1 INFO-VAX	Wed, 19 Apr 2006	Volume 2006 : Issue 217       Contents: Re: Anyone know David Cathey?  Re: Anyone know David Cathey?  Re: DSPP EMEA contact numbers 5 Re: Error codes from Distributed NetBeans for OpenVMS 5 Re: Error codes from Distributed NetBeans for OpenVMS  F$getqui Re: F$getqui Re: F$getqui Re: F$getqui Re: F$getqui Re: Free rz26s and rz28s Re: Free rz26s and rz28s, Re: How does a fix become a published patch?, Re: How does a fix become a published patch?, Re: How does a fix become a published patch?, Re: How does a fix become a published patch?, Re: How does a fix become a published patch? new DST routines for C-RTL Re: new DST routines for C-RTL Re: new DST routines for C-RTL Re: OT: Sparc not dead yet Re: OT: Sparc not dead yet Re: OT: Sparc not dead yet< Re: Partitioned cluster question (reboot during lost quorum)< Re: Partitioned cluster question (reboot during lost quorum)< Re: Partitioned cluster question (reboot during lost quorum)< Re: Partitioned cluster question (reboot during lost quorum)< Re: Partitioned cluster question (reboot during lost quorum)< Re: Partitioned cluster question (reboot during lost quorum)< Re: Partitioned cluster question (reboot during lost quorum)< Re: Partitioned cluster question (reboot during lost quorum)< Re: Partitioned cluster question (reboot during lost quorum)< Re: Run Program with Shareable Image in Non-Default Location$ Re: SoyMail & insufficient privilege. Re: Strange behavior of DEFINE with equal-sign. Re: Strange behavior of DEFINE with equal-sign. Re: Strange behavior of DEFINE with equal-sign! What (WAS) wrong with my network? % Re: What (WAS) wrong with my network? % Re: What (WAS) wrong with my network?  [CSWB] Any news?  F ----------------------------------------------------------------------    Date: 19 Apr 2006 07:30:20 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) & Subject: Re: Anyone know David Cathey?3 Message-ID: <AJB9MZoz5XLB@eisner.encompasserve.org>   f In article <LsudnSZF8rxcA9jZRVn-uw@comcast.com>, "Richard B. Gilbert" <rgilbert88@comcast.net> writes: > H > We all know our VMS systems will run as long as somebody remembers to  > pay the electric bill. > I > When the power goes off, however, my UPS runs down in about 15 minutes   > and that's it.  G    Your UPS lasts longer than my UPS.  But when the power comes back on 2    my systems autoboot even if I'm away for weeks.   ------------------------------    Date: 19 Apr 2006 07:31:27 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) & Subject: Re: Anyone know David Cathey?3 Message-ID: <8vS3zRure7A3@eisner.encompasserve.org>   ^ In article <1145414641.637104.16990@v46g2000cwv.googlegroups.com>, davidc@montagar.com writes: > B > Well, yesterday we've had rolling blackouts through North Texas.E > Keeping OpenVMS up isn't the problem - it's keeping everything else  > around it up.   5    Ah, yes.  The flakey things outside the box.   8-)    ------------------------------    Date: 19 Apr 2006 13:31:01 +01006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)& Subject: Re: DSPP EMEA contact numbers, Message-ID: <44463b95$1@news.langstoeger.at>  e In article <1145366849.046793.80980@v46g2000cwv.googlegroups.com>, "Ian Miller" <ijm@uk2.net> writes:  >>From the UK I have   Not only from the UK. ( It is a international tollfree number...   >Phone: + 800 100 92970  >Fax: + 353 91 75 4445   >Email: dspp.emea@hp.com   Don't know the Fax#    --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------    Date: 19 Apr 2006 01:01:13 -0700 From: jazzfahrer@gmail.com> Subject: Re: Error codes from Distributed NetBeans for OpenVMSC Message-ID: <1145433673.720335.216060@v46g2000cwv.googlegroups.com>    Didn't get any response... ;-(   ------------------------------  # Date: Wed, 19 Apr 2006 13:54:21 GMT & From: John Reagan <john.reagan@hp.com>> Subject: Re: Error codes from Distributed NetBeans for OpenVMS1 Message-ID: <hir1g.6526$WX2.372@news.cpqcorp.net>    jazzfahrer@gmail.com wrote:   > Didn't get any response... ;-( >   # I'll ping Meg to see what happened.    --   John Reagan / HP Pascal/{A|I}MACRO for OpenVMS Project Leader  Hewlett-Packard Company    ------------------------------    Date: 19 Apr 2006 08:30:15 -0700% From: "Eddie" <eddie.camaroto@va.gov>  Subject: F$getqui C Message-ID: <1145460615.357158.196830@u72g2000cwu.googlegroups.com>   D The program below works great till I uncomment out the IF statement.D Then it just loops forever.  I am at wits end.  If anyone has a clue please let me know.   
 Thank you,   Eddie.camaroto@va.gov    $ queue_number = 0 $ job_number = 0 $ TEMP_HOLDER = 0  $ TEMP = F$GETQUI("")  $! $QLOOP1:E $ !IF JOB_NUMBER .gt. 140 THEN MAIL/SUB="EMERGENCY on queue<''qname'> 
 with ''job/ _number' jobs" nl: smtp%"eddie.camaroto@va.gov"  $ JOB_NUMBER = 0 $QLOOP: 4 $ QNAME = F$GETQUI("DISPLAY_QUEUE","QUEUE_NAME","*")! $ QUEUE_NUMBER = QUEUE_NUMBER + 1 * $ IF QNAME .EQS. "" THEN GOTO EXIT_SECTION $ WRITE SYS$OUTPUT "" # $ WRITE SYS$OUTPUT "QUEUE: ", QNAME  $! $JLOOP: 8 $ JNAME = F$GETQUI("DISPLAY_JOB","JOB_NAME",,"ALL_JOBS")$ $ IF JNAME .EQS. "" THEN GOTO QLOOP1 $ JOB_NUMBER = JOB_NUMBER + 1  $ QUEUE_JOB_COUNTER = 4 F$GETQUI("DISPLAY_JOB","JOB_NAME",,"FREEZE_CONTEXT")9 $ WRITE SYS$OUTPUT "TOTAL_NUMBER_JOBS FOR QUEUE<''QNAME'>  ''JOB_NUMBER'" $ GOTO JLOOP $! $exit_section:6 $ write sys$output "There are ''queue_number' queues."   ------------------------------    Date: 19 Apr 2006 10:54:14 -0500 From: briggs@encompasserve.org Subject: Re: F$getqui 3 Message-ID: <8yail2sDUlO3@eisner.encompasserve.org>   k In article <1145460615.357158.196830@u72g2000cwu.googlegroups.com>, "Eddie" <eddie.camaroto@va.gov> writes: F > The program below works great till I uncomment out the IF statement.F > Then it just loops forever.  I am at wits end.  If anyone has a clue > please let me know.  >  > Thank you, >  > Eddie.camaroto@va.gov  >  > $ queue_number = 0 > $ job_number = 0 > $ TEMP_HOLDER = 0  > $ TEMP = F$GETQUI("")  > $!
 > $QLOOP1:G > $ !IF JOB_NUMBER .gt. 140 THEN MAIL/SUB="EMERGENCY on queue<''qname'>  > with ''job1 > _number' jobs" nl: smtp%"eddie.camaroto@va.gov"   @ Speculation:  The smtp transport does delivery using VMS queues.C Accordingly, it interacts with the VMS queue manager in the context ! of the currently running process.   @ [When VMS mail uses a transport (<transportname>%), it does this, by dynamically activating a shareable image,@ SYS$SHARE:<transportname>_MAILSHR in process context and calling" it to do transport-specific stuff]  > This destroys any queue context that you are depending on withA F$GETQUI.  There is only context maintained for the communication 6 between the job controller and any particular process.  A With the context blown, your F$GETQUI goes back and restarts from  the beginning.  G You can confirm this theory by changing the destination address on your   mail message to something local.   ------------------------------  % Date: Wed, 19 Apr 2006 12:40:10 -0400  From: norm.raphael@metso.com Subject: Re: F$getqui Q Message-ID: <OF4AA40203.EDC0D274-ON85257155.005AF935-85257155.005B9114@metso.com>   7 I just ran it with the if-statement reformatted and all 7 three "$!" comments removed, and the count lowered to 3 5 for execution (and my smtp email address, of course).   = The Multinet smtp queue entries build during execution, viz.,  QUEUE: MULTINET_SMTP- TOTAL_NUMBER_JOBS FOR QUEUE <MULTINET_SMTP> 1 - TOTAL_NUMBER_JOBS FOR QUEUE <MULTINET_SMTP> 2 - TOTAL_NUMBER_JOBS FOR QUEUE <MULTINET_SMTP> 3 - TOTAL_NUMBER_JOBS FOR QUEUE <MULTINET_SMTP> 4 - TOTAL_NUMBER_JOBS FOR QUEUE <MULTINET_SMTP> 5 - TOTAL_NUMBER_JOBS FOR QUEUE <MULTINET_SMTP> 6 - TOTAL_NUMBER_JOBS FOR QUEUE <MULTINET_SMTP> 7 - TOTAL_NUMBER_JOBS FOR QUEUE <MULTINET_SMTP> 8   . but everything runs fine and I get the emails.   Have you tried:   @ -*- Charlie Hammond's unsupported DCL checker (Version V3.4) -*- Checking file TEST.COM;1 19-APR-2006 12:27:59.31   # Checking for DCL_CHECK$ logicals... "   "DCL_CHECK$SUPPRESS_BL" = "TRUE"#   "DCL_CHECK$SUPPRESS_CCN" = "TRUE" #   "DCL_CHECK$SUPPRESS_CLD" = "TRUE" #   "DCL_CHECK$SUPPRESS_LFF" = "TRUE" #   "DCL_CHECK$SUPPRESS_LND" = "TRUE" #   "DCL_CHECK$SUPPRESS_LOD" = "TRUE"   . Starting Pass 1 -- 19-APR-2006 12:27:59.67 .... Starting Pass 2 -- 19-APR-2006 12:28:01.30 .... Starting Pass 3 -- 19-APR-2006 12:28:01.47 ...  & Procedure contains:     27 total linesJ                         22 code lines (including 0 lines (0%) w/ comments)8                          2 additional continuation lines5                          0 lines w/i $DECK/$EOD pairs @                          0 comment only lines (0% of code lines)&                          3 blank lines&                          2 diagnostics  #  LINE  CODE  --DIAGNOSTIC MESSAGE--   B     7  ICF-W  possible invalid character found ( % or &) (warning)4    11  LNR-I  label "QLOOP" not referenced (warning)  0 -*- END OF LISTING -*-   19-APR-2006 12:28:01.95   on your version?  @ "Eddie" <eddie.camaroto@va.gov> wrote on 04/19/2006 11:30:15 AM:  F > The program below works great till I uncomment out the IF statement.F > Then it just loops forever.  I am at wits end.  If anyone has a clue > please let me know.  >  > Thank you, >  > Eddie.camaroto@va.gov  >  > $ queue_number = 0 > $ job_number = 0 > $ TEMP_HOLDER = 0  > $ TEMP = F$GETQUI("")  > $!
 > $QLOOP1:# > $!$ IF JOB_NUMBER .gt. 140 THEN - H > $!   MAIL/SUB="EMERGENCY on queue<''qname'> with ''job_number' jobs" -' > $!   nl: smtp%"eddie.camaroto@va.gov"  > $ JOB_NUMBER = 0	 > $QLOOP: 6 > $ QNAME = F$GETQUI("DISPLAY_QUEUE","QUEUE_NAME","*")# > $ QUEUE_NUMBER = QUEUE_NUMBER + 1 , > $ IF QNAME .EQS. "" THEN GOTO EXIT_SECTION > $ WRITE SYS$OUTPUT "" % > $ WRITE SYS$OUTPUT "QUEUE: ", QNAME  > $!	 > $JLOOP: : > $ JNAME = F$GETQUI("DISPLAY_JOB","JOB_NAME",,"ALL_JOBS")& > $ IF JNAME .EQS. "" THEN GOTO QLOOP1 > $ JOB_NUMBER = JOB_NUMBER + 1  > $ QUEUE_JOB_COUNTER = 4 F$GETQUI("DISPLAY_JOB","JOB_NAME",,"FREEZE_CONTEXT")J > $ WRITE SYS$OUTPUT "TOTAL_NUMBER_JOBS FOR QUEUE<''QNAME'> ''JOB_NUMBER'" > $ GOTO JLOOP > $! > $exit_section:8 > $ write sys$output "There are ''queue_number' queues." >    ------------------------------    Date: 19 Apr 2006 18:38:22 +01006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) Subject: Re: F$getqui , Message-ID: <4446839e$1@news.langstoeger.at>  T In article <8yail2sDUlO3@eisner.encompasserve.org>, briggs@encompasserve.org writes:H >You can confirm this theory by changing the destination address on your! >mail message to something local.   ( Or by adding a SPAWN to the MAIL command   --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------    Date: 19 Apr 2006 10:11:05 -0700 From: etmsreec@yahoo.co.uk Subject: Re: F$getqui C Message-ID: <1145466665.761023.266840@i39g2000cwa.googlegroups.com>   E Would it make a difference (I dunno the answer to that) if you didn't A declare smtp% in the address and merely left VMS and the VMS mail . program to decide on the mail protocol to use?  B Alternatively, have your procedure create and output file then getF another job to check for the presence of the file every so often.  Not8 an immediate notification, but a kludge that might work.   ------------------------------    Date: 19 Apr 2006 06:30:39 -0700+ From: "Dr. Dweeb" <comp.os.vms@hotmail.com> ! Subject: Re: Free rz26s and rz28s C Message-ID: <1145453439.291702.145890@g10g2000cwb.googlegroups.com>   E But I think the RZ28 was the largest device supported for the Jensen,  so I would like one please :)    Dweeb    ------------------------------    Date: 19 Apr 2006 07:16:15 -0700' From: "syslost" <wm.reynolds@gmail.com> ! Subject: Re: Free rz26s and rz28s C Message-ID: <1145456175.826146.279640@i39g2000cwa.googlegroups.com>   C I worked late last night and didn't check my email when I got home, D I'll check it tonight.  On the rz28s I have both wide and narrow, so let me know your preference.   ------------------------------  # Date: Wed, 19 Apr 2006 10:49:09 GMT 5 From: rdeininger@mindspringdot.com (Robert Deininger) 5 Subject: Re: How does a fix become a published patch? [ Message-ID: <rdeininger-1904060649070001@dialup-4.233.149.137.dial1.manchester1.level3.net>   3 In article <+PK+6abhVeVy@eisner.encompasserve.org>, , koehler@eisner.aspm.encompasserve.org wrote:  P >In article <op.s7wdb3iizgicya@hyrrokkin>, "Tom Linden" <tom@kednos.com> writes: >>  ) >> BTW,  is 7.3 the last release for VAX?  >  >    Sigh.   Yes.   , HP has NOT said so.  It is an open question.  I The 8.2 VAX release was "postponed".  Unofficially, it's probably fair to J say it's been cancelled at this point.  At this point, a VAX release wouldE probably be V8.3 or later.  Unless it turned out to be V7.4, to avoid " changing the major version number.  H Note that there is NOT a VAX V8.3 currently scheduled to ship with Alpha and Integrity V8.3  F VMS management does revisit the question of another VAX release fairlyI often.  There is virtually no customer demand.  That is of course subject 
 to change.  D There is some demand for feature enhancements within V7.3, mostly toD maintain cluster compatibility with Alpha.  These are addressed on a case-by-case basis.   A LOTS of customers do NOT want a new version, due to worries about E 3rd-party applications that would need to be re-qualified.  While VMS G Engineering could crank up the VAX build machinery, a whole lot of ISVs > (and end users) probably could not do so at a reasonable cost.  F The limited memory and limited address space available on VAXes prettyE much precludes full feature parity with Alpha and/or Integrity.  Some  stuff is just not practical.  L As always, VMS product managers want to hear about customer wants and needs.   ------------------------------  % Date: Wed, 19 Apr 2006 13:21:28 +0100 ' From: "Roger Fraser" <rfraser3@csc.com> 5 Subject: Re: How does a fix become a published patch? 1 Message-ID: <444629c1$1_1@glkas0286.greenlnk.net>   C "Robert Deininger" <rdeininger@mindspringdot.com> wrote in message  U news:rdeininger-1904060649070001@dialup-4.233.149.137.dial1.manchester1.level3.net... 5 > In article <+PK+6abhVeVy@eisner.encompasserve.org>, . > koehler@eisner.aspm.encompasserve.org wrote: > J >>In article <op.s7wdb3iizgicya@hyrrokkin>, "Tom Linden" <tom@kednos.com> 	 >>writes:  >>> * >>> BTW,  is 7.3 the last release for VAX? >> >>    Sigh.   Yes. > . > HP has NOT said so.  It is an open question. > K > The 8.2 VAX release was "postponed".  Unofficially, it's probably fair to L > say it's been cancelled at this point.  At this point, a VAX release wouldG > probably be V8.3 or later.  Unless it turned out to be V7.4, to avoid $ > changing the major version number. > J > Note that there is NOT a VAX V8.3 currently scheduled to ship with Alpha > and Integrity V8.3 > H > VMS management does revisit the question of another VAX release fairlyK > often.  There is virtually no customer demand.  That is of course subject  > to change. > F > There is some demand for feature enhancements within V7.3, mostly toF > maintain cluster compatibility with Alpha.  These are addressed on a > case-by-case basis. M This would definitely be useful for me! We have some applications which must  L stay on VAX as there is no method of re-building. However, they do not rely I on internal data structures and VAX upgrades are not (certainly have not   been since V4.7) a problem.      > C > LOTS of customers do NOT want a new version, due to worries about G > 3rd-party applications that would need to be re-qualified.  While VMS I > Engineering could crank up the VAX build machinery, a whole lot of ISVs @ > (and end users) probably could not do so at a reasonable cost.K But, surely, that could also apply to other architectures. Some sites have  J to re-validate their own code with EVERY upgrade (VAX, Alpha and maybe in  future Integrity)      > H > The limited memory and limited address space available on VAXes prettyG > much precludes full feature parity with Alpha and/or Integrity.  Some  > stuff is just not practical. > H > As always, VMS product managers want to hear about customer wants and  > needs.     Rog    ------------------------------    Date: 19 Apr 2006 07:44:42 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) 5 Subject: Re: How does a fix become a published patch? 3 Message-ID: <1NHtbCxzvFzI@eisner.encompasserve.org>    In article <rdeininger-1904060649070001@dialup-4.233.149.137.dial1.manchester1.level3.net>, rdeininger@mindspringdot.com (Robert Deininger) writes: 5 > In article <+PK+6abhVeVy@eisner.encompasserve.org>, . > koehler@eisner.aspm.encompasserve.org wrote: > Q >>In article <op.s7wdb3iizgicya@hyrrokkin>, "Tom Linden" <tom@kednos.com> writes:  >>> * >>> BTW,  is 7.3 the last release for VAX? >> >>    Sigh.   Yes. > . > HP has NOT said so.  It is an open question. >   G    Let me be clear:  7.3 is the last release that exists for VAX.  I do 1    not claim that will be a permanenet situation.    ------------------------------  % Date: Wed, 19 Apr 2006 06:41:20 -0700 # From: "Tom Linden" <tom@kednos.com> 5 Subject: Re: How does a fix become a published patch? ) Message-ID: <op.s79ca6o1zgicya@hyrrokkin>   6 On Wed, 19 Apr 2006 03:49:09 -0700, Robert Deininger  % <rdeininger@mindspringdot.com> wrote:   F > There is some demand for feature enhancements within V7.3, mostly toF > maintain cluster compatibility with Alpha.  These are addressed on a > case-by-case basis.   I When I posed the question I was (selfishly) thinking about what I would    like+ to see, and HBMM would be nice and TCPIP5.6    ------------------------------    Date: 19 Apr 2006 10:38:59 -0500. From: brooks@cuebid.zko.hp.nospam (Rob Brooks)5 Subject: Re: How does a fix become a published patch? , Message-ID: <oRQPM4cA$cd9@cuebid.zko.hp.com>  % "Tom Linden" <tom@kednos.com> writes: 8 > On Wed, 19 Apr 2006 03:49:09 -0700, Robert Deininger    G >> There is some demand for feature enhancements within V7.3, mostly to G >> maintain cluster compatibility with Alpha.  These are addressed on a  >> case-by-case basis. > K > When I posed the question I was (selfishly) thinking about what I would   2 > like to see, and HBMM would be nice and TCPIP5.6  D I can say with pretty much 100% confidence that you will absolutely % never, ever see HBMM for OpenVMS VAX.   M There are many technical reasons for this; the biggest hurdle is that OpenVMS # VAX cannot maintain master bitmaps.   K This goes back to the minicopy project, where VAXes can contribute to write I bitmaps that exist on Alpha systems, but VAXes themselves cannot have the 7 master copy, nor can they perform minicopy operations.     --    H Rob Brooks    VMS Engineering -- Exec Group     brooks!cuebid.zko.hp.com   ------------------------------  % Date: Wed, 19 Apr 2006 07:51:41 -0400 ) From: "Neil Rieck" <n.rieck@sympatico.ca> # Subject: new DST routines for C-RTL 9 Message-ID: <8vp1g.5670$wK1.284022@news20.bellglobal.com>    Folks,  H HP released a new OpenVMS patch two days ago (VMS732_ACRTL-V0300) which ) contains the following new functionality:   F The Energy Policy Act of 2005 will change the start and end times for I Daylight Savings time in the U.S.  starting in 2007.  Beginning in 2007,  K Daylight Savings time will start on the second Sunday in March (instead of  K the current first Sunday in April) and end on the first Sunday in November  K (instead of the current last Sunday in October).  In addition, there are a  G number of counties in Indiana that will be changing this year from the  7 Eastern U.S.  time zone to the Central U.S.  time zone.   
 Very cool.   Neil   ------------------------------    Date: 19 Apr 2006 07:47:12 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) ' Subject: Re: new DST routines for C-RTL 3 Message-ID: <nfqicZweUmTi@eisner.encompasserve.org>   e In article <8vp1g.5670$wK1.284022@news20.bellglobal.com>, "Neil Rieck" <n.rieck@sympatico.ca> writes:  > Folks, > J > HP released a new OpenVMS patch two days ago (VMS732_ACRTL-V0300) which + > contains the following new functionality:  > H > The Energy Policy Act of 2005 will change the start and end times for K > Daylight Savings time in the U.S.  starting in 2007.  Beginning in 2007,  M > Daylight Savings time will start on the second Sunday in March (instead of  M > the current first Sunday in April) and end on the first Sunday in November  M > (instead of the current last Sunday in October).  In addition, there are a  I > number of counties in Indiana that will be changing this year from the  9 > Eastern U.S.  time zone to the Central U.S.  time zone.  >  > Very cool. >  > Neil  H    And now all of us stuck on other releases will hope that our code was<    never updated to use gmtime() after it became functional.   ------------------------------  % Date: Wed, 19 Apr 2006 09:58:26 -0400  From: norm.raphael@metso.com' Subject: Re: new DST routines for C-RTL Q Message-ID: <OF152248AF.50455D90-ON85257155.004C67CC-85257155.004CC244@metso.com>   D "Neil Rieck" <n.rieck@sympatico.ca> wrote on 04/19/2006 07:51:41 AM:   > Folks, > I > HP released a new OpenVMS patch two days ago (VMS732_ACRTL-V0300) which + > contains the following new functionality:  > G > The Energy Policy Act of 2005 will change the start and end times for J > Daylight Savings time in the U.S.  starting in 2007.  Beginning in 2007,I > Daylight Savings time will start on the second Sunday in March (instead  ofC > the current first Sunday in April) and end on the first Sunday in  NovemberJ > (instead of the current last Sunday in October).  In addition, there are a H > number of counties in Indiana that will be changing this year from the9 > Eastern U.S.  time zone to the Central U.S.  time zone.  >  > Very cool.  E So...given that this is an external environment change that cannot be  ignored,G is there consideration to supplying the needed patch for those still on  Alpha  V7.3-1?  V7.3?  V7.2*? V6.2? 	 VAX V7.2? 	 VAX V6.2?  or earlier versions? >  > Neil >  >  >    ------------------------------    Date: 19 Apr 2006 11:51:26 +01006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)# Subject: Re: OT: Sparc not dead yet , Message-ID: <4446243e$1@news.langstoeger.at>  _ In article <1145364546.552083.297520@u72g2000cwu.googlegroups.com>, bob@instantwhip.com writes: < >sun did not hurt DEC ... DEC hurt DEC with no marketing and   yup, and SUN got the customers  ? >poor business and technology decisions ... and after comparing 8 >CERT counts with solaris and VMS, I do not know why any3 >bank or stock market anywhere would choose sun ...   K Oh, that is simple to answer: Because the applications are (still) there...   K If HP makes VMS visible/viable again, then the ISVs might offer them again. F But in the meantime, VMS is still losing ground (even in the remaining niche^Wkey markets). Sigh.   --   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: 19 Apr 2006 09:26:43 -0700- From: "Andrew" <andrew_harrison@symantec.com> # Subject: Re: OT: Sparc not dead yet C Message-ID: <1145464003.347223.157090@u72g2000cwu.googlegroups.com>    bob@instantwhip.com wrote:= > sun did not hurt DEC ... DEC hurt DEC with no marketing and @ > poor business and technology decisions ... and after comparing9 > CERT counts with solaris and VMS, I do not know why any 4 > bank or stock market anywhere would choose sun ...  @ Sun was giving Digital a pretty good mauling long before DigitalD imploaded. The engineering workstation and server market was a largeE market for DEC but long before Digital started to slide Sun had eased D them out. The arrival of the Sun E3000/4000/6000/10000 range exposedD the weakness of Digitals server range and a big chunk of that market also went Sun's way.  @ And as you know perfectly well there are lies damn lies and CERT counts.  regards  Andrew Harrison    ------------------------------    Date: 19 Apr 2006 10:09:21 -0700- From: "Doug Phillips" <dphill46@netscape.net> # Subject: Re: OT: Sparc not dead yet B Message-ID: <1145466561.450513.67630@v46g2000cwv.googlegroups.com>   Richard B. Gilbert wrote:  > JF Mezei wrote:  >  > > Dave Froble wrote: > > B > >>There you go again.  Has Sun issued a statement that they will. > >>eventually give up on SPARC that I missed? > >  > >  > > K > > No, they have not. But there has also been lots of speculation from the G > > press about Sun eventually going all 8086 once it scales to replace K > > Sparcs. HP has issued FUD about SPARC being abandonned eventually, just W > > like SUN has issued FUD about IA64 being stillborn and eventually to be abandonned.  > >  > C > And what is the maximum number of 8086 family CPUs that will work 2 > together in one box?   I've heard of up to four! > J > One of the good things about SPARC is that you can get up to 64 CPUs perJ > box.  Not to mention that a 440MHz UltraSPARC II will blow the doors off > a 440 MHz pentium   G I think SUN has so far played this well. SPARC is still being advanced, F but any type of realistic projection will put x86 ahead at some point.F HP chopped off the VMS customer base by killing the Alpha before thereG was even a replacement, much less an upgrade path. But then again, *nix = on whatever is an easier sell than VMS on anything, it seems.    ------------------------------  % Date: Wed, 19 Apr 2006 05:23:35 -0400 . From: JF Mezei <jfmezei.spamnot@vaxination.ca>E Subject: Re: Partitioned cluster question (reboot during lost quorum) - Message-ID: <44460193.24E758F7@vaxination.ca>    Michael Moroney wrote: >  Set it to a value that isH >        equal to the sum of the vote parameters of all cluster members,@ >        plus any votes that are contributed by the quorum disk.  G The HELP for EXPECTED_VOTES does not mention that EXPECTED_VOTES/2+1 is D the minimum number of vote needed before a node can proceed with the. boot. That is the real important portion here.  * OK, this may sound like a newbie question:  ; Once a cluster is up and running, does the SYSGEN parameter  EXPECTED_VOTES matter ? H What happens when 3 nodes in a cluster all have different EXPECTED_VOTESF SYSGEN  values ? Wouldn't the cluster software automatically calculate$ the right value for expected votes ?  > I just checked on 3 nodes: 2 have "5" as EXPECTED VOTE and the7 workstation as "3". So obviously I have to review that.         I was under the impression that   F 1-	once the cluster is up, the value of the QUORUM was calculated fromH the maximum number of votes that were seen since the cluster was formed.F (and each node  contributing a fixed number of votes read from SYSGEN.  H 2- But if you use cluster_config to add a new node in the cluster, a newE quorum value is dynamically calculated and the same value is used for 8 all nodes in the cluster irrespective of expected_votes.  M > That's used to adjust quorum downward when some nodes have already left the L > cluster _and are not coming back_, to prevent quorum freezes if more nodes > leave.  G I tried to use it once (SET CLU/EXPECTED) to revise the count downwards G in preparation for shutting down a node that would otherwise cause loss E of quorum. There would have been 1 node left and I knew it would have < been safe vecause the others ndoes were down. (as opposed toD disconnected from ethernet). VMS didn'T complain but didn't actuallyG change the quorum. So I ended up with a frozen workstation to which the A OPA0s of both other VAXes was being displayed, so I couldn't work G properly at the console level fo the other 2 vaxes because there was no E way to keep the remaining VAX unfrozen. I ended up having t up my PDA 9 with its serial port to do the >>> work on the two vaxes.   G If you can only use SET CLU/EXPECTED *after* a node has departed, it is B not very useful since once a node departs, if you lose quorum, you cannot issue that command :-)       B >  If you want to adjust quorum downward when a node is shut down,  > shut it down with REMOVE_NODE.  E Guess what REMOVE_NODE does ? it does a SET CLU/EXPECTED_VOTES (or at ? least, that is what I remember I saw). And in my case, when SET D CLU/EXPECTED didn't work, I did the REMOVE_NODE and the remining VAX" still froze due to lack of quorum.   ------------------------------  % Date: Wed, 19 Apr 2006 05:29:13 -0400 . From: JF Mezei <jfmezei.spamnot@vaxination.ca>E Subject: Re: Partitioned cluster question (reboot during lost quorum) - Message-ID: <444602E4.97131652@vaxination.ca>   / Phillip Helbig---remove CLOTHES to reply wrote: F > Quorum is half + 1 of VOTES or EXPECTED_VOTES, whichever is GREATER.  G Ahh,  this is put into simple and clear, short statement that says it a " lot better than the documentation.    H > Whenever you add a voting node to the cluster, EXPECTED_VOTES goes up,G > and stays up.  You can reduce it manually, if you plan to take a node  > out of the cluster  H Is it 100% correct to state that SET CLUSTER/EXPECTED will not allow you/ to bring the value below the SYSGEN parameter ?   F What is the actual logic that seems to implement a lower limit on $SET CLU/EXPECTED ?   ------------------------------  % Date: Wed, 19 Apr 2006 10:36:18 +0100 * From: "Richard Brodie" <R.Brodie@rl.ac.uk>E Subject: Re: Partitioned cluster question (reboot during lost quorum) 2 Message-ID: <e250ai$oho$1@blackmamba.itd.rl.ac.uk>  < "JF Mezei" <jfmezei.spamnot@vaxination.ca> wrote in message ' news:44460193.24E758F7@vaxination.ca...   I > If you can only use SET CLU/EXPECTED *after* a node has departed, it is D > not very useful since once a node departs, if you lose quorum, you > cannot issue that command :-)   H It depends how big your cluster is; ideally you don't want to start in aF position where removing a single node loses quorum. If you start with,E say 4 (1 vote) nodes with a quorum of 3 and shut one down it's useful ' to use it to recalculate a quorum of 2.   9 Sometimes you do have to go via IPC/AMDS/AvailMan though.   @ > Guess what REMOVE_NODE does ? it does a SET CLU/EXPECTED_VOTES/ > (or at least, that is what I remember I saw).   > No, it defines a logical for OPCCRASH; presumably that sends a* shutdown message forcing a recalculation.    ------------------------------  % Date: Wed, 19 Apr 2006 10:40:38 +0100 * From: "Richard Brodie" <R.Brodie@rl.ac.uk>E Subject: Re: Partitioned cluster question (reboot during lost quorum) 2 Message-ID: <e250im$olj$1@blackmamba.itd.rl.ac.uk>  < "JF Mezei" <jfmezei.spamnot@vaxination.ca> wrote in message ' news:444602E4.97131652@vaxination.ca...   J > Is it 100% correct to state that SET CLUSTER/EXPECTED will not allow you1 > to bring the value below the SYSGEN parameter ?    No.   H > What is the actual logic that seems to implement a lower limit on $SET > CLU/EXPECTED ?  @ You can't lower it below the number of votes currently present.    ------------------------------    Date: 19 Apr 2006 05:20:15 -0700 From: etmsreec@yahoo.co.ukE Subject: Re: Partitioned cluster question (reboot during lost quorum) C Message-ID: <1145449215.629422.111660@g10g2000cwb.googlegroups.com>    JF Mezei wrote:  > B > MC SYSGEN HELP SYS EXPECTED_VOTES mentions the MAXIMUM number of& > expected votes, not the minimum one. >    Hi JF,  F I suspect your confusion comes from the (mis)quote/(mis)reading of the help within SYSGEN.    The text actually says:   A "EXPECTED_VOTES specifies the maximum number of votes that may be ) present in a cluster at any given time. "   B It says maximum number of VOTES, _NOT_ EXPECTED_VOTES.  It's a key difference.   G That said, I know of at least one system manager that specifically sets A his clusters up with an EXPECTED_VOTES of 1 on every node so that E should the network links fail and should they need to reboot one node E all alone in the event of a failure that affects the other nodes they E can do so.  I see his line of reasoning, but don't necessarily agree.    Steve    ------------------------------    Date: 19 Apr 2006 05:45:31 -0700 From: etmsreec@yahoo.co.ukE Subject: Re: Partitioned cluster question (reboot during lost quorum) C Message-ID: <1145450731.907902.264670@i40g2000cwc.googlegroups.com>    > = >Somebody will CLUEXIT, but I'm not sure you can predict who.  >   D Isn't that down to the optimum sub-cluster?  If you've got two votesB one side that are potentially "the" cluster and one vote the other0 side, I would have expected the one vote to die.  " That's probably simplistic though.   ------------------------------    Date: 19 Apr 2006 07:40:04 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) E Subject: Re: Partitioned cluster question (reboot during lost quorum) 3 Message-ID: <YJZR6Dks9N+f@eisner.encompasserve.org>   \ In article <44453DE5.2759B34E@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes:3 > Say you have a 3 node cluster, node1, node2 node3  > < > Each node has 1 vote. You need 2 votes to maintain quorum. > G > You unplug the ethernet from node3. node1 and node2 still have quorum H > and happily chug along. Node3 realises it has lost quorum and freezes. > 4 > What happens if at this point, you reboot node3 ?  > F > Node3 would not see 1 or 2 and thing it was the first node rebooting% > after say a power failure. Right ?  H > Wouldn't NODE3 then form its instance of the cluster with its own vote > and a quorum of 1 ?   F    In this case either you determine that it's OK for node3 to operateE    on its own or you set EXPECTED_VOTES to at least 2 so that it will -    wait for quorum early in the boot process.   A > Or is there anything which would prevent node3 from rebooting ?   9    EXPECTED_VOTES can prevent the reboot from completing.    > E > Once node3 has rebooted, thinking it was alone in the cluster, when G > happens when you plug the ethernet back in and all of a sudden, node3 J > sees nodes 1 and 2 ?   Will nodes 1 and 2 succeed in convinding node3 toE > commit suicide ? Will node3 convince nodes1 and 2 to commit suicide : > (since it is a more recent incarnation of the cluster) ?  ?    Somebody will CLUEXIT, but I'm not sure you can predict who.   J > Or will what is essentially a partitioned cluster continue to exist as 2 > separate cluster instances ?  G    If you can create a partitioned cluster on a single Ethernet, please E    tell VMS Engineering about it.  They've done a lot of work to make     sure that can't happen.   > I > If node3 is i fact allowed to reboot all alone in its own little world, B > it seems to me that shops should have, as part of their operatorC > documentstions, strict guidelines not to reboot machines that are J > frozen. (sort of counter intuitive in a windows environment) since you'd; > want to first fix/investigate the problem of lost quorum.   F    If NODE3 is allowed to reboot in its own little world, you probably    set EXPECTED_VOTES wrong.   ------------------------------  # Date: Wed, 19 Apr 2006 14:36:28 GMT # From: Beach Runner <bob@nospam.com> E Subject: Re: Partitioned cluster question (reboot during lost quorum) = Message-ID: <MVr1g.174370$g47.104849@tornado.tampabay.rr.com>   ? The big thing to note is that help is not a substitute for the  F documentation.  It is very friendly, and does provide syntax and some 
 general help.   B Expected votes should be the total number of votes in the cluster.F My strange way of thinking is that one does modular math, divides the ? total number of votes or expected votes in half, than adds one. H However, to protect against people not setting expected votes correctly,E (very common) it adjusts expected votes based on the number of votes   found in the cluster.   I Set cluster/expected is used when you permanently remove a voting member   of the cluster.   I Now, VMS still can not protect a cheating or unaware system manager that  I sets expected votes too low. Supposed you have a 3 node SAN cluster, and  C the system manager wants to be able to boot either node by itself,  ? perhaps without a quorum disk.  So he sets expected votes to 1. G He can then boot any node by itself.  Once booted they see each other,  0 and quorum will protect against disk corruption.  F But, what if the network is down, one node boots, sees expected votes H are 1, and boots. Another node boots, sees expected votes are 1, and it E too boots.  At that point the system manager has corrupted any disks  / they access.  Hopefully they have good backups!   9 Correctly setting expected votes protects data integrity.    ------------------------------  # Date: Wed, 19 Apr 2006 17:17:48 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>E Subject: Re: Partitioned cluster question (reboot during lost quorum) 2 Message-ID: <0hu1g.6544$Y_2.1582@news.cpqcorp.net>  C    If you would prefer it, I and/or other engineers here at HP can  H certainly subcontract for your coding and your cluster management tasks L here -- HP does offer that and other similar consulting services, of course.   JF Mezei wrote: % > hoffman@xdelta.zko.hp.nospam wrote: E >>   If you would like to learn how to properly set the VOTES and the F >> EXPECTED_VOTES parameters, do please consider reviewing the text on% >> these settings in the OpenVMS FAQ.  >  > 3 >>   It will reboot, but will hang awaiting quorum.  > B > MC SYSGEN HELP SYS EXPECTED_VOTES mentions the MAXIMUM number of& > expected votes, not the minimum one.  E    The FAQ and the manuals also make the same point, and my previous  I posting around the "creative" settings was intended to dissuade you from  G any such "creative" -- read: wrong, dangerous, and potentially capable  B of triggering massive disk data corruptions -- parameter settings.    F > So after a power failure, if you power up your nodes one by one, youJ > mean to say that each node would just get get stuck near the "waiting toD > form or join cluster" until it finds enough votes lying around theC > ethernet to get to a quorum based on the maximum expected votes ?   F    Correct.  As part of this processing, connectivity is established,  and partitioning is avoided.    G > I was under the impression that when node boots and is all alone (eg: I > forms a new cluster), it proceeds with a quorum of its own votes/2 + 1, J > and then the quorum is dynamically adjusted updwards as other nodes join > the cluster.  I    "Being under the impression" can be a somewhat hazardous condition if  H the impression turns out to be incorrect, and the documentation and the ' FAQ can help ameliorate that situation.     I > So EXPECTED_VOTES / 2 + 1 defines the MINIMUM number of votes necessary G > for any node to proceed with booting beyond the awaiting to form/join  > cluster ?   F    No, EXPECTED_VOTES is the number of VOTES expected in the cluster, E whether from voting nodes or from a quorum disk.  (And a quorum disk  F that is not directly accessible through multiple paths and/or through $ multiple hosts is far from optimal.)  D    From this EXPECTED_VOTES value, the initial value of the cluster  quorum is determined.   D    Information on recommended VOTES settings for particular cluster G configurations has been discussed before here in the 'groups, and that  D discussion is certainly fodder for an update to the FAQ.  There are D certainly discussions of that topic within the existing manuals, as C well.  The discussions all tend to devolve into discussions of the  A connections available among the hosts and (when a quorum disk is  H configured) the disk storage, and the particular combination of systems G that should be kept operational for the longest as the availability of  . the particular cluster configuration degrades.  E > Perhaps the documentation should say it this way instead of talking 0 > about MAXIMUM number of votes to be expected.   E    Some folks thought they could re-jigger the old QUORUM setting --  G which is also where you appear headed here -- so the documentation was  I simplified and the configuration was simplified and the whole discussion  G was removed.  There was also some "hardening" around "creative" system  H parameter settings implemented, and work and discussions have continued ? to occur in the area of "hardening" the cluster configurations.   ? > I once tried to ajuset expected_votes downwards with DCL (SET F > CLUSTER/EXPECTED_VOTES and that didn't work. Does this mean that SETD > CLUSTER/EXPECTED_VOTES will not let you being the number below theG > SYSGEN parameter EXPECTED_VOTES ? but If that number does grow beyond I > it, you can then use the DCL command to bring it back down t the SYSGEN 	 > value ?   H    Clustering works very hard to prevent you from corrupting your data. D   The usual means to adjusting the quorum downward is to remove the F node, and to then issue the SET CLUSTER/EXPECTED_VOTES command, or to F use the SHUTDOWN.COM option, or to use the IPC (IPLC) quorum handler. G Clustering goes out of its way to avoid allowing the system manager to   corrupt the data.   H > (I tried to use this while in the process of reconfiguring the clusterE > when I knew that leaving NODE X up and running would not jeoperdize I > integrity but it didn'T have anough votes to keep on working so I ended 3 > up without service with the other node rebooted).   F    And clustering here specifically prevented you from your attempted E "creativity", as it was able to detect and prevent the configuration  F error.  Where "creative" settings really cause problems is when there H are communications problems -- in that case, incorrect system parameter D settings cannot be detected, and the disk data can become corrupted.  B    The OpenVMS documentation and the OpenVMS FAQ are available at:  '     <http://www.hp.com/go/openvms/doc/> '     <http://www.hp.com/go/openvms/faq/>    ------------------------------  # Date: Wed, 19 Apr 2006 17:29:59 GMT , From: Hoff Hoffman <hoff-remove-this@hp.com>E Subject: Re: Run Program with Shareable Image in Non-Default Location 2 Message-ID: <rsu1g.6549$qL2.6014@news.cpqcorp.net>   Craig Dedo wrote:   & > 1.  The command, "$ DEFINE shrimgnamF > ddcu:[dir]othershareimagefilename.exe", worked very well.  Thank youE > very much for this piece of advice.  I am assuming that it does not I > matter at what level or in which table (Process, Job, Group, System, or F > a specialized table) the SHAREIMAGENAME is defined.  The usual rulesD > for logical name resolution would apply.  Am I correct about this?  B    Non-privileged, yes.  Privileged, no.  Privileged or protected I shareables can only be redirected using trusted logical names defined in   trusted logical name tables.  C > 2.  Yes, the sharable image, SHARE_ALL_CTS_MESSAGES.EXE, contains E > site-specific messages built using the OpenVMS Message Utility.  It I > will be used by a lot of our executable programs in the Clinical Trials F > System (CTS).  I have already compiled the messages with the OpenVMSC > Message Utility and linked all of the message facilities into the E > shareable image before I compiled and linked the executable program F > that uses the shareable image.  Are there any other considerations I > need to be aware of?  G    What you specify in the MESSAGE/FILENAME argument is key.  Put just  B the filename there, not the device and directory path.  This then 4 generates am object that gets included in the image.  H    I tend to make heavy use of MESSAGE/SDL and SDL, as well.  If I have I full control over the message file, I also tend to use GNM.  MESSAGE/SDL  G is an undocumented part of OpenVMS, and both SDL and GNM are available  H on the Freeware.  Previous discussions include this, and I've certainly < shown the full sequence in any of my recent hints-and-kinks B presentations -- I'm presently updating that presentation for the  upcoming OpenVMS bootcamp, too.   C    Do also ensure you keep the message condition values consistent.   C > 3.  I just learned this morning that the program I am working on, I > REPORT_CREATOR_V2.EXE, is an installed image.  It is not installed with H > any privileges.  How does that affect what I need to do once I move it* > into Production, and need to install it?  @    Do see the Shareable Image Cookbook for details on the usual F considerations here -- if the program sections are correctly defined, A then installation isn't a big deal.  (If the executable image is  C installed protected or is installed privileged, or if any upstream  @ shareable image is installed protected, then the paranoid image H activation rules do apply, however.  And IIRC the cookbook does discuss  these topics.)    B > 4.  It would be very helpful if most or all of this information,F > including the answers to the above questions and the contents of theF > OpenVMS Shareable Image Cookbook, would be included in a revision ofC > the OpenVMS documentation set.  Putting this information into the E > OpenVMS Programming Concepts Manual, perhaps as a separate chapter, 3 > would make a lot of sense.  What do others think?   H    I wrote the cookbook because the material wasn't centrally available F elsewhere, and because the cookbook gave me a way to cover most of or G all of the core details as the shareable image questions arrived in my  = mailbox -- it is a direct outgrowth of customer and internal  G discussions.  I've submitted it (the cookbook) as an update to various  I of the manuals, but the cookbook itself doesn't fit centrally in any one  G OpenVMS manual, and there really isn't a task-oriented manual anywhere  " in the existing documentation set.   ------------------------------  + Date: Wed, 19 Apr 2006 09:52:56 +0000 (UTC) < From: gartmann@nonsense.immunbio.mpg.de (Christoph Gartmann)- Subject: Re: SoyMail & insufficient privilege ) Message-ID: <e2519o$h0t$2@news.BelWue.DE>   ] In article <124akjl5ngbqr27@corp.supernews.com>, Mark Daniel <mark.daniel@vsm.com.au> writes: 
 >chaos wrote: ) >> Hoping that this will bring more info.  >>  < >> <!-- ***** REPORTING MODULE:LINE IS REQUEST:124 ***** --> >>   >> It doesn't really for me :( >>  
 >> Dumitru > E >In the REQUEST.C module this is the location that is generating the  $ >report and the code looks like this > * >    /* if not subject to authorization */> >    if (!rdptr->RemoteUser[0]) ErrorExit (SS$_NOPRIV, FI_LI); > G >which tends to confirm the dignosis of my previous reply - soyMAIL is  A >not being subject to Apache server authorization.  Id say you'd  2 >definitely be seeing no username/password dialog.  D I had the same problem. There is an "[end]" statement in the defaultE SOYMAIL.CONF file. Any command you place after this [end] will not be D processed. As soon as I placed my "[logout-realm]" before this [end] things worked as expected.   Regards,    Christoph Gartmann    --  E  Max-Planck-Institut fuer      Phone   : +49-761-5108-464   Fax: -452   ImmunbiologieI  Postfach 1169                 Internet: gartmann@immunbio dot mpg dot de   D-79011  Freiburg, Germany 9                http://www.immunbio.mpg.de/home/menue.html    ------------------------------  % Date: Wed, 19 Apr 2006 16:04:39 +0200 / From: Paul Sture <paul.sture.nospam@hispeed.ch> 7 Subject: Re: Strange behavior of DEFINE with equal-sign : Message-ID: <5447c$44464378$50db5015$5699@news.hispeed.ch>   Bob Koehler wrote:y > In article <e16csb$t54$3@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:  > J >>Easy.  None of them are in the FORTRAN 77 character set.  Which reminds I >>me: is the limitation of node names to 6 characters in any way related  ; >>to a similar restriction on variable names in FORTRAN 77?  >> >  > G >    I've never seen such a restriction in any Fortran 77 compiler, and ! >    I'm not sure ANSI allows it.  > H >    6 character variable names were a common restriction in FORTRAN IV, >    a.k.a. Fortran 66.  >   D Also true of DIBOL on RT-11, except that global variable names were G restricted to 5 characters. IIRC the reason for this was that a dollar  8 sign was added to global variable names by the compiler.   ------------------------------  # Date: Wed, 19 Apr 2006 15:12:45 GMT 3 From: hammond@not@peek.ssr.hp.com (Charlie Hammond) 7 Subject: Re: Strange behavior of DEFINE with equal-sign 1 Message-ID: <Nrs1g.6531$bS2.533@news.cpqcorp.net>   J >>>... is the limitation of node names to 6 characters in any way related < >>>to a similar restriction on variable names in FORTRAN 77?   From memory...  E SCSNODE is restricted to 8 characters by the size of the system cell   that contains it.   C When DECnet Phase IV is in use SCSNODE is further restricted to the & 6 character DECnet Phase IV node name.  % i.e., nothing to do with FORTAN, etc.    --  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: 19 Apr 2006 09:46:30 -0700- From: "Doug Phillips" <dphill46@netscape.net> 7 Subject: Re: Strange behavior of DEFINE with equal-sign B Message-ID: <1145465190.650018.44020@t31g2000cwb.googlegroups.com>   Bob Koehler wrote:v > In article <7dd80f60604051001u247dc01ev41d7d69954691125@mail.gmail.com>, "Ken Robinson" <kenrbnsn@gmail.com> writes: > A > > With a little trial and error I discovered that the following ! > > characters do the same thing: 
 > > # > ? ] ^  > > J > > Now if anyone can figure out how these characters are related, perhaps6 > > the solution to this mystery will become apparent. > H >    The answer is:  DCL is full of strange and mysterious code that wasI >    a real pain to port to Alpha.  What is documented to work will work. G >    What is not documented to work is not guaranteed to fail.  In fact ) >    that's true for almost all software.  > 8 >    Try this (note the change from the previous posts): >  >    $a = "hello how are you >    $define test &a >    $show logical a > D >    In fact, what I've just shown you is a well known _unsupported_ >    "feature" of DCL.    E Guess I'm having a dumb day today, but I don't understand what you're  saying:    $a = "hello how are you  $define test &a  $show logical a 1 %SHOW-S-NOTRAN, no translation for logical name A   E What part of that isn't "supported"?  "&" substitution is certainly a B _supported_ "feature" and (at least my copy of) the FM explains itG quite well. Using SHOW LOGICAL to check the value of a symbol must have , been a typo , though. Did you mean to write:   $show logical test3    "TEST" = "hello how are you" (LNM$PROCESS_TABLE)    or:    $show symbol a   A = "hello how are you"    or:    $show symbol test = %DCL-W-UNDSYM, undefined symbol - check validity and spelling   	 or, what?    ------------------------------  % Date: Wed, 19 Apr 2006 08:31:21 +0200 ( From: "Rudolf Wingert" <win@fom.fgan.de>* Subject: What (WAS) wrong with my network?7 Message-ID: <002e01c6637a$df77fb20$994614ac@domina.fom>    Hello,  F Jeff your throughput increased from 9.9kB/s to 790kB/s and 90.1kB/s toG 1.1MB/s. This is still fare away from the possibilities. In our network F I did see a throughput within a range from 85%...95% of the line speedH (e.g. 11.2MB/s for a 100Base-TX and 1.2MB/s for a 10Base-T line). Try toG set the 100Base-TX ports to full duplex and test again. May be you will E get a better performance. Our experience is, if you have a manageable G switch, set both sides to 100Base-TX full duplex. Also it is important, . that you do this also within the console mode.E Also be aware, that the PIPELINE QUOTA is less than the default. This E quota defines the buffer size for outgoing packages. If you send data G form a 100Base-TX node to the 10Base-T node you need a parking place to H store all the packages till the receiver could fetch them. The switch isH this place. But normal switches do not have enough free buffer space, soH you will loose a lot of packages and the throughput slows down to 10% ofF the 10Base-T line. We did set the PIPELINE QUOTA to a low value (aboutC four packages = 6000Bytes). Then we did see the throughput reported  above.   Best regards R. Wingert    ------------------------------  % Date: Wed, 19 Apr 2006 05:14:07 -0700 ( From: Jeff Cameron <roktsci@comcast.net>. Subject: Re: What (WAS) wrong with my network?0 Message-ID: <C06B779F.1E5A4%roktsci@comcast.net>  J On 4/18/06 11:31 PM, in article 002e01c6637a$df77fb20$994614ac@domina.fom,) "Rudolf Wingert" <win@fom.fgan.de> wrote:    > Hello, > H > Jeff your throughput increased from 9.9kB/s to 790kB/s and 90.1kB/s toI > 1.1MB/s. This is still fare away from the possibilities. In our network H > I did see a throughput within a range from 85%...95% of the line speedJ > (e.g. 11.2MB/s for a 100Base-TX and 1.2MB/s for a 10Base-T line). Try toI > set the 100Base-TX ports to full duplex and test again. May be you will G > get a better performance. Our experience is, if you have a manageable I > switch, set both sides to 100Base-TX full duplex. Also it is important, 0 > that you do this also within the console mode.G > Also be aware, that the PIPELINE QUOTA is less than the default. This G > quota defines the buffer size for outgoing packages. If you send data I > form a 100Base-TX node to the 10Base-T node you need a parking place to J > store all the packages till the receiver could fetch them. The switch isJ > this place. But normal switches do not have enough free buffer space, soJ > you will loose a lot of packages and the throughput slows down to 10% ofH > the 10Base-T line. We did set the PIPELINE QUOTA to a low value (aboutE > four packages = 6000Bytes). Then we did see the throughput reported  > above. >  > Best regards R. Wingert  > G Thanks, But They were on Full Duplex and that is when I had the abysmal  performance.   ------------------------------  % Date: Wed, 19 Apr 2006 15:46:12 +0200 ( From: "Rudolf Wingert" <win@fom.fgan.de>. Subject: Re: What (WAS) wrong with my network?7 Message-ID: <000b01c663b7$9f004d80$994614ac@domina.fom>    Hello,F Jeff, I am right, you don't have a manageable switch? But it should beG an additional problem. If I did red your email right, you see different H throughput between the Alphas with 100Base-TX in condition of direction.F This looks strange. May be, there will be another problem too. Did you; check the parameters (PIPELINE QUOTA, RECIEVE BUFFERS,...)?  Best regards R. Wingert    ------------------------------    Date: 19 Apr 2006 18:46:01 +01006 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) Subject: [CSWB] Any news? , Message-ID: <44468569$1@news.langstoeger.at>   To the insiders   J When can we expect CSWB V1.7-13 (MOZILLA V1.7.13 is knocking on the door)?  M When can we expect FIREFOX _and_ THUNDERBIRD (and Chatzilla and HTML editing) / on VMS (MOZILLA is declared dead with V1.7.13)?   q http://developer.mozilla.org/devnews/index.php/2006/04/12/sunset-announcement-for-fxtb-10x-and-mozilla-suite-17x/    Many TIA   --   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   ------------------------------   End of INFO-VAX 2006.217 ************************