1 INFO-VAX	Thu, 18 Aug 2005	Volume 2005 : Issue 459       Contents: Australian DECUS Re: Australian DECUS- Re: Command to find new drives from VMS Level % Re: Help Needed: Tape Backup Save Set % Re: Help Needed: Tape Backup Save Set F Re: How many files can you have in a VMS directory without performanceF Re: How many files can you have in a VMS directory without performanceF Re: How many files can you have in a VMS directory without performanceP Re: How many files can you have in a VMS directory without performance   problemP Re: How many files can you have in a VMS directory without performance   problemP Re: How many files can you have in a VMS directory without performance  problemsP Re: How many files can you have in a VMS directory without performance  problemsP Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance problems?P Re: How many files can you have in a VMS directory without performance roblems? P Re: How many files can you have in a VMS directory withoutperformance problems?  HP 3rd quarter financials  Re: HP 3rd quarter financials  Re: HP 3rd quarter financials  Re: HP 3rd quarter financials  Re: HP 3rd quarter financials  Re: Image restore fails  Re: Is OVMS going Bangalore?  Re: Reading VMS SAS files on PC?  Re: Reading VMS SAS files on PC?  Re: Reading VMS SAS files on PC?  Re: Replacement for VT terminals Re: SAMBA for VMS ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' RE: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!) ' Re: SAMBA for VMS (The saga continues!)  Re: simh define/attach printer Re: Squid for VMS  SYSMAN vs SYSGEN Re: SYSMAN vs SYSGEN Re: SYSMAN vs SYSGEN Re: SYSMAN vs SYSGEN! Re: Turning up an Ethernet & tftp 7 Re: Was -->  Print Queue in "starting" state -- cause??   F ----------------------------------------------------------------------  # Date: Wed, 17 Aug 2005 21:27:33 GMT % From: John Proctor <lost@nowhere.org>  Subject: Australian DECUS 1 Message-ID: <2005081807273216807%lost@nowhereorg>   A I wrote to the current address for the Australian DECUS to get a  H membership as I want to become part of the hobbyist programme. As yet I 0 have had no response and it is over 4 weeks now.  @ I used to work for DEC as a Software Services manager in Sydney E (1977-1979) and a salesman before going to a DEC customer (Alcan) to  H manage their VAX sytems (1980-1985). I have an opportunity to acquire a F 3300 at no cost to me so I am interested in getting it up and running.  ? Is there an alternative way to get a DECUS membership so I can  ' partticipate in the hobbyist programme?    --   Regards, John D Proctor   ------------------------------    Date: 17 Aug 2005 15:41:44 -0700 From: dooleys@snowy.net.au Subject: Re: Australian DECUS C Message-ID: <1124318504.883547.124020@z14g2000cwz.googlegroups.com>   & I think they recently changed address,>  did you get it from http://www.encompass.asn.au/contact.shtml Phil   ------------------------------  % Date: Wed, 17 Aug 2005 18:23:49 -0400 , From: Carl Friedberg <frida.fried@gmail.com>6 Subject: Re: Command to find new drives from VMS Level7 Message-ID: <890539d905081715235e616faa@mail.gmail.com>   L In my experience, it is somewhat more reliable to run autogen from savparam= s,J perhaps to testfiles.  Sometimes the sysgen autoconfigure all has problemsF with hsx (possiblye eva) drives. Autogen runs the configure portion of the startup code as well.   E WARNING: this could crash the machine on which you are running (not a ' common occurance; but if it happens...)   4 On 8/16/05, Phillip Helbig---remove CLOTHES to reply( <helbig@astro.multiclothesvax.de> wrote:D > In article <1124194722.104190.89390@g43g2000cwa.googlegroups.com>,/ > "Shawn M." <shawnm1964@sbcglobal.net> writes:  >=20J > > I am having a brain fart.  I can't remember the command that I can runE > > from the OpenVMS level of a Alpha Server running 7.2-1 which will E > > search the bus(s) and find disk drives.  Cany anyone lend a hand.  >=20J > If you ever are lucky enough to have the joy of VAX (hey, there's a book > title there!), then it is  >=20  > $  MC SYSGEN AUTOCONFIGURE ALL >=20 >    ------------------------------  % Date: Wed, 17 Aug 2005 14:21:27 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> . Subject: Re: Help Needed: Tape Backup Save Set, Message-ID: <4303801E.E6F42C07@teksavvy.com>   Wayne Sewell wrote: K > "If you want to start processing at BOT, and the magnetic tape is already Q > positioned beyond BOT, specify /REWIND. Otherwise, the magnetic tape begins (or A > resumes) processing from the logical end-of-tape (EOT) marker."   F On TK50s, (I used this because any difference in processing results inH large difference in total time), it seemed that doing successive backupsC  would result in the tape being rewound by the second invocation of G backup (I assume in the process of privately mounting the tape to check F its label etc) and then the TK50 would spend much time reading throughG to seek the EOT position because I had specified /NOREWIND. I know much D time was spent doing this because BACKUP was essentially iddle for aG long time while the tape light was blinking away. and it was a very lon 3 time before backup started to do actual processing.   1 Is this a correct interpretation of what I saw ?    B If one is to mount or allocate the tape before first invovation ofH BACKUP, do subsequent invocations still result in a implicity rewind andH seek end-of-tape operation, or would BACKUP be able to see that the tapeC that is already mounted has the right label and is already at EOT ?    ------------------------------  % Date: Wed, 17 Aug 2005 15:25:09 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> . Subject: Re: Help Needed: Tape Backup Save Set, Message-ID: <43038F08.F42614BA@teksavvy.com>  
 AEF wrote:D > Other problems I have are amusing: One drive hurled the tape out a! > couple of feet onto the floor!    F Warning, you may have leftover from a special VMS 5.0 install kit thatD should have never been released to the public. (The one with specialF artificial intelligence shareable image, there was one well documentedF case, documentation  should be available on the net somwhere, look for "VAXORCIST" :-)    ------------------------------    Date: 17 Aug 2005 16:10:02 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) O Subject: Re: How many files can you have in a VMS directory without performance 3 Message-ID: <ej0YGr1dpsuo@eisner.encompasserve.org>   \ In article <4303A36F.55D58659@teksavvy.com>, JF Mezei <jfmezei.spamnot@teksavvy.com> writes: >  > 	[chocolate]cake.dat > H > If the file "[chocolate]cake.dat" located in the current directory, orG > is it the file cake.dat which is located in the chocolate directory ?  > ( > And since : is now allowed, how about: >  > CAKE:[chocolate]mousse.dat  C    These are easy.  See the guide to extended file naming for those B    few characters which are not allowed.  Care to take a first two    guesses?    ------------------------------  # Date: Wed, 17 Aug 2005 23:54:59 GMT * From: "FredK" <fred.nospam@nospam.dec.com>O Subject: Re: How many files can you have in a VMS directory without performance 3 Message-ID: <n7QMe.10548$gx2.6791@news.cpqcorp.net>   H "Bob Koehler" <koehler@eisner.nospam.encompasserve.org> wrote in message- news:ej0YGr1dpsuo@eisner.encompasserve.org... 7 > In article <4303A36F.55D58659@teksavvy.com>, JF Mezei & <jfmezei.spamnot@teksavvy.com> writes: > >  > > [chocolate]cake.dat  > > J > > If the file "[chocolate]cake.dat" located in the current directory, orI > > is it the file cake.dat which is located in the chocolate directory ?  > > * > > And since : is now allowed, how about: > >  > > CAKE:[chocolate]mousse.dat > E >    These are easy.  See the guide to extended file naming for those D >    few characters which are not allowed.  Care to take a first two
 >    guesses?  >   $ Hmmm.  I'll close my eyes and type ^   ------------------------------  % Date: Wed, 17 Aug 2005 21:26:07 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> O Subject: Re: How many files can you have in a VMS directory without performance , Message-ID: <4303E38E.BC700038@teksavvy.com>   FredK wrote:  > > > CAKE:[chocolate]mousse.dat > > G > >    These are easy.  See the guide to extended file naming for those F > >    few characters which are not allowed.  Care to take a first two > >    guesses?  > >  > & > Hmmm.  I'll close my eyes and type ^     The doctumation at:    says otherwise:    ##C In addition, OpenVMS V7.2 on Alpha systems and ODS-5 disks includes < support for the use of file names, and subdirectory and root@ subdirectory  names, that include all possible 8-bit characters,H excluding values 00 through 1F (hexadecimal) and excluding the following characters:                  < > : / \ | ? * "   B  OpenVMS 7.3-1 on Alpha systems and ODS--5 disks includes enhanced< support for the use of file names, and subdirectory and rootH subdirectory names. It supports all possible 8-bit characters, excluding" only the following two characters:   		 ? *     ##    0 So   isn't in the list of forbidden characters.  < I understand how, once the file system knows which part of aE specification is what part, it knows how to encode special characters 
 (using the )   H But what logic is used by f$parse to determine that [ and/or ] should beE considered part of the filename and "escaped" or whether it points to @ the directory portion of the specicication entered by the user ?   ------------------------------  # Date: Thu, 18 Aug 2005 01:29:02 GMT * From: "FredK" <fred.nospam@nospam.dec.com>Y Subject: Re: How many files can you have in a VMS directory without performance   problem 3 Message-ID: <yvRMe.10551$BH2.2696@news.cpqcorp.net>   : "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message& news:4303D75D.E69402A3@teksavvy.com... > FredK wrote:K > > A modern file system design to replace ODS would use a b-tree structure  > > instead of a linear list.  > F > is INDEXF.SYS also a culprit in the "slow" files system for VMS ? or= > just the simple sequential file nature of directory files ?   I I'd suggest that you ask the question of a file system expert, the guy in F the cube next to my NH office seldom writes here, but perhaps the ever  present Bill Todd will chime in.  I My 50,000 ft view is that the ODS file system has a number of limitations J that effect different aspects of performance.  The directory structure has9 an impact on accessing files. If you aren't doing lots of  open/closes/deletes - H then it isn't really much of an issue.  The next big problem is that ODS treatsB the meta-data and data the same, and there is no real write cache.	 Following I that is perhaps the inefficient mapping of UNIX file system calls to RMS.   B UNIX and Windows heavily cache file IO.  Meta-data is written to aK sequential log that allows the file system structure to be reconstructed if L there is a problem - so that the actual file system structures don't have toH be fully completed before IO can occur to the file.  Temporary files mayG never even hit the disk.  Shared IO is much simpler because the concept G of an IO channel is system based, not process based (process based file J structures point off to shared system resources, as opposed to the process centric VMS IO model).  K Can it be fixed?  Yes.  When?  That I really can't say.  The end result has L to be as stable/reliable as ODS-5, fully support share-all cluster semanticsB and be 100% compatible at the RMS interface layer to applications.   ------------------------------  % Date: Wed, 17 Aug 2005 17:57:15 -0700 ( From: Jeff Cameron <roktsci@comcast.net>Y Subject: Re: How many files can you have in a VMS directory without performance   problem 0 Message-ID: <BF292AFB.12ACA%roktsci@comcast.net>  I On 8/17/05 5:34 PM, in article 4303D75D.E69402A3@teksavvy.com, "JF Mezei" % <jfmezei.spamnot@teksavvy.com> wrote:    > FredK wrote:J >> A modern file system design to replace ODS would use a b-tree structure >> instead of a linear list. > F > is INDEXF.SYS also a culprit in the "slow" files system for VMS ? or= > just the simple sequential file nature of directory files ?   L The problem is not with INDEXF.SYS, which is quite efficient. It is with theI fact that the entries in the .DIR file are always sorted. If there was no G care for sorting and reclamation of deleted space in the .DIR file then  these issues would not arise.    ------------------------------  # Date: Wed, 17 Aug 2005 23:57:27 GMT * From: "FredK" <fred.nospam@nospam.dec.com>Y Subject: Re: How many files can you have in a VMS directory without performance  problems 3 Message-ID: <H9QMe.10549$oF2.7049@news.cpqcorp.net>   : "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message& news:4303A0E3.9015CDA7@teksavvy.com...I > Do any operating systems suport directories containing large numbers of F > files efficiently ? If so, in what way do they deal with directories! > differently than VMS now does ?   E Yes.  Many.  The ODS design was done at a time when disks were small, . memory was tight, and simplicity had a virtue.  G A modern file system design to replace ODS would use a b-tree structure  instead of a linear list.    ------------------------------  % Date: Wed, 17 Aug 2005 17:38:32 -0700 ( From: Jeff Cameron <roktsci@comcast.net>Y Subject: Re: How many files can you have in a VMS directory without performance  problems 0 Message-ID: <BF292698.12ABE%roktsci@comcast.net>  I On 8/17/05 1:41 PM, in article 4303A0E3.9015CDA7@teksavvy.com, "JF Mezei" % <jfmezei.spamnot@teksavvy.com> wrote:   I > Do any operating systems suport directories containing large numbers of F > files efficiently ? If so, in what way do they deal with directories! > differently than VMS now does ?   G  Sure, the UNIX file system, FAT-32, NTFS, MAC HFS & EHFS. All of these G files systems don't have the same problem as VMS because when a file is F deleted it's entry in the .dir file is not removed and the file is notI rewritten in sorted order. The section of the file is just no longer used F and it remains that way until some sort of directory sorter or utilityI program comes along and re-optimizes it. New files are simply appended to  the .dir file.  K Of course this is a poor use of file space, but then this problem that this ! thread is about, does not arise.     ------------------------------    Date: 17 Aug 2005 12:50:29 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) Y Subject: Re: How many files can you have in a VMS directory without performance problems? 3 Message-ID: <Aoj2kVx92N6N@eisner.encompasserve.org>   d In article <ddv7ge$n24$1@rdel.co.uk>, "Chris Blackburn" <chris.blackburn@uk.thalesgroup.com> writes:L > I have noticed on our VMS boxes that perfomance begins to degrade when the > number* > of files in a directory exceeds 10,000.   H    Probably depends on ACP_DINDXCACHE and ACP_DIRCACHE.  From experience0    I don't trust AUTOGEN to manage these values.   ------------------------------    Date: 17 Aug 2005 11:39:17 -07003 From: "Vladimir Vershinin" <vershinin-vk@tochka.ru> Y Subject: Re: How many files can you have in a VMS directory without performance problems? B Message-ID: <1124303957.398088.52760@g44g2000cwa.googlegroups.com>  0 You can use VTfm file and directory manager from http:/www.vtfm.narod.ru    ------------------------------  % Date: Wed, 17 Aug 2005 15:39:28 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> Y Subject: Re: How many files can you have in a VMS directory without performance problems? , Message-ID: <43039262.562BD64C@teksavvy.com>   norm.raphael@metso.com wrote: J > Isn't there a penalty paid for activating the DELETE.EXE image for every > file, 5 > instead of handing it a list, or a wildcarded list?     H A long time ago, on my all mighty microvax 2, I had a large directory toE delete. I wrote a command procedure that read list of files sorted in R reverse order, and then built a delete command with 15 or so files per invocation.   Alternatively, you could do 
 DELETE Z*.*;* 
 DELETE Y*.*;*  ... 
 DELETE A*.*;*    (followed by the numbers).   ------------------------------  + Date: Wed, 17 Aug 2005 15:11:05 -0500 (CDT) * From: sms@antinode.org (Steven M. Schweda)Y Subject: Re: How many files can you have in a VMS directory without performance problems? 2 Message-ID: <05081715110563_20A0026A@antinode.org>  G    First, does anyone know why Info-VAX goes nuts on subjects from time ( to time?  (Or is the trouble elsewhere?)  - From: JF Mezei <jfmezei.spamnot@teksavvy.com>    > DELETE Z*.*;*  > DELETE Y*.*;*  > ...  > DELETE A*.*;*  >  > (followed by the numbers).  +    This scheme may be a bit obsolete.  See:   N       http://h71000.www7.hp.com/doc/731FINAL/4506/4506pro_014.html#index_x_890   I quote:   [...]     5.2.3.1.2 Extended Character Set  C In addition, OpenVMS V7.2 on Alpha systems and ODS-5 disks includes < support for the use of file names, and subdirectory and root? subdirectory names, that include all possible 8-bit characters, H excluding values 00 through 1F (hexadecimal) and excluding the following characters:            < > : / \ | ? * "   A OpenVMS 7.3-1 on Alpha systems and ODS--5 disks includes enhanced < support for the use of file names, and subdirectory and rootH subdirectory names. It supports all possible 8-bit characters, excluding& only the following two characters: ? *  F Note that the character set includes both the ISO Latin-1 C1 characterC set (hexadecimal 80 - 9F) as well as graphical and other characters C between 9F and FF. This allows the entire ISO Latin-1 character set D (with the exclusions noted above). In addition, there is support forG names that include any of the defined 16-bit Unicode characters (except * for the character exclusions noted above).   [...]   H    Even ignoring the Unicode problem, perhaps a loop would be worthwhile now.  H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode-org     Saint Paul  MN  55105-2547    ------------------------------  % Date: Wed, 17 Aug 2005 16:52:17 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> Y Subject: Re: How many files can you have in a VMS directory without performance problems? , Message-ID: <4303A36F.55D58659@teksavvy.com>   "Steven M. Schweda" wrote: > > DELETE Z*.*;*  > > DELETE Y*.*;*   - >    This scheme may be a bit obsolete.  See:   " > 5.2.3.1.2 Extended Character Set  J > excluding values 00 through 1F (hexadecimal) and excluding the following
 > characters:  >  >         < > : / \ | ? * "   J > subdirectory names. It supports all possible 8-bit characters, excluding( > only the following two characters: ? *    L Hum, this is starting to look like north american telephone number problems.  E It used to be that the second digit of a number would determine if an E area code was supplied or not. Area codes had either 0 or 1 as second G digit, and telephone number never had them. Then, they decided to ditch B this system in order to reclaim many possible telephone number and/ extend area codes (to allose for 888, 877 etc).   C If [ and ] are  valid characters for file names, how doees the file  system know if   	[chocolate]cake.dat  F If the file "[chocolate]cake.dat" located in the current directory, orE is it the file cake.dat which is located in the chocolate directory ?   & And since : is now allowed, how about:   CAKE:[chocolate]mousse.dat   ------------------------------  % Date: Wed, 17 Aug 2005 16:46:39 -0600 + From: Mark Berryman <mark@theberrymans.com> Y Subject: Re: How many files can you have in a VMS directory without performance problems? % Message-ID: <43035bdf$1@mvb.saic.com>    Steven M. Schweda wrote:I >    First, does anyone know why Info-VAX goes nuts on subjects from time * > to time?  (Or is the trouble elsewhere?)  2 Elsewhere.  Info-VAX does not rewrite the subject.  
 Mark Berryman  Info-VAX admin   ------------------------------  % Date: Wed, 17 Aug 2005 19:15:55 -0400 - From: William Webb <william.w.webb@gmail.com> Y Subject: Re: How many files can you have in a VMS directory without performance problems? 7 Message-ID: <8660a3a1050817161562a5723b@mail.gmail.com>   8 On 8/17/05, Mark Berryman <mark@theberrymans.com> wrote: > Steven M. Schweda wrote:K > >    First, does anyone know why Info-VAX goes nuts on subjects from time , > > to time?  (Or is the trouble elsewhere?) >=204 > Elsewhere.  Info-VAX does not rewrite the subject. >=20 > Mark Berryman  > Info-VAX admin >=20  5 I thought it was reacting to poster sanity bugchecks.    : ^ )    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: Wed, 17 Aug 2005 17:24:56 -0700 ( From: Jeff Cameron <roktsci@comcast.net>Y Subject: Re: How many files can you have in a VMS directory without performance problems? 0 Message-ID: <BF292368.12ABC%roktsci@comcast.net>  J > Isn't there a penalty paid for activating the DELETE.EXE image for every > file, 5 > instead of handing it a list, or a wildcarded list?   F Sure! There are many more image activations but when it comes to largeK directories the rev_del is substantially faster, plus which would you rater K have, lots of mapped page reads for image activation or many more .DIR file L block writes? REV_DEL is not a perfect solution, but it is a better one. TheG best solution would be to have the DELETE.EXE image delete files in the 7 reverse order either by default or by qualifier switch.   " Thanks for the spelling check too!   Jeff   On 8/17/05 6:52 AM, in articleD OF92188C2B.57900971-ON85257060.004C1540-85257060.004C3FEF@metso.com,8 "norm.raphael@metso.com" <norm.raphael@metso.com> wrote:   >  > E > Jeff Cameron <roktsci@comcast.net> wrote on 08/17/2005 09:10:04 AM:  > ! >> On 8/17/05 3:46 AM, in article ? >> 1124275575.684487.287580@z14g2000cwz.googlegroups.com, "RAQ"  >> <_raq_@yahoo.com> wrote:  >>   >>> Hi >>>  >>> Everytime I ask HP aboutG >>> "How many files can you have in a VMS directory without performance  >>> problems?" >>> H >>> they send me an article that is about 10-15 years old for VMS 5.2 orE >>> even 1.0, that describes that more than 300 files in a catalog is 
 >>> bad... >>> J >>> I have a feeling that <25.000 files works ok (no major complains), butF >>> when we reach 100.000 files it gets bad - and with 500.000 it gets >>> really nasty :)  >>>  >>> C >>> Does anyone know the real numbers? Or how I can calculate them? ? >>> Its for VMS 7.x + EVA SAN, and is mainly for file creation.  >>>  >>>  >>> 
 >>> Thanks >>> A >> There are basically three things that can cause directory size 
 > performance 
 >> issues.% >> 1. SYSGTEN parameter ACP_DIRCACHE.  >> 2. The speed of your disk. " >> 3. The speed of your processor.G >> The reason is that VMS keeps the directory entries sorted, and quite  > often J >> the adding and deleting of entries can cause the entire .DIR file to be0 >> flushed from cache and completely re-written. >>  G >> As mentioned by Chris Blackburn this is particularly noticeable when H >> deleting files. When you delete files from a directory, it is done inL >> alphabetical order, or from the top of the list. Depending on how much ofD >> the .DIR file is cached (ACP_DIRCACHE) when the cache needs to be
 > flushed,J >> this causes the entire .DIR file to be rewritten to disk. Since entries > wereI >> removed from the top, and the directory entries are all shifted up and  > eachG >> block of the directory file has now changed because the entries were 	 > shifted  >> up. >>  > >> This issue can be shown by my command procedure REV_DEL.COM> >> (http://support.mti.com/VMS2005cd/supportfiles/rev_del.txt) > J > Isn't there a penalty paid for activating the DELETE.EXE image for every > file, 5 > instead of handing it a list, or a wildcarded list?  > L > (Oh and you spelled deleting as deleteing in one of your comment lines :-| > .  > K >> which deletes files in the reverse order, that is from the bottom of the  > dir J >> file. That has the effect of not having to rewrite the entire .DIR file > withF >> each cache flush, because the only portion of the DIR file that has	 > changed I >> is the end. So if you are deleting files (even a portion of the files)  > fromI >> the bottom or reverse order will take much less time because the upper 5 >> blocks will not be changing with each cache flush.  >>  K >> If you have a large .DIR file and you delete files in the normal fashion L >> with the /LOG qualifier, you will see a hand-full of files being deleted,I >> then a pause, then another hand-full, and another pause, over and over  > again I >> with each pause being when the .DIR file is being flushed back to disk G >> completely re-written. As the .DIR file gets smaller and smaller the  > flush K >> operation becomes shorter and shorter. If you are using MONITOR/FILE you  > can K >> see your directory cache hit rate toggle back and forth from 100% to low K >> percentages (again dependent on ACP_DIRCACHE settings). However deleting  > inJ >> the reverse order, you will not see these pauses, because the .DIR file > is4 >> only being truncated rather than being rewritten. >>  J >> SO back to your question, what is the number? It depends mainly on what > kindI >> of actions you are doing and if the changes are additions or deletions  > and K >> where they are being inserted/removed from the .DIR file. If the changes  > are K >> made near the end of the .DIR file then the consequences are low. If the J >> changes are at the top, then when a cache flush occurs more blocks must > beL >> rewritten to disk. Given this information you can see that increasing any > ofJ >> the the three items listed here (in this order of most effective first) > can 1 >> help you forgo the cost of directory flushing.  >>   >> 1. ACP_DIRCACHE size. >> 2. Disk speed.  >> 3. Processor speed. >>   >> Jeff Cameron  >>   >    ------------------------------  % Date: Wed, 17 Aug 2005 16:41:25 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> Y Subject: Re: How many files can you have in a VMS directory without performance problems? , Message-ID: <4303A0E3.9015CDA7@teksavvy.com>  G Do any operating systems suport directories containing large numbers of D files efficiently ? If so, in what way do they deal with directories differently than VMS now does ?    ------------------------------  % Date: Wed, 17 Aug 2005 17:56:07 -0400 ' From: Dave Froble <davef@tsoft-inc.com> Y Subject: Re: How many files can you have in a VMS directory without performance problems? 0 Message-ID: <11g7cc2145st9ae@corp.supernews.com>   Bob Koehler wrote:f > In article <ddv7ge$n24$1@rdel.co.uk>, "Chris Blackburn" <chris.blackburn@uk.thalesgroup.com> writes: > L >>I have noticed on our VMS boxes that perfomance begins to degrade when the >>number* >>of files in a directory exceeds 10,000.  >  > J >    Probably depends on ACP_DINDXCACHE and ACP_DIRCACHE.  From experience2 >    I don't trust AUTOGEN to manage these values. >   I Actually, the practical answer is 20-22 files.  Any more and they scrool  H off the screen when issuing the DIR command with one file per line.  :-)   --  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: Wed, 17 Aug 2005 15:52:37 -0700  From: Z <Z@no.spam> Y Subject: Re: How many files can you have in a VMS directory without performance problems? , Message-ID: <UcPMe.11806$0E5.10239@fe05.lga>   Colin Butcher wrote:M > You might need to discover DFU (see the Freeware CDs or Freeware download). C > Very useful for mass deletions and removing directory trees. Very  > informative too.  F And totally unsupported ... as in if you use it on a mission critical 4 system and you lose all your data, oh well, too bad.   ------------------------------  % Date: Wed, 17 Aug 2005 20:34:04 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> Y Subject: Re: How many files can you have in a VMS directory without performance problems? , Message-ID: <4303D75D.E69402A3@teksavvy.com>   FredK wrote:I > A modern file system design to replace ODS would use a b-tree structure  > instead of a linear list.   D is INDEXF.SYS also a culprit in the "slow" files system for VMS ? or; just the simple sequential file nature of directory files ?    ------------------------------  % Date: Wed, 17 Aug 2005 21:20:19 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> Y Subject: Re: How many files can you have in a VMS directory without performance problems? + Message-ID: <4303E231.7D05A2C@teksavvy.com>    FredK wrote:I > A modern file system design to replace ODS would use a b-tree structure  > instead of a linear list.   A When you guys built ODS-5, was there consideration to restructure E directory files ? Is all directory file access done through a limited H set of routines, or are there many utilities that bypass system services$ to access directory files directly ?    F Seems to me that there are enough changes to ODS5, especially max fileC name size that any utility that acesses the directory file directly F would need to be updated anyways, both for DEC supplied stuff and userD developped hacks. So if you're going to make a change that require a- change elsewhere, might as well make it big !     F What was/were the reasons for not going for a more efficient directory file structure ?   ------------------------------  + Date: Wed, 17 Aug 2005 16:12:42 -0500 (CDT) * From: sms@antinode.org (Steven M. Schweda)Y Subject: Re: How many files can you have in a VMS directory without performance roblems?  2 Message-ID: <05081716124256_20A0026A@antinode.org>  - From: JF Mezei <jfmezei.spamnot@teksavvy.com>    > "Steven M. Schweda" wrote:  6    He also provided a link to an appropriate document:  N       http://h71000.www7.hp.com/doc/731FINAL/4506/4506pro_014.html#index_x_890  . > [...] Area codes had either 0 or 1 as second3 > digit, and telephone number never had them. [...]   H    The area code map from 1948 looked much more orderly than the currentG one, too.  (I was heartbroken when I was switched from 612 to 651.)  As G I recall, all the details are in the Bell System Technical Journal from 
 that year.  E > If [ and ] are  valid characters for file names, how doees the file  > system know if >  > 	[chocolate]cake.dat > H > If the file "[chocolate]cake.dat" located in the current directory, orG > is it the file cake.dat which is located in the chocolate directory ?   ;    In case your Web browser is broken, here are some hints:   ( ALP $ dire /date /prot /size /sinc [...]   Directory LDA1:[000000]   E chocolate.DIR;1            1  17-AUG-2005 16:17:07.69  (RWE,RWE,RE,E)  ^[chocolate^]cake.dat;1 F                            1  17-AUG-2005 16:17:56.86  (RWED,RWED,RE,)   Total of 2 files, 2 blocks.    Directory LDA1:[chocolate]  F cake.dat;1                 1  17-AUG-2005 16:17:32.88  (RWED,RWED,RE,)   Total of 1 file, 1 block.   0 Grand total of 2 directories, 3 files, 3 blocks.    ( > And since : is now allowed, how about: >  > CAKE:[chocolate]mousse.dat  G    Left as an exercise for the reader.  (You should get out more, too.)   H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode-org     Saint Paul  MN  55105-2547    ------------------------------  % Date: Wed, 17 Aug 2005 21:11:06 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> Y Subject: Re: How many files can you have in a VMS directory withoutperformance problems?  , Message-ID: <4303E009.185D7FB3@teksavvy.com>   Jeff Cameron wrote: I > best solution would be to have the DELETE.EXE image delete files in the 9 > reverse order either by default or by qualifier switch.   ' (rest of full top costing crap deleted)   G I agree entirely. I don't want to have to guess "gee, should I just use E DELETE, or must I setup some symbol pointing to some utility I rarely G use, type help to make sure I understand the syntax properly ? and then  execute it".  E I'd like the brains to be in DELETE so that it automatically kicks in < and doesn't require the user to learn about another utility.   ------------------------------  % Date: Wed, 17 Aug 2005 18:26:43 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> " Subject: HP 3rd quarter financials, Message-ID: <4303B98B.CAA26BFB@teksavvy.com>  & Media are pointing to HP as a success.   Relevant to us:    ## Enterprise Storage and Servers  H   Enterprise Storage and Servers (ESS) reported revenue of $4.0 billion,= up 20% over the prior-year period. On a year-over-year basis, A industry-standard server revenue increased 28%, business-critical E systems (BCS) revenue grew 7% and networked storage revenue grew 15%. H Within BCS, revenue in HP Integrity servers grew 113% year-over-year andH HP-UX revenue grew 8%. ESS reported an operating profit of $150 million,D or 3.8% of revenue, up from a loss of $211 million in the prior-year period.    ##    ! HP-UX is mentioned. Why not VMS ?   E Interesting that IA64 thingies grew 113%, a sign of how low they were F last year but if overall, BCS (and HP-UX) grew 7 and 8% respectively, # IA64 doesn't yet have a huge place.       D Also, why is it Enterprise Storage and Servers instead of EnterpriseF Servers and Storage ? Is this a sign that the storage guy (is it stillE stallard ?) as more weight politically, or that storage is truly more 5 important financially or strategically than servers ?   B Note that HP at least has admitted that those IA64 thingies aren'tK "industry standard" since those are now reported separately as growing 28%.   * Of significance is that if 8086 is growing   ------------------------------  % Date: Wed, 17 Aug 2005 17:02:30 -0600  From: Dan Notov <d9nn0@hp.com>& Subject: Re: HP 3rd quarter financials, Message-ID: <4303c207$1@usenet01.boi.hp.com>   JF Mezei wrote: ( > Media are pointing to HP as a success. >  > Relevant to us:  >  > ##  > Enterprise Storage and Servers > J >   Enterprise Storage and Servers (ESS) reported revenue of $4.0 billion,? > up 20% over the prior-year period. On a year-over-year basis, C > industry-standard server revenue increased 28%, business-critical G > systems (BCS) revenue grew 7% and networked storage revenue grew 15%. J > Within BCS, revenue in HP Integrity servers grew 113% year-over-year andJ > HP-UX revenue grew 8%. ESS reported an operating profit of $150 million,F > or 3.8% of revenue, up from a loss of $211 million in the prior-year
 > period.  >  > ## >  > # > HP-UX is mentioned. Why not VMS ?  > G > Interesting that IA64 thingies grew 113%, a sign of how low they were H > last year but if overall, BCS (and HP-UX) grew 7 and 8% respectively, % > IA64 doesn't yet have a huge place. 9 24% of BCS revenue, which includes Alpha/HP9000/Integrity      > F > Also, why is it Enterprise Storage and Servers instead of EnterpriseH > Servers and Storage ? Is this a sign that the storage guy (is it stillG > stallard ?) as more weight politically, or that storage is truly more 7 > important financially or strategically than servers ?  Black Helicopters circling.   D > Note that HP at least has admitted that those IA64 thingies aren'tM > "industry standard" since those are now reported separately as growing 28%.  > , > Of significance is that if 8086 is growingF Significant growth in ISS platforms. All units are profitable, except 	 software.    ------------------------------  % Date: Wed, 17 Aug 2005 20:00:46 -0400 ' From: Dave Froble <davef@tsoft-inc.com> & Subject: Re: HP 3rd quarter financials0 Message-ID: <11g7jltkee11148@corp.supernews.com>   Dan Notov wrote: > JF Mezei wrote:  > ) >> Media are pointing to HP as a success.  >> >> Relevant to us: >> >> ## ! >> Enterprise Storage and Servers  >>K >>   Enterprise Storage and Servers (ESS) reported revenue of $4.0 billion, @ >> up 20% over the prior-year period. On a year-over-year basis,D >> industry-standard server revenue increased 28%, business-criticalH >> systems (BCS) revenue grew 7% and networked storage revenue grew 15%.K >> Within BCS, revenue in HP Integrity servers grew 113% year-over-year and K >> HP-UX revenue grew 8%. ESS reported an operating profit of $150 million, G >> or 3.8% of revenue, up from a loss of $211 million in the prior-year 
 >> period. >> ##  >> >>$ >> HP-UX is mentioned. Why not VMS ? >>H >> Interesting that IA64 thingies grew 113%, a sign of how low they wereI >> last year but if overall, BCS (and HP-UX) grew 7 and 8% respectively,  & >> IA64 doesn't yet have a huge place. > ; > 24% of BCS revenue, which includes Alpha/HP9000/Integrity   G Half full or half empty issue?  That indicates that 76% is dead Alpha,  * soon to be dead PA-RISC, Tandem, and such.  G >> Also, why is it Enterprise Storage and Servers instead of Enterprise I >> Servers and Storage ? Is this a sign that the storage guy (is it still H >> stallard ?) as more weight politically, or that storage is truly more8 >> important financially or strategically than servers ? >  > Black Helicopters circling.    Think they might take JF?   E >> Note that HP at least has admitted that those IA64 thingies aren't J >> "industry standard" since those are now reported separately as growing  >> 28%.  >>- >> Of significance is that if 8086 is growing  > H > Significant growth in ISS platforms. All units are profitable, except  > software.   I Now if you were a bean counter, you'd suggest discontinuing software and  " just sell the profitable hardware.  I Maybe the software should get some cut from the hardware sale?  How many   could you sell without the OS?   --  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: Wed, 17 Aug 2005 20:22:36 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> & Subject: Re: HP 3rd quarter financials, Message-ID: <4303D4AE.F038EE33@teksavvy.com>   Dan Notov wrote:G > Significant growth in ISS platforms. All units are profitable, except  > software.   " That was the other question I had:  D Does any VMS software/application/compilers belong to the "Software"B division, or is all the stuff a VMS customer would require from HP sold/responsability of BCS ?  G Also, if one looks at a product such as ALL-IN-1/Office Server which in F "maintenance mode" in India, is that a responsability of Ann McQuaid ?J Or is Ann McQuaid responsible only for the VMS operating system and SIPs ?   ------------------------------  % Date: Wed, 17 Aug 2005 20:36:46 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> & Subject: Re: HP 3rd quarter financials, Message-ID: <4303D7FF.88F923D4@teksavvy.com>   Dave Froble wrote: > > Black Helicopters circling.  >  > Think they might take JF?     . Someone else suggested I should go out more...  C So I am about to go outside , ride my bike to the local convenience @ store to get a chocolate bar. If I don't come back, it means theE helicopters have gotten me. If only I could be abducted by aliens, at E least I'd make front page of the World Weekly News. Black helicopters  don't make the news.   ------------------------------  % Date: Wed, 17 Aug 2005 17:24:15 -0400 0 From: Peter Sjoberg <peters38@techwiz.spamno.ca>  Subject: Re: Image restore fails> Message-ID: <pan.2005.08.17.21.24.14.694688@techwiz.spamno.ca>  ; On Mon, 15 Aug 2005 20:56:42 -0500, David J Dachtera wrote:    > Peter Sjoberg wrote: >>  > >> On Fri, 12 Aug 2005 22:25:14 -0500, David J Dachtera wrote: >>   >> > Peter Sjoberg wrote:  >> >> [snip]P >> >> The system I'm trying to restore is VMS 6.x while the lab system I'm doingJ >> >> it on is 7.3-1. I dn't have a VMS 6 cd anywhere here but on the real. >> >> system it was used with the same result. >> > >> > AH! That explains a lot!  >> >M >> > There is a significant change in BACKUP between V6.2 and V7.x that gives  >> > rise to this issue. >> >K >> > If you have a V6.2 CD, boot it up and try the restore again. Should do  >> > better. >> I seen that, E >> http://www3.sympatico.ca/n.rieck/docs/openvms_72_backup_bug.htm is 6 >> basically what I see but in my case it's all Alpha.L >> The hw that has the problem is an AlphaServer 1000A which require 6.2-1H3L >> or later. We could not find a cd with that version and booted up on a 7.2F >> cd (Tried a 6.2 but it failed) and done some of the tests from 7.2. > 8 > E-mail me privately and I can probably help you out...; Thanks everyone for advice etc but I now got it "resolved".   J When digging around I did remember that on a separate disk I had installed? 6.2-1H3 (the last 6.x ?)  specifically for the purpose of image D backup/restore. After booting there and doing a normal image restore8 everything came back to normal and the system booted up.  G The only outstanding question would be why 7.2 & 7.3 fails so badly and < how to avoid that in a future restore? Is there any switch II can add on my 6.x backup to make the 7.2 handle it correctly (/noalias) ?      > / > How to demung the reply-to should be obvious.    ------------------------------  % Date: Wed, 17 Aug 2005 15:21:07 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> % Subject: Re: Is OVMS going Bangalore? , Message-ID: <43038E15.3CDE5729@teksavvy.com>   Rob Young wrote:) >         And the most recent Inqieooops:  > + > http://www.theinquirer.net/?article=23750  > Q > "It seems that the Hitachi Itanium chipset is on it's last legs, if this were a O > medical TV show, they would have already turned up the voltage on the paddles  > twice.  D You could say that about the current Itanium chips too, since a dual) core model is supposed to arrive "soon".    F Now, when you consider the plans in 2007 to have a common chipset withH the 8086 line, it does make sense for chip manufacturer to drop plans toG continue making chipsets proprietary to the IA64 thing. IBM has already $ stopped a long time ago as I recall.  H > http://www.health-itworld.com/newsitems/2005/07-05/07-13-05-news-intel > P > "The company will announce two new Itanium 2 processors on Monday with HitachiN > Ltd., which demonstrated a chipset that supported a 667MHz front-side bus at- > the Spring Intel Developer Forum in March."   @ This is old news. It is the temporary IA64 thing with faster busG interface that will be used by the upcoming dual core model. And again, H it makes sense to state that Hitachi would abandon its older chip set ifD it has a new one for the 667mhz bus. Note that very few are actuallyG building systems with that temporary in-between chip. (HP isn't selling  any such systems).  G This chip probably exists only to test/validate the bus interfaces that & will be used by the dual core version.  / >         It's called agenda driven reporting.    F On the surface, for the first article, i might tend to agree. However,E when you consider that pipelines from starting a  project to having a E product available for customers is long, it is entirely possible that H the hew Hitachi chipsets for the dual core IA64 thing is something which@ was started long ago, while the Inquirer rumour is for strategicF decisions made recently to affect products (or lack thereof) a year or two down the road.  ! So they do not necessarily clash.     L >         Point is... the Inq may never be held accountable like the Pioneer  >         Press is by Powerline.  C I would disagree with this. If Inquirer reports something erroneous G about HP, HP has a the right, and in fact a duty to write back with the F real facts. All too often, HP decides not to argue and lets the rumourE go on. Whether becauise HP is incompetant or because it cannot deny a  true rumour, we don't know.    ------------------------------  % Date: Wed, 17 Aug 2005 16:12:35 -0500 & From: Thomas Wirt <twnews@kittles.com>) Subject: Re: Reading VMS SAS files on PC? E Message-ID: <a534a$4303a844$4367aba2$13680@msgid.meganewsservers.com>    Bernhard Fabricius wrote:    > Dear All,  > I > A long while ago our management decided to ditch SAS on VMS (as did SAS N > Institute a little later :-) and our setinits have long expired. Due warningN > was given. And extended by a year. Now SAS is GONE from VMS here, though SAS; > on Windows is much used especially against our Rdb bases.  > I > Sure enough - some chap turns up with 10 MO disks (yikes) with over 400 M > *.SASEB$DATA files that he insists on being able to read. I find an ancient N > MO drive, I find the right combination of 10 undocumented DIP switches and I- > actually manage to read the stuff onto VMS.  > G > But how do I read the files with SAS (8.x or 9.x) on Windows PCs? SAS K > Institute's on-line help is full of how to EXPORT from VMS. No go. I only N > have the .SASEB$DATA files and need to import/convert under Windows. JudgingJ > from the top of the files, we're talking SAS 6.07 on VAX-VMS 5.5-2, so IH > suppose all sorts of floating point represensation problems will arise  F I have not touched SAS in 7 years, but when I last did we used SAS on A both VMS and Windows every day.  I think that I recall a Windows  G module/proc that could read foreign SAS files such as those on VMS.  I  3 may simply be think of the transport format though.    >  > Any and all help appreciated!  >  > Cheers > 
 > Bernhard > bf at dmu dot dk >  >      --     Thomas Wirt  Operations Manager, IS Dept. Kittle's Home Furnishings  Indianapolis, IN   ------------------------------  % Date: Wed, 17 Aug 2005 19:18:13 -0400 - From: William Webb <william.w.webb@gmail.com> ) Subject: Re: Reading VMS SAS files on PC? 6 Message-ID: <8660a3a10508171618617e8a9@mail.gmail.com>  3 On 8/17/05, Thomas Wirt <twnews@kittles.com> wrote:  > Bernhard Fabricius wrote:  >=20
 > > Dear All,  > > K > > A long while ago our management decided to ditch SAS on VMS (as did SAS L > > Institute a little later :-) and our setinits have long expired. Due wa= rning L > > was given. And extended by a year. Now SAS is GONE from VMS here, thoug= h SAS = > > on Windows is much used especially against our Rdb bases.  > > K > > Sure enough - some chap turns up with 10 MO disks (yikes) with over 400 L > > *.SASEB$DATA files that he insists on being able to read. I find an anc= ientL > > MO drive, I find the right combination of 10 undocumented DIP switches = and I / > > actually manage to read the stuff onto VMS.  > > I > > But how do I read the files with SAS (8.x or 9.x) on Windows PCs? SAS L > > Institute's on-line help is full of how to EXPORT from VMS. No go. I on= lyL > > have the .SASEB$DATA files and need to import/convert under Windows. Ju= dging L > > from the top of the files, we're talking SAS 6.07 on VAX-VMS 5.5-2, so = I J > > suppose all sorts of floating point represensation problems will arise >=20G > I have not touched SAS in 7 years, but when I last did we used SAS on B > both VMS and Windows every day.  I think that I recall a WindowsH > module/proc that could read foreign SAS files such as those on VMS.  I5 > may simply be think of the transport format though.  >=20 > > ! > > Any and all help appreciated!  > > 
 > > Cheers > >  > > Bernhard > > bf at dmu dot dk > >  > >  >=20 >=20 > -- >=20
 > Thomas Wirt  > Operations Manager, IS Dept. > Kittle's Home Furnishings  > Indianapolis, IN >=20  F Whoops, I meant TRANSPORT format, not export format, thus proving once? and for all that my disclaimer about not being a SAS expert was  genuine.   WWWebb   --=20 C NOTE: This email address is only used for noncommerical VMS-related  correspondence. C All unsolicited commercial email will be deemed to be a request for 8 services pursuant to the terms and conditions located at# http://bellsouthpwp.net/w/e/webbww/    ------------------------------  % Date: Wed, 17 Aug 2005 20:16:42 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net>) Subject: Re: Reading VMS SAS files on PC? + Message-ID: <4303E17A.541F0003@comcast.net>    Bernhard Fabricius wrote:  >  > Dear All,  > I > A long while ago our management decided to ditch SAS on VMS (as did SAS N > Institute a little later :-) and our setinits have long expired. Due warningN > was given. And extended by a year. Now SAS is GONE from VMS here, though SAS; > on Windows is much used especially against our Rdb bases.  > I > Sure enough - some chap turns up with 10 MO disks (yikes) with over 400 M > *.SASEB$DATA files that he insists on being able to read. I find an ancient N > MO drive, I find the right combination of 10 undocumented DIP switches and I- > actually manage to read the stuff onto VMS.  > G > But how do I read the files with SAS (8.x or 9.x) on Windows PCs? SAS K > Institute's on-line help is full of how to EXPORT from VMS. No go. I only N > have the .SASEB$DATA files and need to import/convert under Windows. JudgingJ > from the top of the files, we're talking SAS 6.07 on VAX-VMS 5.5-2, so IH > suppose all sorts of floating point represensation problems will arise >  > Any and all help appreciated!    Likewise, I do not know SAS.  G That said, if you have the files on VMS now, what is their file format?   9 Can you post the result of DIRECTORY/FULL on one of them?    --   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: Wed, 17 Aug 2005 15:42:10 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> ) Subject: Re: Replacement for VT terminals , Message-ID: <43039303.2522690C@teksavvy.com>   "Stanley F. Quayle" wrote:& > Funny, my VT220 does the same thing. > E > When connected to a terminal server (DECserver), a SET PORT x BREAK 9 > DISABLED prevents a break signal from going to the VAX.   G Yep, and you still can send a break, but you need to explicitely enable H it on the terminal server port for that session. So it is in fact betterH than a real VT when some inexperienced operator can easily press ofn theA 5th top key thinking it s PF5, when in fact it is the break key !    ------------------------------    Date: 17 Aug 2005 12:55:41 -0500; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: SAMBA for VMS3 Message-ID: <0BPzu11KKu1h@eisner.encompasserve.org>   i In article <1124295005.518903.149560@g14g2000cwa.googlegroups.com>, "WhoDat?" <whohe@whoever.com> writes:  > Bob Koehler wrote:f >> In article <42FDE0E9.11775.59A6E1E@localhost>, "Stanley F. Quayle" <squayle@insight.rr.com> writes: >> >F >> > Shame you can't wait for OpenVMS Engineering to develop their own: >> > Samba implementation.  Should be ready to go Q2 2006. >>E >>    Your a couple decades too late.  And you've got the cart before  >>    the horse. >>D >>    The original author of Samba based it on packet sniffing while >>    running Pathworks. > 0 > You must have missed the news. The horse died.  ?    Yes, but then after the horse we still had the cart running.   >    Somebody must a snuck in one of those new horseless thingys'    while we were looking the other way.    ------------------------------  % Date: Wed, 17 Aug 2005 14:02:17 -0400 6 From: Brad Hamilton <brMadAhamPiltSon@coMmcaAstP.neSt>0 Subject: Re: SAMBA for VMS (The saga continues!)E Message-ID: <C6SdnZ2dnZ0x5raonZ2dnTfmnt6dnZ2dRVn-052dnZ0@comcast.com>    Bill Gunshannon wrote: <snip>I > You know exactly what I meant.  On Unix (any more, at least) you unpack J > the source, run a configure script that autmatically figures out what isF > and is not needed (to include looking at what is available in systemK > libraries) creates a Makefile and you type "make" and it (in the majority  > of cases) builds itself. > E > With this particular package (which is current and one would assume B > therefore state-of-the-art for VMS Freeware) it took a bucnch ofG > manipulating just to get the source unpacked into a working directory A > structure.  The system, BY DEFAULT, did not even provide enough F > resources for SYSTEM to compile the program.  The source as packagedG > was not buildable on a VAX (and I saw no notes or warnings explaining K > what might be required for a VAX as compared to Alpha or IA64.  Everytime N > I fix something I have to re-compile every single piece resulting in several0 > hours before I find out what the next snag is. >   E The build is much simpler on an Alpha; just unzip and build.  I know  I this doesn't help your current situation, but perhaps you could get hold  C of an Alpha for next semester.  The PWS's, XP1000's and DS10's are  ) pretty reasonably priced at islandco.com.   E > Rather than worrying about how to port Unix Open Source apps to VMS I > the community might be better off looking into developing some standard I > ways of configuring stuff so that the build process is less cumbersome. J > Like it or not, you are competing with Unix and Wndows and to an outsideD > observer, VMS does not come off as the easier system to work with. >   D But the end result is much more stable and secure than Windows (and I probably "Unix", as well).  No pain, no gain.	:-)  Actually, you've been  & through enough pain this week.	:-)	:-)   --   Bradford J. Hamilton "All opinions are my own" * "Lose the MAPS, and replace '-at-' with @"   ------------------------------   Date: 17 Aug 2005 18:46:39 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)0 Subject: Re: SAMBA for VMS (The saga continues!), Message-ID: <3mhf0fF16u5dcU1@individual.net>  E In article <C6SdnZ2dnZ0x5raonZ2dnTfmnt6dnZ2dRVn-052dnZ0@comcast.com>, 9 	Brad Hamilton <brMadAhamPiltSon@coMmcaAstP.neSt> writes:  >  > ? > The build is much simpler on an Alpha; just unzip and build.    C Being as the source unziped into a backup save_set I can't see how  ? that part could be any different on Alpha, VAX or IA64.  It was > BACKUP that didn't seem to want to create the proper directoryB structure.  My guess is there was some option that was left out of; the instructions that would have made it all work properly.    F >                                                              I know K > this doesn't help your current situation, but perhaps you could get hold  E > of an Alpha for next semester.  The PWS's, XP1000's and DS10's are  + > pretty reasonably priced at islandco.com.   B You must not read here much.  Like most other schools, there is no? interest whatsoever in the continued use of VMS.  My department A tolerates it so long as it costs them nothing in either budget or , my time (at least the portion they pay for!)   > F >> Rather than worrying about how to port Unix Open Source apps to VMSJ >> the community might be better off looking into developing some standardJ >> ways of configuring stuff so that the build process is less cumbersome.K >> Like it or not, you are competing with Unix and Wndows and to an outside E >> observer, VMS does not come off as the easier system to work with.  >> > ) > But the end result is much more stable    D Not based on comments here about problems people have had with SAMBAD so far.  I am gambling that it is going to work here and at least doH the rudimentary file sharing well enough to keep my users un-complaining if not completely happy.  A >                                        and secure than Windows     This is probable  F >                                                                (and  > probably "Unix", as well).     Much less likely.   K >                             No pain, no gain.	:-)  Actually, you've been  ( > through enough pain this week.	:-)	:-)  C Yeah, but I actually like working with VMS and think there is value C in the students having continued exposure to it.  Rest assured that F this opinion is not common here and I doubt anywhere else in academia.   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: Wed, 17 Aug 2005 15:17:36 -0400 6 From: Brad Hamilton <brMadAhamPiltSon@coMmcaAstP.neSt>0 Subject: Re: SAMBA for VMS (The saga continues!)0 Message-ID: <3a-dnZbCrvvNEJ7eRVn-iQ@comcast.com>   Bill Gunshannon wrote:G > In article <C6SdnZ2dnZ0x5raonZ2dnTfmnt6dnZ2dRVn-052dnZ0@comcast.com>, ; > 	Brad Hamilton <brMadAhamPiltSon@coMmcaAstP.neSt> writes:  >  >>? >>The build is much simpler on an Alpha; just unzip and build.   >  > E > Being as the source unziped into a backup save_set I can't see how  A > that part could be any different on Alpha, VAX or IA64.  It was @ > BACKUP that didn't seem to want to create the proper directoryD > structure.  My guess is there was some option that was left out of= > the instructions that would have made it all work properly.  >     E Forgot the BACKUP; but I just cut-'n-pasted the command from the web  I site, taking care, of course, to substitute dev:[dir...] with the proper  F disk and directory information (don't forget the trailing '...'; it's G necessary to create the directory tree - did you remember to do that?).    <snip>D > You must not read here much.  Like most other schools, there is noA > interest whatsoever in the continued use of VMS.  My department C > tolerates it so long as it costs them nothing in either budget or . > my time (at least the portion they pay for!) >   F Something more than mere "tolerance" must be at work here, if you are 2 allowed to do what you are doing.  Exploit it.	:-)  H Seriously though, David Turner and Sue Skonetski are reasonable folks - < HP and islandco, I smell a marketing opportunity here...	:-) HP just had a good quarter...    <snip>  E > Yeah, but I actually like working with VMS and think there is value E > in the students having continued exposure to it.  Rest assured that H > this opinion is not common here and I doubt anywhere else in academia. >   B "Opinion" be damned - this is the same "opinion" that bleats that I Windows is "good enough", despite every indication to the contrary.  Can  H you say, "Zotob", or "rbot", or...?	:-)  Every time anyone says Windows H is good enough, the correct response should be a chanting of the litany G of over ten years of virii, worms, and spam, the Holy Trinity of "good   enough".	:-)  3 I won't say much about "Unix", not knowing it well.    > bill >    >      --   Bradford J. Hamilton "All opinions are my own" * "Lose the MAPS, and replace '-at-' with @"   ------------------------------  + Date: Wed, 17 Aug 2005 14:18:00 -0500 (CDT) * From: sms@antinode.org (Steven M. Schweda)0 Subject: Re: SAMBA for VMS (The saga continues!)2 Message-ID: <05081714180022_20A0026A@antinode.org>  7 From: John Malmberg <malmberg@dskwld.zko.dec.compaq.hp>   H > >    Why does the use of the funny names constitute "The correct fix"? > D > Because it is the only way to prevent conflicts with the compiler F > optimizer and with shared images that may be loaded on any platform.  ?    My goal is to have source code which will use the system RTL C functions where available, and which will use substitutes where the E system RTL lacks something.  So far as I can tell, because the system B RTL function complement depends on local conditions, the victim is@ always required to compile the calling code (with its __CRTL_VERC dependence).  Given that, the added burden on having to compile the 4 callee (substitute function) code seems small to me.  I > It is only if you are using only object libraries and specific linking  K > instructions and also disabling the compiler optimizer that you can rely  H > on the original UNIX behavior of overlaying operating system supplied B > routines with ones in your program that are in separate modules.  B    I don't expect any overlaying.  I expect %CC-I-IMPLICITFUNC andE %LINK-I-UDFSYM when I try to use something which is missing, and some D kind of MULDEF if I try to supply something which is already there. G (Then I look at this week's header files to figure out how to condition # the code to remove the complaints.)   B    I don't see what sense it would make for the the compiler to beG optimizing somehow with regard to RTL functions which are not available + on the local system, but what do I know?      N > [...] you may not get the result you thought you were getting even on UNIX, B > even if you are not getting compile or link errors, if you were I > intending on having a routine do something different than the standard   > routine with the same name.   G    The intention is simply to substitute for a _missing_ RTL function.  E I'll gladly admit that a differently-working function with a standard  name would be a bad idea.   J > As OpenVMS improves their compliance with the X/Open UNIX specification,J > it will cause more problems with routines that are using these names as  > global symbols.   F    So far as I can see, an improved RTL will cause problems for anyoneF who provides any substitute function with any name, standard or goofy,G unless he can somehow divine what the magic __CRTL_VER value will be to H determine its availability.  I don't see what the goofy names buy you in this case either.   G > And if you want your code to work properly on a wide variety of UNIX  > > like systems you will also want to follow the same practice.  E    Which is why no one there needs to use "configure" scripts to test  everything in situ?   G    I can see where, if you wanted to build an object library containing G substitute functions, and always link with it, it would make some sense @ to use goofy names on the substitutes.  However, if a __CRTL_VERG dependence means that you have to compile almost everything anyway, the H value of such an object library would seem (to me) to be pretty limited.  H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode-org     Saint Paul  MN  55105-2547    ------------------------------  % Date: Wed, 17 Aug 2005 15:49:59 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 0 Subject: Re: SAMBA for VMS (The saga continues!), Message-ID: <430394D8.72AA1D7D@teksavvy.com>   John Malmberg wrote:G > So unless you want your source code or object modules to quit working F > with each improvement with the VMS CRTL in UNIX compliance, you willH > want to avoid using routines with the same names as in X/OPEN, even if. > they are not currently available on OpenVMS.    H One issue I had many years ago was that the VMS C compiler (VAX C at theF time) didn't have the standard "strdup" routine, so I wrote my own andA included it in a site library so that normal code would be linked E properly. Eventually, C did get up to speed and included strdup and I E was stuck now with duplicate definitions. It wasn't a huge deal since D the routine I had written had the same spec and function as the real1 strdup, so removing my own copy was a no brainer.   G But if there is a list of routines not currently implemented on VMS but G which we should consider as reserved words (such as strdup used to be), E perhaps it is the role of the VMS compiler people (or what is left of  them) to publish that list.    ------------------------------  # Date: Wed, 17 Aug 2005 21:17:17 GMT 7 From: John Malmberg <malmberg@dskwld.zko.dec.compaq.hp> 0 Subject: Re: SAMBA for VMS (The saga continues!)2 Message-ID: <xPNMe.10528$ix2.887@news.cpqcorp.net>   JF Mezei wrote:  > John Malmberg wrote: > G >>So unless you want your source code or object modules to quit working F >>with each improvement with the VMS CRTL in UNIX compliance, you willH >>want to avoid using routines with the same names as in X/OPEN, even if. >>they are not currently available on OpenVMS. >   I > But if there is a list of routines not currently implemented on VMS but I > which we should consider as reserved words (such as strdup used to be), G > perhaps it is the role of the VMS compiler people (or what is left of  > them) to publish that list.   G The list of UNIX routines is governed by the X/Open organization and I  4 posted the link to it previously.  Here it is again.  . http://www.unix.org/single_unix_specification/  F Free registration and probably cookies are required for access to the  online version of the standard.   F Of course new routines could be added to this list at any time by the E X/Open. but the process is generally slow.  But because HP is not in  G control of this standard, HP can not predict what will be added in the   future.   ? And not all *NIXes implement all of the X/Open routines either.   I If someone is writing replacement routines for missing functions on VMS,  E I would recommend making sure that the routines behave as closely as  " possible with the X/Open routines.   -John ! malmberg@dskwld.zko.dec.compaq.hp  Personal Opinion Only    ------------------------------  # Date: Wed, 17 Aug 2005 21:27:04 GMT 7 From: John Malmberg <malmberg@dskwld.zko.dec.compaq.hp> 0 Subject: Re: SAMBA for VMS (The saga continues!)3 Message-ID: <IYNMe.10531$bB2.7252@news.cpqcorp.net>    Bill Gunshannon wrote:5 > In article <fPHMe.10496$ng2.3581@news.cpqcorp.net>, < > 	John Malmberg <malmberg@dskwld.zko.dec.compaq.hp> writes: >> >>K > Buld one more time and see if that finally fixes it.  If I make this work J > is there anyone who would actually care about the problems I have had to > fix?  ? The place to post such things is on the SAMBA-VMS mailing list.   C That appears to the public forum where the person maintaining that  ' version of SAMBA for VMS is monitoring.   , He just posted a new version of 2.2.8 today.  C I already submitted fixes like that which were included in a SAMBA  F 2.2.12 upgrade and he decided not to include them in his distribution H and has been making other changes in his 2.2.8 port.  I did not test my 8 changes on anything other than Alpha and eventually I64.   -John ! malmberg@dskwld.zko.dec.compaq.hp  Personal Opinion Only    ------------------------------  # Date: Wed, 17 Aug 2005 22:01:04 GMT 7 From: John Malmberg <malmberg@dskwld.zko.dec.compaq.hp> 0 Subject: Re: SAMBA for VMS (The saga continues!)3 Message-ID: <AsOMe.10538$mA2.2997@news.cpqcorp.net>    Bill Gunshannon wrote: > I > You know exactly what I meant.  On Unix (any more, at least) you unpack J > the source, run a configure script that autmatically figures out what isF > and is not needed (to include looking at what is available in systemK > libraries) creates a Makefile and you type "make" and it (in the majority  > of cases) builds itself.  H That only works if someone has already debugged the configure script on G that version of UNIX so that it produces the right answer or something  , close enough that the bugs are not apparent.  F I have been monitoring the mailing lists and newsgroups of the people G maintaining those configure scripts.  One person makes a change to the  G configure script and then it takes at least a week before they get the  C configure script working again on all the *NIXes that are using it.   C If you are on a reasonably main-stream UNIX then it is likely that  E someone debugged the script ahead of you.  If not then the configure  J only has a slightly better chance of succeeding on it then it does on VMS.  I If you are one of the maintainers for a project on a specific *NIX, then  I anytime some changes the scripts, you have to debug the configure output  D to make sure that it is not doing something wrong.  And many of the . configure errors are only visible at run-time.  I In order to get the configure script to work on VMS or any platform, you  I have to know what the possible correct answers are.  And if you have the  , correct answers, you do not need the script.  E > With this particular package (which is current and one would assume . > therefore state-of-the-art for VMS Freeware)  H That version of SAMBA is far from current or even complete.  The person F that ported it did it for their personal/company use and then put the D result on his web site for other people to work.  They only did the E minimum to get the older SAMBA program working and did not implement   many of the features.   E > Rather than worrying about how to port Unix Open Source apps to VMS I > the community might be better off looking into developing some standard I > ways of configuring stuff so that the build process is less cumbersome. J > Like it or not, you are competing with Unix and Wndows and to an outsideD > observer, VMS does not come off as the easier system to work with.  ? The Open Source community has made the choice on how to do the  H configuration, and that method is one that requires many tools that are H simply not available on VMS and many of them only seem to be needed for  the configure process.  @ With the *NIX platforms there are volunteers that just maintain I build/test machines for the open source projects where they download the  1 daily changes and make sure that they still work.   H Currently the tools needed for the automatic download of the source are G not available on VMS, so those systems would have to be coupled with a    *NIX system to fetch the source.  F There is more than enough for people to do at any skill level if they  want to volunteer.   -John ! malmberg@dskwld.zko.dec.compaq.hp  Personal Opinion Only    ------------------------------  % Date: Wed, 17 Aug 2005 18:05:43 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 0 Subject: Re: SAMBA for VMS (The saga continues!), Message-ID: <4303B4A0.D8569994@teksavvy.com>   John Malmberg wrote:H > The list of UNIX routines is governed by the X/Open organization and I6 > posted the link to it previously.  Here it is again. > 0 > http://www.unix.org/single_unix_specification/  5 Fine. So I registered. And where is that list again ?    Which one of these ?   Base Definitions        1.Introduction       2.Conformance      3.Definitions      4.General Concepts       5.File Format         Notation      6.Character Set 
     7.Locale       8.Environment         Variables  
     9.Regular         Expressions      10.Directory         Structure and        Devices      11.General Terminal        Interface  
    12.Utility         Conventions      13.Headers         Or is it in some other section ?   ------------------------------  % Date: Wed, 17 Aug 2005 18:13:50 -0400 ' From: Dave Froble <davef@tsoft-inc.com> 0 Subject: Re: SAMBA for VMS (The saga continues!)0 Message-ID: <11g7ddbk0g56v2b@corp.supernews.com>    From 2 of John's posts:   John Malmberg wrote:  C I already submitted fixes like that which were included in a SAMBA  F 2.2.12 upgrade and he decided not to include them in his distribution H and has been making other changes in his 2.2.8 port.  I did not test my 8 changes on anything other than Alpha and eventually I64.    I > The list of UNIX routines is governed by the X/Open organization and I  6 > posted the link to it previously.  Here it is again. > 0 > http://www.unix.org/single_unix_specification/   So, check me on this.   E The X/Open organization, which I know nothing about, is setting some  G standards with respect to naming conventions, and specific names, with  > the purpose I'd guess to avoid conflicts on various platforms.  E The person recognized as a/the maintainer of SAMBA refused to accept  F changes that implemented usage of the standard names that would avoid ' conflicts, and are specified by X/Open.   G I may not have the above correct, but if I do, I have to again ask, is  5 this chaos called 'Open Software' really a good idea?    --  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: Wed, 17 Aug 2005 23:13:59 GMT 7 From: John Malmberg <malmberg@dskwld.zko.dec.compaq.hp> 0 Subject: Re: SAMBA for VMS (The saga continues!)3 Message-ID: <XwPMe.10542$Fw2.6546@news.cpqcorp.net>    Steven M. Schweda wrote:9 > From: John Malmberg <malmberg@dskwld.zko.dec.compaq.hp>  > G >>>   Why does the use of the funny names constitute "The correct fix"?  >>D >>Because it is the only way to prevent conflicts with the compiler F >>optimizer and with shared images that may be loaded on any platform. >  > A >    My goal is to have source code which will use the system RTL E > functions where available, and which will use substitutes where the G > system RTL lacks something.  So far as I can tell, because the system D > RTL function complement depends on local conditions, the victim isB > always required to compile the calling code (with its __CRTL_VERE > dependence).  Given that, the added burden on having to compile the 6 > callee (substitute function) code seems small to me.  C The __CRTL_VER should be in the CONFIG.H or equivalent.  Most Open  $ Source packages now use that flavor.  E The stuff I did with rsync (still slowly making progress) uses a DCL  C procedure to generate the CONFIG.H and the automatically generated  I CONFIG.H contains #include "config_vms.h".  The config_vms.h is manually  ? edited for the things that the DCL procedure can not calculate.   I If you are manually generating a CONFIG.H file, then you can tune it for  ( the VMS versions that you are targeting.  G The __CRTL_VER symbol is one thing that you can use at compile time to  G determine that the OS routine is now available.  The alternative is to  K have a configure script that tries to figure out what is or is not present.   I >>It is only if you are using only object libraries and specific linking  K >>instructions and also disabling the compiler optimizer that you can rely  H >>on the original UNIX behavior of overlaying operating system supplied B >>routines with ones in your program that are in separate modules. >   D >    I don't expect any overlaying.  I expect %CC-I-IMPLICITFUNC andG > %LINK-I-UDFSYM when I try to use something which is missing, and some F > kind of MULDEF if I try to supply something which is already there. I > (Then I look at this week's header files to figure out how to condition % > the code to remove the complaints.)   H The MULDEF is what causes angry posters to show up on comp.os.vms after I they upgrade VMS and then try to install an existing package.  And it is  K easily avoided by treating the UNIX library names as reserved symbol names.   : When a problem is easy to prevent, it should be prevented.  D >    I don't see what sense it would make for the the compiler to beI > optimizing somehow with regard to RTL functions which are not available - > on the local system, but what do I know?      H The compiler is maintained on a different release schedule than for the H operating system, and has components that are common to other operating  systems.  H Therefore the compiler may have knowledge of things that are not yet in F the operating system.  For some calls like the "printf" family, there ; are common optimizations that apply to all of the variants.   A In the case of snprintf(), the compiler knew about it and how to  > optimize it for while before it finally showed up in the CRTL.   > N >>[...] you may not get the result you thought you were getting even on UNIX, B >>even if you are not getting compile or link errors, if you were I >>intending on having a routine do something different than the standard   >>routine with the same name.  >  > I >    The intention is simply to substitute for a _missing_ RTL function.  G > I'll gladly admit that a differently-working function with a standard  > name would be a bad idea.   6 But what is missing today may not be missing tomorrow.  G And substituting differently working functions is common for debugging  3 malloc() type issues or adding profiling libraries.   I In these cases the public symbol name needs to be unique, and a macro is  < used to cause it to be used instead of the standard library.  A In some cases, I have seen replacement routines used because the  I configure script determines that the supplied run-time routine has a bug  G in it.  Of course the configure scripts usually do not know how to set  H the options needed for a platform to generate a call to a fixed routine  that it supplies.    > J >>As OpenVMS improves their compliance with the X/Open UNIX specification,J >>it will cause more problems with routines that are using these names as  >>global symbols.  >   H >    So far as I can see, an improved RTL will cause problems for anyoneH > who provides any substitute function with any name, standard or goofy,I > unless he can somehow divine what the magic __CRTL_VER value will be to J > determine its availability.  I don't see what the goofy names buy you in > this case either.   H If you need to provide a routine because the CRTL is missing foo(), you . provide the routine with a name like my_foo().  % Then in the config.h file you put in:    #define foo my_foo  H Now your source code will compile/link/run with the current versions of 8 VMS and any following updates for the CRTL that show up.  H Now if a new version of VMS shows up with foo() implemented on it, then G you can either put in a test in your configure script to detect it, or  G if you are like most people that are manually maintaining the config.h  H files you can put the __CRTL_VER test in so that it does not substitute ! the routine on the newer version.   H This allows you to have one source code distribution that will continue , to work on older version of VMS or the CRTL.  G >>And if you want your code to work properly on a wide variety of UNIX  > >>like systems you will also want to follow the same practice. >   G >    Which is why no one there needs to use "configure" scripts to test  > everything in situ?   D I do not understand your statement.  The duplicate symbol issue has @ shown up on the SAMBA-TECHNICAL list a few times because it has B prevented the building of SAMBA on a particular UNIX that was not ! participating in the daily build.   G In this case the replacement routines in SAMBA were needed because the  B ones supplied on the platform did not pass the configure test for F correct results, but the link failed because of the duplicate symbols.  C And it seems to take at least a week for even minor changes to the  G configure scripts to start producing apparently correct results on all  , platforms participating in the daily builds.  E And many errors in configure scripts simply go unnoticed until it is  2 discovered that some routine is not working right.  I >    I can see where, if you wanted to build an object library containing I > substitute functions, and always link with it, it would make some sense B > to use goofy names on the substitutes.  However, if a __CRTL_VERI > dependence means that you have to compile almost everything anyway, the J > value of such an object library would seem (to me) to be pretty limited.  G The __CRTL_VER is used when you are manually editing the config.h file  D so that it can support older versions of VMS as well as the current.  G If you are using a configure script to test what is there, then you do   not need to use the __CRTL_VER.   H In the case of SAMBA 2.2.8, the config.h file is manually generated, so H using the __CRTL_VER is the way to determine if the substitute routines ' should be used by the rest of the code.   H The COMPILE.COM and LINK.COM can also be conditionalized not to include @ the module with the substitute routines if a full build is done.  D But by having the goofy names, object modules that are built on the H oldest version of VMS supported can be relinked on the newest and still C work the same way.  And SAMBA 2.2.x needs to be linked against the  3 system image, so .EXE files can not be distributed.   H If you use the "goofy" names, then your if your source code builds on a I given version of VMS, it will continue to build on VMS unless someone at  G X/Open just happens to decide to name a function with the "goofy" name  G that you picked and that function was needed enough that it got put in   to VMS.   E If you use the same names for missing routines, then you may have to  H edit your source code and recompile before you can relink after any ECO  or VMS upgrade.   E We have been having people showing up complaining about these easily  D preventable multiple global symbol warnings on comp.os.vms at least * every 6 months for the past several years.  @ For people that use freeware that do not have a compiler or the G programming skills to do a rebuild of the project, then this becomes a  F show-stopper bug until they can contact who ever supplied the project  to get a new kit.   H And some freeware with this problem has been only distributed in object  module form.   -John ! malmberg@dskwld.zko.dec.compaq.hp  Personal Opinion Only    ------------------------------  # Date: Wed, 17 Aug 2005 23:19:43 GMT 7 From: John Malmberg <malmberg@dskwld.zko.dec.compaq.hp> 0 Subject: Re: SAMBA for VMS (The saga continues!)3 Message-ID: <jCPMe.10543$4E2.6368@news.cpqcorp.net>    JF Mezei wrote:  > John Malmberg wrote: > H >>The list of UNIX routines is governed by the X/Open organization and I6 >>posted the link to it previously.  Here it is again. >>0 >>http://www.unix.org/single_unix_specification/ >  > 7 > Fine. So I registered. And where is that list again ?  >    "System Interfaces"   H On my browser it is showing up in the upper left corner near the search " box in the section labled "INDEX".   -John ! malmberg@dskwld.zko.dec.compaq.hp  Personal Opinion Only    ------------------------------  # Date: Wed, 17 Aug 2005 23:43:32 GMT 7 From: John Malmberg <malmberg@dskwld.zko.dec.compaq.hp> 0 Subject: Re: SAMBA for VMS (The saga continues!)3 Message-ID: <EYPMe.10546$MF2.3483@news.cpqcorp.net>    Dave Froble wrote: >  From 2 of John's posts: >  > John Malmberg wrote: > E > I already submitted fixes like that which were included in a SAMBA  H > 2.2.12 upgrade and he decided not to include them in his distribution J > and has been making other changes in his 2.2.8 port.  I did not test my : > changes on anything other than Alpha and eventually I64. >   J >> The list of UNIX routines is governed by the X/Open organization and I 7 >> posted the link to it previously.  Here it is again.  >>1 >> http://www.unix.org/single_unix_specification/  >    > So, check me on this.  > G > The X/Open organization, which I know nothing about, is setting some  I > standards with respect to naming conventions, and specific names, with  @ > the purpose I'd guess to avoid conflicts on various platforms.  I Apparently you have something better to do that track who owns UNIX this  G week.  X/Open owns the UNIX trademark and the rights to define what is  G or is not UNIX.  I have lost track of everyone that has owned it since   ATT sold it.  H The X/Open not only defines the names, they define what each routine is E required to do, and what error codes it must return, and what it can   optionally do.  G > The person recognized as a/the maintainer of SAMBA refused to accept  H > changes that implemented usage of the standard names that would avoid ) > conflicts, and are specified by X/Open.   6 The official maintainers of SAMBA did put in that fix.  E The person that ported the SAMBA 2.2.8 code to VMS never asked to be  H recognized as a maintainer of SAMBA for VMS, and only put what they did H on the www to help others.  He has equal standing with anyone else that + wants to put up a version of SAMBA for VMS.   G He puts up fixes and changes as he finds time for, and the priority is  H apparently for problems that are bothering him or have the potential to  bother him.   I > I may not have the above correct, but if I do, I have to again ask, is  7 > this chaos called 'Open Software' really a good idea?   @ When done correctly it works pretty well.  Multiple testers and . developers find bugs and fixes rather quickly.  E When someone works by them self doing a quick port for their own for  F primarily their own use based on a period snapshots of an Open Source # product, the results are different.    -John ! malmberg@dskwld.zko.dec.compaq.hp  Personal Opinion Only    ------------------------------  % Date: Wed, 17 Aug 2005 19:44:06 -0400 ' From: Dave Froble <davef@tsoft-inc.com> 0 Subject: Re: SAMBA for VMS (The saga continues!)0 Message-ID: <11g7imn35su9jf6@corp.supernews.com>   John Malmberg wrote: > Dave Froble wrote: >  >>  From 2 of John's posts:  >> >> John Malmberg wrote:  >>F >> I already submitted fixes like that which were included in a SAMBA I >> 2.2.12 upgrade and he decided not to include them in his distribution  H >> and has been making other changes in his 2.2.8 port.  I did not test > >> my changes on anything other than Alpha and eventually I64. >>   >>I >>> The list of UNIX routines is governed by the X/Open organization and  : >>> I posted the link to it previously.  Here it is again. >>> 2 >>> http://www.unix.org/single_unix_specification/ >> >>   >> So, check me on this. >>H >> The X/Open organization, which I know nothing about, is setting some J >> standards with respect to naming conventions, and specific names, with A >> the purpose I'd guess to avoid conflicts on various platforms.  >  > K > Apparently you have something better to do that track who owns UNIX this   > week.   
 Hopefully.  C >  X/Open owns the UNIX trademark and the rights to define what is  I > or is not UNIX.  I have lost track of everyone that has owned it since   > ATT sold it. > J > The X/Open not only defines the names, they define what each routine is G > required to do, and what error codes it must return, and what it can   > optionally do. > H >> The person recognized as a/the maintainer of SAMBA refused to accept I >> changes that implemented usage of the standard names that would avoid  * >> conflicts, and are specified by X/Open. >  > 8 > The official maintainers of SAMBA did put in that fix. > G > The person that ported the SAMBA 2.2.8 code to VMS never asked to be  J > recognized as a maintainer of SAMBA for VMS, and only put what they did J > on the www to help others.  He has equal standing with anyone else that - > wants to put up a version of SAMBA for VMS.  > I > He puts up fixes and changes as he finds time for, and the priority is  J > apparently for problems that are bothering him or have the potential to 
 > bother him.  > J >> I may not have the above correct, but if I do, I have to again ask, is 8 >> this chaos called 'Open Software' really a good idea? >  > B > When done correctly it works pretty well.  Multiple testers and 0 > developers find bugs and fixes rather quickly. > G > When someone works by them self doing a quick port for their own for  H > primarily their own use based on a period snapshots of an Open Source % > product, the results are different.  >  > -John # > malmberg@dskwld.zko.dec.compaq.hp  > Personal Opinion Only   + It sounds reasonable.  Thanks for the info.    --  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: 18 Aug 2005 00:23:16 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)0 Subject: Re: SAMBA for VMS (The saga continues!), Message-ID: <3mi2nkF161jmmU1@individual.net>  0 In article <3a-dnZbCrvvNEJ7eRVn-iQ@comcast.com>,9 	Brad Hamilton <brMadAhamPiltSon@coMmcaAstP.neSt> writes:  > Bill Gunshannon wrote:H >> In article <C6SdnZ2dnZ0x5raonZ2dnTfmnt6dnZ2dRVn-052dnZ0@comcast.com>,< >> 	Brad Hamilton <brMadAhamPiltSon@coMmcaAstP.neSt> writes: >>   >>> @ >>>The build is much simpler on an Alpha; just unzip and build.  >>   >>  F >> Being as the source unziped into a backup save_set I can't see how B >> that part could be any different on Alpha, VAX or IA64.  It wasA >> BACKUP that didn't seem to want to create the proper directory E >> structure.  My guess is there was some option that was left out of > >> the instructions that would have made it all work properly. >>   > G > Forgot the BACKUP; but I just cut-'n-pasted the command from the web  K > site, taking care, of course, to substitute dev:[dir...] with the proper  H > disk and directory information (don't forget the trailing '...'; it's I > necessary to create the directory tree - did you remember to do that?).   C Of course not.  I'm used to archive packages doing that themselves. F Remember the computer is supposed to serve us not us them. And anyway,E how would I pre-create a directory structure when I have no idea what  it is?   >  > <snip>E >> You must not read here much.  Like most other schools, there is no B >> interest whatsoever in the continued use of VMS.  My departmentD >> tolerates it so long as it costs them nothing in either budget or/ >> my time (at least the portion they pay for!)  >>   > H > Something more than mere "tolerance" must be at work here, if you are 4 > allowed to do what you are doing.  Exploit it.	:-)  E Trust me, it is little more than tolerance. (actually, my interest in F legacy system like RSTS, RSX and VMS is seen as eccentricity.  welcome to the real world!)    > J > Seriously though, David Turner and Sue Skonetski are reasonable folks -   B Dave has already been instrumental in helping me get the VAX goingB by providing me with a Star Coupler for just the cost of shipping.  Something I greatly appreciated.  > > HP and islandco, I smell a marketing opportunity here...	:-) > HP just had a good quarter...   > HP has no real interest in academia.  I've been down the road.   >  > <snip> > F >> Yeah, but I actually like working with VMS and think there is valueF >> in the students having continued exposure to it.  Rest assured thatI >> this opinion is not common here and I doubt anywhere else in academia.  >> > D > "Opinion" be damned - this is the same "opinion" that bleats that K > Windows is "good enough", despite every indication to the contrary.  Can  J > you say, "Zotob", or "rbot", or...?	:-)  Every time anyone says Windows J > is good enough, the correct response should be a chanting of the litany I > of over ten years of virii, worms, and spam, the Holy Trinity of "good   > enough".	:-)  D Easy for you to say, but that "opinion" controls the budget and paysA my salary.  Some people here really need to get out more.  VMS is C irrelevant in the modern computing world.  Virii, "good enough" who A cares.  Windows has the industry by b***s and it would take a lot A to change that now.  Sadly, VMS is doing nothing and becomes more D irrelevant every day.  Some of us try to reverse that rend, but thatC is more wishful thinking than anything else (even I will admit that 0 some of my actions are not grounded in reality).   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: Wed, 17 Aug 2005 20:18:37 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 0 Subject: Re: SAMBA for VMS (The saga continues!), Message-ID: <4303D3BF.DA1600AA@teksavvy.com>   John Malmberg wrote: > "System Interfaces"   B Thanks. I would have never found it, I would have been looking for  something like "reserved names".   For the record, it is at: @ http://www.opengroup.org/onlinepubs/000095399/idx/functions.html   ------------------------------  % Date: Wed, 17 Aug 2005 21:27:30 -0400 ' From: "Main, Kerry" <Kerry.Main@hp.com> 0 Subject: RE: SAMBA for VMS (The saga continues!)R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB6B1F4D@tayexc19.americas.cpqcorp.net>   > -----Original Message-----$ > From: bill@triangle.cs.uofs.edu=20A > [mailto:bill@triangle.cs.uofs.edu] On Behalf Of Bill Gunshannon  > Sent: August 17, 2005 8:23 PM  > To: Info-VAX@Mvb.Saic.Com 2 > Subject: Re: SAMBA for VMS (The saga continues!) >=20  	 [snip...]    >=20@ > HP has no real interest in academia.  I've been down the road. >=20 > >=20
 > > <snip> > >=20H > >> Yeah, but I actually like working with VMS and think there is valueH > >> in the students having continued exposure to it.  Rest assured thatA > >> this opinion is not common here and I doubt anywhere else=20  > in academia. > >> > >=20H > > "Opinion" be damned - this is the same "opinion" that bleats that=20@ > > Windows is "good enough", despite every indication to the=20 > contrary.  Can=20 A > > you say, "Zotob", or "rbot", or...?	:-)  Every time anyone=20  > says Windows=20 @ > > is good enough, the correct response should be a chanting=20 > of the litany=20< > > of over ten years of virii, worms, and spam, the Holy=20 > Trinity of "good=20  > > enough".	:-) >=20F > Easy for you to say, but that "opinion" controls the budget and paysC > my salary.  Some people here really need to get out more.  VMS is E > irrelevant in the modern computing world.  Virii, "good enough" who C > cares.  Windows has the industry by b***s and it would take a lot C > to change that now.  Sadly, VMS is doing nothing and becomes more F > irrelevant every day.  Some of us try to reverse that rend, but thatE > is more wishful thinking than anything else (even I will admit that 2 > some of my actions are not grounded in reality). >=20 > bill >=20    G Reminds me of all those in 1988 who stated that there was absolutely no G way that anything but WordPerfect would ever be the std word processing H pkg - especially in the Govt. If you were to suggest that something elseE might come along, those "experienced techies and analysts" would have 8 smiled knowingly and thought "he just doesn't get it .."   Things change.=20   H Heck, 5 years ago, Linux was hardly anything but a lab experiment in theD back labs. 7 Years ago it was not even on analysts radar as an OS to look at.   Things change.   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: Wed, 17 Aug 2005 20:14:32 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net>0 Subject: Re: SAMBA for VMS (The saga continues!)+ Message-ID: <4303E0F8.C385E1F3@comcast.net>    Bill Gunshannon wrote: > [SNIP]J > So, one last time.  Is there anywhere I can find a version of SAMBA thatF > will actually build and run on a VAX or am I really wasting my time?  ? Last I looked, the JYC kits came with .OLB files and a LINK.COM # procedure. Dunno if that's changed.   H I'm going to try my Samba repackaging proc. again in the hope that I canB make some CD-Rs to take to N'awlins with me. The resulting installF procedure is intended to install (including LINKing) directly from theF CD without copying all kinds of stuff to your system's disk(s) just to link and install.   G I don't have a VAX running right now, but if time permits, I'll try the @ install procedure on OpenVMS-VAX V7.3 and see if it still works.   --   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: 18 Aug 2005 01:57:15 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)0 Subject: Re: SAMBA for VMS (The saga continues!), Message-ID: <3mi87rF16m5v6U1@individual.net>  + In article <4303E0F8.C385E1F3@comcast.net>, 5 	David J Dachtera <djesys.nospam@comcast.net> writes:  > Bill Gunshannon wrote:	 >> [SNIP] K >> So, one last time.  Is there anywhere I can find a version of SAMBA that G >> will actually build and run on a VAX or am I really wasting my time?  > A > Last I looked, the JYC kits came with .OLB files and a LINK.COM % > procedure. Dunno if that's changed.   D There are both .OLB and source.  The .OLB did not build successfullyF and I chose to try to fix the source distribution rather than the .OLB stuff.   > J > I'm going to try my Samba repackaging proc. again in the hope that I canD > make some CD-Rs to take to N'awlins with me. The resulting installH > procedure is intended to install (including LINKing) directly from theH > CD without copying all kinds of stuff to your system's disk(s) just to > link and install.  > I > I don't have a VAX running right now, but if time permits, I'll try the B > install procedure on OpenVMS-VAX V7.3 and see if it still works.  H I have successfully built 2.2.8 from sources.  I will attempt to installD it tomorrow, if I have the time, and will report on wether or not itH actually works.  If it does, I will make an attempt to make it available+ to any other VAXophiles who are interested.    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, 18 Aug 2005 01:55:41 GMT   From: John Santos <john@egh.com>0 Subject: Re: SAMBA for VMS (The saga continues!)* Message-ID: <xURMe.7468$Al5.5115@trnddc04>   Bill Gunshannon wrote:G > In article <C6SdnZ2dnZ0x5raonZ2dnTfmnt6dnZ2dRVn-052dnZ0@comcast.com>, ; > 	Brad Hamilton <brMadAhamPiltSon@coMmcaAstP.neSt> writes:  >  >>? >>The build is much simpler on an Alpha; just unzip and build.   >  > E > Being as the source unziped into a backup save_set I can't see how  A > that part could be any different on Alpha, VAX or IA64.  It was @ > BACKUP that didn't seem to want to create the proper directoryD > structure.  My guess is there was some option that was left out of= > the instructions that would have made it all work properly.  >     @ Worked fine last night on my VAX, except it didn't supersede old? versions of files (files in the save set with identical version C numbers to existing files on the disk), so I re-executed the backup G after adding a /REPLACE.  (Last time I did this, I hit the same thing,  ? but let it finish and followed up with a BACKUP/COMPARE, and it B turned out the old versions that didn't get replaced (i.e. because? they had identical version numbers) were identical so it didn't + matter.  I didn't do that check this time.)   = However, if you don't execute the correct backup command, you A will see exactly this symptom.  You need to use disk:[dir...] for > the output spec, substituting your real disk and directory for. "disk" and "dir", just like the web page says.     > F >>                                                             I know K >>this doesn't help your current situation, but perhaps you could get hold  E >>of an Alpha for next semester.  The PWS's, XP1000's and DS10's are  + >>pretty reasonably priced at islandco.com.  >  > D > You must not read here much.  Like most other schools, there is noA > interest whatsoever in the continued use of VMS.  My department C > tolerates it so long as it costs them nothing in either budget or . > my time (at least the portion they pay for!) >  > F >>>Rather than worrying about how to port Unix Open Source apps to VMSJ >>>the community might be better off looking into developing some standardJ >>>ways of configuring stuff so that the build process is less cumbersome.K >>>Like it or not, you are competing with Unix and Wndows and to an outside E >>>observer, VMS does not come off as the easier system to work with.  >>>  >>) >>But the end result is much more stable   >  > F > Not based on comments here about problems people have had with SAMBAF > so far.  I am gambling that it is going to work here and at least doJ > the rudimentary file sharing well enough to keep my users un-complaining > if not completely happy. >  > A >>                                       and secure than Windows   >  >  > This is probable >  > F >>                                                               (and  >>probably "Unix", as well).   >  >  > Much less likely.  >  > K >>                            No pain, no gain.	:-)  Actually, you've been  ( >>through enough pain this week.	:-)	:-) >  > E > Yeah, but I actually like working with VMS and think there is value E > in the students having continued exposure to it.  Rest assured that H > this opinion is not common here and I doubt anywhere else in academia. >  > bill >     F P.S. the compile is still running, but I have encountered some warningH messages about pointer type mismatches (sent to the SAMBA mailing list),E which might cause problems with the link, or may be fine.  Didn't get F these compiling it on Alpha two days ago.  And in the meantime, JY has; posted a new version, so my compile is all obsolete anyway!      --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  # Date: Thu, 18 Aug 2005 03:01:43 GMT   From: John Santos <john@egh.com>0 Subject: Re: SAMBA for VMS (The saga continues!)+ Message-ID: <rSSMe.18896$Rp5.6118@trnddc03>    David J Dachtera wrote:  > Bill Gunshannon wrote: >  >>[SNIP]J >>So, one last time.  Is there anywhere I can find a version of SAMBA thatF >>will actually build and run on a VAX or am I really wasting my time? >  > A > Last I looked, the JYC kits came with .OLB files and a LINK.COM % > procedure. Dunno if that's changed.   E It has.  According to the web site, he no longer has access to a VAX, % so is no longer providing VAX .OLB's.    > J > I'm going to try my Samba repackaging proc. again in the hope that I canD > make some CD-Rs to take to N'awlins with me. The resulting installH > procedure is intended to install (including LINKing) directly from theH > CD without copying all kinds of stuff to your system's disk(s) just to > link and install.  > I > I don't have a VAX running right now, but if time permits, I'll try the B > install procedure on OpenVMS-VAX V7.3 and see if it still works. >   D I have a VAX but no CDROM on it.  I also have an Alpha with a CDROM A drive that I planned to cluster with the VAX, but the CDrom drive  recently broke. :-(    --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  + Date: Thu, 18 Aug 2005 03:58:02 +0000 (UTC) % From: Dan Foster <usenet@evilphb.org> 0 Subject: Re: SAMBA for VMS (The saga continues!)5 Message-ID: <slrndg81rb.pfq.usenet@zappy.catbert.org>   M In article <rSSMe.18896$Rp5.6118@trnddc03>, John Santos <john@egh.com> wrote:  > G > It has.  According to the web site, he no longer has access to a VAX, ' > so is no longer providing VAX .OLB's.   H That's unfortunate; he could use SIMH or CHARON-VAX to work around this.   -Dan   ------------------------------    Date: 17 Aug 2005 11:55:53 -0700# From: "WhoDat?" <whohe@whoever.com> ' Subject: Re: simh define/attach printer C Message-ID: <1124304953.119728.185730@o13g2000cwo.googlegroups.com>    phil@rephil.org wrote: > Hi,  > G > I'm trying to set up simh to output text on a file, when printing.  I I > have looked, but am unable to find the info in a form I can comprehend. H >  Pardon the newbie-ish questions.  Also pardon the lengthy msg; I haveF > included what I hope is the necessary output from the system if someH > kind soul can help me straighten my head out.  (I'm sure the system is# > doing just what I tell it to!) ;)  >  > Some pertinent info: >    SIMH 3.4 VAX under Linux , >    VMS 7.2 (working well enough otherwise) > H > Q:  Under simh 3.4, is it necessary to 'attach lpt <output_file>' when+ > configuring the emulator?  (Did, no joy.)  > C > I am new at configuring queues, so this may be my problem, but on  > startup I get: > 
 > --------C > The OpenVMS VAX system is now executing the site-specific startup  > commands.  > : > %%%%%%%%%%%  OPCOM  16-AUG-2005 13:46:59.85  %%%%%%%%%%%* > Message from user QUEUE_MANAGE on ERSATZC > %QMAN-I-QUENOTSTART, queue SYS$PRINT could not be started on node  > ERSATZ > : > %%%%%%%%%%%  OPCOM  16-AUG-2005 13:46:59.90  %%%%%%%%%%%* > Message from user QUEUE_MANAGE on ERSATZ/ > -SYSTEM-W-NOSUCHDEV, no such device available  > / > %SYSTEM-W-NOSUCHDEV, no such device available ! > %JBC-E-NOSUCHQUE, no such queue H > %SET-I-INTSET, login interactive limit = 64, current interactive value > = 0 : >   SYSTEM       job terminated at 16-AUG-2005 13:47:02.72 > -----  >  > G > I ran START/QUEUE/MANAGER/NEW initially, so that seems to be alright. ' > The lines from SYSTARTUP_VMS.COM are:  >  > ! > ------------------------------- A > $ DEFINE /SYSTEM $PRINTER  _LPA0:     ! if using a line printer 0 > $ SET PRINTER /NOWRAP /NOTRUNCATE /CR $PRINTER< > $ SET DEVICE /SPOOLED=(SYS$PRINT, SYS$SYSDEVICE:) $PRINTER > $ ENABLE AUTOSTART /QUEUES > $ START /QUEUE SYS$PRINT! > -------------------------------  >   E Well, since you START /QUEUE SYS$PRINT it appears not be an autostart B queue. Are you INITIALIZEing the queue someplace before you try to	 start it?    ------------------------------  % Date: Wed, 17 Aug 2005 20:05:25 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net> Subject: Re: Squid for VMS+ Message-ID: <4303DED5.194ED82F@comcast.net>    Volker Englisch wrote: >  > Hi,  > L > I'm wondering if there is something like squid for OpenVMS? There seems toL > be no port of squid to this OS. Is there maybe another http-proxy software > for OpenVMS?  # Standard question: What is "squid"?   D UN*X-land programs are not guaranteed to be familiar or known to theG OpenVMS crowd. Please describe what the program does so that a parallel @ facility, if any, on VMS can be cited as a possible replacement.   --   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: Wed, 17 Aug 2005 14:26:04 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>  Subject: SYSMAN vs SYSGEN , Message-ID: <43038132.EDC10FB9@teksavvy.com>  1 Something that I have meant to ask for some time.   J on Alpha, SYSMAN has all mighty capabilities such as IO AUTOCONFIGURE/ALL.  D Is this implemented smiply by SYSMAN spawning a subprocess that runsA SYSGEN and does the AUTOCONFIGURE/ALL as well as all other SYSGEN B commands that SYSMAN emulates ? Or is there a "callable SYSGEN" onF Alphas that SYSMAN can use more directly, or does SYSMAN actually have4 all the guts/code necessary to do what SYSGEN does ?   ------------------------------  # Date: Wed, 17 Aug 2005 23:43:25 GMT * From: "FredK" <fred.nospam@nospam.dec.com> Subject: Re: SYSMAN vs SYSGEN 3 Message-ID: <xYPMe.10545$nE2.3276@news.cpqcorp.net>   : "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message& news:43038132.EDC10FB9@teksavvy.com...3 > Something that I have meant to ask for some time.  > L > on Alpha, SYSMAN has all mighty capabilities such as IO AUTOCONFIGURE/ALL. > F > Is this implemented smiply by SYSMAN spawning a subprocess that runsC > SYSGEN and does the AUTOCONFIGURE/ALL as well as all other SYSGEN D > commands that SYSMAN emulates ? Or is there a "callable SYSGEN" onH > Alphas that SYSMAN can use more directly, or does SYSMAN actually have6 > all the guts/code necessary to do what SYSGEN does ?    Invest in a source listings kit.  J The IO configuration capability is not in SYSGEN on Alpha, nor does SYSMANG invoke SYSGEN.  The IO command under SYSMAN invokes the IOGEN shareable L image, which contains the logic to configure devices and invokes one or moreH ICBMs, which contain the bus-specific configuration logic.  The original
 design wasH (typically) over-engineered, and not very friendly - see notes on ICBM's
 (they wereI originally platform-specific, and run in an elevated mode - making them a 
 lot of funH to debug).  The refinement of this was to create a general ICBM, so that	 only very L unusual cases would need a custom ICBM.  The general-purpose one today knowsJ how to configure all bus types in use, and uses a plain text ASCII file to define devices. I Some devices, such as SCSI and FC disks have additional support needed to F probe their busses, and this logic is called from the ICBM when such a	 device is  discovered.   G The Alpha/Itanium design is a much, much different design than the VAX.   I The details of the boot time and run time (plus cluster) configuration is  complex,C obscure, and for the practical user or driver provider - not really  important to know.  H If there are circumstances where a user application needs to call any of	 this code 0 by hand, I have yet to think of (or hear of) it.  L _Fred (who is responsible for the most part in the change to make it simple)   ------------------------------  # Date: Wed, 17 Aug 2005 23:53:14 GMT * From: "FredK" <fred.nospam@nospam.dec.com> Subject: Re: SYSMAN vs SYSGEN 3 Message-ID: <K5QMe.10547$RB2.2902@news.cpqcorp.net>   H "Bob Koehler" <koehler@eisner.nospam.encompasserve.org> wrote in message- news:4NwaXRGmTf98@eisner.encompasserve.org... 7 > In article <43038132.EDC10FB9@teksavvy.com>, JF Mezei & <jfmezei.spamnot@teksavvy.com> writes:5 > > Something that I have meant to ask for some time.  > > ; > > on Alpha, SYSMAN has all mighty capabilities such as IO  AUTOCONFIGURE/ALL. > > H > > Is this implemented smiply by SYSMAN spawning a subprocess that runsE > > SYSGEN and does the AUTOCONFIGURE/ALL as well as all other SYSGEN F > > commands that SYSMAN emulates ? Or is there a "callable SYSGEN" onJ > > Alphas that SYSMAN can use more directly, or does SYSMAN actually have8 > > all the guts/code necessary to do what SYSGEN does ? > F >   That code is in SYSMAN.  ALpha's SYSGEN doesn't have it.  Only VMS >   Engineering knows why. >   L Because the engineer designed the Alpha IO configuration in V1.0 decided to?    We are all grateful that he did.   ------------------------------  % Date: Wed, 17 Aug 2005 20:32:06 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>  Subject: Re: SYSMAN vs SYSGEN , Message-ID: <4303D6E7.3725D9BD@teksavvy.com>   FredK wrote:L > The IO configuration capability is not in SYSGEN on Alpha, nor does SYSMANI > invoke SYSGEN.  The IO command under SYSMAN invokes the IOGEN shareable N > image, which contains the logic to configure devices and invokes one or more= > ICBMs, which contain the bus-specific configuration logic.    F Thanks. So if I understand correctly, everytime you export an Alpha orD IA64 thing running VMS to another country, you send ICBMs with it...  H The reason I asked was that SYSMAN also has the ability to change SYSGEN, parameters which are less "device" oriented.  D So, you've explained well how/why you moved the AUTOCONFIGURE (and I3 suspect SHOW DEVICE/DRIVERS) from SYSGEN to SYSMAN.   E But what about SYSMAN's managing the SYSGEN parameters ? What was the * background/reason/implementation on this ?B Is there a sysgen sharable image used by both sysgen and sysman to handle those parameters ?   G And what about in a sysboot environment, is that still sysgen under the  hood ?      J > If there are circumstances where a user application needs to call any of > this code 2 > by hand, I have yet to think of (or hear of) it.  E Not really, but when doing system management, one wants to understand D some of the reasons and implementation of changes, especially if youF know that you may be called upon to change in-house documentstion, DCL scripts etc.   ------------------------------  % Date: Wed, 17 Aug 2005 20:01:25 -0500 2 From: David J Dachtera <djesys.nospam@comcast.net>* Subject: Re: Turning up an Ethernet & tftp+ Message-ID: <4303DDE5.95C08C39@comcast.net>    Neil Cherry wrote: > 8 > On Tue, 16 Aug 2005 08:25:18 -0500, Neil Cherry wrote:? > > On Sat, 13 Aug 2005 23:08:26 -0500, David J Dachtera wrote:  > >> Alan Greig wrote: > >>>  > >>> David J Dachtera wrote:  > >>> >  > >>> >   > >>> > Do you get a link light? > 1 > Both interfaces had the link lights up (green).   9 If these are 10/100 cards, are the high-speed lights lit?   ( What does the console think about them?    Post the output of:    P00>>> show ew*   4 ...if possible, and, with VMS booted, the output of:   $ MC LANCP SHOW DEV/ALL/CHAR   --   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: Wed, 17 Aug 2005 15:03:51 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> @ Subject: Re: Was -->  Print Queue in "starting" state -- cause??, Message-ID: <43038A0B.EBF953BA@teksavvy.com>  O > > /AUTOSTART_ON=(XXXXX1::"ISAB:9100",XXXXX2::"ISAB:9100",XXXXX3::"ISAB:9100")     " Based on HELP INIT/QUEUE/AUTOSTART  G doing the ENABLE AUTOSTART/QUEUES will start the queue on the node with 7 the highest rating (eg: the first one listed, XXXXX1).    D Does node XXXXX1 have what it takes to run the queue ? (TCPIP up and running etc etc) ?  F From which node are you issuing ENABLE AUTOSTART/QUEUE command and the START/QUEUE which hangs ?   E To answer the specific question, because you use CTRL-Y to cancel the H START/QUEUE, it explains why it stays in STARTING stte forever since you9 didn't give it a chance to complete the starting process.   H Have you cosnidered letting it START/QUEUE overnight and see what outputA may have be sent the next morning ? OPCOM may have a QUEUE_MANAGE E message about the queue once some timeout occurs. (and you may have a E message on the terminal that issued the START/QUEUE explaining why it + failed after X time). You may also wish to    
 $SHOW TIME $START/QUEUE xxxx  $ SHOW TIME   F ((eg: enter a SHOW TIME command in the typeahead buffer so you have an4 idea of when the command has completed (with error).   ------------------------------   End of INFO-VAX 2005.459 ************************