1 INFO-VAX	Mon, 08 Aug 2005	Volume 2005 : Issue 439       Contents:" Re: "the normal ordering process"?" Re: "the normal ordering process"?" Re: "the normal ordering process"?; Alpha Server 1200: first install VMS then upgrade firmware? ? Re: Alpha Server 1200: first install VMS then upgrade firmware? ? Re: Alpha Server 1200: first install VMS then upgrade firmware? ? Re: Alpha Server 1200: first install VMS then upgrade firmware? ; RE: HP Pearl - HP Says Over 5,000 Applications on Integrity ; Re: HP Pearl - HP Says Over 5,000 Applications on Integrity 7 Re: Killing a process that has allocated the tape drive 7 Re: Killing a process that has allocated the tape drive 7 Re: Killing a process that has allocated the tape drive 7 Re: Killing a process that has allocated the tape drive  Re: Newbie DCL Question F Re: OpenVMS under the umbrella - or umbrella is supported at the time?+ Re: strange terminal-characteristic problem + Re: strange terminal-characteristic problem + Re: strange terminal-characteristic problem + Re: strange terminal-characteristic problem + Re: strange terminal-characteristic problem + Re: strange terminal-characteristic problem + Re: strange terminal-characteristic problem + Re: strange terminal-characteristic problem + Re: strange terminal-characteristic problem 8 Re: Why is disk I/O operation rate .GT. direct I/O rate?8 Re: Why is disk I/O operation rate .GT. direct I/O rate?8 Re: Why is disk I/O operation rate .GT. direct I/O rate?8 Re: Why is disk I/O operation rate .GT. direct I/O rate?8 Re: Why is disk I/O operation rate .GT. direct I/O rate?8 Re: Why is disk I/O operation rate .GT. direct I/O rate?8 Re: Why is disk I/O operation rate .GT. direct I/O rate?  F ----------------------------------------------------------------------  # Date: Sun, 07 Aug 2005 23:20:00 GMT   From: John Santos <john@egh.com>+ Subject: Re: "the normal ordering process"? ) Message-ID: <AGwJe.10204$WQ.467@trnddc03>    Larry Kilgallen wrote:] > In article <BF1681F1.11C2A%roktsci@comcast.net>, Jeff Cameron <roktsci@comcast.net> writes:  > L >>On 8/2/05 1:57 PM, in article 00A47B0B.AE905A81@SendSpamHere.ORG, "VAXman-8 >>@SendSpamHere.ORG" <VAXman-  @SendSpamHere.ORG> wrote: >> >>J >>>OK... I need to break down and get myself a copy of the OpenVMS Itanium7 >>>source listings.  On one of the /DSPP pages it says:  >>> H >>>DSPP does not provide listings kits.  Partners may order them from HPI >>>through the normal ordering process.  Part numbers are provided below.  >>>  >>> H >>>OK, what *is* a normal ordering process with HP.  My experiences haveG >>>not left me to believe that there is such a beast.  Anybody that has A >>>obtained the Itanium source listings, please help me out here.  >>> H >>>I'm sincerely hoping this doesn't turn into the rack mounting kit de- >>>bacle of several months ago.  >>I >>When you are a registered OpenVMS partner with HP your "normal ordering J >>process" is spelled out in the contract, and is different in details for >>each partner.  >  >  > Nice theory. >  > H >>Are you a registered partner? If so your agreement should specify your1 >>process. If you are not then you must register.  >  > G > Presumably that includes DSPP, for which no such solution is provided # > (which is what Brian was saying).   8 Not exactly a followup to this post, but to this thread:  H My DSPP Alpha VMS SDK showed up on Friday.  (Naturally, I was out of theB office...)  It included yet another copy of Alpha VMS V8.2 and theD missing "OpenVMS Alpha Software Products Library Q2CY2005", i.e. the' June 2005 SPL.  Also the June 2005 ODL.   B According to the shipping labels, as best I can figure out, it wasE shipped UPS Ground from Nashua on Aug 4 and arrived here in Lexington # (about 25 miles away) the next day.          --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  # Date: Mon, 08 Aug 2005 00:52:53 GMT " From:   VAXman-  @SendSpamHere.ORG+ Subject: Re: "the normal ordering process"? 0 Message-ID: <00A47F1A.71A0325A@SendSpamHere.ORG>  L In article <AGwJe.10204$WQ.467@trnddc03>, John Santos <john@egh.com> writes: >Larry Kilgallen wrote: ^ >> In article <BF1681F1.11C2A%roktsci@comcast.net>, Jeff Cameron <roktsci@comcast.net> writes: >>  M >>>On 8/2/05 1:57 PM, in article 00A47B0B.AE905A81@SendSpamHere.ORG, "VAXman- 9 >>>@SendSpamHere.ORG" <VAXman-  @SendSpamHere.ORG> wrote:  >>>  >>> K >>>>OK... I need to break down and get myself a copy of the OpenVMS Itanium 8 >>>>source listings.  On one of the /DSPP pages it says: >>>>I >>>>DSPP does not provide listings kits.  Partners may order them from HP J >>>>through the normal ordering process.  Part numbers are provided below. >>>> >>>>I >>>>OK, what *is* a normal ordering process with HP.  My experiences have H >>>>not left me to believe that there is such a beast.  Anybody that hasB >>>>obtained the Itanium source listings, please help me out here. >>>>I >>>>I'm sincerely hoping this doesn't turn into the rack mounting kit de-   >>>>bacle of several months ago. >>> J >>>When you are a registered OpenVMS partner with HP your "normal orderingK >>>process" is spelled out in the contract, and is different in details for  >>>each partner.   >>   >>   >> Nice theory.  >>   >>  I >>>Are you a registered partner? If so your agreement should specify your 2 >>>process. If you are not then you must register. >>   >>  H >> Presumably that includes DSPP, for which no such solution is provided$ >> (which is what Brian was saying). > 9 >Not exactly a followup to this post, but to this thread:  > I >My DSPP Alpha VMS SDK showed up on Friday.  (Naturally, I was out of the C >office...)  It included yet another copy of Alpha VMS V8.2 and the E >missing "OpenVMS Alpha Software Products Library Q2CY2005", i.e. the ( >June 2005 SPL.  Also the June 2005 ODL. > C >According to the shipping labels, as best I can figure out, it was F >shipped UPS Ground from Nashua on Aug 4 and arrived here in Lexington$ >(about 25 miles away) the next day.  G Great news!  Even if Q3 is just around the corner at least I can expect 0 to see the Q2 CDs soon.  Better late than never!  H Now the only hurdle is the Itanium source listings.  I have some contactG numbers and names but I'm still not at easy with the lack of any update G service.  I simply can't afford to outlay $2284 for each release of VMS G on Itanium.  Anyway, as soon as I have waded my way through this septic < quagmire of obtaining these listings, I'll report back here. --  K VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM              5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------  $ Date: Sun, 7 Aug 2005 23:03:39 -0400- From: William Webb <william.w.webb@gmail.com> + Subject: Re: "the normal ordering process"? 7 Message-ID: <8660a3a1050807200330e685d7@mail.gmail.com>   E On 8/7/05, VAXman-@sendspamhere.org <VAXman-@sendspamhere.org> wrote: L > In article <AGwJe.10204$WQ.467@trnddc03>, John Santos <john@egh.com> writ= es:  > >Larry Kilgallen wrote: L > >> In article <BF1681F1.11C2A%roktsci@comcast.net>, Jeff Cameron <roktsci= @comcast.net> writes:  > >>L > >>>On 8/2/05 1:57 PM, in article 00A47B0B.AE905A81@SendSpamHere.ORG, "VAX= man-; > >>>@SendSpamHere.ORG" <VAXman-  @SendSpamHere.ORG> wrote:  > >>>  > >>> L > >>>>OK... I need to break down and get myself a copy of the OpenVMS Itani= um: > >>>>source listings.  On one of the /DSPP pages it says: > >>>>K > >>>>DSPP does not provide listings kits.  Partners may order them from HP L > >>>>through the normal ordering process.  Part numbers are provided below= .  > >>>> > >>>>K > >>>>OK, what *is* a normal ordering process with HP.  My experiences have J > >>>>not left me to believe that there is such a beast.  Anybody that hasD > >>>>obtained the Itanium source listings, please help me out here. > >>>>K > >>>>I'm sincerely hoping this doesn't turn into the rack mounting kit de- " > >>>>bacle of several months ago. > >>> L > >>>When you are a registered OpenVMS partner with HP your "normal orderin= g L > >>>process" is spelled out in the contract, and is different in details f= or > >>>each partner. > >> > >> > >> Nice theory.  > >> > >>K > >>>Are you a registered partner? If so your agreement should specify your 4 > >>>process. If you are not then you must register. > >> > >>J > >> Presumably that includes DSPP, for which no such solution is provided& > >> (which is what Brian was saying). > > ; > >Not exactly a followup to this post, but to this thread:  > > K > >My DSPP Alpha VMS SDK showed up on Friday.  (Naturally, I was out of the E > >office...)  It included yet another copy of Alpha VMS V8.2 and the G > >missing "OpenVMS Alpha Software Products Library Q2CY2005", i.e. the * > >June 2005 SPL.  Also the June 2005 ODL. > > E > >According to the shipping labels, as best I can figure out, it was H > >shipped UPS Ground from Nashua on Aug 4 and arrived here in Lexington& > >(about 25 miles away) the next day. >=20I > Great news!  Even if Q3 is just around the corner at least I can expect 2 > to see the Q2 CDs soon.  Better late than never! >=20J > Now the only hurdle is the Itanium source listings.  I have some contactI > numbers and names but I'm still not at easy with the lack of any update I > service.  I simply can't afford to outlay $2284 for each release of VMS I > on Itanium.  Anyway, as soon as I have waded my way through this septic > > quagmire of obtaining these listings, I'll report back here. > --L > VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)C= OM >=206 >   "Well my son, life is like a beanstalk, isn't it?" >=20  $ We got our 3Q2005 VAX SPD last week.   ?    WWWebb --=20 C NOTE: This email address is only used for noncommerical VMS-related  correspondence. C All unsolicited commercial email will be deemed to be a request for 8 services pursuant to the terms and conditions located at# http://bellsouthpwp.net/w/e/webbww/    ------------------------------  * Date: Sun, 7 Aug 2005 18:32:17 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)D Subject: Alpha Server 1200: first install VMS then upgrade firmware?$ Message-ID: <dd5k3h$ji9$3@online.de>  D I want to do a fresh install of VMS 7.3-2 on an Alpha Server 1200.  I However, after booting the installation CD I am warned that I don't have  F the recommended firmware version, but rather that my version is older H than desired (but apparently still usable).  Since this is a standalone H machine and I don't have another system disk I can swap into it quickly ? (nor a freeware CD which is new enough), I'm thinking of first  > installing VMS then, as soon as everything is up and running, I downloading the latest firmware to the system disk then rebooting into a   firmware upgrade.   H Is there any reason not to do this?  Is there any reason to upgrade the  firmware before installing VMS?    ------------------------------   Date: 7 Aug 2005 20:50:00 +0100 6 From: peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER)H Subject: Re: Alpha Server 1200: first install VMS then upgrade firmware?, Message-ID: <42f673f8$1@news.langstoeger.at>  w In article <dd5k3h$ji9$3@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: E >I want to do a fresh install of VMS 7.3-2 on an Alpha Server 1200.   J >However, after booting the installation CD I am warned that I don't have G >the recommended firmware version, but rather that my version is older  I >than desired (but apparently still usable).  Since this is a standalone  I >machine and I don't have another system disk I can swap into it quickly  @ >(nor a freeware CD which is new enough), I'm thinking of first ? >installing VMS then, as soon as everything is up and running,  J >downloading the latest firmware to the system disk then rebooting into a  >firmware upgrade.  	 Go ahead.   I >Is there any reason not to do this?  Is there any reason to upgrade the    >firmware before installing VMS?  G Iff the firmware version is below MINIMUM firmware version for VMS, you I can't install VMS (only an older version of it). Iff the firmware version K is below RECOMMENDED firmware version, you can do what you planned to do...   M btw: Alpha firmware is not on the VMS freeware CD (or why was it mentioned ?) E Newest Alpha Firmware can be downloaded as an CD image which you burn K (eg. on a PC) and do the firmware upgrade even before the VMS installation. @ Unfortunately, the 1200 is too old for the latest firmware CD...  A 	ftp://ftp.digital.com/pub/DEC/Alpha/firmware/archive/as1200.html    --   Peter "EPLAN" LANGSTOEGER % Network and OpenVMS system specialist  E-mail  peter@langstoeger.atF A-1030 VIENNA  AUSTRIA              I'm not a pessimist, I'm a realist   ------------------------------  * Date: Sun, 7 Aug 2005 19:50:35 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)H Subject: Re: Alpha Server 1200: first install VMS then upgrade firmware?$ Message-ID: <dd5omb$ji9$8@online.de>  w In article <dd5k3h$ji9$3@online.de>, helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:   F > I want to do a fresh install of VMS 7.3-2 on an Alpha Server 1200.  K > However, after booting the installation CD I am warned that I don't have  H > the recommended firmware version, but rather that my version is older J > than desired (but apparently still usable).  Since this is a standalone J > machine and I don't have another system disk I can swap into it quickly A > (nor a freeware CD which is new enough), I'm thinking of first    # Should be "firmware CD", of course.    ------------------------------  * Date: Sun, 7 Aug 2005 19:51:37 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)H Subject: Re: Alpha Server 1200: first install VMS then upgrade firmware?$ Message-ID: <dd5oo8$ji9$9@online.de>  H In article <42f673f8$1@news.langstoeger.at>, peter@langstoeger.at (Peter 'EPLAN' LANGSTOEGER) writes:    O > btw: Alpha firmware is not on the VMS freeware CD (or why was it mentioned ?) G > Newest Alpha Firmware can be downloaded as an CD image which you burn M > (eg. on a PC) and do the firmware upgrade even before the VMS installation. B > Unfortunately, the 1200 is too old for the latest firmware CD...  I What I normally do (when I have a working system disk) is download it to  H the system disk and boot from the system disk with flags appropriate to  a firmware upgrade.    ------------------------------  $ Date: Sun, 7 Aug 2005 15:45:39 -0400' From: "Main, Kerry" <Kerry.Main@hp.com> D Subject: RE: HP Pearl - HP Says Over 5,000 Applications on IntegrityR Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB6B1C17@tayexc19.americas.cpqcorp.net>   > -----Original Message------ > From: John Smith [mailto:a@nonymous.com]=20  > Sent: August 6, 2005 9:35 PM > To: Info-VAX@Mvb.Saic.Com F > Subject: Re: HP Pearl - HP Says Over 5,000 Applications on Integrity >=20 > Main, Kerry wrote: > >> -----Original Message----- 4 > >> From: Bill Todd [mailto:billtodd@metrocast.net]! > >> Sent: August 6, 2005 4:06 AM  > >> To: Info-VAX@Mvb.Saic.ComB > >> Subject: Re: HP Pearl - HP Says Over 5,000 Applications on=20 > Integrity  > >>' > >> susan_skonetski@hotmail.com wrote:   > >>> -----Original Message----- > >>> From: Skonetski, Susan+ > >>> Sent: Friday, August 05, 2005 6:23 PM * > >>> To: VMS-SIG@LISTSERV.ENCOMPASSUS.ORGF > >>> Subject: HP Pearl - HP Says Over 5,000 Applications on Integrity > >>> Servers - External Use > >>>  > >>>   	 [snip ..]    > > H > > Given that OpenVMS on Integrity has only "officially" been available> > > since January timeframe, I would say that the number of=20 > OpenVMS ports G > > (planned and available) has been going very well. Keep in mind that G > > some ISV's and Cust's prefer to start with official code - not beta 
 > > releases.  > >  > > Check out the following:@ > > http://h71000.www7.hp.com/solutions/matrix/i64partner_A.html3 > > (click on each letter for different vendor's..)  >=20 >=20? > How many of these VMS apps are from companies that used to=20  > have VMS apps inG > the past and have come back to VMS because of the "industry standard"  > processor it now runs on?  >=209 > Think Sybase as an example that you might care to cite.  >=20  G While I would love to see Sybase back on OpenVMS, the fact is that from H an enterprise perspective, even with some of its latest features, SybaseG ASE is struggling against Oracle, SQL Server and the likes of MySQL and  other freeware DB's.  
 Reference:H http://www.it-director.com/article.php?articleid=3D12797 (July 08, 2005)H "Given that Sybase is no longer regarded as one of the 800lb gorillas inF the database market, it is an ongoing surprise as to just how good ASE is."  B "With comparable or better performance than the big three databaseG vendors and a lower cost of ownership, you can see why Sybase remains a @ cause of surprise: why is it that more companies do not think of implementing ASE? Beats me."   Regards   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------  % Date: Sun, 07 Aug 2005 19:28:26 -0700 # From: "Tom Linden" <tom@kednos.com> D Subject: Re: HP Pearl - HP Says Over 5,000 Applications on Integrity( Message-ID: <opsu53tokpzgicya@hyrrokkin>  I On Sun, 7 Aug 2005 15:45:39 -0400, Main, Kerry <Kerry.Main@hp.com> wrote:   I > While I would love to see Sybase back on OpenVMS, the fact is that from J > an enterprise perspective, even with some of its latest features, SybaseI > ASE is struggling against Oracle, SQL Server and the likes of MySQL and  > other freeware DB's.  H Well, I am aware of one wall street powerhouse that was lost as a result therof.    ------------------------------  % Date: Sun, 07 Aug 2005 14:31:32 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> @ Subject: Re: Killing a process that has allocated the tape drive, Message-ID: <42F6537C.1D21EE46@teksavvy.com>   > MUSTAFA ATAKAN wrote: N > deallocate on dismount, mounted, file-oriented device, available to cluster,   > $ stop /id=20602CF3   6 > SYSTEM TILOT3 SYSTEM 20602CF3 TNA346: (disconnected)    H > What is the reason that i can not stop the above process? Is there anyL > command like "kill -9" in *nix environments? Is there any way to stop this; > process without rebooting or waiting some timeout period?     H File oriented device probably means that the user  didn't mount the tapeF /foreign, so the first operation he did to it would have generated theN creation of a tape ACP process (for instance MUAACP). Such processes are peskyL little buggers in that they refuse to let go of the tape and go RWAST if you try to delete them.   I RWAST is the VMS's satan. It is dreaded, disliked, and yet, the engineers  refise to remove it :-)   F The reason it is there is simple: with the process having issued an IOG request, the VMS system will insist on the process remaining there with I whatever allocated buffers the IO request point to so that if/when the IO L completes, it will be writing to a location in memory that isn't being uisedL by some other process which used that memory location if you had succesfully
 killed it.    N if you SHOW SYSTEM/proc=*ACP  you will probably see one ACP process related to	 the tape.    ------------------------------  # Date: Sun, 07 Aug 2005 19:12:20 GMT ( From: Alan Greig <greigaln@netscape.net>@ Subject: Re: Killing a process that has allocated the tape drive< Message-ID: <o2tJe.20323$ia4.1082@fe1.news.blueyonder.co.uk>   JF Mezei wrote:    > K > RWAST is the VMS's satan. It is dreaded, disliked, and yet, the engineers  > refise to remove it :-)   F Anywhere I've seen this problem in the past (and found no easy way to B alter procedures to prevent it) it has been possible to obtain an G unsupported utility from DEC/Compaq/HP specific to each version of VMS  I which can attempt to free the drive. You can also go in and zap the bits  * yourself if you feel really, really lucky.    H > The reason it is there is simple: with the process having issued an IOI > request, the VMS system will insist on the process remaining there with K > whatever allocated buffers the IO request point to so that if/when the IO N > completes, it will be writing to a location in memory that isn't being uisedN > by some other process which used that memory location if you had succesfully > killed it. >  > P > if you SHOW SYSTEM/proc=*ACP  you will probably see one ACP process related to > the tape.    --  
 Alan Greig   ------------------------------  # Date: Sun, 07 Aug 2005 21:44:49 GMT   From: John Santos <john@egh.com>@ Subject: Re: Killing a process that has allocated the tape drive' Message-ID: <lhvJe.5620$z%.96@trnddc02>    MUSTAFA ATAKAN wrote:  > Hi,  >   I >   I have a problem in killing a process that has already allocated the  J > tape drive (that is, the system user mounted the tape drive, but closed J > the session without deallocating the drive). I want to the kill process H > since I want to mount the tape drive (as a system user, of course) in ) > another telnet session. In other words:  >    > -----------------------------------------------------------------------------------------------------------------------------------------------------------------------  > $mount $2$mga1: /ov=id; > %MOUNT-I-OPRQST, device already allocated to another user  >    > $dismount $2$mga1:> > %SYSTEM-W-DEVALLOC, device already allocated to another user >    > $ show dev mga1 /full F > Magtape $2$MGA1: (TILOT3), device type COMPAQ SuperDLT1, is online,  > allocated,N > deallocate on dismount, mounted, file-oriented device, available to cluster,N > device has multiple I/O paths, error logging is enabled, controller supportsF > compaction (compaction disabled), device supports fastskip (per_io).- > Error count 2 Operations completed 33626407 - > Owner process *"SYSTEM"* Owner UIC [SYSTEM] B > Owner process ID *20602CF3* Dev Prot S:RWPL,O:RWPL,G:RWPL,W:RWPL > ...  > .... >     B You didn't post *all* of the output.  Is the drive in Mount-verify? state?  If so, replace the original tape in the driver, put the 1 drive back online, and let it finish cleaning up.   = If the original tape has been written to after it was removed < from the drive, it won't pass the mount verification checks.  A Maybe with enough privileges, you can "$ dismount/abort $2$MGA1:" . and get it to abort all the outstanding I/O's.       > $ stop /id=20602CF3  > $ stop /id=20602CF3  > $ stop /id=20602CF3  >    > $ show user system /full > .... > ....6 > SYSTEM TILOT3 SYSTEM 20602CF3 TNA346: (disconnected) > .... > .... > -----------------------------------------------------------------------------------------------------------------------------------------------------------------------  >   I > What is the reason that i can not stop the above process? Is there any  H > command like "kill -9" in *nix environments? Is there any way to stop @ > this process without rebooting or waiting some timeout period? >    > Thanks in advance...  H There is a very good reason for this.  It is to keep the tape controllerD from writing all over memory that no longer belongs to it.  The tapeE drive has been instructed to read or write data, but hasn't completed F yet.  If VMS just let you kill the process, eventually the drive wouldH trash someone else's memory.  Therefore, you can't delete a process withF pending I/O.  STOP/ID just sets a "delete pending" flag in the process? control block, and when all the I/O completes, it will go away.    --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Sun, 07 Aug 2005 21:17:13 -0700 ( From: Jeff Cameron <roktsci@comcast.net>@ Subject: Re: Killing a process that has allocated the tape drive0 Message-ID: <BF1C2AD9.11F2F%roktsci@comcast.net>  L > This message is in MIME format. Since your mail reader does not understand< this format, some or all of this message may not be legible.   --B_3206294236_42506786  Content-type: text/plain;  	charset="ISO-8859-1"  Content-transfer-encoding: 8bit   J The problem of tape drive processes going into the RWAST (which I call theG you R WASTed state has to do with the way the VMS engineers wrote the I Magtape driver. It is different than every other driver in that it allows F the REWIND (and a few other) IOs to return  with complete status evenF though the physical operation has not completed and further operationsH cannot occur until the physical operation comes to actual completion. An; article describing the details of this can be located here:   $ http://www.jcameron.com/vms/em14.htm  G Sometimes (less than 25%) the RWAST can be cleared without rebooting by J mounting a tape forward spacing it manually and then rewinding and waitingI for the tape to come online at BOT. This can be difficult with newer tape L drives such as DLTs or LTOs, but I have had success by placing in a cleaningJ cartridge, followed by a blank tape and allowing it to come online. But inE most cases this does not work. But the effort is ususally better than 
 rebooting.     On 8/7/05 10:32 AM, in articleH F014DACB8BE63442993543B780A2F018014A62DC@asteriks.inteltek.ist, "MUSTAFA( ATAKAN" <matakan@inteltek.com.tr> wrote:   > Hi,  >   M >   I have a problem in killing a process that has already allocated the tape P > drive (that is, the system user mounted the tape drive, but closed the sessionM > without deallocating the drive). I want to the kill process since I want to O > mount the tape drive (as a system user, of course) in another telnet session.  > In other words:  >   P > ------------------------------------------------------------------------------P > ------------------------------------------------------------------------------
 > -----------  > $mount $2$mga1: /ov=id; > %MOUNT-I-OPRQST, device already allocated to another user  >    > $dismount $2$mga1:> > %SYSTEM-W-DEVALLOC, device already allocated to another user >    > $ show dev mga1 /full P > Magtape $2$MGA1: (TILOT3), device type COMPAQ SuperDLT1, is online, allocated,N > deallocate on dismount, mounted, file-oriented device, available to cluster,N > device has multiple I/O paths, error logging is enabled, controller supportsF > compaction (compaction disabled), device supports fastskip (per_io).- > Error count 2 Operations completed 33626407 + > Owner process "SYSTEM" Owner UIC [SYSTEM] @ > Owner process ID 20602CF3 Dev Prot S:RWPL,O:RWPL,G:RWPL,W:RWPL > ...  > .... >    > $ stop /id=20602CF3  > $ stop /id=20602CF3  > $ stop /id=20602CF3  >    > $ show user system /full > .... > ....6 > SYSTEM TILOT3 SYSTEM 20602CF3 TNA346: (disconnected) > .... > ....P > ------------------------------------------------------------------------------P > ------------------------------------------------------------------------------
 > -----------  >   P > What is the reason that i can not stop the above process? Is there any commandL > like "kill -9" in *nix environments? Is there any way to stop this process3 > without rebooting or waiting some timeout period?  >    > Thanks in advance... >          --B_3206294236_42506786  Content-type: text/html; 	charset="ISO-8859-1" + Content-transfer-encoding: quoted-printable    <HTML> <HEAD>F <TITLE>Re: Killing a process that has allocated the tape drive</TITLE> </HEAD>  <BODY>P <FONT FACE=3D"Verdana, Helvetica, Arial"><SPAN STYLE=3D'font-size:12.0px'>The p=M roblem of tape drive processes going into the RWAST (which I call the &#8220= M ;you R WASTed&#8221; state has to do with the way the VMS engineers wrote th= M e Magtape driver. It is different than every other driver in that it allows = M the REWIND (and a few other) IO&#8217;s to return &nbsp;with complete status= M  even though the physical operation has not completed and further operations= M  cannot occur until the physical operation comes to actual completion. An ar= = ticle describing the details of this can be located here:<BR>  <BR>N <a href=3D"http://www.jcameron.com/vms/em14.htm">http://www.jcameron.com/vms/= em14.htm</a><BR> <BR>L Sometimes (less than 25%) the RWAST can be cleared without rebooting by mou=M nting a tape forward spacing it manually and then rewinding and waiting for = M the tape to come online at BOT. This can be difficult with newer tape drives= M  such as DLTs or LTOs, but I have had success by placing in a cleaning cartr= M idge, followed by a blank tape and allowing it to come online. But in most c= M ases this does not work. But the effort is ususally better than rebooting.<B=  R> <BR> <BR>L On 8/7/05 10:32 AM, in article F014DACB8BE63442993543B780A2F018014A62DC@ast=M eriks.inteltek.ist, &quot;MUSTAFA ATAKAN&quot; &lt;matakan@inteltek.com.tr&g= 
 t; wrote:<BR>  <BR>P </SPAN></FONT><BLOCKQUOTE><SPAN STYLE=3D'font-size:12.0px'><FONT FACE=3D"Arial"= >Hi,<BR>5 </FONT><FONT FACE=3D"Verdana, Helvetica, Arial"> <BR> N </FONT><FONT FACE=3D"Arial"> &nbsp;I have a problem in killing a process that=M  has already allocated the tape drive (that is, the system user mounted the = M tape drive, but closed the session without deallocating the drive). I want t= M o the kill process since I want to mount the tape drive (as a system user, o= 8 f course) in another telnet session. In other words:<BR>5 </FONT><FONT FACE=3D"Verdana, Helvetica, Arial"> <BR> N </FONT><FONT FACE=3D"Arial">-------------------------------------------------=M ----------------------------------------------------------------------------= . ------------------------------------------<BR> $mount $2$mga1: /ov=3Did<BR>= %MOUNT-I-OPRQST, device already allocated to another user<BR> 5 </FONT><FONT FACE=3D"Verdana, Helvetica, Arial"> <BR> 2 </FONT><FONT FACE=3D"Arial">$dismount $2$mga1:<BR>@ %SYSTEM-W-DEVALLOC, device already allocated to another user<BR>5 </FONT><FONT FACE=3D"Verdana, Helvetica, Arial"> <BR> 5 </FONT><FONT FACE=3D"Arial">$ show dev mga1 /full<BR> L Magtape $2$MGA1: (TILOT3), device type COMPAQ SuperDLT1, is online, allocat= ed,<BR> L deallocate on dismount, mounted, file-oriented device, available to cluster= ,<BR> L device has multiple I/O paths, error logging is enabled, controller support= s<BR> H compaction (compaction disabled), device supports fastskip (per_io).<BR>/ Error count 2 Operations completed 33626407<BR> N Owner process <B>&quot;<FONT COLOR=3D"#FF0000">SYSTEM&quot;</FONT></B> Owner = UIC [SYSTEM]<BR>N Owner process ID <FONT COLOR=3D"#FF0000"><B>20602CF3</B></FONT> Dev Prot S:RW= PL,O:RWPL,G:RWPL,W:RWPL<BR>  ...<BR>  ....<BR>5 </FONT><FONT FACE=3D"Verdana, Helvetica, Arial"> <BR> 5 </FONT><FONT FACE=3D"Arial">$ stop /id=3D20602CF3<BR>  $ stop /id=3D20602CF3<BR>  $ stop /id=3D20602CF3<BR> 5 </FONT><FONT FACE=3D"Verdana, Helvetica, Arial"> <BR> 8 </FONT><FONT FACE=3D"Arial">$ show user system /full<BR> ....<BR> ....<BR>8 SYSTEM TILOT3 SYSTEM 20602CF3 TNA346: (disconnected)<BR> ....<BR> ....<BR>L ---------------------------------------------------------------------------=M ----------------------------------------------------------------------------=  ----------------<BR>5 </FONT><FONT FACE=3D"Verdana, Helvetica, Arial"> <BR> N </FONT><FONT FACE=3D"Arial">What is the reason that i can not stop the above =M process? Is there any command like &quot;kill -9&quot; in *nix environments?= M  Is there any way to stop this process without rebooting or waiting some tim=  eout period?<BR>5 </FONT><FONT FACE=3D"Verdana, Helvetica, Arial"> <BR> 4 </FONT><FONT FACE=3D"Arial">Thanks in advance...<BR>4 </FONT><FONT FACE=3D"Verdana, Helvetica, Arial"><BR>P </FONT></SPAN></BLOCKQUOTE><SPAN STYLE=3D'font-size:12.0px'><FONT FACE=3D"Verda= na, Helvetica, Arial"><BR> <BR> </FONT></SPAN> </BODY>  </HTML>      --B_3206294236_42506786--    ------------------------------  % Date: Sun, 07 Aug 2005 15:40:43 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net>  Subject: Re: Newbie DCL Question+ Message-ID: <42F671CB.2ECB987A@comcast.net>    Kevin wrote: >  > Hi > H > If this isn't an appropriate group to post this question, I apologise. > H > I am prompting the user for a nickname using read/prompt from within aG > login script in a captive account.   I will use this name to create a # > subdirectory later in the script.  > E > Apart from using f$length to check size, and f$edit to collapse and H > convert to uppercase, how can I validate against users entering quoted: > strings and various punctuation that will upset cre/dir?  H Well, one approach would be to use F$EXTRACT() in a loop, and test everyE character in the string against a known set of valid characters (A-Z,  0-9, $, _, -).  / $ VALID := ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789 
 $ PNTR = 0 $LOOP:! $ CHR = F$EXTR( PNTR, 1, string ) : $ IF F$LOCATE( CHR, VALID ) .NES. F$LENGTH( VALID ) THEN -: $ insert code here to issue the complaint and prompt again $ PNTR = PNTR + 1 ( $ IF PNTR .LT. F$LENGTH( string ) THEN - $ GOTO LOOP   - Other than that, it gets a bit more involved.    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  % Date: Sun, 07 Aug 2005 15:32:32 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net>O Subject: Re: OpenVMS under the umbrella - or umbrella is supported at the time? + Message-ID: <42F66FE0.5F7234C6@comcast.net>    sean@obanion.us wrote: > < > My balls still flash when I bounce them against the floor,3 > but I realize that not everybody is in to that...   A I'm sure I'd see stars, flashing lights and more if I were to ...    Nah, don't go there.   --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  * Date: Sun, 7 Aug 2005 18:24:16 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)4 Subject: Re: strange terminal-characteristic problem$ Message-ID: <dd5jkg$ji9$1@online.de>  5 In article <42F5A0E1.BE7EDDA1@teksavvy.com>, JF Mezei ' <jfmezei.spamnot@teksavvy.com> writes:    1 > Phillip Helbig---remove CLOTHES to reply wrote: 1 > > Everything appears to work OK EXCEPT editing.  > ? > Then it is a 7bit vs 8 bit problem and/or flow control issue.   H That's what I thought.  However, I compared the output of SHOW TERMINAL G with that of another machine, it and its console terminal located less  " than a meter away.  No difference.  K > > I don't yet have DECnet running.  If I set host via LAT or TELNET, then   > > everything works fine again. > N > Ok, perhaps you need to remind me again how the terminal is connected to oneN > machine and from that machine to the suspect vax. I was under the impression' > it was with DECNET (SET HOST <node>).   B All my machines have VT320 terminals as the console.  If I log in F interactively at the problem machine, I see the problem.  If, after I H log in, I set host to the same machine via LAT (or connect via TELNET),  the problem goes away.  D > If you SET HOST/DTE to the console port, remember that there are 3 > ports involved.   C No, it's a "real console", i.e. a VT plugged in via a serial cable.  > < > [TERMINAL]----[TXA1:][NODE1][TXA2:]---------[OPA0:][NODE2]  I While I'm at it, another machine in the cluster has a strange problem as  I well: SHOW CLUSTER/CONTINUOUS cannot be exited with CTRL-Z.  One can get  F out with CTRL-Y then typing EXIT, but then the terminal is screwed up C and needs a SET TERM/INQUIRE.  Interestingly, this problem is also  B independent of the physical terminal and physical machine: I just D replaced a VAX 4000/100A with a VAX 4000/105A.  The problem remains.   ------------------------------  * Date: Sun, 7 Aug 2005 18:26:11 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)4 Subject: Re: strange terminal-characteristic problem$ Message-ID: <dd5jo2$ji9$2@online.de>  < In article <11fcfrjq3bqlkf9@corp.supernews.com>, Dave Froble <davef@tsoft-inc.com> writes:   1 > Phillip Helbig---remove CLOTHES to reply wrote:  > > > > In EDT.  Not at the console prompt, not at the VMS prompt. > >  > I > Not been following this closly.  Figured it was a HW problem until you  C > stated that you moved the disk to another system and the problem   > followed the disk. > J > Does that disk have an EDT ini file (or whatever it's called)?  Perhaps C > you're configuring EDT in some manner that's causing the problem?   H The system disks are more or less identical.  They are both 7.3 with allF patches (one might have a couple more layered products installed).  ItH is a "shared everything" cluster; the idea is one shouldn't need to know? or care which node one is logged into.  All files (SYS$SYLOGIN, C EDTINI.EDT etc) are the same (not just the same contents, the same  $ physical file on a non-system disk).   ------------------------------  % Date: Sun, 07 Aug 2005 15:32:59 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 4 Subject: Re: strange terminal-characteristic problem, Message-ID: <42F661E0.A65B3D78@teksavvy.com>  / Phillip Helbig---remove CLOTHES to reply wrote: C > All my machines have VT320 terminals as the console.  If I log in G > interactively at the problem machine, I see the problem.  If, after I I > log in, I set host to the same machine via LAT (or connect via TELNET),  > the problem goes away.   Oh ! OK. That is different.    So terminal connected to OPA0:  I And you checked the SYSGEN parameter TTY_DEFCHAR and TTY_DEFCHAR2 on that 8 system against that of a system where OPA0: works fine ?  M I know that not all parameters are applicable to OPA0: (for instance, you may D not be able to change the serial port speed on the port with the SETN TERM/SPEED command, even though SHOW TERM will display whatever speed you have4 set- and this may vary from VAX model to VAX model).      M Does the VT320 itself have the right setup in terms of speed, parity and flow K control ? Does it emit xoff at 64 or 128 characters ? (get it to emit at at K 128, giving more time for any data to stop flowing before buffer overflow).     L Something else to test is to bring down the machine, change the OPA0: serialN port speed on the hardware switch (or >>> approciate command), change terminalN to match and then start VMS again. Bring it down to 300 and see if the problem happens again.    M There is some funky stuff that happens because OPA0: doesn't use the standard K TT driver, it has the "OPERATOR" driver on top of it as well as being quite J dependant on the hardware itself which is different from a standard serial- card since it is embedded in the motherboard.    ------------------------------  * Date: Sun, 7 Aug 2005 19:48:55 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)4 Subject: Re: strange terminal-characteristic problem$ Message-ID: <dd5oj7$ji9$7@online.de>  5 In article <42F661E0.A65B3D78@teksavvy.com>, JF Mezei ' <jfmezei.spamnot@teksavvy.com> writes:    K > And you checked the SYSGEN parameter TTY_DEFCHAR and TTY_DEFCHAR2 on that : > system against that of a system where OPA0: works fine ?  9 Both are the same and equal to the defaults on all nodes.   O > Does the VT320 itself have the right setup in terms of speed, parity and flow M > control ? Does it emit xoff at 64 or 128 characters ? (get it to emit at at M > 128, giving more time for any data to stop flowing before buffer overflow).   G All of my VTs have identical setups.  (Remember, I swapped the VTs and  G the problem stayed with the node.  I also swapped the computer and the  I problem stayed with the system disk, though I did swap it with a similar   model. > N > Something else to test is to bring down the machine, change the OPA0: serialP > port speed on the hardware switch (or >>> approciate command), change terminalP > to match and then start VMS again. Bring it down to 300 and see if the problem > happens again.    	 It's 300.   O > There is some funky stuff that happens because OPA0: doesn't use the standard M > TT driver, it has the "OPERATOR" driver on top of it as well as being quite L > dependant on the hardware itself which is different from a standard serial/ > card since it is embedded in the motherboard.   < That could explain why the problem seems to happen on all my0 VAXstations, but not on VAXes and not on ALPHAs.   ------------------------------  % Date: Sun, 07 Aug 2005 15:28:26 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net>4 Subject: Re: strange terminal-characteristic problem+ Message-ID: <42F66EEA.308DA0B9@comcast.net>   / Phillip Helbig---remove CLOTHES to reply wrote:  > > > In article <42F55010.FF777D70@comcast.net>, David J Dachtera% > <djesys.nospam@comcast.net> writes:  > @ > > Does the data cable run close to anything that might produceE > > electromagnetic "noise"? ...*ANY* kind of noise? (A P.C. monitor, G > > another terminal, some other piece of electronic gear)? Does it run H > > parallel to a UTP cable? ...another network cable or high-speed dataF > > line (even E1/T1, ADSL or ISDN)? ...a power cable (since you're inG > > Europe (.de), the 240V power line throws more EMI than a US cord at  > > 120VAC)? > H > Remember the original Star Trek series?  Picture Scotty cramped insideI > the Jeffries tube.  [Scottisch accent]  "You're straining the dilithium F > crystals, Captain.  I can give you power to the shields, but I can'tI > give you warp drive."  My machine room looks similar.  Behind the shelf I > where the computers are are of course several cables.  What puzzles me, H > though, is that I see this problem only on the VAXstations, not on theI > other machines.  And at another location, with a standalone machine and H > almost no cables, I also see the problem---but only with a VAXstation. > K > > You said that the problem stays with the system disk. That is, when you L > > boot the troubled machine up as a different node, the problem goes away? > J > I replaced a VAXstation 4000/60 with a 4000/90A but with the same system > disk.  Problem stays.  > I > > Does the VAX support an alternate console port, like some of the uVAX  > > 3100's do? >  > Not that I know of.  > G > > Do you see this same behavior outside of TPU? ...at the DCL prompt?  > > ....at the console prompt? > < > In EDT.  Not at the console prompt, not at the VMS prompt.  G Well, other situations result in some characters being ignored. CTRL+K, 2 for example, has no impact on DCL, that I can see.   O.k. Next dumb questions:   3 Is this a new development, or how it's always been?   ' How long are these async. serial lines?   B What is the cable type? Have you tried - or can you try - Cat-5 orD better twisted pair? ...shielded cable (each twisted pair shielded)?  B Is the distance between the tube(s) and the machine(s) enough that5 ground potential may drift between the two locations?   # There's GOTTA be a reason for it...    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  * Date: Sun, 7 Aug 2005 22:02:26 +0000 (UTC)P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)4 Subject: Re: strange terminal-characteristic problem$ Message-ID: <dd60dh$6vd$1@online.de>  < In article <42F66EEA.308DA0B9@comcast.net>, David J Dachtera$ <djesys.nospam@comcast.net> writes:   5 > Is this a new development, or how it's always been?   H It's been that way for a while; perhaps it's always been the case.  It'sE not a big problem, since most of the time I work remotely from 500 km F away (and am logged in via telnet, so everything is fine) and when I'mE there, when I do need to edit from a console-login session, I can use B another console.  Thus, it's been low on my list of priorities.  IF recently acquired a VAXstation 4000/90A with 80 MB and a VAX 4000/100AG with 128 MB.  Since the hardware was otherwise similar, I used these to H replace a VAXstation 4000/60 with 32 MB and a VAX 4000/100A with 64 MB. D When the problem didn't go away, I started thinking about it again.   ) > How long are these async. serial lines?   G 1--2 meters.  However, consider the standalone VAXstation 4000/60 from  I which I am now logged into my main cluster.  It has a very short (50 cm)  B MMJ cable, very little EMF in its vicinity---and exactly the same  problem.  D > What is the cable type? Have you tried - or can you try - Cat-5 orF > better twisted pair? ...shielded cable (each twisted pair shielded)?  F The problem is with the console port.  Only an MMJ (DEC-423) plugs in I there.  An RS232 cable connected to the same VT and to the RS232 port of  G the same machine doesn't show the problem, but of course it is not the  B console.  Of course, when I log in via LAT or TELNET, the logical F terminal is no longer OPA0:.  Thus, it appears to be a combination of " OPA0: and the VAXstation hardware.  D > Is the distance between the tube(s) and the machine(s) enough that7 > ground potential may drift between the two locations?  > % > There's GOTTA be a reason for it...   + Yes, but I don't think it's an obvious one.   D I suppose the next step is to collect all the information about the H terminal setting---perhaps it's not all in SHOW TERMINAL.  F$GETDVI has I many TT* arguments.  Unfortunately, I am not aware of anything which can  ' collect all this information at one go.    ------------------------------  # Date: Sun, 07 Aug 2005 22:33:46 GMT   From: John Santos <john@egh.com>4 Subject: Re: strange terminal-characteristic problem) Message-ID: <e%vJe.10090$WQ.587@trnddc03>   / Phillip Helbig---remove CLOTHES to reply wrote: 7 > In article <42F44A57.97E28951@teksavvy.com>, JF Mezei ) > <jfmezei.spamnot@teksavvy.com> writes:   >  > 1 >>Phillip Helbig---remove CLOTHES to reply wrote:  >>C >>>I have a node in my cluster where I can't properly edit from the J >>>console.  I see "line noise" with EDT (and with TPU), bogus blank linesH >>>etc.  If I connect to the same node with LAT or TELNET from a consoleG >>>login, then all is OK.  Other console terminals on other nodes don't  >>>have this problem.  >>J >>Does the target now have a SET TERM/INQUIRE during the login sequence  ? >  > I > Yes, that's in SYS$SYLOGIN.  As I said, the output of SHOW TERMINAL is  = > the same; all interactive processes go through SYS$SYLOGIN.  >  > P >>Make sure that both the originating and destination node have similar terminalF >>characteristics in terms of SET TERM/EIGHTBIT or SET TERM/NOEIGHTBIT >  >  > Yes, they are the same.  > H > What I get are bogus blank lines, boldface (backwards?) question marksK > (similar to what TPU sometimes intentionally displays, but I see them in  K > EDT as well), overwritten lines, cursor position in "reality" and on the  G > display out of sync etc.  Sometimes, CTRL-W makes things look better.  > G > What I really think is puzzling is that, as far as I can tell, I get  < > this only with the VAXstation but not with other hardware! > I > Are there terminal characteristics which SHOW TERMINAL doesn't display?   H I have lots of problems with TECO in screen mode (4,7:w) on *all* serialE console terminals (AS1200, MicroVAX 4000-2000, AS2000, AS4100, ES40.) C This wasn't always true, some of the systems broke at about VMS 7.0 C timeframe which had previously worked fine.  Consoles a re VT320's, , VT420's, VT520's, it doesn't seem to matter.  G W$$ (refresh screen) often but doesn't always fix it, or cleans up part F of the screen and messes up another part of it.  I assumed the consoleF port couldn't keep up with the I/O traffic.  Maybe network connections< slow down the I/O enough that the console UARTs can keep up?      G (Graphics consoles, once DECWindows is up, seem to work fine using TECO  in screen mode under DECTERM.)   --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Sun, 07 Aug 2005 20:56:46 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 4 Subject: Re: strange terminal-characteristic problem, Message-ID: <42F6ADCD.54A2380E@teksavvy.com>   David J Dachtera wrote: ) > How long are these async. serial lines?   L Based on what he said, that isn't the problem. If he uses the same terminal,N same cable, same system/OPA0: and from $ he TELNETS or SET HOST/LAT to anotherI host, then the session on the other host works fine without OPA0: and the J serial cable scrambling characters into an omelette, but once back at that, node's $ prompt, OPA0:.s behaviour is funky.   Mr remove-clothes-to-reply:   N You need to TYPE a large file with aligned data, hopefully not greater than 80N characters in record length, and use the hold-screen button a few times to seeM if there might be flow control issues. With aligned data, you will easily see N any missing stuff since the data will be screwed up after you release the hold and it starts scrolling again.    N Out of curiosity, did you say Vaxstation 4000xx ?  If so, does the serial portJ really appear as OPA0: ? Is there a second serial port on those machines ?L Once booted, if you plug the terminal into the other port and login, does itK exhibit the same behaviour ? Or is that behaviour limited to ports that are P seen as OPA0: (eg: the OPERATOR device driver acting above the terminal driver).  I If you use the graphics display without decwindows, you get a truly basic / character cell interface without VT100 support.    ------------------------------  % Date: Sun, 07 Aug 2005 21:02:12 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 4 Subject: Re: strange terminal-characteristic problem, Message-ID: <42F6AF13.A28C1E8B@teksavvy.com>  / Phillip Helbig---remove CLOTHES to reply wrote: C > console.  Of course, when I log in via LAT or TELNET, the logical G > terminal is no longer OPA0:.  Thus, it appears to be a combination of $ > OPA0: and the VAXstation hardware.    M From OPA0:, if you use EDT, it is bad. Right ? If from OPA0: you SET HOST/LAT O to another node and on that other node you use EDT, is that output also flaky ?    ------------------------------   Date: 7 Aug 2005 10:56:31 -0700 @ From: "Hein (RMS) van den Heuvel" <heinvandenheuvel@hotmail.com>A Subject: Re: Why is disk I/O operation rate .GT. direct I/O rate? C Message-ID: <1123437391.012515.250800@f14g2000cwb.googlegroups.com>    AEF,1       - Why leave your name as a complete riddle? /       - Why pose a question mostly as a riddle? G Are you concerned about the difference of 4.55  disk IO/sec versus 3.62  Direct IO over 10 hours?>       - What problem are you really trying to solve? Just idle speculation?@ There is no IO to worry about, not from a average point of view,% neither from a max Ioad point of view  .  >> Is this a bug?  What? 1 Are some I/O operations to disk not direct I/O's?  Yes, notably paging IO.  If so, how does that work?( That works very well. Thanks for asking. Is it something else? % Is what something else? Yes probably. $ Nothing is what it seems these days.  @ Paul wrote: "Disk IO done by the XQP is counted as buffered IO."A An ambiguish answer, welll in line with the question. To expand a  little: G Some file operations like OPEN, CREATE, EXTEND, CLOSE cause one or more D physical IOs but are counted as buffered IO because it does not meet the definition of direct IO.   fwiw,  Hein.    ------------------------------   Date: 7 Aug 2005 12:16:49 -0700 $ From: "AEF" <spamsink2001@yahoo.com>A Subject: Re: Why is disk I/O operation rate .GT. direct I/O rate? B Message-ID: <1123442209.720362.26690@g47g2000cwa.googlegroups.com>    Hein (RMS) van den Heuvel wrote: > AEF,3 >       - Why leave your name as a complete riddle?     : It was less typing. A determined person could find it out.    1 >       - Why pose a question mostly as a riddle?     D It's not a riddle. I previously thought that all disk I/O was directC I/O. But this new information showed me that it is probably not the  case. So I asked.     I > Are you concerned about the difference of 4.55  disk IO/sec versus 3.62  > Direct IO over 10 hours?    E No. But I noticed that this is consistent across my systems and I was ) just wondering what was really going oin.     @ >       - What problem are you really trying to solve? Just idle > speculation?    6 I am trying to better understand VMS and how it works.    B > There is no IO to worry about, not from a average point of view,' > neither from a max Ioad point of view      OK.      > >> Is this a bug?  > What?      You heard me! :-)     3 > Are some I/O operations to disk not direct I/O's?  > Yes, notably paging IO.    OK   > If so, how does that work?* > That works very well. Thanks for asking.   Cool.    > Is it something else? ' > Is what something else? Yes probably. & > Nothing is what it seems these days.  : Is there something wrong with the universe these days? :-)   > B > Paul wrote: "Disk IO done by the XQP is counted as buffered IO."C > An ambiguish answer, welll in line with the question. To expand a 	 > little: I > Some file operations like OPEN, CREATE, EXTEND, CLOSE cause one or more F > physical IOs but are counted as buffered IO because it does not meet > the definition of direct IO.    C OK. Resovled: Some disk I/O's are not direct I/O's, but in fact are  buffered I/O's.   ? I was surprised by this because the performance manual seems to F indicate that buffered I/O's are anything but disk I/O's. At least theE BIO sections in said manual never refer to disks as a possible source  of BIO.   B But upon re-checking the manual I just discovered that "system I/OE activity to the disk is not reflected in the direct I/O count MONITOR F provides ..."  Sorry I overlooked that when I posed my question. I was1 concentrating on the "understanding" chapter. OK.    > fwiw,  > Hein.    Thanks for your help.    ------------------------------  % Date: Sun, 07 Aug 2005 15:38:47 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> A Subject: Re: Why is disk I/O operation rate .GT. direct I/O rate? , Message-ID: <42F6633B.636AA604@teksavvy.com>  
 AEF wrote:F > It's not a riddle. I previously thought that all disk I/O was directE > I/O. But this new information showed me that it is probably not the  > case. So I asked.   N Consider the case where you do an IO to read 128 bytes. Results in a direct ION that reads a block (512 bytes). Next IO for 128 bytes reads from the 512 stillL in the buffer.  (And I assume there are plenty of process and RMS parameters that tune this behaviour).   ------------------------------   Date: 7 Aug 2005 12:53:36 -0700 $ From: "AEF" <spamsink2001@yahoo.com>A Subject: Re: Why is disk I/O operation rate .GT. direct I/O rate? B Message-ID: <1123444416.752047.73070@z14g2000cwz.googlegroups.com>    Hein (RMS) van den Heuvel wrote: [...]  > B > Paul wrote: "Disk IO done by the XQP is counted as buffered IO."C > An ambiguish answer, welll in line with the question. To expand a 	 > little:   D How is the question ambiguous? I asked about the componencts that goB into disk I/O, asking why some of it is not counted as direct I/O.  E Others have asked ambiguous questions but people here are often happy B to dive in and speculate what the questioner has in mind, like theB recent question about data file search order in some applications.  ( Hey, at least my questions are on topic!   [...]    Thanks again for your help.    ------------------------------   Date: 7 Aug 2005 13:25:31 -0700 @ From: "Hein (RMS) van den Heuvel" <heinvandenheuvel@hotmail.com>A Subject: Re: Why is disk I/O operation rate .GT. direct I/O rate? B Message-ID: <1123446331.637318.29450@g43g2000cwa.googlegroups.com>  
 AEF wrote:" > Hein (RMS) van den Heuvel wrote: > > AEF,5 > >       - Why leave your name as a complete riddle? < > It was less typing. A determined person could find it out.  " Ok mr Feldman found name and such.   > 3 > >       - Why pose a question mostly as a riddle? F > It's not a riddle. I previously thought that all disk I/O was direct > I/O.  6 I failed to look back at the title which was explicit.@ The topic text left me wondering what things you were looking at prompting the question.   8 > I am trying to better understand VMS and how it works.  
 Same here :-)   D > > Paul wrote: "Disk IO done by the XQP is counted as buffered IO."E > > An ambiguish answer, welll in line with the question. To expand a   5 I was just expanding a little on Paul's short answer.    Cheers,  Hein.    ------------------------------   Date: 7 Aug 2005 15:00:38 -0700 $ From: "AEF" <spamsink2001@yahoo.com>A Subject: Re: Why is disk I/O operation rate .GT. direct I/O rate? C Message-ID: <1123452038.384099.264870@g14g2000cwa.googlegroups.com>     Hein (RMS) van den Heuvel wrote: > AEF wrote:$ > > Hein (RMS) van den Heuvel wrote:
 > > > AEF,7 > > >       - Why leave your name as a complete riddle? > > > It was less typing. A determined person could find it out. > $ > Ok mr Feldman found name and such.   I've been found out! :-)  5 > > >       - Why pose a question mostly as a riddle? H > > It's not a riddle. I previously thought that all disk I/O was direct > > I/O. > 8 > I failed to look back at the title which was explicit.B > The topic text left me wondering what things you were looking at > prompting the question.    OK.    [...] F > > > Paul wrote: "Disk IO done by the XQP is counted as buffered IO."G > > > An ambiguish answer, welll in line with the question. To expand a  > 7 > I was just expanding a little on Paul's short answer.  > 	 > Cheers,  > Hein.   8 Your help is greatly appreciated. Thanks for chiming in.   ------------------------------  % Date: Sun, 07 Aug 2005 20:59:10 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> A Subject: Re: Why is disk I/O operation rate .GT. direct I/O rate? , Message-ID: <42F6AE5D.D762E843@teksavvy.com>  
 AEF wrote:& > > Ok mr Feldman found name and such. >  > I've been found out! :-)    L Not completely. I still haven't figured out what the E stands for.  (I guessN Edward, but it could be anything. I had only remember the first name Allan and the last name.).   ------------------------------   End of INFO-VAX 2005.439 ************************