0 INFO-VAX	Wed, 08 Feb 2006	Volume 2006 : Issue 78      Contents:
 Access issuse  Re: Access issuse  Re: Access issuse  Re: Another sad day for VMS... Re: Another sad day for VMS...* Re: Cisco to sink: names Capellas to board& Different console prompt "_>" on AS600& Re: Hobbyist kit installation problems& Re: Hobbyist kit installation problems& Re: Hobbyist kit installation problems& Re: Hobbyist kit installation problems& Re: Hobbyist kit installation problems& Re: Hobbyist kit installation problems& Re: Hobbyist kit installation problems$ Re: is this a bug or feature of ssh?$ Re: is this a bug or feature of ssh?( Most recent firmware version for HSZ80 ? Re: null terminated strings  Re: null terminated strings  Re: null terminated strings  RE: null terminated strings  Re: null terminated strings  Re: null terminated strings  Re: null terminated strings  Re: null terminated strings  Re: null terminated strings  RE: null terminated strings  Re: null terminated strings  Re: null terminated strings  Re: null terminated strings  Re: null terminated strings P OpenVMS.org Contributing Editor Robert Gezelter to speak on Software ArchitecturP Re: OpenVMS.org Contributing Editor Robert Gezelter to speak on Software ArchiteP Re: OpenVMS.org Contributing Editor Robert Gezelter to speak on Software Archite> Re: OT: English grammar and the state of today's education :-)> Re: OT: English grammar and the state of today's education :-) Re: PDPs in the news!  Re: PDPs in the news!  Re: PDPs in the news!  Re: PDPs in the news! ( Re: perl build error (& subscribe error)! Question about Greece/EU Warranty ! Re: sneak preview - BACKUP/REPAIR ! Re: sneak preview - BACKUP/REPAIR ) Re: Using SDA to view Global Section Data ) Re: Using SDA to view Global Section Data   Re: VAXstation 2000 boot devices Re: vaxstation 4000vlc Re: vaxstation 4000vlc= Re: VS3100 SCSI disks (was: re: VAXstation 2000 boot devices)   F ----------------------------------------------------------------------   Date: 8 Feb 2006 03:36:32 -0800  From: santhoo.jay@gmail.com  Subject: Access issuseB Message-ID: <1139398592.768317.85890@g44g2000cwa.googlegroups.com>  E I have directory in of the VMS server  . whose details are below ....   i ---------------------------------------------------------------------------------------------------------  NRC002::$ dir FEED.DIR;1/ful    Directory PW_DISK:[000000.SHARE]  1 FEED.DIR;1                    File ID:  (848,3,0) 6 Size:          217/324        Owner:    [PWRK$DEFAULT]" Created:   19-NOV-1998 16:13:36.64( Revised:    7-FEB-2006 12:57:23.28 (253) Expires:   <None specified> " Backup:     4-FEB-2006 14:03:59.34 Effective: <None specified>  Recording: <None specified>  File organization:  Sequential Shelved state:      OnlineG File attributes:    Allocation: 324, Extend: 0, Global buffer count: 0, 4 No default version limit, Contiguous, Directory fileC Record format:      Variable length, maximum 512 bytes, longest 512  bytes 4 Record attributes:  No carriage control, Non-spanned RMS attributes:     None Journaling enabled: NoneA File protection:    System:RWED, Owner:RWED, Group:RWED, World:RE  Access Cntrl List:> (IDENTIFIER=APP_LP_FEEDSTOCK,ACCESS=READ+WRITE+EXECUTE+DELETE)  N (IDENTIFIER=APP_LP_FEEDSTOCK,OPTIONS=DEFAULT,ACCESS=READ+WRITE+EXECUTE+DELETE)/                     Suppressed 3 PATHWORKS ACES     Total of 1 file, 217/324 blocks. -----------------------------------------------------------------------------------------------------------------------------------   F Now i have  a particular user "BOSCH1"  who  is  neither a owner nor a group member .> He belongs to world . I have to give  him RWED access  to thisD directory . But i do not want make him system user or owner . also i: cannot add him to group because i have to change his UIC .D So my question is how can i give him RWED access keeping in mind the above conditions? .  Isn't it  tricky ? .... 0 It could be a great help from you guys .........   ------------------------------  $ Date: Wed, 8 Feb 2006 11:45:15 -0000* From: "Richard Brodie" <R.Brodie@rl.ac.uk> Subject: Re: Access issuse2 Message-ID: <dsclkb$e0k$1@blackmamba.itd.rl.ac.uk>  ) <santhoo.jay@gmail.com> wrote in message  < news:1139398592.768317.85890@g44g2000cwa.googlegroups.com...  F > So my question is how can i give him RWED access keeping in mind the > above conditions? .   = Giving him the APP_LP_FEEDSTOCK right fulfils the conditions. @ If that isn't a fine grained enough right then make another one.E Or you could just add an ACL giving him permissions explicitly to his A UIC: it's usually considered good practice to not do that though.    ------------------------------  # Date: Wed, 08 Feb 2006 17:03:19 GMT 7 From: John Malmberg <malmberg@dskwld.zko.hp.compaq.dec>  Subject: Re: Access issuse2 Message-ID: <rvpGf.2821$b01.2503@news.cpqcorp.net>   santhoo.jay@gmail.com wrote:G > I have directory in of the VMS server  . whose details are below ....  > k > ---------------------------------------------------------------------------------------------------------  > NRC002::$ dir FEED.DIR;1/ful > " > Directory PW_DISK:[000000.SHARE] > 3 > FEED.DIR;1                    File ID:  (848,3,0) 8 > Size:          217/324        Owner:    [PWRK$DEFAULT]C > File protection:    System:RWED, Owner:RWED, Group:RWED, World:RE  <snip> > Access Cntrl List:@ > (IDENTIFIER=APP_LP_FEEDSTOCK,ACCESS=READ+WRITE+EXECUTE+DELETE) > P > (IDENTIFIER=APP_LP_FEEDSTOCK,OPTIONS=DEFAULT,ACCESS=READ+WRITE+EXECUTE+DELETE)1 >                     Suppressed 3 PATHWORKS ACES  > " > Total of 1 file, 217/324 blocks. > -----------------------------------------------------------------------------------------------------------------------------------  > H > Now i have  a particular user "BOSCH1"  who  is  neither a owner nor a > group member .@ > He belongs to world . I have to give  him RWED access  to thisF > directory . But i do not want make him system user or owner . also i< > cannot add him to group because i have to change his UIC .F > So my question is how can i give him RWED access keeping in mind the > above conditions? .  > Isn't it  tricky ? .... 2 > It could be a great help from you guys .........  E Is user "BOSCH1" needing access through a system that is a client of  E Advanced Server or Pathworks, or is the access through their OpenVMS  	 username.   G The answers to that directly affect the actions that need to be taken,   if any.   G Also you may need an additional pair of ACE's on the directory for the  I PWRK$DEFAULT identifier if the PWRK$DEFAULT account has not been granted  H    the APP_LP_FEEDSTOCK right.  Otherwise things may work a bit strange.   -John ! malmberg@dskwld.zko.hp.compaq.dec  Personal Opinion Only    ------------------------------   Date: 8 Feb 2006 02:18:22 -0800  From: etmsreec@yahoo.co.uk' Subject: Re: Another sad day for VMS... C Message-ID: <1139393902.146491.118440@g44g2000cwa.googlegroups.com>    David Froble wrote: H >>>If I was the #1 person in VMS management, I'd make sure I had people,E don't care who, sales, ambassadors, whatever, who had a complete list  ofD everyone using VMS, or at least those still with any type of contactB such as maintenance contracts.  I'd demand that every customer wasD contacted, in person, by phone, whatever, to see how the customer is= doing, and asking what HP could do to help that customer. <<<   G You're assuming that HP know who their customers are.  Experience shows F that they don't because data are spread out between different sectionsG of the company and one doesn't communicate very well with another since E that would be either a) client confidentiality; 2) data protection or   3) why would we want to do that?  E A few, key, high profile customers get seen and viewed and are known. D They're typically the handful that deal direct with HP for the wholeC solution or are, for one reason or another, brought to management's F attention.  For example, a company that was replacing all of their VAX= and/or Alpha systems with Integrity would probably come to HP A management's attention.  Your average company with one or two VMS C systems that just tick along and occasionally have a need for Field D Service to come out and feed and water them don't even get lifted asC far as HP's horizon.  They (HP) just don't know who these customers  are!  F The fundamental (no BS) reason for this is the sales model.  ResellersG hold the maintenance contracts (just think of when Metrologie went bust > in 2000 - Compaq dropped all the maintenance contracts becauseD Metrologie weren't there any more) and resellers supply stuff to endE users and prefer to keep their customer names to themselves.  I would D expect that even when HP are asked to do things like relocations forG customers, these data are kept within the Customer Services section and ) don't make it to the rest of the company.   D The classic sales model within Digital and Compaq was referred to asF Stove Pipes - rigid paths to sales for VMS or Unix or Wintel and neverD the twain shall meet.  Stove pipes are still around from what I see.A They've just moved so that although sales people may see variety, 9 that's where the variety and sharing of information ends.    That's my opinion though.    Steve    ------------------------------  % Date: Wed, 08 Feb 2006 13:34:50 -0500 ' From: Dave Froble <davef@tsoft-inc.com> ' Subject: Re: Another sad day for VMS... 9 Message-ID: <_uWdnTA3hsI7pHfenZ2dnUVZ_sudnZ2d@libcom.com>    etmsreec@yahoo.co.uk wrote:  > David Froble wrote:  > I >>>>If I was the #1 person in VMS management, I'd make sure I had people,  > G > don't care who, sales, ambassadors, whatever, who had a complete list  > ofF > everyone using VMS, or at least those still with any type of contactD > such as maintenance contracts.  I'd demand that every customer wasF > contacted, in person, by phone, whatever, to see how the customer is? > doing, and asking what HP could do to help that customer. <<<  > I > You're assuming that HP know who their customers are.  Experience shows H > that they don't because data are spread out between different sectionsI > of the company and one doesn't communicate very well with another since G > that would be either a) client confidentiality; 2) data protection or " > 3) why would we want to do that? > G > A few, key, high profile customers get seen and viewed and are known. F > They're typically the handful that deal direct with HP for the wholeE > solution or are, for one reason or another, brought to management's H > attention.  For example, a company that was replacing all of their VAX? > and/or Alpha systems with Integrity would probably come to HP C > management's attention.  Your average company with one or two VMS E > systems that just tick along and occasionally have a need for Field F > Service to come out and feed and water them don't even get lifted asE > far as HP's horizon.  They (HP) just don't know who these customers  > are! > H > The fundamental (no BS) reason for this is the sales model.  ResellersI > hold the maintenance contracts (just think of when Metrologie went bust @ > in 2000 - Compaq dropped all the maintenance contracts becauseF > Metrologie weren't there any more) and resellers supply stuff to endG > users and prefer to keep their customer names to themselves.  I would F > expect that even when HP are asked to do things like relocations forI > customers, these data are kept within the Customer Services section and + > don't make it to the rest of the company.  > F > The classic sales model within Digital and Compaq was referred to asH > Stove Pipes - rigid paths to sales for VMS or Unix or Wintel and neverF > the twain shall meet.  Stove pipes are still around from what I see.C > They've just moved so that although sales people may see variety, ; > that's where the variety and sharing of information ends.  >  > That's my opinion though.  >  > Steve  >   H There may be all the obsticles you mention, and more.  However, go back H to the original statement.  "If I was the #1 person in VMS management". G   In that situation, I would do whatever it took to gather the contact  E information for as many VMS customers as possible.  Whatever it took!    --  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, 08 Feb 2006 09:03:41 +0100 + From: Karsten Nyblad <nospam@nospam.nospam> 3 Subject: Re: Cisco to sink: names Capellas to board = Message-ID: <43e9a5db$0$67262$157c6196@dreader2.cybercity.dk>    JF Mezei wrote: H > I wonder who will be the suitor. Would Microsoft buy Cisco ? They coldK > extend their monopoly to networking and teak standards the way they want.   F Would the European Union and USA allow such a merger?  I have a hunch I that there would be so many conditions on the deal, that Cisco would not   be interesting to M$.    ------------------------------   Date: 8 Feb 2006 09:09:50 -0800 ( From: "Rich Jordan" <jordan@ccs4vms.com>/ Subject: Different console prompt "_>" on AS600 C Message-ID: <1139418590.258999.162020@z14g2000cwz.googlegroups.com>   G My home webserver is an AS600 5/333 running VMS V8.2, SWS 2.1, PHP 1.3, D etc (all upgraded from V7.2-2 and associated a few days ago, currentF ECOs too).  Today I was telnet'ed into it (I'm out of the office, away@ from my docs, console, etc) when the system went down; my telnet0 session hung, webserver stopped responding, etc.  B I connected to my other Alpha which has the webserver console portB plugged into a local serial port (which let me work on the VMS and@ layered products upgrades while away from home), and got a "_> "G prompt, whichI can't recall ever seeing before.  Anything I type at the @ prompt echos, but nothing happens when I hit return except a newE prompt.  Disconnect/reconnect makes no difference, nor does sending a  break.  D I can't get to docs, and I can't do a lot of websearching right now.C Can anyone tell me what that _> prompt is?  And if there's a way to F break out of it to reboot the system so I can find out whats going on?   Thanks!    ------------------------------  % Date: Wed, 08 Feb 2006 01:58:49 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> / Subject: Re: Hobbyist kit installation problems , Message-ID: <43E9967B.F8E69FE1@teksavvy.com>   "greg.chabala@gmail.com" wrote:  > BIND Resolver Parameters   This looks good.       > SHOW INTERFACE SE0/FULL  >  > Interface:  SE0 T >    IP_Addr:   192.168.1.205     NETWRK:  255.255.255.0      BRDCST:  192.168.1.255  H If your router is at 192.168.0.*, this host will not be able to reach itG because it will be considered to be outside its subnet.  Network shoudl / be 255.255.0.0 and broadcast at 192.168.255.255   H > Type      Destination                                          Gateway > I > AH        127.0.0.1                                           127.0.0.1  > AH        192.168.1.0/24 > 192.168.1.205  > AH        192.168.1.205  > 192.168.1.205   " You are missing the default route.  6 TCPIP> SET ROUTE/DEFAULT/PERMANENT/GATEWAY=ip.ad.re.ss, TCPIP> SET ROUTE/DEFAULT/GATEWAY=ip.ad.re.ss  @ But the ip.ad.re.ss must be within the same subnet as one of theJ interfaces of your host, otherwise your VAX won't knwo how to route to it.  E Also, about the transceiver. The switch doesn't turn off power to the D transceiver when the VAX expects to talk via the coax.  You may have% been mislead into thinking it worked.    ------------------------------  % Date: Wed, 08 Feb 2006 02:04:51 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> / Subject: Re: Hobbyist kit installation problems , Message-ID: <43E997E5.58BCD047@teksavvy.com>   BTW,     TCPIP>  HELP SET INTERFACE TCPIP>  HELP SET CONF INTERFACE  TCPIP>  HELP SET ROUTE TCPIP>  HELP SET NAME  etc etc etc   * HELO is your friend. IT also has examples.   ------------------------------   Date: 7 Feb 2006 23:35:51 -0800 7 From: "greg.chabala@gmail.com" <greg.chabala@gmail.com> / Subject: Re: Hobbyist kit installation problems C Message-ID: <1139384151.023565.231440@g47g2000cwa.googlegroups.com>   A AHHHH, sweet goodness. It's finally running. Screw the DHCP, this G works, and I don't have to sit in the cold dusty basement and type on a 	 terminal.   ( I now have the wonder of copy and paste.  E Thanks to you Mr. Mezei, for sticking with me to the end, and all the ) helpful suggestions, late into the night.    ------------------------------   Date: 8 Feb 2006 07:12:33 -0600 - From: Kilgallen@SpamCop.net (Larry Kilgallen) / Subject: Re: Hobbyist kit installation problems 3 Message-ID: <i2zZUkCkY6sJ@eisner.encompasserve.org>   } In article <1139372552.966466.315630@z14g2000cwz.googlegroups.com>, "greg.chabala@gmail.com" <greg.chabala@gmail.com> writes: D > Okay, scratch the DHCP then. Do you have a step by step plan I canB > follow? I think I've already tried to do it the 'intuitive' way.  F This post makes no sense, based on the order in which messages arrived at my newsreader.   D Please always quote enough context, in this case to indicate to what0 the "Okay, scratch the DHCP then" is a response.  : Do not assume everybody's newsreader is the same as yours.  < Do not assume messages arrive in the same order to everyone, or with the same certainty.    ------------------------------  $ Date: Wed, 8 Feb 2006 08:42:23 -0500- From: William Webb <william.w.webb@gmail.com> / Subject: Re: Hobbyist kit installation problems I Message-ID: <8660a3a10602080542g40acd2f0g3d32065388694cb7@mail.gmail.com>   4 On 7 Feb 2006 23:35:51 -0800, greg.chabala@gmail.com <greg.chabala@gmail.com> wrote: C > AHHHH, sweet goodness. It's finally running. Screw the DHCP, this I > works, and I don't have to sit in the cold dusty basement and type on a  > terminal.  > * > I now have the wonder of copy and paste. > G > Thanks to you Mr. Mezei, for sticking with me to the end, and all the + > helpful suggestions, late into the night.  >  >   B Nobody mentioned to you that TCPIP SET whatever tweaks the active,B live, volatile information, while TCP SET CONFIGURATION tweaks theB permanent, stored, non-volatile database, whatever terminology you feel like employing.  E And TCPIP SHOW and TCPIP SHOW CONF are the same except you're looking  instead of tweaking.  C And the command syntax isn't always exactly the same for comparable = commands in TCPIP SET/SHOW blahblah vs. TCPIP SET/SHOW CONFIG 	 blahblah.   F This is particularly true if you are using a command procedure to readD what's out there and then reset it to something else (found this out) while moving EVERYTHING to a new network)   * For what it's worth, I've either felt that  4 1) DHCP is evil and a favorite tool of the devil, orF 2) DHCP is a sign from God that you need to create additional subnets.  F depending on what the particular problem du jour was that caused me to ponder the issue.    WWWebb   --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, 08 Feb 2006 16:47:43 +0100 ( From: Paul Sture <paul.sture@bluewin.ch>/ Subject: Re: Hobbyist kit installation problems + Message-ID: <44ui51F4205qU1@individual.net>    William Webb wrote:   
 <ker-snip>   > , > For what it's worth, I've either felt that > 6 > 1) DHCP is evil and a favorite tool of the devil, orH > 2) DHCP is a sign from God that you need to create additional subnets. > H > depending on what the particular problem du jour was that caused me to > ponder the issue.   @ My (consumer grade*) cable router seems to encourage the use of:   a) wireless rather than cable # b) DHCP rather than fixed addresses   D and the salesman thought I was clearly mad for going for the router " rather than a USB modem. <chuckle>  F * El cheapo router though - if I reboot my Alpha, the relevant LED on G the router goes out and I need to pull the cable between the Alpha and  , router then push it back in, then all is OK.   ------------------------------  $ Date: Wed, 8 Feb 2006 10:59:46 -0500- From: William Webb <william.w.webb@gmail.com> / Subject: Re: Hobbyist kit installation problems I Message-ID: <8660a3a10602080759k4a26f6a3p6c43135e7c22e410@mail.gmail.com>   4 On 2/8/06, Paul Sture <paul.sture@bluewin.ch> wrote: > William Webb wrote:  >  > <ker-snip> >  > > . > > For what it's worth, I've either felt that > > 8 > > 1) DHCP is evil and a favorite tool of the devil, orJ > > 2) DHCP is a sign from God that you need to create additional subnets. > > J > > depending on what the particular problem du jour was that caused me to > > ponder the issue.  > B > My (consumer grade*) cable router seems to encourage the use of: >  > a) wireless rather than cable % > b) DHCP rather than fixed addresses  > E > and the salesman thought I was clearly mad for going for the router $ > rather than a USB modem. <chuckle> > G > * El cheapo router though - if I reboot my Alpha, the relevant LED on H > the router goes out and I need to pull the cable between the Alpha and. > router then push it back in, then all is OK. >   A Hmmm.  Wireless.  Now *that* would be a nice thing for the device  driver guys to work on.    WWWebb   --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: 8 Feb 2006 02:00:37 -0800 / From: "Volker Halle" <volker_halle@hotmail.com> - Subject: Re: is this a bug or feature of ssh? C Message-ID: <1139392837.176833.143050@g47g2000cwa.googlegroups.com>    Jouk,   ? it's a feature. It's the purpose of SSH to prevent the need for F cleartext passwords to be stored in files and it's quite successful in prohibiting such use ;-)  9 I've recently described the necessary setup for hostbased  authentication in ITRC:   K http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=985928    Hope this helps,   Volker.    ------------------------------  % Date: Wed, 08 Feb 2006 06:09:11 -0500 ' From: Rich Whalen <whalenr@process.com> - Subject: Re: is this a bug or feature of ssh? 8 Message-ID: <b6kju19nk1p7v4pn5888nl4fbsd70ucp13@4ax.com>  , On 8 Feb 2006 02:00:37 -0800, "Volker Halle"! <volker_halle@hotmail.com> wrote:    >Jouk, > @ >it's a feature. It's the purpose of SSH to prevent the need forG >cleartext passwords to be stored in files and it's quite successful in  >prohibiting such use ;-)  > : >I've recently described the necessary setup for hostbased >authentication in ITRC: > L >http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=985928 >  >Hope this helps,  >  >Volker.  9 Another option for unattened authentication is public key A authentication, where the user creates his own public/private key E pair, puts the public key on the remote system and sets up some files D to point to the keys.  On VMS these files are [.SSH2]IDENTIFICATION.D on the client system and [.SSH2]AUTHORIZATION. on the server system.   ------------------------------  % Date: Wed, 08 Feb 2006 14:47:30 +0000  From: Roy Omond <Roy@Omond.net> 1 Subject: Most recent firmware version for HSZ80 ? 4 Message-ID: <dsd0a3$lqj$1$8300dec7@news.demon.co.uk>   Gentle colleagues,  5 can someone please inform me what the most up-to-date # firmware version is for the HSZ80 ?    Is it V832-0 ?  8 What is version V83Z, and how does it differ from V832 ?   Many thanks in advance,   	 Roy Omond  Blue Bubble Ltd.   ------------------------------  % Date: Wed, 08 Feb 2006 08:28:27 +0100 + From: Karsten Nyblad <nospam@nospam.nospam> $ Subject: Re: null terminated strings= Message-ID: <43e99d99$0$78279$157c6196@dreader1.cybercity.dk>    JF Mezei wrote:  > sol gongola wrote: > A >> program does no data validation (or length checking). Null (or D >> other) terminated strings are blamed because they were being usedC >> in the programs involved with the overrun situation. It's really  >> the code that is at fault.  >  > G > My first reaction is to agree with the above. When you write routines G >  that will write to a buffer, you should require the caller to supply E > a maximum length for that buffer and make sure you don't exceed it.  > E > And the io commands in C such as fread  do provide a maximum length  > argument.  > G > However, strcpy doesn't. And I guess this is where the C run time has F >  flaws because it can go bezerk and write 2 megs worth of data in an > 80 byte buffer.   C C has had strncpy for years.  There are two problems.  First people G being lazy who use strcpy where strncpy should be used.  Yes it happens H again and again.  In too many cases free software does not include rangeF checks.  The programmer has not used the last one or two days it takes! to put in primitive range checks.   C Secondly even when you use strncpy, there are lots of possibilities H programming errors, errors that are unlikely if the programming languageF and/or libraries do not automate range checks.  In practice it is veryH difficult to write a C program doing serious string manipulation withoutE at least one bug in length checks on strings.  Checks that could have  been automated.   D > Consider a customer name record field that starts out at 30 bytes,G > but later in the project, it is increased to 40 bytes.  By that time, F > you may have written code that copies the name to temporary buffers,E > and at this point, when you increase the field size, you need to do D > due diligence and check to see that any temporary buffers to which' > you copy the name are also increased.   G That is also a result of bad coding practices.  Of course the length of : such a field should be declared as a constant or #defined.   ------------------------------  % Date: Wed, 08 Feb 2006 08:42:50 +0100 + From: Karsten Nyblad <nospam@nospam.nospam> $ Subject: Re: null terminated strings= Message-ID: <43e9a0f7$0$67263$157c6196@dreader2.cybercity.dk>    Bill Todd wrote: > Roger Ivie wrote:  > 6 >> On 2006-02-07, sol gongola <sol@adldata.com> wrote: >>G >>> Null terminated strings are convenient to pass to a subroutine. It   >>> saves a 4 >>> parameter and (in 'olden' days) it saved memory. >> >>I >> It only saves memory if strings can be longer than 255 characters. For I >> strings consisting of a count byte followed by the string, you're just , >> trading the count byte for the null byte. >  > D > Well, you also save the code (and in some architectures even more J > importantly the register) that handles the count, since the termination G > condition is simply a condition-code check for null in the move loop.   F The moment you need range checks you end up having two conditions for H terminating loops:  one for the zero terminating the string and one for D the range check.  If you use counted strings is normally trivial to H calculate the number of characters to process before executing the loop.  F Don't tell me that you could do without range checks.  In the days of F 16-bit computers the size of buffers were always a compromise between 3 being big enough and how much memory you could use.    ------------------------------  % Date: Wed, 08 Feb 2006 08:57:52 -0500 # From: sol gongola <sol@adldata.com> $ Subject: Re: null terminated strings0 Message-ID: <1139407044.370340@nntp.acecape.com>   Bill Gunshannon wrote:. > In article <43E9029A.8EC832CC@teksavvy.com>,2 > 	JF Mezei <jfmezei.spamnot@teksavvy.com> writes: >> sol gongola wrote: I >>> program does no data validation (or length checking). Null (or other) N >>> terminated strings are blamed because they were being used in the programsO >>> involved with the overrun situation. It's really the code that is at fault. H >> My first reaction is to agree with the above. When you write routinesI >> that will write to a buffer, you should require the caller to supply a D >> maximum length for that buffer and make sure you don't exceed it. >>P >> And the io commands in C such as fread  do provide a maximum length argument. >>H >> However, strcpy doesn't. And I guess this is where the C run time hasI >> flaws because it can go bezerk and write 2 megs worth of data in an 80  >> byte buffer.  >>I >> Consider a customer name record field that starts out at 30 bytes, but H >> later in the project, it is increased to 40 bytes.  By that time, youJ >> may have written code that copies the name to temporary buffers, and atC >> this point, when you increase the field size, you need to do due J >> diligence and check to see that any temporary buffers to which you copy >> the name are also increased.  >>K >> Languages such as COBOL would automaically pad or truncate fields during E >> a MOVE operation to ensure no data overrun (and may issue compiler J >> warnings, but C's lack of a formal string type prevents it from issuingL >> compiler warnings (although it does when you move a long to a short etc). > H > Don't be so sure of that.  While I am sure COBOL2006 (or whatever theyI > are using today) does this the COBOL compilers that were contemporaries F > of the time when C was being developed did not do range checking andE > overwriting one working storage entry while referencing another was   F The language compilers I used generally did no range checking unless aK debug/range check option is specified during compile. This adds substantial I overhead to program execution to check for string and array bounds. It is D normally removed for production execution and you lose the automatic protection.   J This range checking for all code movement is not needed. Range checking isI only really needed at the point where data from 'strangers' is introduced  to the program.   G > trivial.  I actually used to demonstrate it in two ways when I taught I > COBOL.  One was using array out of bounds examples and one was by doing E > arithmetic on the values contained in an alpha field.  COBOL was no G > saint.  I even know someone who made considerable use of these "capa- K > bilities" in order to embed non-printing characters into declared strings K > in order to create a terminal handling routine back before we had DISPLAY  > AT verbs.   J Fortran, PL/1, Cobol have an equivalent or redefines capability which doesH let you overlay different data type and sizes over an area. Depending onJ the language, you can do really weird things with it. One old language letO you equivalence a statement label with a numeric array. You deposit binary code L into the array and branch to the label defined as the beginning of the code.  H >> So it takes more experienmce and more self discipline to program in CG >> properly. Problem is when you unleash young windows weenies and they I >> start to write C code, they don't see the problems and they don't have @ >> enough experience and self discipline to write armoured code. > K > I have heard people here advocate doing things that are very bad examples K > of programming practice, too.  In any case, it is not the languages fault J > no matter what.  The responsibility falls on the shoulders of managementG > or the programmer.  Choose the right tool for the job and know how to H > use that tool properly.  if you are trying to turn out hex allen screwG > with a phillips screwdriver, don't blame either of them when the head  > gets all chewed up.  >  > bill >    sol    ------------------------------  $ Date: Wed, 8 Feb 2006 09:28:29 -0500# From: "Dan Allen" <dallen@nist.gov> $ Subject: RE: null terminated strings: Message-ID: <JFEPKAPBPMDFDBOIANGDKEMLHDAA.dallen@nist.gov>  L > the language, you can do really weird things with it. One old language letF > you equivalence a statement label with a numeric array. You deposit 
 > binary code N > into the array and branch to the label defined as the beginning of the code. >    Sounds like C to me ;-)    Dan       ------------------------------  % Date: Wed, 08 Feb 2006 06:37:20 -0800 # From: "Tom Linden" <tom@kednos.com> $ Subject: Re: null terminated strings( Message-ID: <ops4nr8iqczgicya@hyrrokkin>  H On Wed, 08 Feb 2006 08:57:52 -0500, sol gongola <sol@adldata.com> wrote:  H > The language compilers I used generally did no range checking unless aC > debug/range check option is specified during compile. This adds   
 > substantial K > overhead to program execution to check for string and array bounds. It is F > normally removed for production execution and you lose the automatic
 > protection. L >  This range checking for all code movement is not needed. Range checking   > isK > only really needed at the point where data from 'strangers' is introduced  > to the program.   I Then you are not using the right language.  There is range checking and    thenK there is range checking.  The one you are referring to is a shotgun blast   	 approach, L which resolves to generalized system exceptions.  The correct way to do it   is to H code in ON-conditions into your code, in other words. failure tolerant   designJ and implementation.  The problem with languages like C, because of the lowL abstraction level is you have to write so much more code, which of course,   is notK only reinventing the wheel each time, but is a source of potential bugs.     WithK such facilities built in to the compiler you lower your development costs    and you K improve your reliability aand reduce your maintenance.  Now, why wouldn't    you do that?  ! ('you' is intended to be generic)    ------------------------------  % Date: Wed, 08 Feb 2006 09:44:29 -0500 # From: sol gongola <sol@adldata.com> $ Subject: Re: null terminated strings0 Message-ID: <1139409841.714732@nntp.acecape.com>   Dan Allen wrote:M >> the language, you can do really weird things with it. One old language let G >> you equivalence a statement label with a numeric array. You deposit   >> binary codeO >> into the array and branch to the label defined as the beginning of the code.  >> >  > Sounds like C to me ;-)  >  > Dan  >     Michigan Algorithm Decoder (MAD)   ------------------------------   Date: 8 Feb 2006 09:02:29 -0600 - From: Kilgallen@SpamCop.net (Larry Kilgallen) $ Subject: Re: null terminated strings3 Message-ID: <l+GrjeIjJpRK@eisner.encompasserve.org>   V In article <1139407044.370340@nntp.acecape.com>, sol gongola <sol@adldata.com> writes:  L > Fortran, PL/1, Cobol have an equivalent or redefines capability which doesJ > let you overlay different data type and sizes over an area. Depending onL > the language, you can do really weird things with it. One old language letQ > you equivalence a statement label with a numeric array. You deposit binary code N > into the array and branch to the label defined as the beginning of the code.  ? Even Ada has similar capabilities, but such dangerous practices  are _not_ the default.  D The fact that the neighborhood store offers cigarettes does not mean I have to buy them.    ------------------------------   Date: 8 Feb 2006 15:05:08 GMT ( From: bill@cs.uofs.edu (Bill Gunshannon)$ Subject: Re: null terminated strings+ Message-ID: <44ufl4F40ibfU1@individual.net>   : In article <JFEPKAPBPMDFDBOIANGDKEMLHDAA.dallen@nist.gov>,& 	"Dan Allen" <dallen@nist.gov> writes:M >> the language, you can do really weird things with it. One old language let G >> you equivalence a statement label with a numeric array. You deposit   >> binary codeO >> into the array and branch to the label defined as the beginning of the code.  >> >  > Sounds like C to me ;-)   C Sorry to disappoint you, but the same can be done in COBOL, Fortran C and even Pascal. (I have an terminal program written in UCSD Pascal @ for the PDP-11 which loads integer values into an array and then@ installs the address of that array in the LTC interupt.  It thenC runs as an interupt driven serial input routine grabbing characters C and putting them into a ring buffer, which is then accessed as, you C guessed it, another Pascal array.)  I find it interesting that even C those languages that brag about how strongly typed and bullet-proof C they are all have backdoors to do the things that C does naturally. 
 Even Ada!!   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, 08 Feb 2006 10:16:50 -0500 # From: sol gongola <sol@adldata.com> $ Subject: Re: null terminated strings0 Message-ID: <1139411783.638239@nntp.acecape.com>   Tom Linden wrote: J > On Wed, 08 Feb 2006 08:57:52 -0500, sol gongola <sol@adldata.com> wrote: > I >> The language compilers I used generally did no range checking unless a C >> debug/range check option is specified during compile. This adds  E >> substantial overhead to program execution to check for string and  E  >> array bounds. It is normally removed for production execution and &  >> you lose the automatic protection.B >>  This range checking for all code movement is not needed. RangeK >> checking is only really needed at the point where data from 'strangers'  !  >> is introduced to the program.  > J > Then you are not using the right language.  There is range checking and K > then there is range checking.  The one you are referring to is a shotgun  H > blast approach, which resolves to generalized system exceptions.  The L  > correct way to do it is to code in ON-conditions into your code, in otherG  > words. failure tolerant design and implementation.  The problem with   E They are not generalized system exceptions. The compiler adds code to G the boundary checking each time the a string or array is used. I am not A sure which language is involved (basic, dcl, etc.) but the global G ON-conditions also involve code being inserted into the compile output.   H On-conditions are not necessarily 'quick' commands to wait for somethingB to happen. There really are two kinds. One waits for and handles aF system exception (like memory addressing, under/overflow, protection).A The other involves adding code (or function calls) to check for a ; condition each time the potential for the situation occurs.   H  > languages like C, because of the low abstraction level is you have toH  > write so much more code, which of course, is not only reinventing the<  > wheel each time, but is a source of potential bugs.  WithF > such facilities built in to the compiler you lower your development G  > costs and you improve your reliability aand reduce your maintenance. !  > Now, why wouldn't you do that?   G Compilers don't necessarily have the ability to intelligently determine D when the added code should be invoked so it is done each time. DoingF it yourself allows you to perform these checks when you determine thatD it is necessary. Even with faster machines, automatic bound checking" overhead can still be significant.   > # > ('you' is intended to be generic)    ------------------------------  $ Date: Wed, 8 Feb 2006 11:07:17 -0500# From: "Dan Allen" <dallen@nist.gov> $ Subject: RE: null terminated strings: Message-ID: <JFEPKAPBPMDFDBOIANGDEEMOHDAA.dallen@nist.gov>   > -----Original Message-----F > From: bill@triangle.cs.uofs.edu [mailto:bill@triangle.cs.uofs.edu]On > Behalf Of Bill Gunshannon - > Sent: Wednesday, February 08, 2006 10:05 AM  > To: Info-VAX@Mvb.Saic.Com & > Subject: Re: null terminated strings >  > < > In article <JFEPKAPBPMDFDBOIANGDKEMLHDAA.dallen@nist.gov>,( > 	"Dan Allen" <dallen@nist.gov> writes:O > >> the language, you can do really weird things with it. One old language let H > >> you equivalence a statement label with a numeric array. You deposit > >> binary codeG > >> into the array and branch to the label defined as the beginning of  > the code.  > >> > >  > > Sounds like C to me ;-)  > E > Sorry to disappoint you, but the same can be done in COBOL, Fortran   N Not disappointed and not surprised - just stirring the pot Bill. Been around aA long time and done it in several languages - usually by accident.    Dan    ------------------------------   Date: 8 Feb 2006 16:43:08 GMT ( From: bill@cs.uofs.edu (Bill Gunshannon)$ Subject: Re: null terminated strings+ Message-ID: <44ulcsF42an0U1@individual.net>   + In article <44ukdgF42en5U1@individual.net>, + 	Paul Sture <paul.sture@bluewin.ch> writes:  > Bill Gunshannon wrote: >  > <snip> >  > ? >>  I actually used to demonstrate it in two ways when I taught J >> COBOL.  One was using array out of bounds examples and one was by doingF >> arithmetic on the values contained in an alpha field.  COBOL was no	 >> saint.  > J > The array out of bounds problem was solved in VAX-COBOL many years ago. A > In the late 1980s I was compiling with /CHECK=(BOUNDS,PERFORM),   C But that's not COBOL, that was a VAX extension.  Most compilers had D the option, but that doesn't change the fact that COBOL specificallyE did not include range checking.  And precisely for the reasons others C here are just now pointing out.  The overhead was too great so they E opted to specifically leave it out of the early standards.  They knew B what it was and knew it was a good thing, but opted to leave it upB to the programmer to decide when he dould and could not afford the	 overhead.   K >                                                                 and when  K > a customer reported an ACCVIO error it took me straight to the offending  K > line of code. The overhead was imperceptible, since those apps tended to   > be I/O bound.   D Yeah, that was one of the things I liked back in my VM370 days.  IBMC had the easiest to read and most informative dumps of any machine I B ever used.  How much the overhead involved probably depended a lotF ont he machine you were running COBOL on.  And as I said, the standardD of the time thought it netter top leave choice of overhead up to theD programmer.  Maybe it was just that we were a lot better programmersF back int he days when we had to be concerned about so many things that+ today's programmers don't even know exists.    > A > > I even know someone who made considerable use of these "capa- E > > bilities" in order to embed non-printing characters into declared J > > strings in order to create a terminal handling routine back before we  > > had DISPLAY AT verbs.  > >  > J > It was the only means to to do it way back when, so that it what I did. ? > And use $QIOW to interpret input escape sequences "manually".   C Yes, but $QIOW is also not COBOL.  I thought we were addressing the D capabilities and shortcomings of different languages.  I would guessD if you wrote all your C on VMS to use only $Qwhatever calls it would* be much safer, but then, it wouldn't be C.   > E > Mind you, in the days when I was modifying programs to output on a  K > terminal as well as a printer, I could have cheerfully throttled the guy  J > who thought it a good idea to embed one of more <CR> characters into an  > output record.   bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------   Date: 8 Feb 2006 16:46:45 GMT ( From: bill@cs.uofs.edu (Bill Gunshannon)$ Subject: Re: null terminated strings+ Message-ID: <44uljkF42an0U2@individual.net>   : In article <JFEPKAPBPMDFDBOIANGDEEMOHDAA.dallen@nist.gov>,& 	"Dan Allen" <dallen@nist.gov> writes: >  >  >> -----Original Message----- G >> From: bill@triangle.cs.uofs.edu [mailto:bill@triangle.cs.uofs.edu]On  >> Behalf Of Bill Gunshannon. >> Sent: Wednesday, February 08, 2006 10:05 AM >> To: Info-VAX@Mvb.Saic.Com' >> Subject: Re: null terminated strings  >> >>= >> In article <JFEPKAPBPMDFDBOIANGDKEMLHDAA.dallen@nist.gov>, ) >> 	"Dan Allen" <dallen@nist.gov> writes: P >> >> the language, you can do really weird things with it. One old language letI >> >> you equivalence a statement label with a numeric array. You deposit  >> >> binary code H >> >> into the array and branch to the label defined as the beginning of >> the code. >> >>  >> > >> > Sounds like C to me ;-) >>F >> Sorry to disappoint you, but the same can be done in COBOL, Fortran > P > Not disappointed and not surprised - just stirring the pot Bill. Been around aC > long time and done it in several languages - usually by accident.   F Oh, I know a lot of us have been around for a long time.  I think mostD of this is more for those who haven't.  I have done it frequently onF purpose and, I am sure, once or twice by accident. :-)  The problem isI just like in other endeavors people tend to look at the past with today's C eyes and can't understand that there actually were reasons why some A things that seem so bad today were done.  Like 2 character dates.    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, 8 Feb 2006 17:05:47 +0000 (UTC)- From: klewis@OMEGA.MITRE.ORG (Keith A. Lewis) $ Subject: Re: null terminated strings. Message-ID: <dsd8db$po8$1@newslocal.mitre.org>  s "Tom Linden" <tom@kednos.com> writes in article <ops4nr8iqczgicya@hyrrokkin> dated Wed, 08 Feb 2006 06:37:20 -0800: I >On Wed, 08 Feb 2006 08:57:52 -0500, sol gongola <sol@adldata.com> wrote:  > I >> The language compilers I used generally did no range checking unless a D >> debug/range check option is specified during compile. This adds   >> substantialL >> overhead to program execution to check for string and array bounds. It isG >> normally removed for production execution and you lose the automatic  >> protection.M >>  This range checking for all code movement is not needed. Range checking    >> is L >> only really needed at the point where data from 'strangers' is introduced >> to the program. >   J >Then you are not using the right language.  There is range checking and  I >then there is range checking.  The one you are referring to is a shotgun H >blast   approach, which resolves to generalized system exceptions.  TheF >correct way to do it   is to code in ON-conditions into your code, inH >other words. failure tolerant   design and implementation.  The problemK >with languages like C, because of the low abstraction level is you have to H >write so much more code, which of course,   is not only reinventing theA >wheel each time, but is a source of potential bugs.    With such K >facilities built in to the compiler you lower your development costs   and E >you improve your reliability aand reduce your maintenance.  Now, why ; >wouldn't   you do that?  ('you' is intended to be generic)   I PL/I works the way he describes, Tom.  The /CHECK option defaults to off.    STRINGRANGE                     Source Listing                   8-FEB-2006 17:02:12    DEC PL/I V4.2A-112                  Page   1z 01                                                               8-FEB-2006 17:02:04    USER:[LEWIS.TEST]stringrange.PLI;5  F    |	      1 /* STRINGRANGE.PLI Keith Lewis  8-FEB-2006 16:53:25.25 */	 	      2  / 	      3 stringrange: procedure options (main);         1      4          1      5 dcl 1 structure,*        1      6         2 s	char (10) var,?        1      7         2 t	(100) fixed bin (7) init ((*)(65));         1      8 dcl c	char;         1      9 )        1     10     on stringrange begin; @        1     11         put skip list ('Stringrange condition');        1     12     end;        1     13         1     14     s = 'TEST'; )        1     15     c = substr(s, 20, 1); C        1     16     put skip list ('20th char of string is ' || c);         1     17          1     18 end stringrange;    
  COMMAND LINE 
  ------- ----    PLI stringrange/LIS    $ run stringrange    20th char of string is A  = (Notice that the STRINGRANGE ON-condition was not triggered.)   ' p.s.  Any word on PL/I for Itanium VMS?   0 --Keith Lewis              klewis {at} mitre.org> The above may not (yet) represent the opinions of my employer.   ------------------------------  % Date: Wed, 08 Feb 2006 19:00:37 +0100 + From: Karsten Nyblad <nospam@nospam.nospam> $ Subject: Re: null terminated strings= Message-ID: <43ea31c1$0$78284$157c6196@dreader1.cybercity.dk>    Bill Gunshannon wrote:5 > In article <l+GrjeIjJpRK@eisner.encompasserve.org>, 2 > 	Kilgallen@SpamCop.net (Larry Kilgallen) writes: > X >>In article <1139407044.370340@nntp.acecape.com>, sol gongola <sol@adldata.com> writes: >> >>M >>>Fortran, PL/1, Cobol have an equivalent or redefines capability which does K >>>let you overlay different data type and sizes over an area. Depending on M >>>the language, you can do really weird things with it. One old language let R >>>you equivalence a statement label with a numeric array. You deposit binary codeO >>>into the array and branch to the label defined as the beginning of the code.  >>A >>Even Ada has similar capabilities, but such dangerous practices  >>are _not_ the default. >  > G > That's true, but what's worse, telling someone the languiage they are H > using is dangerous and that it is their responsibility to allow for itH > or to tell them it's safe and then let them do bad things thinking the > language will protect them?  >  > F >>The fact that the neighborhood store offers cigarettes does not mean >>I have to buy them.  >  > K > Sticking to your analogy, it's like selling cigarettes in the drug store.  > If the pharmacist, who is a doctor, thinks their OK, then they must besafe, right?  (Think of your average unskilled porgrammer as the same as that teen- @ > agerwho doesn't have the knowledge background to know better.)  B No it is more like ordinary drugs.  If you use them for the right I disease they will do you good, but if you use them for the wrong reason,   they will harm you.   E The dirty language features should be so that everybody can see that  D something unusual is going when they see code were the features are F used.  All those unsafe features of Ada are implemented in such a way I that a single look at the code will make you know that you had better be  I careful.  In older languages you can too often make really weird things,  D but people reading your your code will note notice that before they  study the code very carefully.   ------------------------------   Date: 8 Feb 2006 04:01:35 -0800 ) From: "Bob Gezelter" <gezelter@rlgsc.com> Y Subject: OpenVMS.org Contributing Editor Robert Gezelter to speak on Software Architectur C Message-ID: <1139399097.244656.213290@g47g2000cwa.googlegroups.com>   F Robert Gezelter, CSA, CSE, CDP, a Contributing Editor for OpenVMS.org,? will present "Architectural Techniques for Interoperability and F Coexistence" on Wednesday, February 8, 2006 at 12:15 PM in Middletown,E New Jersey. This session is hosted by the New Jersey Coast Chapter of C the IEEE Computer Society. The IT community in the New Jersey Coast  area is invited to attend.  G Mr. Gezelter will examine how systems and software architectures enable F long system life. Good architectures eliminate the need for expensive,G on-going revisions. Good architecture also enables systems to evolve to ; additional tasks and missions without incompatible changes. E Systems can operate efficiently for decades without any architectural D revision. When revisions are needed, they are easily integrated with the existing structure.   D The full abstract and location of this presentation can be found at:  B http://www.rlgsc.com/ieee/CoastalNewJersey/2006-02/ann-swarch.html  F This presentation is made possible by the support of the IEEE Computer* Society's Distinguished Visitor's Program.   ------------------------------  # Date: Wed, 08 Feb 2006 14:52:07 GMT " From:   VAXman-  @SendSpamHere.ORGY Subject: Re: OpenVMS.org Contributing Editor Robert Gezelter to speak on Software Archite 0 Message-ID: <00A5101D.C918D5E9@SendSpamHere.ORG>  o In article <1139399097.244656.213290@g47g2000cwa.googlegroups.com>, "Bob Gezelter" <gezelter@rlgsc.com> writes:  >  > G >Robert Gezelter, CSA, CSE, CDP, a Contributing Editor for OpenVMS.org, @ >will present "Architectural Techniques for Interoperability andG >Coexistence" on Wednesday, February 8, 2006 at 12:15 PM in Middletown, F >New Jersey. This session is hosted by the New Jersey Coast Chapter ofD >the IEEE Computer Society. The IT community in the New Jersey Coast >area is invited to attend.  > H >Mr. Gezelter will examine how systems and software architectures enableG >long system life. Good architectures eliminate the need for expensive, H >on-going revisions. Good architecture also enables systems to evolve to< >additional tasks and missions without incompatible changes.F >Systems can operate efficiently for decades without any architecturalE >revision. When revisions are needed, they are easily integrated with  >the existing structure. > E >The full abstract and location of this presentation can be found at:  > C >http://www.rlgsc.com/ieee/CoastalNewJersey/2006-02/ann-swarch.html  > G >This presentation is made possible by the support of the IEEE Computer + >Society's Distinguished Visitor's Program.     G You are only speaking for 45 minutes.  It would take me longer to drive ( to Middletown than you will be speaking.   --  K VAXman- A Bored Certified VMS Kernel Mode Hacker   VAXman(at)TMESIS(dot)COM              5   "Well my son, life is like a beanstalk, isn't it?"     ------------------------------   Date: 8 Feb 2006 09:05:14 -0600 - From: Kilgallen@SpamCop.net (Larry Kilgallen) Y Subject: Re: OpenVMS.org Contributing Editor Robert Gezelter to speak on Software Archite 3 Message-ID: <qptKxQggoNeZ@eisner.encompasserve.org>   U In article <00A5101D.C918D5E9@SendSpamHere.ORG>,   VAXman-  @SendSpamHere.ORG writes: q > In article <1139399097.244656.213290@g47g2000cwa.googlegroups.com>, "Bob Gezelter" <gezelter@rlgsc.com> writes:  >> >>H >>Robert Gezelter, CSA, CSE, CDP, a Contributing Editor for OpenVMS.org,A >>will present "Architectural Techniques for Interoperability and H >>Coexistence" on Wednesday, February 8, 2006 at 12:15 PM in Middletown,G >>New Jersey. This session is hosted by the New Jersey Coast Chapter of E >>the IEEE Computer Society. The IT community in the New Jersey Coast  >>area is invited to attend.  I > You are only speaking for 45 minutes.  It would take me longer to drive * > to Middletown than you will be speaking.  H But some comp.os.vms readers are in California.  Should Robert's talk beB extended to last 36 hours to match the required drive for them :-)   ------------------------------  % Date: Wed, 08 Feb 2006 12:19:36 +0100 ( From: Paul Sture <paul.sture@bluewin.ch>G Subject: Re: OT: English grammar and the state of today's education :-) + Message-ID: <44u2edF401ahU1@individual.net>    William Webb wrote:    > My own personal take on this:  > : > "Population Density" is a phrase which has two meanings. > E > And that goes right past about 98 out of 100 people, which verifies  > its veracity.  > L > I'll bet that you didn't know that the plural of "idiot" is "legislature". >   H I don't think the above examples are quite what Benjamin Franklin meant 9 when he said "We hold these truths to be self-evident"...   # ... however applicable they may be.    ;-)      > & > My contract just ended, I'm looking. > 
 Good luck.   ------------------------------  * Date: Wed, 8 Feb 2006 09:49:51 -0600 (CST)* From: sms@antinode.org (Steven M. Schweda)G Subject: Re: OT: English grammar and the state of today's education :-) 2 Message-ID: <06020809494870_20331674@antinode.org>  ( From: Paul Sture <paul.sture@bluewin.ch>  J > I don't think the above examples are quite what Benjamin Franklin meant ; > when he said "We hold these truths to be self-evident"...   8 http://usinfo.state.gov/products/pubs/hrintro/wehold.htm  G       Although the first battles of the American Revolution occurred in G       1775, the Declaration of Independence by the American colonies in G       1776 formally announced the revolution.  Its principal author was C       Thomas Jefferson, who later became the third president of the        United States.  H President Franklin may have said this at some time, but he did not write> the document.  (I'm sure that he was president of _something_,E somewhere, sometime.)  On the bright side, the folks at state.gov can 5 use "its" properly, which is getting to be a novelty.   H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode-org     Saint Paul  MN  55105-2547    ------------------------------  $ Date: Wed, 8 Feb 2006 05:43:22 -0500, From: Carl Friedberg <frida.fried@gmail.com> Subject: Re: PDPs in the news!H Message-ID: <890539d90602080243j27520c65ga1adae2cb4ea161@mail.gmail.com>  ) On 6 Feb 2006 08:03:59 -0600, Bob Koehler 0 <koehler@eisner.nospam.encompasserve.org> wrote:L > In article <dq0cg3$1db$1@rumours.uwaterloo.ca>, dfevans@bcr10.uwaterloo.c= a (David Evans) writes:  > 8 > >  The author also seems a bit confused about what PDP
 > > means. > I >    The author may have read about them, but has probably never actually " >    seen a Purple Data Processor. >  > 7 Bob, Please, the PDP1,5,6,7,10s  were NOT purple.. Carl    ------------------------------   Date: 8 Feb 2006 07:41:01 -0600 ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)  Subject: Re: PDPs in the news!3 Message-ID: <nT8dMqN1+LF4@eisner.encompasserve.org>   w In article <890539d90602080243j27520c65ga1adae2cb4ea161@mail.gmail.com>, Carl Friedberg <frida.fried@gmail.com> writes: + > On 6 Feb 2006 08:03:59 -0600, Bob Koehler 2 > <koehler@eisner.nospam.encompasserve.org> wrote:M >> In article <dq0cg3$1db$1@rumours.uwaterloo.ca>, dfevans@bcr10.uwaterloo.c=  > a (David Evans) writes:  >>9 >> >  The author also seems a bit confused about what PDP  >> > means.  >>J >>    The author may have read about them, but has probably never actually# >>    seen a Purple Data Processor.  >> >>9 > Bob, Please, the PDP1,5,6,7,10s  were NOT purple.. Carl   B   Niether is the 12 down at the Smithsonian (or is that a 15?), or!   the Micro PDP-8 I used to have.   E   We all know that PDP really stands for Programmable Data Processor, E   except that it started out as Programmed Data Processor.  But lots  E   of us have spent time in a room with a PDP-11/70 and I don't think  -   the author has spent time with any of them.    ------------------------------   Date: 8 Feb 2006 14:01:15 GMT ( From: bill@cs.uofs.edu (Bill Gunshannon) Subject: Re: PDPs in the news!+ Message-ID: <44ubtbF3taeiU1@individual.net>   3 In article <nT8dMqN1+LF4@eisner.encompasserve.org>, > 	koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:y > In article <890539d90602080243j27520c65ga1adae2cb4ea161@mail.gmail.com>, Carl Friedberg <frida.fried@gmail.com> writes: , >> On 6 Feb 2006 08:03:59 -0600, Bob Koehler3 >> <koehler@eisner.nospam.encompasserve.org> wrote: N >>> In article <dq0cg3$1db$1@rumours.uwaterloo.ca>, dfevans@bcr10.uwaterloo.c= >> a (David Evans) writes: >>> : >>> >  The author also seems a bit confused about what PDP >>> > means. >>> K >>>    The author may have read about them, but has probably never actually $ >>>    seen a Purple Data Processor. >>>  >>> : >> Bob, Please, the PDP1,5,6,7,10s  were NOT purple.. Carl > D >   Niether is the 12 down at the Smithsonian (or is that a 15?), or# >   the Micro PDP-8 I used to have.  > G >   We all know that PDP really stands for Programmable Data Processor, G >   except that it started out as Programmed Data Processor.  But lots  G >   of us have spent time in a room with a PDP-11/70 and I don't think  / >   the author has spent time with any of them.   D Some of us are still spending time in rooms with PDP-11's of various configurations.  :-)   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, 08 Feb 2006 15:00:03 GMT + From: Allison-nospam@nouce.bellatlantic.net  Subject: Re: PDPs in the news!8 Message-ID: <ln1ku1178342l419eeosa3qe4s9000137j@4ax.com>  E On 8 Feb 2006 07:41:01 -0600, koehler@eisner.nospam.encompasserve.org  (Bob Koehler) wrote:  x >In article <890539d90602080243j27520c65ga1adae2cb4ea161@mail.gmail.com>, Carl Friedberg <frida.fried@gmail.com> writes:, >> On 6 Feb 2006 08:03:59 -0600, Bob Koehler3 >> <koehler@eisner.nospam.encompasserve.org> wrote: N >>> In article <dq0cg3$1db$1@rumours.uwaterloo.ca>, dfevans@bcr10.uwaterloo.c= >> a (David Evans) writes: >>> : >>> >  The author also seems a bit confused about what PDP >>> > means. >>> K >>>    The author may have read about them, but has probably never actually $ >>>    seen a Purple Data Processor. >>>  >>> : >> Bob, Please, the PDP1,5,6,7,10s  were NOT purple.. Carl > C >  Niether is the 12 down at the Smithsonian (or is that a 15?), or " >  the Micro PDP-8 I used to have. > F >  We all know that PDP really stands for Programmable Data Processor,F >  except that it started out as Programmed Data Processor.  But lots F >  of us have spent time in a room with a PDP-11/70 and I don't think . >  the author has spent time with any of them.  ' :)  Bob, your being kind to the writer.   ? Those of us that live with them and enjoy the features are hard @ pressed to find so called modern machines however fast they are * as satisfying or plain hard core reliable.   Allison    ------------------------------   Date: 8 Feb 2006 09:19:13 +0100 / From: huber@NIRWANA-mppmu.mpg.de (Joseph Huber) 1 Subject: Re: perl build error (& subscribe error) + Message-ID: <wFAx4iSxG8Sh@vms.mppmu.mpg.de>   } In article <craigberry-C5B8A5.23174907022006@news.isp.giganews.com>, "Craig A. Berry" <craigberry@mac.com.spamfooler> writes: - > In article <CAR+F7vzBnPw@vms.mppmu.mpg.de>, 3 >  huber@NIRWANA-mppmu.mpg.de (Joseph Huber) wrote: F >> In article <dsa4jl$nl6$1$8300dec7@news.demon.co.uk>, Chris Sharman ' >> <chris.sharman@sorry.nospam> writes:  >> >  L >> > That's a bit chicken and egg, if that's the case - I've never seen any 6 >> > other product that couldn't install from scratch.. >> Yeah it is, You need Perl to build Perl ... > H > You absolutely do not need Perl in order to build Perl.  Nor is there H > any likelihood that building Perl will conflict with an existing Perl E > installation unless Perl's library search list has been explicitly   > modified via logical names.   C Maybe not needed, but actually the build procedure seems to use it: E the starting point of this thread was an error message along the line E " error in something.PM - need 5.005 or newer perl ..." when building  5.8.8 .   F Anyhow, I am gratefull You provide ready built VMS Perl distributions,
 Thank You.    --  @    Joseph Huber , Muenchen,Germany:  http://www.huber-joseph.de/   ------------------------------  $ Date: Wed, 8 Feb 2006 10:22:14 -0500C From: "David Turner, Island Computers US Corp" <dbturner@icusc.com> * Subject: Question about Greece/EU Warranty6 Message-ID: <1UnGf.1737$UD1.60@bignews2.bellsouth.net>  E A customer of mine in Greece has a DS25 that freezes up after 4 hours @ HP told him that they think that the problem is the motherboard.  I He purchased it in 2004 through an Authorised solutions provider in Italy  (durst) K I told him that it should be under warranty for THREE years (HP told him it  was 1 year) L Now being in the USA I know it is only 12 months on US supplied systems, but" I am POSITIVE that all EU supplied NEW systems carry three years   $ Can anyone in the EU confirm this???  L The poor guy is at a mid sized printing company and will be forced to buy an7 $8500 part when HP should be replacing it at no charge.    David      --     David B Turner Island Computers US Corp 2700 Gregory St, Suite 180 Savannah GA 31404  Tel: 912 447 6622 X201 Cell: 912 447 6622 X252  Fax: 912 201 0402  Email: dbturner@icusc.com  Web: http://www.islandco.com% ===================================== < All orders are subject to the following terms and conditions. of sale. These should be read before ordering.% http://www.islandco.com/warranty.html    ------------------------------  % Date: Wed, 08 Feb 2006 03:29:20 -0500 ' From: Dave Froble <davef@tsoft-inc.com> * Subject: Re: sneak preview - BACKUP/REPAIR/ Message-ID: <EYOdnSeDPZ8kNnTeRVn-gQ@libcom.com>    David J Dachtera wrote:  > Dave Froble wrote: >  >>David J Dachtera wrote:  >>D >>>...and, of course, we'd like to see that back-ported further than< >>>possible (V5.5-2, V6.2[-1H3], V7.2-2, V7.3-2, V8.2[-1?]). >>>  >>, >>Here's the diagnose part of your backport. >  > I > Of course, I meant back-port the additional cod to the earlier versions G > of BACKUP. There's already a DCL proc. on the freeware CD to do this.  >   I I know what you meant.  My post was a not subtle way of saying "go pound  F salt".  I really don't understand these types of requests.  It's hard H enough keeping the current version of any software working.  To ask for I a feature to be placed in an older version assumes that it can easily be  H plugged in.  This assumption too many times will be invalid.  To expect I an in-depth analysis of every prior version of any utility is asking way  	 too much.    --  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, 08 Feb 2006 16:25:38 GMT & From: John Reagan <john.reagan@hp.com>* Subject: Re: sneak preview - BACKUP/REPAIR0 Message-ID: <6YoGf.2818$jZ.190@news.cpqcorp.net>   David J Dachtera wrote:  > Dave Froble wrote: >  >>David J Dachtera wrote:  >>D >>>...and, of course, we'd like to see that back-ported further than< >>>possible (V5.5-2, V6.2[-1H3], V7.2-2, V7.3-2, V8.2[-1?]). >>>  >>, >>Here's the diagnose part of your backport. >  > I > Of course, I meant back-port the additional cod to the earlier versions G > of BACKUP. There's already a DCL proc. on the freeware CD to do this.  >   F We have some rather strict guidelines about backporting.  In general, H customers who have stayed with older versions of OpenVMS, do so because  they are stable.  E Backporting new features, while a worthy goal, increases the risk of  @ instability (Guy can't be perfect all of the time).  BACKUP, in F particular, demands even more paranoia since it is complex, and it is ! hard to test in all combinations.   E Unless there is a very compelling business reason (ie, LOTS of money  < riding on the backported feature), it probably won't happen.   --   John Reagan / HP Pascal/{A|I}MACRO for OpenVMS Project Leader  Hewlett-Packard Company    ------------------------------   Date: 8 Feb 2006 01:17:18 -0800  From: stuie_norris@yahoo.com.au 2 Subject: Re: Using SDA to view Global Section DataC Message-ID: <1139390238.321243.231380@z14g2000cwz.googlegroups.com>   
 Hi Volker,  + Thanks for the response.  Worked perfectly.   @ How about the case when I am not lucky enough to know the return addresses from sys$crmpsc?  , Is it possible to work out from MAP and SDA?   Thanks   Stuart   ------------------------------   Date: 8 Feb 2006 01:51:06 -0800 / From: "Volker Halle" <volker_halle@hotmail.com> 2 Subject: Re: Using SDA to view Global Section DataC Message-ID: <1139392266.501866.108950@g47g2000cwa.googlegroups.com>    Stuart,   E you can obtain the GSTX (Global Section Table Index) for your section  with SDA> SHOW GSD  , Then SDA> SET PROC/IND=<pid-of-your-process>F SDA> SHOW PROC/PAGE/GSTX=<your-section-gstx> ! should work with recent versions of OpenVMS.  D You will see the mapped addresses of your section within the process. and you can examine them with SDA> EXA address   Volker.    ------------------------------  $ Date: Wed, 8 Feb 2006 01:37:39 -0500, From: "Richard Tomkins" <tomkinsr@istop.com>) Subject: Re: VAXstation 2000 boot devices = Message-ID: <43e99290$0$27797$6d36acad@titian.nntpserver.com>   K You'd also have to create a new queue manager as the node name is embedded.   ? "David J Dachtera" <djesys.nospam@comcast.net> wrote in message % news:43E94D35.6944E84B@comcast.net...  > Bob Koehler wrote: > > 9 > > In article <43E7C010.63F8AF2B@teksavvy.com>, JF Mezei & <jfmezei.spamnot@teksavvy.com> writes: > > > L > > > OR: once the 2000 is booted as satellite of the 4000 (make sure it hasJ > > > no local pagefiles), you can use backup/IMAGE to transfer the 4000'sJ > > > disk onto the 2000's disk and then boot the 2000 outside the cluster and $ > > > personalise node name etc etc. > > E > >    sys$manager:cluster_config.com has a procedure for cloning the K > >    system disk that you might want to use instead of doing backup/image  > >    yourself. > G > I looked at the code for that and I'm not sure I'd want to do it that I > way. Granted, it is likely both documented and supported. I just didn't I > like what I was looking at after I thought about things like the DECnet C > databases, UCX service files that are node-specific and the like.  > K > >    If you simply do backup/image your 2000 will try to boot as the same L > >    DECnet, SCS, LAT, and IP addreses the 4000 is using.  Which means youI > >    want to disconnect it from the network and fix all that before you  > >    reconnect it. > G > ...or boot up minimal, reconfigure the nodename, SCSSystemId, DECnet, F > UCX, etc., then CAUTIOUSLY boot up (which you'd want to do in either > case). >  > --   > David J Dachtera > dba DJE Systems  > http://www.djesys.com/ > ( > Unofficial OpenVMS Marketing Home Page# > http://www.djesys.com/vms/market/  > * > Unofficial Affordable OpenVMS Home Page:! > http://www.djesys.com/vms/soho/  > $ > Unofficial OpenVMS-IA32 Home Page:! > http://www.djesys.com/vms/ia32/  > + > Unofficial OpenVMS Hobbyist Support Page: $ > http://www.djesys.com/vms/support/   ------------------------------  % Date: Wed, 08 Feb 2006 03:30:45 -0500 ' From: Dave Froble <davef@tsoft-inc.com>  Subject: Re: vaxstation 4000vlc / Message-ID: <EYOdnSaDPZ-QMXTeRVn-gQ@libcom.com>    GreyCloud wrote:1 > Phillip Helbig---remove CLOTHES to reply wrote:  > F >> In article <7sZFf.16152$wl.1194@text.news.blueyonder.co.uk>, "Colin? >> Butcher" <colin_DOT.butcher_AT@xdelta_DOT.co_DOT.uk> writes:  >>C >>> There's may be a block size limit in the boot ROM code for the   >>> maximum sizeL >>> of disc you can boot from. Many of the earlier VAXes had trouble bootingL >>> from discs greater than around 1Gbyte. Data discs should be OK, providedH >>> that you're running a more recent version of VMS - there was a size 	 >>> limit ( >>> change around V5.5-2 / V6.2 I think. >> >> >>F >> I believe this was only for the VAXstation 3100, not the VAXstation >> 4000, series. >> > K > I've got a 2Gb drive in mine.  And it spins at 10k.  A bit pricey but it  , > did speed up things like the help command. >   < Do you have the help library decompressed?  That is helpful.   --  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    ------------------------------  $ Date: Wed, 8 Feb 2006 08:28:10 -0500- From: William Webb <william.w.webb@gmail.com>  Subject: Re: vaxstation 4000vlc I Message-ID: <8660a3a10602080528n7e6e8150s8085243e4baa6889@mail.gmail.com>   3 On 2/8/06, Dave Froble <davef@tsoft-inc.com> wrote:  > GreyCloud wrote:3 > > Phillip Helbig---remove CLOTHES to reply wrote:  > > H > >> In article <7sZFf.16152$wl.1194@text.news.blueyonder.co.uk>, "ColinA > >> Butcher" <colin_DOT.butcher_AT@xdelta_DOT.co_DOT.uk> writes:  > >>D > >>> There's may be a block size limit in the boot ROM code for the > >>> maximum sizeL > >>> of disc you can boot from. Many of the earlier VAXes had trouble boot= ing L > >>> from discs greater than around 1Gbyte. Data discs should be OK, provi= ded I > >>> that you're running a more recent version of VMS - there was a size  > >>> limit * > >>> change around V5.5-2 / V6.2 I think. > >> > >> > >>H > >> I believe this was only for the VAXstation 3100, not the VAXstation > >> 4000, series. > >> > > L > > I've got a 2Gb drive in mine.  And it spins at 10k.  A bit pricey but i= t . > > did speed up things like the help command. > >  > > > Do you have the help library decompressed?  That is helpful. >  > --6 > David Froble                       Tel: 724-529-04506 > Dave Froble Enterprises, Inc.      Fax: 724-529-0596@ > DFE Ultralights, Inc.              E-Mail: davef@tsoft-inc.com > 170 Grimplin Road  > Vanderbilt, PA  15486  > - Y'all are talking about two different things.   A One is the apx 1.07GB system disk limitation for certain MicroVAX  3100s and VAXstation 3100s.   F The other is the apx 8.3 GB limitation for disk volumes under VMS 5.x.  F I referenced the particular sections of the FAQ that discuss these TWO issues in a very recent post.   J (Out THREE issues are surprise, fear, and a system disk no larger than...)   WWWebb --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, 8 Feb 2006 11:27:08 -0000= From: "Alan Scott" <alan dot scott at dial dot pipex dot com> F Subject: Re: VS3100 SCSI disks (was: re: VAXstation 2000 boot devices). Message-ID: <TYCdnVaIkfwVSHTeRVnytA@pipex.net>  L Thanks for the correction.    Our people who were trying to use larger SCSI K disks with old VS3100s/4000s were trying to use a single 4GB external boot  L device to replace the internal drive (RZ24?), which they found hard to swap M quickly if/when a drive failed.    I got them to buy vintage RZ55/6 external  K enclosures and disks eventually, these booted fine, though fussiness about  H the external cables and terminators still made reconfiguration a little B problematic.    They're still running, with VMS 5.1 or .2 I think.  9 <Allison-nospam@nouce.bellatlantic.net> wrote in message  2 news:152du1hpc3vpv99ul29cf0iefkgs13e0vu@4ax.com...D > On Sun, 5 Feb 2006 22:08:07 -0000, "Alan Scott" <alan dot scott at  > dial dot pipex dot com> wrote:   > Incorrect. > H > The disk size limitations for the VS2000 is around 200mb (no mfm disksE > larger were made).  There were RLL and EDSI disks but they were not  > VS2000 compatable. > ... E > The disk limitations for the 3100s was 1.07mb and only for the boot G > disk (boot rom limitation).  The SCSI in the 3100s is SCSI-II and any A > disk I've tried (up to 4.3gb) has worked as second/third/fourth @ > drives.  PC style partitioning is not compatable with VAX fileD > boot as the boot is limited in addressing.  Later 3100s could boot > larger disks.  > ... 	 > Allison  >    ------------------------------   End of INFO-VAX 2006.078 ************************