emulated vax can't connect to the linux box, and the linux boxI > can't connect to the emulated vax. Both, the linux box and the emulatedt vaxu4 > can connect to anything else without any problems. >dL > Physically, both use the same network card, although I use a different MACL > address on the emulated vax. The emulator is running in superuser mode and. > puts the network card into promiscuous mode. >-G > The rest of the network (decnet) works fine, set host to the 4000 VLCh works. >,$ > Any clues and suggestions welcome. >u > Peter Sutter >y >? run DECnet on the simhVAX.   ------------------------------  # Date: Mon, 14 Jul 2003 18:13:54 GMTA# From: "John Smith" <a@nonymous.com> / Subject: OT: HP's PeeCee customer statisfactionkI Message-ID: <CpCQa.63555$sI91.13096@news04.bloor.is.net.cable.rogers.com>u  F http://story.news.yahoo.com/news?tmpl=story&ncid=1211&e=10&u=/ap/200302 710/ap_on_hi_te/tech_s_best_and_worst&sid=95573371  & Survey: HP, Compaq Flunk PC Users Test8 Thu Jul 10, 4:17 PM ET  Add Technology - AP to My Yahoo!    & By BRIAN BERGSTEIN, AP Business Writer  E NEW YORK - The biggest merger in high-tech history may be a financial E success for Hewlett-Packard Co. and the former Compaq Computer Corp., E but it hasn't made buyers of their computers any happier. Both brandsuB picked up terrible grades in an annual PC user satisfaction survey released Thursday.  ; Dell Computer Corp., on the other hand, continued to shine.J     more....  E Meanwhile, HP and Compaq desktops both got an E - the lowest possiblefC grade - along with those by Acer Inc., eMachines Inc. and NEC Corp.a  D HP had scored a D-minus last year, while Compaq had the scarlet E inB 2002, as well. When PC Magazine subscribers were asked about theirF experiences in getting technical support or repair help, HP and Compaq5 scored worse than average in 12 out of 15 categories.   B To be sure, HP's $19 billion acquisition of Compaq last year neverC really was billed as a boon to consumers. The companies' main goalsiA were to streamline their PC divisions and strengthen products andc! services for corporate customers.u  E Still, the failing grades indicate how little progress HP has made in E answering to competition from Dell, which regained the worldwide lead ) in PC sales from HP in the first quarter.    more...e    D I guess this means that any money for VMS advertising/marketing willB be shifted to the PeeCee side of the house to bolster HP's saggingD image there....after all, a satisfied PeeCee buyer is also a printer? and printer ink buyer, and that's where HP's bread is buttered.y   ------------------------------  # Date: Mon, 14 Jul 2003 19:12:46 GMTa" From:   VAXman-  @SendSpamHere.ORG3 Subject: Re: OT: HP's PeeCee customer statisfactiong0 Message-ID: <00A22DA2.DC8D4D1B@SendSpamHere.ORG>  o In article <CpCQa.63555$sI91.13096@news04.bloor.is.net.cable.rogers.com>, "John Smith" <a@nonymous.com> writes: G >http://story.news.yahoo.com/news?tmpl=story&ncid=1211&e=10&u=/ap/20030e3 >710/ap_on_hi_te/tech_s_best_and_worst&sid=95573371  > ' >Survey: HP, Compaq Flunk PC Users Test 9 >Thu Jul 10, 4:17 PM ET  Add Technology - AP to My Yahoo!r >  >i' >By BRIAN BERGSTEIN, AP Business Writerw >iF >NEW YORK - The biggest merger in high-tech history may be a financialF >success for Hewlett-Packard Co. and the former Compaq Computer Corp.,F >but it hasn't made buyers of their computers any happier. Both brandsC >picked up terrible grades in an annual PC user satisfaction survey  >released Thursday.i >a< >Dell Computer Corp., on the other hand, continued to shine. >e >U	 >more....c >nF >Meanwhile, HP and Compaq desktops both got an E - the lowest possibleD >grade - along with those by Acer Inc., eMachines Inc. and NEC Corp. >AE >HP had scored a D-minus last year, while Compaq had the scarlet E intC >2002, as well. When PC Magazine subscribers were asked about their G >experiences in getting technical support or repair help, HP and Compaq 6 >scored worse than average in 12 out of 15 categories. >0C >To be sure, HP's $19 billion acquisition of Compaq last year neverTD >really was billed as a boon to consumers. The companies' main goalsB >were to streamline their PC divisions and strengthen products and" >services for corporate customers. > F >Still, the failing grades indicate how little progress HP has made inF >answering to competition from Dell, which regained the worldwide lead* >in PC sales from HP in the first quarter. >a >more... >t >rE >I guess this means that any money for VMS advertising/marketing will C >be shifted to the PeeCee side of the house to bolster HP's sagginghE >image there....after all, a satisfied PeeCee buyer is also a printerp@ >and printer ink buyer, and that's where HP's bread is buttered.  C A PeeCee is a PeeCee.  They all use an intel or AMD chip capable of2D executing BillyBloatWarez.  How is any one different than the other?  E Comparing systems with the same memory, CPU, and drive hardware, whatn is the difference? --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMu            h5   "Well my son, life is like a beanstalk, isn't it?" N   ------------------------------  % Date: Mon, 14 Jul 2003 12:55:52 -0700s* From: "Jack Peacock" <peacock@simconv.com>3 Subject: Re: OT: HP's PeeCee customer statisfactiony2 Message-ID: <oY6cndo6kOJVl46iRTvUqw@mpowercom.net>  , <VAXman- @SendSpamHere.ORG> wrote in message* news:00A22DA2.DC8D4D1B@SendSpamHere.ORG...E > A PeeCee is a PeeCee.  They all use an intel or AMD chip capable ofgF > executing BillyBloatWarez.  How is any one different than the other? >nG > Comparing systems with the same memory, CPU, and drive hardware, whatl > is the difference? >tK I have occasion to install Win2K servers for both Dell and Compaq (customeruK buys whatever hardware they want).  Often this involves migrating from somenI old machine.  The Compaq servers have this nice surprise where IDE drivesaH are not supported in the boot rom.  The Dell still retains that feature.H Irrelevant?  Well no, because imaging an old drive to migrate to the newK machine is far faster than using tape backup.  On the Compaqs I have to putfL in an external SCSI controller and box of SCSI drives.  On the Dell I use anL 40GB IDE internally.  Its also handy to make checkpoint disk images on thoseG cheap and plentiful large IDE drives while upgrading.  Given a choice IJ recomend Dell over Compaq.  H Oh yeah, there's also those neato "server agents" from Compaq that causeK nightmares when upgrading.  Best of all, some of them can't be uninstalled.sE Others leave open HTTP backdoors on high ports.  Our policy now is ton9 *never* install those agents unless customer requests it.s  C The parts may be commodities but how they bolt together does make a K difference.  Not that Dell is all that outstanding.  Trying to make a phoneUJ call into "Dell Hell" tech support is worse than Compaq (but not by much).    Jack PeacockL   ------------------------------    Date: 14 Jul 2003 15:00:33 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)h3 Subject: Re: OT: HP's PeeCee customer statisfaction 3 Message-ID: <7Ag6OAC$leCt@eisner.encompasserve.org>E  U In article <00A22DA2.DC8D4D1B@SendSpamHere.ORG>,   VAXman-  @SendSpamHere.ORG writes:  > E > A PeeCee is a PeeCee.  They all use an intel or AMD chip capable ofoF > executing BillyBloatWarez.  How is any one different than the other? > G > Comparing systems with the same memory, CPU, and drive hardware, whatg > is the difference?  D    Some of them actually do run worse than others.  I've had PC thatC    won't run properly until they've warmed up.  You can get really  )    flaky parts assembled by some vendors.D   ------------------------------    Date: 14 Jul 2003 16:21:35 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)-3 Subject: Re: OT: HP's PeeCee customer statisfactioni3 Message-ID: <uxRwEnNz5Hsn@eisner.encompasserve.org>v  _ In article <oY6cndo6kOJVl46iRTvUqw@mpowercom.net>, "Jack Peacock" <peacock@simconv.com> writes:e  E > The parts may be commodities but how they bolt together does make ahM > difference.  Not that Dell is all that outstanding.  Trying to make a phonetL > call into "Dell Hell" tech support is worse than Compaq (but not by much).  J Dell's ditzy interns are playing office politics and making TV commercialsJ about it.  (Probably a US-only joke.  In fact _hopefully_ a US-only joke.)   ------------------------------  # Date: Tue, 15 Jul 2003 00:15:26 GMTm From: Rob.Buxton@wcc.govt.nz3 Subject: Re: OT: HP's PeeCee customer statisfactionr% Message-ID: <3f134713.102096046@news>   2 On Mon, 14 Jul 2003 12:55:52 -0700, "Jack Peacock" <peacock@simconv.com> wrote:  - ><VAXman- @SendSpamHere.ORG> wrote in messagen+ >news:00A22DA2.DC8D4D1B@SendSpamHere.ORG... F >> A PeeCee is a PeeCee.  They all use an intel or AMD chip capable ofG >> executing BillyBloatWarez.  How is any one different than the other?H >>H >> Comparing systems with the same memory, CPU, and drive hardware, what >> is the difference?m >>L >I have occasion to install Win2K servers for both Dell and Compaq (customerL >buys whatever hardware they want).  Often this involves migrating from someJ >old machine.  The Compaq servers have this nice surprise where IDE drivesI >are not supported in the boot rom.  The Dell still retains that feature.tI >Irrelevant?  Well no, because imaging an old drive to migrate to the new L >machine is far faster than using tape backup.  On the Compaqs I have to putM >in an external SCSI controller and box of SCSI drives.  On the Dell I use anrM >40GB IDE internally.  Its also handy to make checkpoint disk images on thoseDH >cheap and plentiful large IDE drives while upgrading.  Given a choice I >recomend Dell over Compaq.  >eI >Oh yeah, there's also those neato "server agents" from Compaq that causetL >nightmares when upgrading.  Best of all, some of them can't be uninstalled.F >Others leave open HTTP backdoors on high ports.  Our policy now is to: >*never* install those agents unless customer requests it.E We use those "neato server agents" extensively and I find updates arec	 a doddle. E We use Insight Manager and there are any number of ways of performing- the updates.2 As for HTTP Backdoors, the Agents are now all SSL.@ You're advising your clients to do without an excellent and free Server Management tool.D >ID >The parts may be commodities but how they bolt together does make aL >difference.  Not that Dell is all that outstanding.  Trying to make a phoneK >call into "Dell Hell" tech support is worse than Compaq (but not by much).a >   Jack Peacock >s >.   ------------------------------  % Date: Mon, 14 Jul 2003 17:05:08 -0700t, From: Ken Fairfield <My.Full.Name@intel.com>F Subject: Re: Q: How to close and reopen detached process's Sys$Output?) Message-ID: <3F134534.E8788CF4@intel.com>C  ! VAXman-, @SendSpamHere.ORG wrote:h > Z > In article <3F0F3557.2750455D@intel.com>, Ken Fairfield <My.Full.Name@intel.com> writes:3 > >Using: Compaq C V6.5-001 on OpenVMS Alpha V7.3-1t > >aG > >I'm trying to enhance an existing program, written in C and run as atD > >detached process, to periodically close it's stdout/stderr (i.e.,G > >Sys$Output as specified in the /Output qualifier to RUN), and reopen  > >a new version of the file.t > > D > >In order that the output (log) file be readable while the processA > >is running (and it may run for months at a time), I've already D > >added code to the beginning of main that looks roughly like this: > >nD > >----------------------------------------------------------------- > >            fclose(stderr); > >            fclose(stdout);@ > >            stdout = fopen("sys$output", "w", "shr=get,put"); >  > Try: >  >     char logfilenm[255]; > " >     fgetname(stdout, logfilenm); >     fclose(stdout);  > 2 >     stdout = fopen(logfilenm,"w","shr=get,put");  $ 	Thanks, Brian, that worked a champ!  2 	I did have to make the following change, i.e., to; strip off the file version before feeding it back to fopen:n  2 --------------------------------------------------     char logfilnm[255];1     char *lfnptr;   "     fgetname(stdout, logfilnm, 1);#     lfnptr = strtok(logfilnm, ";");r     strcpy(logfilnm, lfnptr);        fclose(stdout); 1     stdout = fopen(logfilnm, "w", "shr=get,put");23 ---------------------------------------------------p  8 If I were more facile with C, I probably could have done; that better, but the above does, indeed, open a new versione; of the output file each time the fclose/fopen pair is done,y. at least for a detached process without a CLI.  8 When run under the CLI (as a batch job, or detached from7 a command file with SYS$SYSTEM:LOGINOUT as the image), r4 'stdout' points to SYS$OUTPUT, and that doesn't get  translated back to a file name.i  7 I may attempt solving this under the CLI, but at least e5 you've given a good solution for the non-CLI problem.-  
 	Thanks!  Ken- --6 I don't speak for Intel, Intel doesn't speak for me...  
 Ken Fairfields" D1C Automation VMS System Support ) kenneth[dt]h[dt]fairfield[ta]intel[dt]com1   ------------------------------  % Date: Mon, 14 Jul 2003 21:23:45 +0200b9 From: Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <aaa@aaa.com> 0 Subject: Re: Questions from Pat St. Laurent @ HP' Message-ID: <3F130341.A7D03536@aaa.com>o  < Hm, doesn't HP reject mails based on IP addresses or domains6 the's in "black-lists" that HP gets from some 3-part ?	 Jan-Erik.s   Larry Kilgallen wrote: > l > In article <O5BQa.157$X4.44@news.cpqcorp.net>, "Fred Kleinsorge" <my-last-name@stardotzko.dec.com> writes: > >mO > > I sent you mail with her address.  In general just take the name and hp.coma# > > for example - first.last@hp.comr > ' > Brian's issue is not email addresses.  > I > There is a problem in that HP's email system rejects email from Brian's J > IP address, and no HP employee who deals with Brian has been abl to find# > who at HP could fix that problem.-   ------------------------------    Date: 14 Jul 2003 20:33:53 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)e0 Subject: Re: Questions from Pat St. Laurent @ HP3 Message-ID: <2GyOioZc0UMZ@eisner.encompasserve.org>e  [ In article <3F1358E9.7798D0CB@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:  > Jan-Erik Sderholm wrote:y >> i? >> Hm, doesn't HP reject mails based on IP addresses or domainsc9 >> the's in "black-lists" that HP gets from some 3-part ?t > G > As I remember VAXman's explanation, his IP address back-translates tooC > something other than the sending domain. As a result, hp.com's MX  > bounces the mail.t  & That seems like a defective algorithm.  D I believe a correct algorithm would take the domain name obtained byE a reverse DNS lookup and ensure a forward lookup returned the same IPc7 address.  An IP address can have multiple domain names.   > I just ran tmesis.com through http://www.dnsreport.com and got   	0 Failuress) 	7 Warnings (none related to reverse DNS)e   ------------------------------  % Date: Mon, 14 Jul 2003 20:29:13 -0500"1 From: "David J. Dachtera" <djesys.nospam@fsi.net>t0 Subject: Re: Questions from Pat St. Laurent @ HP' Message-ID: <3F1358E9.7798D0CB@fsi.net>u   Jan-Erik Sderholm wrote:t > > > Hm, doesn't HP reject mails based on IP addresses or domains8 > the's in "black-lists" that HP gets from some 3-part ?  E As I remember VAXman's explanation, his IP address back-translates toxA something other than the sending domain. As a result, hp.com's MX  bounces the mail.w   -- m David J. Dachterad dba DJE Systemsv http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------   Date: 14 Jul 2003 19:58:37 GMT" From: rufusvii@aol.com (Rufus VII)" Subject: Re: RF31 hardware problem: Message-ID: <20030714155837.07182.00000278@mb-m25.aol.com>  ; hey, no hardware gurus or field service reading this board? N I still need to know, what the meaning of error code A508 is, and if there's a) change to "revitalize" the two RF31 disks    Rufusb   ------------------------------    Date: 15 Jul 2003 00:08:45 +0200B From: holitska_a@cut-it-outludens.elte.hu (Holi - Holitska Andrs)" Subject: Re: RF31 hardware problem! Message-ID: <d1JtfGtvQZR2@ludens>i  _ In article <20030714155837.07182.00000278@mb-m25.aol.com>, rufusvii@aol.com (Rufus VII) writes: = > hey, no hardware gurus or field service reading this board?eP > I still need to know, what the meaning of error code A508 is, and if there's a+ > change to "revitalize" the two RF31 disksd  ? Maybe you should try comp.sys.dec and comp.sys.dec.micro too... ' I can't help you, but good luck anyway!i   <Holi>   > RufusT   ------------------------------  % Date: Mon, 14 Jul 2003 19:47:35 -0500h From: "m68k" <m68k@comcast.net>r, Subject: Re: Serial Connection over Ethernet0 Message-ID: <O6CcnaMsR77e046iXTWJhw@comcast.com>  F If you want real cheap you could set up a workstation (Unix, Linux, orG Windows) attached to the console with a terminal emulation program. YouwA could then use graphical remote administration software like VNC,cJ PcAnywhere, etc. to control the workstation over IP and thus get access to4 the VAX console via the terminal emulation software.  K You could also do the same thing "text based" using ssh, telnet, or rlogin. E I did something similar to this once on a non-mission critical VAX. IaK telnet'ed into a cheap Linux box and used an open-source terminal emulationFJ program. This setup was totally free! I used a spare 486, Redhat 6.2 LinuxJ distro, and Minicom (also worked with Kermit) terminal emulation software.  L Of course, these options open you up for some major security threats, so you want to consider those as well!e   -- MARVIN JOHNSON* "B.Eckstein" <eck@ivu.de> wrote in message' news:bendqe$79i$1@swifty.westend.com...s* > Rolona schrub im Jahre 11.07.2003 21:05: > H > > I need to do some work on a system miles away and need access to the consoletI > > from where I am. I have Wan access to the same room. Anyone know of a > > > "cheap" way to rig up the console of a VAX over an IP WAN? >uF > DecServer700. Or something like that. There are also loads of single1 > port servers to find on the net e.g. www.wut.de' >a > --J > B.Eckstein, eck@ivu.de          Cheap, Fast, Good - pick any two of them; > Die FAQ zu de.comp.hardware.netzwerke: http://how.to/dchnaJ > Tips und Tricks zum Mozilla & Netscape 6/7: http://www.holgermetzger.de/J > I want your VAX http://monster-island.org/tinashumor/humor/iwantyou.html >    ------------------------------   Date: 14 Jul 2003 19:20:38 GMT/ From: Chris Scheers <chris@applied-synergy.com>s/ Subject: Re: VAXstation 4000m90 with no monitorn3 Message-ID: <3F3BE10B.F852EE0C@applied-synergy.com>r   "Lars Holmstrm" wrote:a > G > I have a VAXstation 4000m90 and would like to use it with no monitor.h0 > I connected a VT terminal to the console port. > L > There is a switch at the fronpanel called S3. SHould this switch be in the > up or down position ?e > ! > Should the console be 9k6 8N1 ?b  G There is one other thing to be aware of.  You must leave the video cardeE in the 4000m90, even if you aren't using it.  If you remove the videosD card, VMS will not be able to identify the machine and your licenses will not work.  G -----------------------------------------------------------------------r$ Chris Scheers, Applied Synergy, Inc.  C Voice: 817-237-3360            Internet: chris@applied-synergy.com     Fax: 817-237-3074r   ------------------------------    Date: 14 Jul 2003 12:15:06 -0700$ From: rcbryan@hotmail.com (RC Bryan): Subject: What color were the VMS manuals for old versions?= Message-ID: <fbcf38dc.0307141115.75bbbca0@posting.google.com>n  C I was having a discussion in the office with an old timer and theree? was some controversy about the color of the binders for the olds versions of VMS.   I dimly remember having:   V1 (before my time with VMS) V2 (before my time with VMS) V3 Bluer	 V4 Orangea V5 Grayn! V6 White (paperback, not binders)f' V7 Do they have printed manuals for V7?e   Thanks,o	 /RC Bryan    ------------------------------  # Date: Mon, 14 Jul 2003 20:01:21 GMT ; From: Charlie McCutcheon <charlie.mccutcheonspam@nothp.com>n> Subject: Re: What color were the VMS manuals for old versions?) Message-ID: <3F130540.B63FC012@nothp.com>f   >V5 Gray  A My memory says China Red, which looks to most people like orange.h  ( >V7 Do they have printed manuals for V7?  C Yes, non-binder manuals.  Mine are from 7.1, Various colors, mostly & white.  Also manuals available online!   Charlie-     RC Bryan wrote:e  E > I was having a discussion in the office with an old timer and there.A > was some controversy about the color of the binders for the old3 > versions of VMS. >r > I dimly remember having: >i > V1 (before my time with VMS) > V2 (before my time with VMS)	 > V3 Bluel > V4 OrangeX	 > V5 Graya# > V6 White (paperback, not binders)N) > V7 Do they have printed manuals for V7?o >o	 > Thanks,  > /RC BryanE   ------------------------------  % Date: Mon, 14 Jul 2003 15:49:11 -0400 * From: "Stanley F. Quayle" <stan@stanq.com>> Subject: Re: What color were the VMS manuals for old versions?. Message-ID: <3F12D0F7.17900.A0C4020@localhost>  ( On 14 Jul 2003 at 12:15, RC Bryan wrote:E > I was having a discussion in the office with an old timer and theretA > was some controversy about the color of the binders for the olda > versions of VMS. > # > V6 White (paperback, not binders)s) > V7 Do they have printed manuals for V7?u   Wasn't that:  1 V6 Paperback, not binders, color-coded by subjecte* V7 Paperback, not binders, not color-coded  D Paper manuals are available for V7.  However, the CD set is so much 
 lighter...  
 --Stan Quaylec Quayle Consulting Inc.  
 ----------C Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363  Fax: +1 614 868-1671S1 8572 North Spring Ct. NW, Pickerington, OH  43147o= Preferred address:  stan@stanq.com       http://www.stanq.coms   ------------------------------  % Date: Mon, 14 Jul 2003 12:44:02 -0700 ( From: Alan Frisbie <Abuse@NelsonUSA.com>> Subject: Re: What color were the VMS manuals for old versions?, Message-ID: <3F130802.5060102@NelsonUSA.com>   RC Bryan wrote:sE > I was having a discussion in the office with an old timer and there.A > was some controversy about the color of the binders for the olde > versions of VMS. >  > I dimly remember having: >  > V1 (before my time with VMS) > V2 (before my time with VMS)  # V2 Light Blue (I still have my set)    Alan   ------------------------------    Date: 14 Jul 2003 15:03:30 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)m> Subject: Re: What color were the VMS manuals for old versions?3 Message-ID: <LICiOTCpyCOW@eisner.encompasserve.org>   d In article <fbcf38dc.0307141115.75bbbca0@posting.google.com>, rcbryan@hotmail.com (RC Bryan) writes:  @    IIRC V0.5 was the first ship to customers, somebody must have    seen the manuals.   > V1 (before my time with VMS)B    Manuals?  Nobody ever let us see the manuals when we had VMS 1.   > V2 (before my time with VMS)    Blueo  	 > V3 Bluet > V4 Orange     aka "Chinese Red"   ------------------------------    Date: 14 Jul 2003 16:19:12 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)c> Subject: Re: What color were the VMS manuals for old versions?3 Message-ID: <aDkMOk$ahAsg@eisner.encompasserve.org>o  [ In article <3F12D0F7.17900.A0C4020@localhost>, "Stanley F. Quayle" <stan@stanq.com> writes:   F > Paper manuals are available for V7.  However, the CD set is so much  > lighter...  , Which can be an advantage or a disadvantage.  A For V3 or thereabouts I actually looked into the price of bindersrA that lock into a bench as used for telephone books near a bank of/
 phone booths.n   ------------------------------  % Date: Mon, 14 Jul 2003 18:26:12 -0400e* From: "Stanley F. Quayle" <stan@stanq.com>> Subject: Re: What color were the VMS manuals for old versions?. Message-ID: <3F12F5C4.10421.A9C00A8@localhost>  / On 14 Jul 2003 at 16:19, Larry Kilgallen wrote: C > For V3 or thereabouts I actually looked into the price of bindersmC > that lock into a bench as used for telephone books near a bank ofs > phone booths.e  = Right.  I remember using binders like that, back in the 70's..  @ These days, you'd install the CD's on a file server that's in a < locked room, and let everyone access them via the network...  
 --Stan Quayleh Quayle Consulting Inc.  
 ----------C Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363  Fax: +1 614 868-1671h1 8572 North Spring Ct. NW, Pickerington, OH  43147e= Preferred address:  stan@stanq.com       http://www.stanq.com    ------------------------------    Date: 14 Jul 2003 20:03:46 -0700$ From: rcbryan@hotmail.com (RC Bryan)> Subject: Re: What color were the VMS manuals for old versions?= Message-ID: <fbcf38dc.0307141903.50303524@posting.google.com>e  E > > For V3 or thereabouts I actually looked into the price of binderssE > > that lock into a bench as used for telephone books near a bank oft > > phone booths.  > ? > Right.  I remember using binders like that, back in the 70's.: > B > These days, you'd install the CD's on a file server that's in a > > locked room, and let everyone access them via the network...  C Those were the good old days... Naaah, they were the bad old days. v@ High priced Manuals were a big pain.  The idea that I can have a> hundred pounds of manuals on a CD at my disposal is fantastic.  > I have a set of V6 manuals on the shelf over my desk (with theA overflow in the file cabinet) that I refer to endlessly when I ambF working on VMS problems.  They were going out on my last job so I tookE them with me.  Anything subtle or V7 specific, requires accessing the=> online manuals.  I have the CD but the DEC/HP web site is moreB convenient but nothing beats the paper.  I can prop my feet up andE flip through the manuals, keeping my fingers in the interesting pagesl/ as I flip. Browser bookmarks just don't cut it.    In the original message I said,e< > I was having a discussion in the office with an old timer F It occurs to me that, perhaps I should have said, another old timer...   Regards,	 /RC BryanB   ------------------------------   End of INFO-VAX 2003.387 ************************