1 INFO-VAX	Fri, 07 Apr 2006	Volume 2006 : Issue 192       Contents:$ Re: "Mount Verify" messages in OPCOM$ Re: "Mount Verify" messages in OPCOM$ Re: "Mount Verify" messages in OPCOM$ Re: "Mount Verify" messages in OPCOM$ Re: "Mount Verify" messages in OPCOM6 Re: Can you use Avail Manager w/o openview or decamds?? Re: Can't get rid of trailing separator page with CMU LPR queue  Re: DECWindows not starting  Re: DECWindows not starting  Re: DECWindows not starting  Re: DECWindows not starting  Re: DECWindows not starting  Re: DECWindows not starting  Re: GLIB and GTK for OpenVMS ? monitoring I/O Channels  Re: monitoring I/O Channels  Now in stock DS15 systems ' Re: OpenVMS is the worlds best desktop! ' Re: OpenVMS is the worlds best desktop! ' Re: OpenVMS is the worlds best desktop! ' Re: OpenVMS is the worlds best desktop! ' Re: OpenVMS is the worlds best desktop! ' Re: OpenVMS is the worlds best desktop! ' Re: OpenVMS is the worlds best desktop! ' Re: OpenVMS is the worlds best desktop! ' Re: OpenVMS is the worlds best desktop! ' Re: OpenVMS is the worlds best desktop! . Re: Opinion: I was just trying to sell OpenVMS. Re: Opinion: I was just trying to sell OpenVMS. Re: Opinion: I was just trying to sell OpenVMS. Re: Opinion: I was just trying to sell OpenVMS. Re: Opinion: I was just trying to sell OpenVMS= Sign Up for a FREE Porting and Manageability Worskhop from HP  Using the Serial Terminal  Re: Using the Serial Terminal P VMS in healthcare - a user's perspective (was:Re: Opinion: I was just trying to  Re: VXT 2000+ monitor options G Re: Why are my interactive sessions limited to 6176 pages working set ? G Re: Why are my interactive sessions limited to 6176 pages working set ?   F ----------------------------------------------------------------------  % Date: Thu, 06 Apr 2006 14:14:40 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> - Subject: Re: "Mount Verify" messages in OPCOM , Message-ID: <44355A8E.26C9B5FB@teksavvy.com>   Rob Brooks wrote: K > In general, we suppress mount verification messages if the reason for the Q > mount verification can be "fixed" almost immediately.  The suppression work was H > done in response to user feedback that spurious, inconsequential mount5 > verification messages were troubling to operators,      F Shouldn't any mount verification message be investigated ? Seems to me3 that they are an indication of a problem somewhere.    ------------------------------   Date: 6 Apr 2006 15:26:44 -0500 . From: brooks@cuebid.zko.hp.nospam (Rob Brooks)- Subject: Re: "Mount Verify" messages in OPCOM , Message-ID: <L76P0zgQKX4P@cuebid.zko.hp.com>  / JF Mezei <jfmezei.spamnot@teksavvy.com> writes: 
 > I wrote:  L >> In general, we suppress mount verification messages if the reason for theN >> mount verification can be "fixed" almost immediately.  The suppression workM >> was done in response to user feedback that spurious, inconsequential mount 6 >> verification messages were troubling to operators,   H > Shouldn't any mount verification message be investigated ? Seems to me5 > that they are an indication of a problem somewhere.   H As I said, manual path switching of a mounted device will cause a mount J verification.  Depending on configuration, you may wind up with one mount I verification message per node in the cluster.  On a large cluster, that's F pretty noisy.  Most importanntly, it's not an indication of a problem.  I With complex SAN configurations involving multiple fabrics and switches,  N there can be enough SAN "noise" to periodically cause mount verifications that* are not an indication of failing hardware.  K Overloaded HSG80's will apply backpressure to the fibre channel port driver H if it's getting more I/O than it can handle  (the notorious "queue full"L problem).  That'll also cause mount verifications that are not indicative of8 failing hardware.  Underconfigured yes, but broken, no.   C Mount verification suppression can be disabled by setting either of $ MVSUPMSG_INTVL or MVSUPMSG_NUM to 0.  G If I was worried about failing hardware, I'd keep track of error counts  on the device(s) in question.    --    L Rob Brooks    VMS Engineering -- I/O Exec Group     brooks!cuebid.zko.hp.com   ------------------------------   Date: 6 Apr 2006 16:07:51 -0700 " From: "Jose Baars" <peut@peut.org>- Subject: Re: "Mount Verify" messages in OPCOM C Message-ID: <1144364871.815184.226360@i39g2000cwa.googlegroups.com>    Rob Brooks wrote:   I > If I was worried about failing hardware, I'd keep track of error counts  > on the device(s) in question.  >  > -- >   F >From a few experiences in the past few years : check the error countsB on your fibre channel switch as well. We had a few cases of faulty GBIC's in the switch and on F an HSG80, some of them causing these suppressed mount verifies, others causing @ a path switch to a path with working GBIC's ( thereby hiding the problem from VMS!).    ------------------------------   Date: 6 Apr 2006 19:33:39 -0500 . From: brooks@cuebid.zko.hp.nospam (Rob Brooks)- Subject: Re: "Mount Verify" messages in OPCOM , Message-ID: <YaRUxOPQ1cN5@cuebid.zko.hp.com>  $ "Jose Baars" <peut@peut.org> writes: > Rob Brooks wrote:  > J >> If I was worried about failing hardware, I'd keep track of error counts  >> on the device(s) in question.  G > From a few experiences in the past few years : check the error counts D > on your fibre channel switch as well. We had a few cases of faulty > GBIC's in the switch and on H > an HSG80, some of them causing these suppressed mount verifies, othersJ > causing a path switch to a path with working GBIC's ( thereby hiding the > problem from VMS!).   K If there was a path switch due to a failing path, there likely was an OPCOM M message from the multipath poller indicating that the fibre channel path from F which the various devices switched has gone bad.  The multipath pollerH can be disabled by setting MPDEV_POLLER = 0, but it should really always= be left on, for various reasons.  MPDEV_POLLER defaults to 1.    --    H Rob Brooks    VMS Engineering -- Exec Group     brooks!cuebid.zko.hp.com   ------------------------------  % Date: Thu, 06 Apr 2006 21:34:26 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>- Subject: Re: "Mount Verify" messages in OPCOM 6 Message-ID: <4435CFB2.4CF434D4@NeOaSrPtAhMlNiOnWk.net>   Rob Brooks wrote:  > 1 > JF Mezei <jfmezei.spamnot@teksavvy.com> writes:  > > I wrote: > N > >> In general, we suppress mount verification messages if the reason for theP > >> mount verification can be "fixed" almost immediately.  The suppression workO > >> was done in response to user feedback that spurious, inconsequential mount 7 > >> verification messages were troubling to operators,  > J > > Shouldn't any mount verification message be investigated ? Seems to me7 > > that they are an indication of a problem somewhere.  > I > As I said, manual path switching of a mounted device will cause a mount K > verification.  Depending on configuration, you may wind up with one mount K > verification message per node in the cluster.  On a large cluster, that's H > pretty noisy.  Most importanntly, it's not an indication of a problem. > J > With complex SAN configurations involving multiple fabrics and switches,P > there can be enough SAN "noise" to periodically cause mount verifications that, > are not an indication of failing hardware. > M > Overloaded HSG80's will apply backpressure to the fibre channel port driver J > if it's getting more I/O than it can handle  (the notorious "queue full"N > problem).  That'll also cause mount verifications that are not indicative of9 > failing hardware.  Underconfigured yes, but broken, no.   H We're seeing the same kind of OPCOM messages as Dave, but when I look onH the console of the HSG associated with the underlying volume I'm gettingG tremendous numbers of "command aborted" messages, much faster than 9600 > baud can accommodate - the screen scrolls constantly at times.  6 Any ideas where to start looking for a possible cause?  G Without intervention, the HSGs will eventually choke and either hang or F stop responding to the console forcing a manual reset. If the HSG80 is@ not hung, sometimes units go off-line (MVTMO). V8.8-4 of the HSG firmware, OVMS V7.3-2 + ECOs.   E > Mount verification suppression can be disabled by setting either of & > MVSUPMSG_INTVL or MVSUPMSG_NUM to 0. > I > If I was worried about failing hardware, I'd keep track of error counts  > on the device(s) in question.   @ We don't see eror counts increase until an HSG hangs or units go	 off-line.    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  & Unofficial OpenVMS Marketing Home Page! http://www.djesys.com/vms/market/   ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/   ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/   ------------------------------  % Date: Thu, 06 Apr 2006 21:27:46 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>? Subject: Re: Can you use Avail Manager w/o openview or decamds? 6 Message-ID: <4435CE22.C7AC1FFF@NeOaSrPtAhMlNiOnWk.net>   Ian Miller wrote:  > [snip]I > Availability Manager runs on VMS and Windows and requires somewhat more  > resources than DECAMDS.    That's Java for ya...    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  & Unofficial OpenVMS Marketing Home Page! http://www.djesys.com/vms/market/   ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/   ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/   ------------------------------   Date: 6 Apr 2006 16:06:10 -0700  From: mcbill20@yahoo.comH Subject: Re: Can't get rid of trailing separator page with CMU LPR queueC Message-ID: <1144364770.721041.219490@i39g2000cwa.googlegroups.com>   C Is this a documented part of the LPR protocol? Someone here at work G found the BLISS source and found where these four lines are appended to F the stream at the end of printing a file. We have another old VAX that@ is running Multinet instead and the trailing pages do not print.  D I am confused about what you refer to as the "printer command port."  C I don't see any documentation about this. We are running the latest C version of HP WEBjetadmin and there is no setting like this listed. @ Also, if it is a printer setting, then why would jobs from VAXen; running Multinet not print the trailer but the CMU ones do?    Bill   ------------------------------  % Date: Thu, 06 Apr 2006 16:42:28 -0700 # From: "Tom Linden" <tom@kednos.com> $ Subject: Re: DECWindows not starting) Message-ID: <op.s7l1g2ylzgicya@hyrrokkin>   H On Thu, 06 Apr 2006 09:05:48 -0700, FredK <fred.nospam@nospam.dec.com>   wrote:  7 > "Bill Gunshannon" <bill@cs.uofs.edu> wrote in message ' > news:49kpjqFolooiU1@individual.net... + >> In article <e13agc$4u$1@news.belwue.de>, A >> gartmann@nonsense.immunbio.mpg.de (Christoph Gartmann) writes: = >> > In article <op.s7lbjuj9zgicya@hyrrokkin>, "Tom Linden"    >> <tom@kednos.com> 	 > writes: J >> >>DS10L running 7.3-1, no graphics device,  didn't start on boot, tried, >>                        ^^^^^^^^^^^^^^^^^^
 >> >>manually  >> >> % >> >>NORNS> @sys$manager:decw$startup 8 >> >>%DECW$DEVICE-I-NODEVICE, no graphics devices found.G >> >>%DCL-I-SUPERSEDE, previous value of DECW$SERVER_TRANSPORT has been  >> >>superseded  >> >> 0 >> >>WINDOW_SYSTEM is set to 1, logicals look OK >> >>  >> >>What am I missing?  >> >H >> > Somehow your system doesn't find a graphics device. You could try a >>. >> It didn't find one because there isn't one.J >> That can't be the problem because I run DECwindows on headless machines >> with no problem.  >> > J > Well.  What exactly are you looking to "start"?  Are you saying that youG > can't do a SET/DISPLAY/TRANSPORT FOO and run X11 applications?  The   	 > message  > is just an informational.  > 4 > Try editing sys$manager:systartup_vms.com and add: > 0 > $ define/sys/exec DECW$IGNORE_WORKSTATION TRUE > = > That should prevent it from trying to start local graphics.  >   + It does , but still can't start DECW$SERVER   3 NORNS> define/sys/exec DECW$IGNORE_WORKSTATION TRUE   NORNS> @sys$manager:decw$startupD %DCL-I-SUPERSEDE, previous value of DECW$SERVER_TRANSPORT has been  
 superseded NORNS> sho display3 %DECW-W-OPENIN, error opening DECW$DISPLAY as input ( NORNS> pipe sho sys | sear sys$pipe decw' %SEARCH-I-NOMATCHES, no strings matched    ------------------------------  * Date: Fri, 7 Apr 2006 00:48:41 +0000 (UTC)7 From: moroney@world.std.spaamtrap.com (Michael Moroney) $ Subject: Re: DECWindows not starting( Message-ID: <e14ct9$3ti$1@pcls4.std.com>  % "Tom Linden" <tom@kednos.com> writes:   K >>> >>DS10L running 7.3-1, no graphics device,  didn't start on boot, tried - >>>                        ^^^^^^^^^^^^^^^^^^     , >It does , but still can't start DECW$SERVER  4 >NORNS> define/sys/exec DECW$IGNORE_WORKSTATION TRUE! >NORNS> @sys$manager:decw$startup E >%DCL-I-SUPERSEDE, previous value of DECW$SERVER_TRANSPORT has been    >superseded  >NORNS> sho display 4 >%DECW-W-OPENIN, error opening DECW$DISPLAY as input) >NORNS> pipe sho sys | sear sys$pipe decw ( >%SEARCH-I-NOMATCHES, no strings matched  F Ummm, if there is no graphics device, what do you expect a DECW$SERVER process to do?  - If you are trying to display on another node, I $ SET DISPLAY/CREATE/NODE=<othernode>/TRANSPORT=whatever is what you need  to do.   ------------------------------  $ Date: Thu, 6 Apr 2006 20:47:24 -0400* From: "FredK" <fred.nospam@nospam.dec.com>$ Subject: Re: DECWindows not starting* Message-ID: <4435b69d@usenet01.boi.hp.com>  . "Tom Linden" <tom@kednos.com> wrote in message# news:op.s7l1g2ylzgicya@hyrrokkin... H > On Thu, 06 Apr 2006 09:05:48 -0700, FredK <fred.nospam@nospam.dec.com> > wrote: > 9 > > "Bill Gunshannon" <bill@cs.uofs.edu> wrote in message ) > > news:49kpjqFolooiU1@individual.net... - > >> In article <e13agc$4u$1@news.belwue.de>, C > >> gartmann@nonsense.immunbio.mpg.de (Christoph Gartmann) writes: = > >> > In article <op.s7lbjuj9zgicya@hyrrokkin>, "Tom Linden"  > >> <tom@kednos.com>  > > writes: L > >> >>DS10L running 7.3-1, no graphics device,  didn't start on boot, tried. > >>                        ^^^^^^^^^^^^^^^^^^ > >> >>manually  > >> >> ' > >> >>NORNS> @sys$manager:decw$startup : > >> >>%DECW$DEVICE-I-NODEVICE, no graphics devices found.I > >> >>%DCL-I-SUPERSEDE, previous value of DECW$SERVER_TRANSPORT has been  > >> >>superseded  > >> >> 2 > >> >>WINDOW_SYSTEM is set to 1, logicals look OK > >> >>  > >> >>What am I missing?  > >> >J > >> > Somehow your system doesn't find a graphics device. You could try a > >>0 > >> It didn't find one because there isn't one.L > >> That can't be the problem because I run DECwindows on headless machines > >> with no problem.  > >> > > L > > Well.  What exactly are you looking to "start"?  Are you saying that youG > > can't do a SET/DISPLAY/TRANSPORT FOO and run X11 applications?  The  > > message  > > is just an informational.  > > 6 > > Try editing sys$manager:systartup_vms.com and add: > > 2 > > $ define/sys/exec DECW$IGNORE_WORKSTATION TRUE > > ? > > That should prevent it from trying to start local graphics.  > >  > - > It does , but still can't start DECW$SERVER  >   J Uh, a server needs hardware?  So why would you expect it to start a server if you have no graphics?  5 > NORNS> define/sys/exec DECW$IGNORE_WORKSTATION TRUE " > NORNS> @sys$manager:decw$startupD > %DCL-I-SUPERSEDE, previous value of DECW$SERVER_TRANSPORT has been > superseded > NORNS> sho display5 > %DECW-W-OPENIN, error opening DECW$DISPLAY as input * > NORNS> pipe sho sys | sear sys$pipe decw) > %SEARCH-I-NOMATCHES, no strings matched  >   C So exactly where do you think the display should be set by default?   J If there *is* a local display, then we can make that the default (at leastJ for things like DECterms that are started *inside* the session).  If thereJ isn't a display - what do you expect DECW$DISPLAY to point at?  If you logI into a workstation via TELNET - do you expect to have a DECW$DISPLAY set?   8 You need to do SET DISPLAY/CREATE/TRANSPORT=xxx/NODE=xxx   ------------------------------  $ Date: Thu, 6 Apr 2006 20:50:38 -0400* From: "FredK" <fred.nospam@nospam.dec.com>$ Subject: Re: DECWindows not starting, Message-ID: <4435b75e$1@usenet01.boi.hp.com>  5 "FredK" <fred.nospam@nospam.dec.com> wrote in message $ news:4435b69d@usenet01.boi.hp.com... >    > > >> > > > J > > > Well.  What exactly are you looking to "start"?  Are you saying that you I > > > can't do a SET/DISPLAY/TRANSPORT FOO and run X11 applications?  The 
 > > > message  > > > is just an informational.  > > > 8 > > > Try editing sys$manager:systartup_vms.com and add: > > > 4 > > > $ define/sys/exec DECW$IGNORE_WORKSTATION TRUE > > > A > > > That should prevent it from trying to start local graphics.  > > >  > > / > > It does , but still can't start DECW$SERVER  > >  > L > Uh, a server needs hardware?  So why would you expect it to start a server > if you have no graphics? >   G To follow up my own post as an aside, I do plan to check in the Virtual J Color Frame Buffer in a future release.  I can probably give Steve Hoffman> V7.3-1 and V8.2 and Itanium version of it for the freeware CD.  L With the virtual frame buffer you could start a server that has no hardware.   ------------------------------  # Date: Fri, 07 Apr 2006 01:01:03 GMT " From: reed@forge.{fe}.net (Brian )$ Subject: Re: DECWindows not starting, Message-ID: <84qgg3-1eu.ln1@server.iron.net>  > >>> > In article <op.s7lbjuj9zgicya@hyrrokkin>, "Tom Linden"   >>> <tom@kednos.com>
 >> writes:K >>> >>DS10L running 7.3-1, no graphics device,  didn't start on boot, tried - >>>                        ^^^^^^^^^^^^^^^^^^  >>> >>manually >>> >>& >>> >>NORNS> @sys$manager:decw$startup9 >>> >>%DECW$DEVICE-I-NODEVICE, no graphics devices found. H >>> >>%DCL-I-SUPERSEDE, previous value of DECW$SERVER_TRANSPORT has been >>> >>superseded >>> >>1 >>> >>WINDOW_SYSTEM is set to 1, logicals look OK  >>> >> >>> >>What am I missing? >>> > I >>> > Somehow your system doesn't find a graphics device. You could try a  >>> / >>> It didn't find one because there isn't one. K >>> That can't be the problem because I run DECwindows on headless machines  >>> with no problem. >>>  >>K >> Well.  What exactly are you looking to "start"?  Are you saying that you H >> can't do a SET/DISPLAY/TRANSPORT FOO and run X11 applications?  The  
 >> message >> is just an informational. >>5 >> Try editing sys$manager:systartup_vms.com and add:  >>1 >> $ define/sys/exec DECW$IGNORE_WORKSTATION TRUE  >>> >> That should prevent it from trying to start local graphics. >> > , >It does , but still can't start DECW$SERVER > 4 >NORNS> define/sys/exec DECW$IGNORE_WORKSTATION TRUE! >NORNS> @sys$manager:decw$startup E >%DCL-I-SUPERSEDE, previous value of DECW$SERVER_TRANSPORT has been    >superseded   T It's been many years, but isn't the server attached to the display?  Therefore there+ wouldn't be a server process at this point.   M What does sys$manager:decw$server_0_error.log say?  That yields many answers.    >NORNS> sho display 4 >%DECW-W-OPENIN, error opening DECW$DISPLAY as input  @ Have you set the display?  I didn't see the command anywhere.     S I may be missing exactly what you are trying to do, and the commands used to do it.    ------------------------------  % Date: Thu, 06 Apr 2006 18:57:59 -0700 # From: "Tom Linden" <tom@kednos.com> $ Subject: Re: DECWindows not starting) Message-ID: <op.s7l7qx0rzgicya@hyrrokkin>   H On Thu, 06 Apr 2006 17:47:24 -0700, FredK <fred.nospam@nospam.dec.com>   wrote:   > 0 > "Tom Linden" <tom@kednos.com> wrote in message% > news:op.s7l1g2ylzgicya@hyrrokkin... I >> On Thu, 06 Apr 2006 09:05:48 -0700, FredK <fred.nospam@nospam.dec.com> 	 >> wrote:  >>: >> > "Bill Gunshannon" <bill@cs.uofs.edu> wrote in message* >> > news:49kpjqFolooiU1@individual.net.... >> >> In article <e13agc$4u$1@news.belwue.de>,D >> >> gartmann@nonsense.immunbio.mpg.de (Christoph Gartmann) writes:> >> >> > In article <op.s7lbjuj9zgicya@hyrrokkin>, "Tom Linden" >> >> <tom@kednos.com> >> > writes:I >> >> >>DS10L running 7.3-1, no graphics device,  didn't start on boot,    >> tried/ >> >>                        ^^^^^^^^^^^^^^^^^^  >> >> >>manually >> >> >>( >> >> >>NORNS> @sys$manager:decw$startup; >> >> >>%DECW$DEVICE-I-NODEVICE, no graphics devices found. J >> >> >>%DCL-I-SUPERSEDE, previous value of DECW$SERVER_TRANSPORT has been >> >> >>superseded >> >> >>3 >> >> >>WINDOW_SYSTEM is set to 1, logicals look OK  >> >> >> >> >> >>What am I missing? >> >> > K >> >> > Somehow your system doesn't find a graphics device. You could try a  >> >> 1 >> >> It didn't find one because there isn't one. F >> >> That can't be the problem because I run DECwindows on headless   >> machines  >> >> with no problem. >> >>  >> >K >> > Well.  What exactly are you looking to "start"?  Are you saying that    >> youH >> > can't do a SET/DISPLAY/TRANSPORT FOO and run X11 applications?  The >> > message >> > is just an informational. >> >7 >> > Try editing sys$manager:systartup_vms.com and add:  >> >3 >> > $ define/sys/exec DECW$IGNORE_WORKSTATION TRUE  >> >@ >> > That should prevent it from trying to start local graphics. >> > >>. >> It does , but still can't start DECW$SERVER >> > G > Uh, a server needs hardware?  So why would you expect it to start a    > server > if you have no graphics? > 6 >> NORNS> define/sys/exec DECW$IGNORE_WORKSTATION TRUE# >> NORNS> @sys$manager:decw$startup E >> %DCL-I-SUPERSEDE, previous value of DECW$SERVER_TRANSPORT has been 
 >> superseded  >> NORNS> sho display 6 >> %DECW-W-OPENIN, error opening DECW$DISPLAY as input+ >> NORNS> pipe sho sys | sear sys$pipe decw * >> %SEARCH-I-NOMATCHES, no strings matched >> > E > So exactly where do you think the display should be set by default?  > H > If there *is* a local display, then we can make that the default (at   > least H > for things like DECterms that are started *inside* the session).  If   > there J > isn't a display - what do you expect DECW$DISPLAY to point at?  If you   > log K > into a workstation via TELNET - do you expect to have a DECW$DISPLAY set?  > : > You need to do SET DISPLAY/CREATE/TRANSPORT=xxx/NODE=xxx > L My misunderstanding,  when I tried that before the above define, I got the  	 the error K opening display message, didn't realize the that it was dynamic.  Problem    solved and thank you.   ------------------------------  # Date: Thu, 06 Apr 2006 20:00:58 GMT 7 From: John Malmberg <malmberg@dskwld.zko.hp.compaq.dec> ' Subject: Re: GLIB and GTK for OpenVMS ? 1 Message-ID: <_reZf.5995$Th.4267@news.cpqcorp.net>    John Malmberg wrote: > JOUKJ wrote: >  >>G >> hmm... you seem to get better results then me. Did you do something  H >> special? All png-tests fail on my machine and the X11-tests need the  >> Xrender-extention >  > F > A search of google reveals that this package includes a client side G > implementation of the Xrender extension for X-11 servers that do not  
 > have it. >  > http://www.fvwm.org/authors/ > C > I have not found a separate source for this code or if it can be  " > separated into it's own library.  G Having now looked at it, it does not have a client side implementation  C of Xrender, just some macros that allow the package to be compiled  1 without Xrender present at reduced functionality.   G > It appears that while the Cairo package does not require the Xrender  D > extension, it will not display anything with out it and the PANGO G > library as part of GTK+ is depending on Cairo doing the X-11 display.   H The Cairo package does require Xrender to be built to display for X-11, D but is not supposed to require that the X-11 Server support Xrender.  E According to the comments in the tests, for most of them if the test  G does not detect Xrender support in the server, the test is skipped and  : the tests for the fallback behavior have not been written.  I With the Xrender package present, I can get the Cairo package to pass 35  H out of the 60 tests attempted.  None of the failures seem to be OpenVMS 9 specific, and at least two of them are bugs in the tests.    -John ! malmberg@dskwld.zko.hp.compaq.dec  Personal Opinion Only    ------------------------------   Date: 6 Apr 2006 14:54:24 -0700 " From: dave.baxter@bannerhealth.com  Subject: monitoring I/O ChannelsC Message-ID: <1144360464.305834.278190@u72g2000cwu.googlegroups.com>   E    The system parameter CHANNELCNT determines how many "permanent I/O F Channels" are available on the system.     Is there a way to determine how many are actually in use??  @ At the moment I am monitoring the number of TCPIP device socketsD because when I get up close to the value of CHANNELCNT (9999) my app starts experiencing problems.   B The maximum number of Telnet Sockets (BG devices) is set at 32,767D (using sysconfig parameter ovms_unit_maximum).     However it cannot exceed CHANNELCNT.  G I have the parameter tcp_keepidle = 3600.    Is this causing the system  to hold onto idle connections ? E If I reduce this value (say to 300) will this relieve the pressure on  I/O channels ??   ( Any help/sugestions greatly appreciated.   Dave.   > p.s. TCPIP S/W is TCPIP services fro OpenVMS Vers 5.4 ECO 5 (+ patchkit),  VMS is 7.3-2   ------------------------------  # Date: Fri, 07 Apr 2006 04:46:52 GMT   From: John Santos <john@egh.com>$ Subject: Re: monitoring I/O Channels, Message-ID: <09mZf.20953$dU3.17956@trnddc01>  # dave.baxter@bannerhealth.com wrote: G >    The system parameter CHANNELCNT determines how many "permanent I/O H > Channels" are available on the system.     Is there a way to determine  > how many are actually in use??    E CHANNELCNT is per process, not per system.  It determines the maximum E number of channels any one process can have open.  I don't understand E the help text, which is confusing.  I think it determines the size of ( the channel table in the process header.   > B > At the moment I am monitoring the number of TCPIP device socketsF > because when I get up close to the value of CHANNELCNT (9999) my app > starts experiencing problems.   C On Alpha V7.3-2, maximum value of channel count is 65536, according   to SYSMAN PARAM SHOW CHANNELCNT.     > D > The maximum number of Telnet Sockets (BG devices) is set at 32,767F > (using sysconfig parameter ovms_unit_maximum).     However it cannot > exceed CHANNELCNT. >   A I don't know with TCPIP (aka UCX).  With TCPWARE, telnet sessions @ are limited by the process limit (one process per inbound telnetC connection), not by any limit to the telnet server's channel count. B There is a setting to allow multiple telnet listeners, but this isF to improve responsiveness.  Maybe it is different with TCPIP services.C There could be a logical name or setting that allow multiple telnet ; server processes, but I couldn't find it at a quick glance.       I > I have the parameter tcp_keepidle = 3600.    Is this causing the system ! > to hold onto idle connections ? G > If I reduce this value (say to 300) will this relieve the pressure on  > I/O channels ??  >     C Isn't that an hour?  Is this how long it keeps around idle sessions F (connected, but no input or output is occurring), or how long it holdsB on to a socket after a connection is broken?  If the latter, a fewH seconds to a minute should be enough.  If the former, than yeah, killingD idle processes will reduce the number of processes on the system, ifE there are any idle processes, but is that really what you want to do?     * > Any help/sugestions greatly appreciated. >  > Dave.  > @ > p.s. TCPIP S/W is TCPIP services fro OpenVMS Vers 5.4 ECO 5 (+ > patchkit),  VMS is 7.3-2 >   @ If you really have close to 9999 telnet sessions, and want more,@ then maybe your limit is process count or interactive job limit, or something like that.        --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  $ Date: Thu, 6 Apr 2006 22:01:33 -0400* From: "d b turner" <dbturner@islandco.com>" Subject: Now in stock DS15 systems7 Message-ID: <xHjZf.9083$Sf.8783@bignews6.bellsouth.net>   - We now have Refurbished DS15 systems in stock I We have a limited quantity so we must offer these as a "First come first   serve" offer8 And yes, they carry our same as new 1 Year warranty too!  & Call or email us at sales@islandco.com   Tel: USA 912-447-6622   # Thanks for all your support, y'all     ------------------------------  % Date: Thu, 06 Apr 2006 11:46:16 -0600 " From: GreyCloud <mist@cumulus.com>0 Subject: Re: OpenVMS is the worlds best desktop!: Message-ID: <cOGdnb-ctK93zqjZnZ2dnUVZ_sSdnZ2d@bresnan.com>   bob@instantwhip.com wrote: > Bill Gunshannon wrote: > . >>He has obviously never priced a VMS license. >  > 3 > how much does a hobbyist license cost these days?  >   H Actually, the license fees to use an operating system these days should @ be made obsolete.  I see it as nothing more than an obstacle to D embracing the o/s.  Too many CEOs wanting to get on the Great Gravy  Train money ride.    --   Where are we going?   And why am I in this handbasket?   ------------------------------   Date: 6 Apr 2006 18:21:41 GMT ( From: bill@cs.uofs.edu (Bill Gunshannon)0 Subject: Re: OpenVMS is the worlds best desktop!+ Message-ID: <49l4hlFpbchdU2@individual.net>   : In article <cOGdnb-ctK93zqjZnZ2dnUVZ_sSdnZ2d@bresnan.com>,% 	GreyCloud <mist@cumulus.com> writes:  > bob@instantwhip.com wrote: >> Bill Gunshannon wrote:  >>  / >>>He has obviously never priced a VMS license.  >>   >>  4 >> how much does a hobbyist license cost these days? >>   > J > Actually, the license fees to use an operating system these days should B > be made obsolete.  I see it as nothing more than an obstacle to F > embracing the o/s.  Too many CEOs wanting to get on the Great Gravy  > Train money ride.   B So, how would you propose paying VMS Engineering?  With the excessD profits HP gets for selling Wintel hardware?  (Sure sounds like more creeping Stallmanism to me!!)    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Thu, 06 Apr 2006 19:46:49 -0400 ' From: Dave Froble <davef@tsoft-inc.com> 0 Subject: Re: OpenVMS is the worlds best desktop!9 Message-ID: <XqGdnbt85MJFOqjZnZ2dneKdnZydnZ2d@libcom.com>    GreyCloud wrote: > bob@instantwhip.com wrote: >  >> Bill Gunshannon wrote:  >>0 >>> He has obviously never priced a VMS license. >> >> >>4 >> how much does a hobbyist license cost these days? >> > J > Actually, the license fees to use an operating system these days should B > be made obsolete.  I see it as nothing more than an obstacle to F > embracing the o/s.  Too many CEOs wanting to get on the Great Gravy  > Train money ride.  >   H Well I for one do not understand free software.  You don't get free gas H with that car you bought.  You don't get free food with that dinnerware  you bought.   I However, once one does it, the rest are at a disadvantage if they don't.  I   I just don't see how software people can pay their bills, buy gas, buy   food, ........   --  4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596> DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com 170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------   Date: 6 Apr 2006 16:48:12 -0700  From: paco.linux@gmail.com0 Subject: Re: OpenVMS is the worlds best desktop!C Message-ID: <1144367292.054966.137130@e56g2000cwe.googlegroups.com>   C How are the Solaris, AIX, IRIX, Microsoft :) OS engineering payed ? F In this times, seems that the software must be free, what do you thinkF about a free-vesion of VMS that everybody can enhance?, something like* solaris, you can save money on engineering   just kidding Paco Bill Gunshannon wrote:< > In article <cOGdnb-ctK93zqjZnZ2dnUVZ_sSdnZ2d@bresnan.com>,' > 	GreyCloud <mist@cumulus.com> writes:  > > bob@instantwhip.com wrote: > >> Bill Gunshannon wrote:  > >>1 > >>>He has obviously never priced a VMS license.  > >> > >>6 > >> how much does a hobbyist license cost these days? > >> > > K > > Actually, the license fees to use an operating system these days should C > > be made obsolete.  I see it as nothing more than an obstacle to G > > embracing the o/s.  Too many CEOs wanting to get on the Great Gravy  > > Train money ride.  > D > So, how would you propose paying VMS Engineering?  With the excessF > profits HP gets for selling Wintel hardware?  (Sure sounds like more > creeping Stallmanism to me!!)  >  > bill >  > --L > Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesF > bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. > University of Scranton   |@ > Scranton, Pennsylvania   |         #include <std.disclaimer.h>   ------------------------------   Date: 7 Apr 2006 00:22:36 GMT ( From: bill@cs.uofs.edu (Bill Gunshannon)0 Subject: Re: OpenVMS is the worlds best desktop!+ Message-ID: <49lpmcFp5d0qU1@individual.net>   C In article <1144367292.054966.137130@e56g2000cwe.googlegroups.com>,  	paco.linux@gmail.com writes:  > How are the Solaris,    A Sun makes money selling hardware.  They are not just a low margin @ Wintel box company.  Solaris just went free so we haven't reallyC had a chance to see the effect this will have on either the company  or the OS yet.  2 >                      AIX, IRIX, Microsoft :) OS    None of these are free.   F .                                                  engineering payed ?7 > In this times, seems that the software must be free,    G Why?  Are you another Stallman who thinks there is no inherent value in % work programmers do, so why pay them?   I >                                                       what do you think H > about a free-vesion of VMS that everybody can enhance?, something like, > solaris, you can save money on engineering  ? When it is the only product you have left, if you give it away,  why stay in business?    >  > just kidding   Icertainly hope so!    bill     --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Thu, 06 Apr 2006 21:02:25 -0400 6 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= <arne@vajhoej.dk>0 Subject: Re: OpenVMS is the worlds best desktop!. Message-ID: <pSiZf.40381$C85.26353@dukeread10>   Bill Gunshannon wrote:C > Sun makes money selling hardware.  They are not just a low margin  > Wintel box company.   4 Not now. But they may end as one within a few years.  " (low margin x86-64 boxes at least)   Arne   ------------------------------   Date: 7 Apr 2006 02:40:00 GMT ( From: bill@cs.uofs.edu (Bill Gunshannon)0 Subject: Re: OpenVMS is the worlds best desktop!+ Message-ID: <49m1o0Fpdbv7U1@individual.net>   / In article <e14hb2$f82$00$1@news.t-online.com>, + 	Michael Kraemer <M.Kraemer@gsi.de> writes:  > Bill Gunshannon schrieb:F >> In article <1144367292.054966.137130@e56g2000cwe.googlegroups.com>,  >> 	paco.linux@gmail.com writes: >>   >>>How are the Solaris,  >>   >>  D >> Sun makes money selling hardware.  They are not just a low marginC >> Wintel box company.  Solaris just went free so we haven't really F >> had a chance to see the effect this will have on either the company >> or the OS yet.  >>   >>  3 >>>                     AIX, IRIX, Microsoft :) OS   >>   >>   >> None of these are free. > ' > Depends on what you mean with "free". $ > Certainly they aren't open source,2 > but neither IBM nor Sun charges an arm and a leg > for using their base OS.< > E.g. last time I bought an IBM box AIX media came with it, > as does the license.  J Of course it does.  It was included in the price.  Try getting AIX without buying hardware.  A > Solaris is free for download and Sun is even bragging about it.   E I said that.  But it is a new concept for Sun and we haven't seen the H impact on quality of the OS yet.  And, they make money selling hardware.   >  >>  J >> Why?  Are you another Stallman who thinks there is no inherent value in( >> work programmers do, so why pay them? > - > Oh come on, I think we had this one before. = > Maybe you should shoot out your personal feud with that guy  > at some other place ?   D Nothing personal.  I just think that what programmers put into theirI work is worth at least as much as what other occupations put into theirs. K So, when people say software isn't worth anything and programmers shouldn't F be paid for their work I tend to take umbrage.  I will admit that mostF of the crap coming out of people's basements isn't worth anything, but1 then they're not professional programmers either.    >  >  >>  B >> When it is the only product you have left, if you give it away, >> why stay in business? > 5 > Well, if the business model of companies is to sell / > complete solutions rather than lines of code, J > giving away e.g. an OS for free (or at media costs) isn't necessarily a 5 > bad move, if it helps to sell hardware and support.   A How does that apply tp HP?  They killed Alpha.  They no longer do E hardware.  They sell other people's hardware.  At razor thin margins. @ VMS is about the only thing they have left with any real value.    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Thu, 06 Apr 2006 22:27:02 -0600 " From: GreyCloud <mist@cumulus.com>0 Subject: Re: OpenVMS is the worlds best desktop!: Message-ID: <KLCdnZmiuqyDd6jZnZ2dnUVZ_tSdnZ2d@bresnan.com>   Michael Kraemer wrote:F > In article <cOGdnb-ctK93zqjZnZ2dnUVZ_sSdnZ2d@bresnan.com>, GreyCloud > <mist@cumulus.com> writes: > J >>Actually, the license fees to use an operating system these days should B >>be made obsolete.  I see it as nothing more than an obstacle to F >>embracing the o/s.  Too many CEOs wanting to get on the Great Gravy  >>Train money ride.  >  > I > With its high license costs just for the base OS and the PAK weirdness  < > even for hobbyist use VMS is a singularity among the OSes.A > HP(UX), IBM, Sun etc charge little more than the media cost for L > their respective OS CDs. And they don't care if you run them as a hobbyistD > on their boxes at home. Sun probably even *wants* you to do so.     C I have Solaris 10 x86 installed on my other machine.  They are now  I allowing free downloads of their o/s and also SunStudio that has C, C++,  B and Fortran for the x86.  It's ok, but you have to make sure your H hardware is supported on their HCL.  Their installer leaves a lot to be  desired.@ Now if only IBM will make a free AIX download for the apple G5s. ( not likely )     --   Where are we going?   And why am I in this handbasket?   ------------------------------  % Date: Thu, 06 Apr 2006 22:30:19 -0600 " From: GreyCloud <mist@cumulus.com>0 Subject: Re: OpenVMS is the worlds best desktop!: Message-ID: <9IqdnUaAQ5BDd6jZnZ2dnUVZ_vGdnZ2d@bresnan.com>   Bill Gunshannon wrote:  < > In article <cOGdnb-ctK93zqjZnZ2dnUVZ_sSdnZ2d@bresnan.com>,' > 	GreyCloud <mist@cumulus.com> writes:  >  >>bob@instantwhip.com wrote: >> >>>Bill Gunshannon wrote:  >>>  >>> 0 >>>>He has obviously never priced a VMS license. >>>  >>> 4 >>>how much does a hobbyist license cost these days? >>>  >>J >>Actually, the license fees to use an operating system these days should B >>be made obsolete.  I see it as nothing more than an obstacle to F >>embracing the o/s.  Too many CEOs wanting to get on the Great Gravy  >>Train money ride.  >  > D > So, how would you propose paying VMS Engineering?  With the excessF > profits HP gets for selling Wintel hardware?  (Sure sounds like more > creeping Stallmanism to me!!)   B The same way that Sun and the others are doing it... thru support  service contracts.G I'm not exactly for Stalls methods, but at least a bit of common sense  F wouldn't hurt.  But as soon as a new trend starts, like free software I and such, then unfortunately the rate of innovation will be slown down a  F bit.  For example, I can't see paying over a $1000 for a good fortran : compiler as a hobbyist.  As a professional I could see it.     --   Where are we going?   And why am I in this handbasket?   ------------------------------  % Date: Thu, 06 Apr 2006 22:38:30 -0600 " From: GreyCloud <mist@cumulus.com>0 Subject: Re: OpenVMS is the worlds best desktop!0 Message-ID: <I8udnVWXCKpTcajZRVn-vg@bresnan.com>   Dave Froble wrote:   > GreyCloud wrote: >  >> bob@instantwhip.com wrote:  >> >>> Bill Gunshannon wrote: >>> 1 >>>> He has obviously never priced a VMS license.  >>>  >>>  >>>  >>> 5 >>> how much does a hobbyist license cost these days?  >>>  >>D >> Actually, the license fees to use an operating system these days J >> should be made obsolete.  I see it as nothing more than an obstacle to G >> embracing the o/s.  Too many CEOs wanting to get on the Great Gravy   >> Train money ride. >> > J > Well I for one do not understand free software.  You don't get free gas J > with that car you bought.  You don't get free food with that dinnerware 
 > you bought.  > K > However, once one does it, the rest are at a disadvantage if they don't.  J >  I just don't see how software people can pay their bills, buy gas, buy  > food, ........ >   E I remember that I used to get free glass mugs or dinner ware after a  - fill up.  Of course those days are long gone.   F What I probably should have said "exhorbitant license fees".  I don't D mind paying a reasonable amount, but I remember the gov. had to pay I around $5000 per year for the single machine vms license.  Back then DEC  F could get it, as everyone else was doing it.  Today, the situation is G different... M$ is there and the most dominant.  The cost of XP Pro is  F reasonable in comparison to $5k or even $1k.  I remember back in 1993 H where Solaris x86 was priced around $750, but retiring around that time H I wasn't about to blow that kind of money on a hobby, so I settled then - on just win3.1 which was a lot lower in cost.      --   Where are we going?   And why am I in this handbasket?   ------------------------------   Date: 6 Apr 2006 18:18:00 GMT ( From: bill@cs.uofs.edu (Bill Gunshannon)7 Subject: Re: Opinion: I was just trying to sell OpenVMS + Message-ID: <49l4aoFpbchdU1@individual.net>   , In article <4435592d$1@usenet01.boi.hp.com>, 	"Jilly" <jilly@hp.com> writes: L > How big is your 'around here'.  I know that just north of you where I am, O > the Guthrie Healthcare and Robert Packer Hospital folks still have plenty of  4 > VMS systems in use for various hospital functions.  > Yeah, and they still drive horse & buggies up there, too.  :-)  B Sorry, I meant in the big city where most of the patients and real hospitals are.  F If they are still using legacy systems up there it is much more likelyD due to the lack of funds to modernize anything than to any perceived value in staying with VMS.   billI (For those who may have thought I was joking in the first line, I'm not.  G There a number of thriving Amish communities in that area of rural PA.)    --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Thu, 06 Apr 2006 20:52:28 -0400 6 From: =?ISO-8859-1?Q?Arne_Vajh=F8j?= <arne@vajhoej.dk>7 Subject: Re: Opinion: I was just trying to sell OpenVMS . Message-ID: <4JiZf.40379$C85.17633@dukeread10>   GreyCloud wrote:J > Interesting bit of news.  I suppose one can fantasize and see HP almost G > handing VMS on a silver platter to Bill Gates in the not too distant  I > future.  Seeing that Vista is having problems getting out of the door,  , > VMS may well be what he needs to continue.  : Bill has done fine without VMS so far. He probably will in the future.   > And I can not see anything in VMS he will want. If he had then> Windows NT/2K/XP/2K3 would look like VMS now. They do not. TheB obvious conclusion is that the VMS style of an operating system is not what he want.    Arne   ------------------------------  % Date: Thu, 06 Apr 2006 21:19:54 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>7 Subject: Re: Opinion: I was just trying to sell OpenVMS 6 Message-ID: <4435CC4A.7C373A36@NeOaSrPtAhMlNiOnWk.net>   Bill Gunshannon wrote: > 2 > In article <BPidnZbKNJQhHKnZRVn-gQ@bresnan.com>,. >         GreyCloud <mist@cumulus.com> writes: > > David J. Dachtera wrote: > >  > >> GreyCloud wrote:  > >> > >>>David J. Dachtera wrote:  > >>>  > >>>>GreyCloud wrote: > >>>> > >>>> > >>>>>Alan Greig wrote: > >>>>>  > >>>>>  > >>>>>>GreyCloud wrote: > >>>>>> > >>>>>> > >>>>>> > >>>>>>>JF Mezei wrote:	 > >>>>>>> 	 > >>>>>>> 	 > >>>>>>> Q > >>>>>>>>NSK was never designed to be versatile. It is very specialised. VMS was P > >>>>>>>>not designed to be specialised. It is very versatile. And the owner is5 > >>>>>>>>purposefully not leveraging this potential. 	 > >>>>>>> 	 > >>>>>>> 	 > >>>>>>> > > >>>>>>>It is the purpose that we need to know about and why. > >>>>>> > >>>>>>M > >>>>>>A friend of mine who works for a major UK bank with NSK tells me that P > >>>>>>they play on moving off the platform because they do not have confidence( > >>>>>>in HP. IBM have lured them away. > >>>>>> > >>>>>  > >>>>>Interesting bit of news.  > >>>> > >>>>K > >>>>The news in my little corner of the healthcare world is that as sites J > >>>>begin to max out their 32-CPU GS1280s with huge Oracle databases andH > >>>>ever increasing transaction counts, and with last Alpha sale datesL > >>>>looming large on the horizon, a move away from VMS-capable hardware inL > >>>>search of 64+ CPU performance will lead to one inescapable conclusion. > >>>>L > >>>>I think I see now the method in their madness: in a world where takingN > >>>>responsibility for one's own actions is almost mortifying as speaking inK > >>>>public or admitting a mistake, VMS has been left to perish at Intel's H > >>>>hand, what with Alpha last dates just around the corner and ItanicL > >>>>foundering ever faster. Makes them appear to "come out smelling like a > >>>>rose". > >>>> > >>> A > >>>Elucidate further...  the situation is making my brain hurt.  > >> > >>M > >> Well, I tried to be clear, but let me say it another way: the opinion is L > >> held that to get to 64-CPU capability - which is deemd necessary, being5 > >> able to run VMS is not considered a requirement.  > >> > > G > > Thnx much.  It won't be long till M$ usurps that particular market.  > J > Where have you been hiding?  MS replaced VMS in all the hospitals around > here more than a decade ago.  @ If you looked around where I work, that's exactly what you would think...  H ...until you came up to the data center where we're about to switch fromB two GS160s(2 instances each) and three GS1280s to two GS1280s plusH another four GS1280s. Mostly 16-CPU, 64GB except one (guess which) which is 24-CPU, 96GB.   --   David J Dachtera dba DJE Systems  http://www.djesys.com/  & Unofficial OpenVMS Marketing Home Page! http://www.djesys.com/vms/market/   ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/   ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/   ------------------------------  % Date: Thu, 06 Apr 2006 21:25:04 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>7 Subject: Re: Opinion: I was just trying to sell OpenVMS 6 Message-ID: <4435CD80.B585FD61@NeOaSrPtAhMlNiOnWk.net>   Bill Gunshannon wrote: > . > In article <4435592d$1@usenet01.boi.hp.com>,( >         "Jilly" <jilly@hp.com> writes:M > > How big is your 'around here'.  I know that just north of you where I am, P > > the Guthrie Healthcare and Robert Packer Hospital folks still have plenty of6 > > VMS systems in use for various hospital functions. > @ > Yeah, and they still drive horse & buggies up there, too.  :-) > D > Sorry, I meant in the big city where most of the patients and real > hospitals are. > H > If they are still using legacy systems up there it is much more likelyF > due to the lack of funds to modernize anything than to any perceived > value in staying with VMS.  G Well, yes and no. This outfit is not penny-pinching about acquiring new H hardware, but they're not doing it optimally, either. The benefit is theH preservation of the existing investment and experience of the staff. TheG balance is shifting because demand for processing power is estimated to H outstrip the capacity of the final Alphas to be sold, and I64 has yet toH approach the starting gate on the Enterprise scale as compared to Alpha.A So, to achieve the needed processing power, it's very likely that H they'll chuck VMS in favor of something mainstream that runs on machines with 64+ CPUs.   --   David J Dachtera dba DJE Systems  http://www.djesys.com/  & Unofficial OpenVMS Marketing Home Page! http://www.djesys.com/vms/market/   ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/   ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/   ------------------------------  % Date: Thu, 06 Apr 2006 22:48:59 -0600 " From: GreyCloud <mist@cumulus.com>7 Subject: Re: Opinion: I was just trying to sell OpenVMS 0 Message-ID: <fJqdndSEYdGhcqjZRVn-rg@bresnan.com>   Arne Vajhj wrote:   > GreyCloud wrote: > D >> Interesting bit of news.  I suppose one can fantasize and see HP G >> almost handing VMS on a silver platter to Bill Gates in the not too  H >> distant future.  Seeing that Vista is having problems getting out of 7 >> the door, VMS may well be what he needs to continue.  >  > < > Bill has done fine without VMS so far. He probably will in
 > the future.  > @ > And I can not see anything in VMS he will want. If he had then@ > Windows NT/2K/XP/2K3 would look like VMS now. They do not. TheD > obvious conclusion is that the VMS style of an operating system is > not what he want.   @ Didn't M$ try in their first run of NT and get caught trying to  incorporate VMS code? E Also, David Cutler works for M$ now.  Davids slant is that he really  H doesn't like the Pentium line from Intel and favors the new AMD 64 line.) He likes a lot of registers to work with.      --   Where are we going?   And why am I in this handbasket?   ------------------------------   Date: 6 Apr 2006 13:43:16 -0700 % From: "Laura" <lmcgaughey@parsec.com> F Subject: Sign Up for a FREE Porting and Manageability Worskhop from HPC Message-ID: <1144356196.160582.277500@e56g2000cwe.googlegroups.com>   C Do you need help and advice for porting your source code to OpenVMS % I64? HP experts are here to help you!    WHEN: June 13 - 15, 2006  ' WHERE: HP's Nashua, NH OpenVMS Facility    WHAT: 7 =B7 Hands-on porting of your source code to OpenVMS I64 > =B7 Hands-on installation and customization of CA's management	 solutions ( =B7 Get help and advice from the experts =B7 See how easy it can be  G TO PARTICIPATE: Contact Russ Baldermann at 888-472-7732, extension 228, - or e-mail rbaldermann@parsec.com, to sign up.    ------------------------------   Date: 6 Apr 2006 16:48:12 -0700 / From: "doypapio@gmail.com" <doypapio@gmail.com> " Subject: Using the Serial TerminalB Message-ID: <1144367292.903055.81850@j33g2000cwa.googlegroups.com>  = Hi. Was wondering if anyone had information on using a serial D connection/terminal (say a VT)  to output certain values like from a0 digital weighing scale or a digital thermometer.  F My problem is on how the serial interface actually takes input and how3 will vms take in the input, say after a key stroke.   " Would appreciate any info on this.   Thanks in advance.   ------------------------------   Date: 6 Apr 2006 17:22:04 -0700  From: mcbill20@yahoo.com& Subject: Re: Using the Serial TerminalC Message-ID: <1144369324.941788.128070@z34g2000cwc.googlegroups.com>   F Hello. In order to help you we need more information. With VMS, serialF port communications are generally very easy. However, a lot depends onF the device you are trying to attach. I had one device a few years back4 that hardware-wise was absolute hell to get working.  D First, you will need to find out what the basic serial communicationF settings are for the device: speed, parity, stop bits, etc. Also, whatE kind of flow control (if any)? What hardware lines are required? This G could be just the basic 2,3 and 7 or can get more complex if the device G needs to see other lines held high or low and moreso if the device will @ be using hardware handshaking. Will the device be connected to aD terminal server or a serial port on the VMS box? What type of serial. port (that is, does it support modem control)?  D For software, what is required from the VMS side? Will the device beE polled or will it just send the raw data at certain time intervals or ' when a button on the device is pressed?   A Is the data the device sends just plain ASCII text or does it use E control characters or strange terminators? Does it require a response 	 from VMS?   A Also, I was a little confused about why you asked how VMS handles F keystrokes. Can you give the group a more specific picture of what you are trying to accomplish?    Bill doypapio@gmail.com wrote: ? > Hi. Was wondering if anyone had information on using a serial F > connection/terminal (say a VT)  to output certain values like from a2 > digital weighing scale or a digital thermometer. > H > My problem is on how the serial interface actually takes input and how5 > will vms take in the input, say after a key stroke.  > $ > Would appreciate any info on this. >  > Thanks in advance.   ------------------------------  % Date: Thu, 06 Apr 2006 17:42:09 -0400 % From: BRAD <bradhamilton@comcast.net> Y Subject: VMS in healthcare - a user's perspective (was:Re: Opinion: I was just trying to  * Message-ID: <44358B31.3080904@comcast.net>   Bill Gunshannon wrote:. > In article <4435592d$1@usenet01.boi.hp.com>,! > 	"Jilly" <jilly@hp.com> writes:  > L >>How big is your 'around here'.  I know that just north of you where I am, O >>the Guthrie Healthcare and Robert Packer Hospital folks still have plenty of  4 >>VMS systems in use for various hospital functions. >  > @ > Yeah, and they still drive horse & buggies up there, too.  :-) > D > Sorry, I meant in the big city where most of the patients and real > hospitals are.  F I've just started working in the healthcare field.  The hospital with G which our group practice is afilliated does have some VMS systems left  E (IDX Scheduler and IDX Rad), but the IDX system that I use (GPMS) is  F AIX-based.  The system is scheduled to be "upgraded" soon - while I'm G not privy to the details, the front end will be a GUI, rather than the  D current character-based interface, and the back end will remain AIX.  H At a medical software firm that I worked for several years ago, VMS was G perceived to be "on the way out" by our hospital/clinic customers, and  D our owner and sales force.  There were still a few larger hospitals I hanging on to their VMS systems, and the Canadian customers were solidly  G VMS, but I can tell you from my own (limited?) experience that VMS use  G is limited, considered legacy, and not considered for new applications.   H > If they are still using legacy systems up there it is much more likelyF > due to the lack of funds to modernize anything than to any perceived > value in staying with VMS.    H Hard to say what all the motivations are, but certainly "lack of funds" C is a contributing factor.  Those of us who know and use VMS in the  E healthcare field realize the value of its stability and scalability,  I (the AIX-based system that I use slows down at the beginning of the day,  C and just after lunch is done, as all users log on at once - I'd be  I willing to be you wouldn't see that on any well-tuned VMS-based system!)  G but the strategic outlook says further attrition in the installed base   is a certainty.    ------------------------------   Date: 6 Apr 2006 10:40:25 -0700 # From: "AndyZ" <afzillins@yahoo.com> & Subject: Re: VXT 2000+ monitor optionsC Message-ID: <1144345225.285214.302280@t31g2000cwb.googlegroups.com>   B There will be two unused BNC connectors.  Only need to connect RedA Green and Blue.  Monitors automatically determine that Sync is on  Green.  C See you figured out the gender issue.  I ordered both both male and 2 female versions because I wasn't sure at the time.   ------------------------------  $ Date: Thu, 6 Apr 2006 14:00:07 -0400* From: "Syltrem" <syltremzulu@videotron.ca>P Subject: Re: Why are my interactive sessions limited to 6176 pages working set ?0 Message-ID: <123aloop7vie88c@corp.supernews.com>  8 <etmsreec@yahoo.co.uk> a crit dans le message de news: 7 1144256218.690953.18510@i40g2000cwc.googlegroups.com... H > How is the batch queue set up?  Does that have working sets applied to > it? G > Have you tried the SET WORKING_SET command on the interactive jobs to # > see if this makes any difference?    Hi !  G I didn't know about the SET WORKING_SET command until I read your post.   
 That was it !   M There was a SET WORKING_SET in the SYS$SYLOGIN. It's been there for at least  H 15 years, and didn't matter for all but those processes I was trying to J tune. It was put there by the previous sysadmin back when we had VAX 8650  and 4100 for production.   Greetings !  Syltrem  .    ------------------------------  $ Date: Thu, 6 Apr 2006 14:11:13 -0400* From: "Syltrem" <syltremzulu@videotron.ca>P Subject: Re: Why are my interactive sessions limited to 6176 pages working set ?0 Message-ID: <123amd8t2dc17bb@corp.supernews.com>  I "Peter Weaver" <newsonly@weaverconsulting.ca> a crit dans le message de  4 news: SFYYf.2207$sh3.147570@news20.bellglobal.com...8 > "Syltrem" <syltremzulu@videotron.ca> wrote in message , > news:12351atgct0dg96@corp.supernews.com... >>... K >> This is an interactive session. *All* interactive sessions have a WS of  K >> 6176 pages in the WS and some do a lot of pagefaults, regardless of the   >> WS* values in UAF. J >> All other process types (detached, batch) with the same user accounts, % >> don't suffer from this limitation.  >>...  > B > Based on what I remember with previous postings you are running M > applications with an Oracle database. Today I looked at one of our systems  I > running Oracle and see that a lot of the users have rather low Working  M > Sets. Have you tried logging into an account and running some standard VMS  D > command that would require a large working set to see if there is I > something specific about Oracle. Maybe try editing a 20,000 block file  4 > with EVE and see what happens to your Working Set? >    Hi Peter  + There are 2 components when you use Oracle. K For each user there is the interactive process running the client software  J (could be Powerhouse, Gembase, or it could even be a Cobol or C program), J and there is the Oracle server process that handles the database requests  made by this client.  K It's the client process I had problems with. Clients running Powerhouse or  F Cobol programs accessing Oracle did not require as much memory as the M Gembase clients. I needed to tune Gembase clients and I couln't see why only  / those could not get all the memory they needed. L Problem was, only those Gembase clients did not have enough memory with the J SET WORKING_SET setting that existed in the SYS$LOGIN. The rest were fine.  7 I'll lower the WSDEF and get rid of the SET WORKING_SET    Syltrem    ------------------------------   End of INFO-VAX 2006.192 ************************