1 INFO-VAX	Sat, 15 Apr 2000	Volume 2000 : Issue 211       Contents: DEC BASIC Open w/RRL set Re: DEC BASIC Open w/RRL set Re: DEC BASIC Open w/RRL set Re: DEC BASIC Open w/RRL set Re: DECC 6.n BUG? /OPTIMIZE  Re: DECC 6.n BUG? /OPTIMIZE ! Re: Dropping DECnet..don't do it! ! Re: Dropping DECnet..don't do it! ! Re: Dropping DECnet..don't do it! ! Re: Dropping DECnet..don't do it!  Re: how test the network ? Re: Maximum Password Lengths Re: Maximum Password Lengths Re: Maximum Password Lengths Re: OpenVMS Certification  Re: OpenVMS Certification  Re: OpenVMS Times ! Re: Scheduler & PSDC alternatives % Re: Shock, horror! An OpenVMS advert!  Re: So who will buy VMS ?   F ----------------------------------------------------------------------  # Date: Sat, 15 Apr 2000 16:09:19 GMT / From: "Tom Simpson" <simpsont@xxx.mediaone.net> ! Subject: DEC BASIC Open w/RRL set E Message-ID: <Pu0K4.30102$h6.142832@typhoon.jacksonville.mediaone.net>   C Does anyone have an example of how to open a file in DEC Basic with K the RRL (read regardless of lock bit set?)?   I think that's what I want...   F I am trying to open a program log file so it can be examined while theH program is still running, similar to how you can examine a batch job log file.   L I tried, but have not been able to get it to work so far...   Many moons agoJ I had a routine that would do this in Pascal but I can't locate the source code for it.   Thanks in advance.   Regards, Tom    ------------------------------  % Date: Sat, 15 Apr 2000 09:43:02 -0700 5 From: "cstranslations" <cstranslations@email.msn.com> % Subject: Re: DEC BASIC Open w/RRL set ) Message-ID: <ezw0nmvp$GA.206@cpmsnbbsa03>   J I don't have the BASIC manuals here at home. The set runs about $235 whichJ is a bit more than I want to spend (and I'd do it in C even though there'sC more involved in reading and writing to the file - I've done it for L applications before and I'm doing it for an application I'm developing right now).   * Nevertheless, to answer the question . . .  C Dig out the BASIC manual. What you are looking for is the "USEROPEN J subroutine"  specifier on the OPEN statement. When you use this BASIC doesJ some preliminary set up of the RMS FAB and RAB control structures and thenH executes the subroutine you specify. The subroutine that you specify canH fiddle around with the control structures and hypothetically set the fabH fab$b_shr field to allow shared reading. Control then returns to BASIC's open statement.   L That's the idea anyway. I've never written code to write to a log file usingB BASIC. Keep in mind that you are going to have to flush the outputI periodically or you're not going to see anything. I suppose your useropen K subroutine could store the address of the RAB in a global variable that you  could then pass to sys$flush.    Joe     8 Tom Simpson <simpsont@xxx.mediaone.net> wrote in message? news:Pu0K4.30102$h6.142832@typhoon.jacksonville.mediaone.net... E > Does anyone have an example of how to open a file in DEC Basic with E > the RRL (read regardless of lock bit set?)?   I think that's what I  want...  > H > I am trying to open a program log file so it can be examined while theJ > program is still running, similar to how you can examine a batch job log > file.  > J > I tried, but have not been able to get it to work so far...   Many moons ago L > I had a routine that would do this in Pascal but I can't locate the source > code for it. >  > Thanks in advance. > 
 > Regards, > Tom  >  >  >  >    ------------------------------  # Date: Sat, 15 Apr 2000 16:58:24 GMT ) From: "Neil Rieck" <n.rieck@sympatico.ca> % Subject: Re: DEC BASIC Open w/RRL set : Message-ID: <Qc1K4.29779$2D6.915437@news20.bellglobal.com>  8 Tom Simpson <simpsont@xxx.mediaone.net> wrote in message? news:Pu0K4.30102$h6.142832@typhoon.jacksonville.mediaone.net... E > Does anyone have an example of how to open a file in DEC Basic with E > the RRL (read regardless of lock bit set?)?   I think that's what I  want...  > H > I am trying to open a program log file so it can be examined while theJ > program is still running, similar to how you can examine a batch job log > file.  >   * open "my_file.ext" for input as #100,    & ,access read    & 
 ,allow modify   
 when error in    while 1=1 K     get #100, regardless ! READ or LINPUT if a MAP was not used in the open 	 statement      gosub do_something   next use    my_error = err end when  J I hope this helps. BTW, if the file has been opened by a process that doesH the BASIC equivalent of "ALLOW NONE", then you may have to open the fileK with a USEROPEN statement and then directly manipulate the FAB (file access K block) and RAB (record access block) bits directly before continuing. Also, G if the other file is open for write with the BASIC equivalent of "ALLOW H NONE", be advised that data isn't written to the file until that process) either closes or flushes its I/O channel.   
 Neil Rieck* Kitchener(New Berlin?)/Waterloo/Cambridge, Ontario, Canada.! http://www3.sympatico.ca/n.rieck/ 6 http://www3.sympatico.ca/n.rieck/docs/alpha_diary.html   ------------------------------  % Date: Sat, 15 Apr 2000 10:43:34 -0700 3 From: Jeff Coffield <Jeffrey@DigitalSynergyInc.com> % Subject: Re: DEC BASIC Open w/RRL set 5 Message-ID: <38F8AA46.FCAEB887@DigitalSynergyInc.com>    Tom Simpson wrote:  E > Does anyone have an example of how to open a file in DEC Basic with M > the RRL (read regardless of lock bit set?)?   I think that's what I want...  > H > I am trying to open a program log file so it can be examined while theJ > program is still running, similar to how you can examine a batch job log > file.  > N > I tried, but have not been able to get it to work so far...   Many moons agoL > I had a routine that would do this in Pascal but I can't locate the source > code for it. >  > Thanks in advance. > 
 > Regards, > Tom        PROGRAM READ_LOG       DECLARE STRING STRING.DATA       WHEN ERROR IN 1         OPEN 'XYZ.LOG' FOR INPUT AS FILE 1%,    &               RECORDTYPE ANY,    &!             ACCESS READ,        &              ALLOW MODIFY     USE          PRINT ERT$(ERR)          GOTO FINISH      END WHEN   LOOP:      WHEN ERROR IN #         INPUT LINE #1%, STRING.DATA      USE %         PRINT ERT$(ERR) IF ERR <> 11%          GOTO FINISH      END WHEN        PRINT EDIT$(STRING.DATA, 4%)
     GOTO LOOP    FINISH:      END    ------------------------------  % Date: Sat, 15 Apr 2000 19:12:02 +0930 / From: Mark Daniel <mark.daniel@wasd.vsm.com.au> $ Subject: Re: DECC 6.n BUG? /OPTIMIZE/ Message-ID: <38F8396A.9C96447C@wasd.vsm.com.au>   ` Yes, sorry can't get to this system again for a couple of weeks ... I'm off-site, which is why I_ didn't report it via CSC (or whatever Q calls it these days) - wouldn't be around to handle the P call.  Should have grabbed the CRTL image ident though, must be out of practice.  c Although, I wouldn't have thought differences in the CRTL would have produced such a result.  Seems V like it should be just integer arithmetic instructions for such elementary operations.    Thanks for trying the code, Ken.  9 "Ken Fairfield; SLAC: 650-926-2924; FAX: 926-3515" wrote:   J >         OBTW,  my  SYS$LIBARARY:DECC$SHR.EXE  show  this  for  the image >     identification:  > * >         Image Identification Information > ( >                 image name: "DECC$SHR"6 >                 image file identification: "X07.2-0"D >                 image file build identification: "X6TF-0050120000"9 >                 link date/time: 28-MAY-1999 23:13:51.98 1 >                 linker identification: "A11-39"  >  >         -Ken > --O >  Kenneth H. Fairfield            |  Internet: Fairfield@SLC.Slac.Stanford.Edu < >  SLAC, 2575 Sand Hill Rd, MS 46  |  Voice:    650-926-2924< >  Menlo Park, CA  94025           |  FAX:      650-926-3515  E +-------------------------------------------------------------------+ D  Mark Daniel            Opinions my own ... and on loan from others.E  mailto:Mark.Daniel@wasd.vsm.com.au (Mark.Daniel@dsto.defence.gov.au) E +-------------------------------------------------------------------+    ------------------------------  % Date: Sat, 15 Apr 2000 09:19:14 -0700 5 From: "cstranslations" <cstranslations@email.msn.com> $ Subject: Re: DECC 6.n BUG? /OPTIMIZE) Message-ID: <#Jp1UZvp$GA.231@cpmsnbbsa03>   I I tried out the code - with and with out the "alert" in each fprintf. Its 6 presence or absence didn't seem to affect things . . .  .     dec 3000 model 400 running OpenVMS 7.1-1H14     Compaq C V6.2-003 [of course on OpenVMS 7.1-1H1]  J I compiled it (no qualifiers, /opt qualifier, /noopt qualifier, even threwK in /debug on the off hand chance that that might "shift things around") and K linked (no qualifiers, /debug, /nodebug). All of these combinations gave me D pretty much the same results (the exceptions being where /noopt whatK specified and the resulting effect was that the output was generated slower  by a factor 100 or so.  L I would tend to think that there's something else going on (other than a bugI in the compiler). That's some pretty basic math and I would have to think 1 that someone would have tripped over this before.   G Something like this in Visual C++ I'd believe.  OpenVMS - come on . . .    :-)    Joe   : Mark Daniel <mark.daniel@wasd.vsm.com.au> wrote in message( news:38F75CDE.D282B81@wasd.vsm.com.au...E > /OPTIMIZE seems to BREAK SIMPLE INTEGER ARITHMETIC in the following  > environment: > - >   DEC 3000 Model 500 running OpenVMS V7.2-1 - >   Compaq C V6.2-003 on OpenVMS Alpha V7.2-1  > @ > The included demonstration of the problem produces the resultsE > following the code, vastly different with /OPT (and incorrect) than C > /NOOPT (correct).  Essentially 1,100,000 divided by 1,000,000 can  > produce a result of 0!!    ------------------------------  # Date: Sat, 15 Apr 2000 12:51:59 GMT 8 From: Veli =?iso-8859-1?Q?K=F6rkk=F6?= <korkko@decus.fi>* Subject: Re: Dropping DECnet..don't do it!( Message-ID: <38F85795.41F453B5@decus.fi>  F I'have installed just about every version of DECnet since DECnet Phase IVB from VMS version V4.5. And that therefore includes DECnet Phase IV	 Extension G products, various DECnet/OSI, DECnet/Plus etc levels including most of  F field test versions. I've used them in many various environments and I can G say with reasonable confidence that DECnet Phase V works. MAny of those A systems have had DECnotes also. I have used NSP, TP4 and RFC1006+  methods  of communicating.   B DECnotes V2.5 works reasonably well and is very straightforward to install G on top of DECnet be it Phase IV or Phase V. Whilst there certainly was  9 strange problem on your system, I simply refuse to accept   J > figured it. That caused different errors. While I was away at a session,K > somebody fixed it, but nobody knows what it was (it was a case of "fiddle ! > with it until it gets better").   H Troubleshooting DECnet/Plus has been and still is fairly straightforward and logical.   _veli      Terry Kennedy wrote: > , > Richard Brodie <R.Brodie@rl.ac.uk> writes:E > > I'd be interested to hear why you think DECnet V is an issue. The A > > whole point of the IV->V change was that Phase V supported IP ? > > (and another protocol family whose name escapes me) as well   > > as the native network layer. > K >   My reaction to Phase IV is "over my dead body!". That used to be an un- L > informed opinion, but as part of the DECUServe migration test, I installedK > VMS V7.2 and DECnet-Plus on an AS2100. I selected the default options for F > the DECnet-Plus install. I then installed Notes on that same system. > M >   While it was possible to access a remote DECnet Notes conference, inbound L > Notes access to the AS2100 caused DECnet to fail (the first access attemptN > would give a connection reject, and accesses after that got "dead air" until > the system was rebooted).  > M >   This happened a week or two before the Providence '99 DECUS Symposium. At K > the Symposium, we had the Compaq DECnet-Plus engineers as well as a bunch K > of very knowledgable DECUS members look at it, and nobody could figure it L > out. Finally one of the Compaq folks deleted the configuration and re-con-J > figured it. That caused different errors. While I was away at a session,K > somebody fixed it, but nobody knows what it was (it was a case of "fiddle ! > with it until it gets better").  > L >   Needless to say, the new production DECUServe system will be using PhaseI > IV - all the folks who were saying "use Phase V" have been suitably un-  > impressed by this fiasco.  > N >   For the original poster - if you're a MultiNet shop, you should be able toN > use Phase/IV (originally called DECnot) to encapsulate DECnet over Phase IV. > 6 >         Terry Kennedy             http://www.tmk.com7 >         terry@tmk.com             Jersey City, NJ USA    ------------------------------  # Date: Sat, 15 Apr 2000 13:54:55 GMT $ From: Ed Wilts <ewilts@mediaone.net>* Subject: Re: Dropping DECnet..don't do it!, Message-ID: <38F874AE.A8B51475@mediaone.net>   Terry Kennedy wrote: > K >   I suspect that the Compaq Phase IV code is only touched to fix critical K > bugs, and they don't want to make enhancements to it. Not to mention that 2 > Phase IV improvements would undermine Phase V...  F Not even critical bugs are fixed.  According to the CSC, if you have aD Phase IV routing Alpha and turn a circuit off, you have have a 50/50F chance of bugchecking the host.  Engineering will not commit resourcesE to fixing this, stating that you must upgrade to Phase V.  Routing is ! simply not supported in Phase IV.   H One would have thought that a bugcheck would be considered critical, but then what do I know?   --   Ed Wilts Mounds View, MN, USA mailto:ewilts@mediaone.net   ------------------------------  # Date: Sat, 15 Apr 2000 14:18:14 GMT $ From: Ed Wilts <ewilts@mediaone.net>* Subject: Re: Dropping DECnet..don't do it!, Message-ID: <38F87A23.375CEF7C@mediaone.net>  9 "Ken Fairfield; SLAC: 650-926-2924; FAX: 926-3515" wrote:  > U > In article <38F67D13.6E280361@mediaone.net>, Ed Wilts <ewilts@mediaone.net> writes: = > > "Ken Fairfield; SLAC: 650-926-2924; FAX: 926-3515" wrote:  > >>M > >>         BTW, I made a _large_ effort  testing with Decnet-Plus and LANCP M > >>     on  a DS10 (at VMS 7.2-1) prior to upgrading the rest of my  cluster M > >>     from VMS 7.1-1H2 to 7.2-1 last December.  I had two primary  reasons M > >>     for  doing  so.   First,  so  that I could run DECNET over two of my M > >>     AS4100s' four network interfaces instead  of the single interface at . > >>     a  time as restricted under Phase IV. > > L > > This is a myth.  I spent some time with Tony in the CSC listening to himJ > > tell me that DECnet would only select the first functioning interface,B > > only to have me prove him wrong and clearly demonstrate a node3 > > simultaneously using DECnet on FOUR interfaces.  > J >         I'd like you to explain to  me  how  you did that.  Our AS4100'sJ >     have  4 network devices each, two of which are configured for DECNETJ >     (Phase IV).  So in NCP, I have four LINEs defined (so I  can  gatherJ >     network  statistics on all four interfaces) and two CIRCUITs defined >     and enabled.  H I've got 4 lines and 4 circuits for my interfaces (2 Ethernet, 2 FDDI). ? I have verified that the 2 Ethernet segments are definitely not H bridged.  A host on Ethernet 1 can definitely communicate with a host on( Ethernet 2 via my Phase IV Routing node.   $ show network/old  F will give you information about routing similar to what we used to getE before somebody decided to trim all the useful information out of the G display.  Most of my hosts have dual FDDI connections, but not all.  My E router will talk to hosts that have FDDI using FDDI, and will talk on A Ethernet 1 when it has to, or Ethernet 2 if it has to.  I haven't H verified if it will actually use both FDDI paths to a single host.  FDDID here is really used a cluster interconnect, not a network transport.   > J >         Do your Alphas show non-zero packets  sent and received on _all_J >     (or  even  just more than one) DECNET circuits you've  configured???2 >     If so, I'd sure like to know how you did it!  # I'll send you the proof via e-mail.    	.../Ed  -- C Ed Wilts Mounds View, MN, USA mailto:ewilts@mediaone.net   ------------------------------  # Date: Sat, 15 Apr 2000 14:12:47 GMTC( From: Terry Kennedy <terry@gate.tmk.com>* Subject: Re: Dropping DECnet..don't do it!' Message-ID: <Ft2A5B.65M@spcuna.spc.edu>   % Veli K?rkk? <korkko@decus.fi> writes:iD > DECnotes V2.5 works reasonably well and is very straightforward to	 > installsI > on top of DECnet be it Phase IV or Phase V. Whilst there certainly was e; > strange problem on your system, I simply refuse to accept   H   Remember, the other end of this was a VAX running VMS V5.5-2 with V1.x Notes.  J > Troubleshooting DECnet/Plus has been and still is fairly straightforward > and logical.  E   That may be, but given that neither Engineering nor various skilled.H Symposium attendees could figure it out, and that unpaid volunteers have/ to support the system, we'll be using Phase IV.   E   Oh, and there were other problems - the OSU web server had various hF problems, and apparently (I didn't participate in the discussions) theA server's author said that Phase V was problematic for the server.w  - 	Terry Kennedy             http://www.tmk.comw5         terry@tmk.com             Jersey City, NJ USAt   ------------------------------  # Date: Sat, 15 Apr 2000 15:47:06 GMT / From: "Tom Simpson" <simpsont@xxx.mediaone.net>l# Subject: Re: how test the network ?nE Message-ID: <_90K4.30099$h6.142783@typhoon.jacksonville.mediaone.net>a  K One thing you might want to keep in mind is that the fastest NIC card speeda) may not turn out to be the optimal speed./  I I set the NIC cards to 100mb FD on my ES40 systems and it actually slowedCK down the network by generating lots of framing errors...  I had to back offnK the full duplex setting to get it work properly.  It should have worked buta
 I have not* had any spare time to figure out why, yet.  L I have seen references in this conference to using DTSEND for testing, but II have not been able to try it yet myself.   Sorry I can't be more helpful.u   Regards, Tomi  7 "Pedro Viena" <contracer2@hotmail.com> wrote in messagee- news:20000415021728.6515.qmail@hotmail.com...a >s >u6 > In my job we have three alphaservers DS20 in network >tE >   (not in cluster !), and we use 100Mb network cards (full-duplex).d >aB > We are have problems in the network (slow network). my question: >_7 > Is there any command to verify if the network is ok ?o >S- > We don't use muxservers, only switches IBM.y >' > Thank you in advance ! >e8 > ______________________________________________________8 > Get Your Private, Free Email at http://www.hotmail.com >i   ------------------------------  # Date: Sat, 15 Apr 2000 12:52:01 GMTa8 From: Veli =?iso-8859-1?Q?K=F6rkk=F6?= <korkko@decus.fi>% Subject: Re: Maximum Password Lengths ( Message-ID: <38F858B4.7E6EB1C0@decus.fi>  @ I think it would 14 on NT. Cannot say about W2K. I sure rememberA that we used to have minimum password length of 15 for privilegedCA accounts on VMS and then the policy got changed, i.e. new minimumnC lenght was 14 to accommodate NT restrictions, i.e. their maximum ofi 14.    _veli  x   "Boyle, Darren" wrote: > & > VMS is 32 characters, NT & W2K is 15 >  > > ----------B > > From:         Michael D. Ober[SMTP:mdo.@.wakeassoc.com.nospam]3 > > Sent:         Thursday, April 13, 2000 11:46 AMt > > To:   Info-VAX@Mvb.Saic.Comn* > > Subject:      Maximum Password Lengths > >sM > > What is the maximum user password length on VMS 7.2, and NT 4 Domain, and- > > a A > > Win 2000 Active Directory?  I need information for all three.@ > >t > > Thanks,j > > Mike Ober. > >e > >R > >  > H > **********************************************************************E > This message and any files transmitted with it are confidential andtL > may be privileged and/or subject to the provisions of privacy legislation.O > They are intended solely for the use of the individual or entity to whom they M > are addressed. If the reader of this message is not the intended recipient, D > please notify the sender immediately and then delete this message.K > You are notified that reliance on, disclosure of, distribution or copyingp  > of this message is prohibited. >  > Bank of BermudahH > **********************************************************************   ------------------------------  % Date: Sat, 15 Apr 2000 07:35:14 -0400 * From: David A Froble <davef@tsoft-inc.com>% Subject: Re: Maximum Password Lengthse- Message-ID: <38F853F2.764C0251@tsoft-inc.com>t   d.webb@mdx.ac.uk wrote:m > 0 > In article <8d53qp$42u$2@info.service.rug.nl>, >   helbig@astro.rug.nl wrote:F > > In article <8d515o$t27$1@nnrp1.deja.com>, d.webb@mdx.ac.uk writes: > >oG > > > > For OpenVMS, Passwords can be from 0 to 32 characters in lengthv > andf> > > > > can include alphanumeric characters, dollar signs, and > underscores. > > >hJ > > > It's high time that this restricted set of characters was increased.B > > > It should include virtually any character inputable from the
 > standard > > > keyboard at leaste > >c > > Why? > 6 > To increase the size of the possible password space. > = > There are two possible ways to increase the password space.g > ! > 1) Increase length of password. D > We can already go upto 32 characters but I doubt very much whetherF > anyone (unless forced to in a military/goverment establishment) uses( > more than 8 or 10 character passwords. > E > 2) Increase the range of characters which can be used in passwords.n > G > This can have it's downside. Those of our students who use Unix ofteniI > forget whether they have upper or lowercase letters in their passwords. = > However no such confusion exists with the other characters.a > ( > Why increase the password space size ? > E > Security.  The processing power available nowadays means that brutei? > force cracking of 6 to 8 character passwords is now feasible.  > B > VMS's other security features which prevent users getting at theH > SYSUAF file and prevent repeated login attempts restrict this problem.H > However I would rather have both this and an increased password space. > I > Another reason (Which if handled incorrectly can worsen security but is D > still a valid requirement for some users) is to allow for passwordG > synchronisation with other systems NT, UNIX etc which use much largerT > charactersets for passwords. >  > > > ie > > > 8 > > > !"$%^&*()_-+={}[]:;@'~#<,>.?/|\ + alphanumerics. > >2, > > Err, where's the pound sign on my LKxxx? > >. > ? > Replace  with local currency character if outside US and UK..E > If in US - well thats what you get for having forced everybody else $ > to have a $ on their keyboard.  :) >  > David Webb > VMS and Unix team leader > CCSS > Middlesex University > ( > Sent via Deja.com http://www.deja.com/ > Before you buy.t  L While there may be valid arguments for such a modification to VMS, there are other considerations.n  K In just looking at the issue, I for one do not believe that any brute forcetL method for guessing passwords will work on a VMS system with properly set upO security and reasonably obscure passwords of up to 32 characters.  The break-in N detection will usually extend the time required for such a task far beyond theO normal expected life of the computer system.  In addition, the correct password P may be guessed during a time the port is disabled by the break-in detection, andJ thus if the break-in attempt doesn't re-try passwords, something that willO lengthen the process, then the attempt will always fail, regardless of the timei allowed.  P Also, with the exception of known user accounts, the attempt must also guess theK username.  This will extend the requirements to a point where it's just not4M feasible to attempt a break-in.  As for SYSTEM, it's not hard to disable such5L accounts and manage the VMS system from another user account set up for this purpose.  H In a time where "less is better" :-( the VMS development team has finiteM resources and time.  I for one would rather these resources be devoted to the-P most useful work.  For example, DECnet IV over TCP/IP would rate far higher thanP password enhancements, and then another might say that Phase V already does that! and suggest a more pressing need.:   Just another opinion.    Dave   -- p4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596; 170 Grimplin Road               E-Mail: davef@tsoft-inc.come Vanderbilt, PA  15486g   ------------------------------   Date: 15 Apr 2000 10:01 CST-' From: carl@gerg.tamu.edu (Carl Perkins)t% Subject: Re: Maximum Password Lengthsw- Message-ID: <15APR200010012824@gerg.tamu.edu>g  D In article <8d6svc$usm$1@nnrp1.deja.com>, d.webb@mdx.ac.uk writes.../ }In article <8d53qp$42u$2@info.service.rug.nl>,  }  helbig@astro.rug.nl wrote:sE }> In article <8d515o$t27$1@nnrp1.deja.com>, d.webb@mdx.ac.uk writes:h }>F }> > > For OpenVMS, Passwords can be from 0 to 32 characters in length }and= }> > > can include alphanumeric characters, dollar signs, andm
 }underscores.s }> >I }> > It's high time that this restricted set of characters was increased. A }> > It should include virtually any character inputable from the.	 }standard  }> > keyboard at least }> }> Why?k }  } 5 }To increase the size of the possible password space.F } < }There are two possible ways to increase the password space. }   }1) Increase length of password.C }We can already go upto 32 characters but I doubt very much whether E }anyone (unless forced to in a military/goverment establishment) usesn' }more than 8 or 10 character passwords.l } D }2) Increase the range of characters which can be used in passwords. } F }This can have it's downside. Those of our students who use Unix oftenH }forget whether they have upper or lowercase letters in their passwords.< }However no such confusion exists with the other characters. }  } ' }Why increase the password space size ?. } D }Security.  The processing power available nowadays means that brute> }force cracking of 6 to 8 character passwords is now feasible. }  }David Webb.  G I don't think any of this is relevant. The password is not stored, only D a hashed version of it which is always a 32 bit integer. To get intoB an account you don't need the "correct" password, you only need toF use something that produces the same hash. In this case, the characterE set restriction should actually make it *harder* to break in since itdF means there are fewer combinations of valid input characters that workG out to any given hash value. (Unless I'm forgetting something - I don'tVC think there is any "length of password" data being stored, just thep hashed value.)  H There are a some known pairs of passwords that are interchangeable - butG I don't know any of 'em. (I've seen some, but promptly forgotten them.)o   --- Carl   ------------------------------  % Date: Sat, 15 Apr 2000 07:52:31 -0400a* From: David A Froble <davef@tsoft-inc.com>" Subject: Re: OpenVMS Certification- Message-ID: <38F857FF.ECD68D52@tsoft-inc.com>w   Phillip Helbig wrote:s > > > In article <38F6833B.C7D3E7F7@tsoft-inc.com>, David A Froble > <davef@tsoft-inc.com> writes:a > E > > > That was my goal, but Real Work always gets in the way of doing L > > > everything I'd like to do.  If I could figure out how to make a living$ > > > doing freeware, I'd love that. > >n > > Charge for it. > $ > Then it's no longer freeware.  :-) > K > > I'll confess, I really don't understand freeware, at least for productslD > > that were/are not trivial.  Your time and effort should be worth( > > something.  Just the view from here. > I > I think it's basically down to people saying "I've done this, it works,hD > it's useful---I'm sure other folks would like it as well" and thusH > making it available.  Sure, they could come up with even more stuff ifC > they were paid to do it, but this is really quite a hassle and iseG > probably not a "secure job".  So it's a question of make it availableD+ > for free or not make it available at all.f  I I guess it might depend on whether you work for a company, and do not getuP (directly) paid for each thing you do, or are a vendor and get paid directly forM whatever you do.  In the former case, you still get your salary regardless offM whether you do something extra.  In the latter, any work you do which doesn'tiP produce revenue would be time better spent on billable work, that is if you careO at all about making the mortgage payment each month.  The independantly wealthyn" would be an exception to this. :-)  I > I think the thing to do would be for Compaq to hire Hunter, Dave Jones,oF > Matt Madison, Richard Levitte and the like and integrate their stuffJ > into the operating system or at least layered products level.  No reasonF > to increase VMS prices as a result of this---hey, we're only talkingJ > about a few salaries, peanuts to Compaq---since they would more than payH > for themselves in increased sales.  (Some stuff is VMS-specific and soE > on, but some of the stuff is to provide "basic functionality" which J > people in some sense get "for free" with other operating systems but notI > with VMS---OK, some stuff like VMSTAR is mainly for unix compatibility,tG > and wouldn't be needed if everyone used VMS, but some stuff, like webiH > servers, newsreaders etc is just stuff which many people need nowadaysF > and to me has just as much reason to be a layered product as, say, a > compiler.) > F > Guys, what do you say?  If you could make your current salary or theJ > same salary as, say, Hoff (whichever is higher) and Compaq would pay youH > to continue your VMS development you've been doing up until now in the' > context of freeware, would you do it?n  L Gotta be careful here.  What if the 'current salary' is more than any of theO current VMS developers are making?  I can imagine significant disgruntlement ifjP people were brought in and paid more than the current people are making, or, youN face the prospect of raising everyone's salary to match the highest.  ProbablyI not much disgruntlement in that case. :-) (Except for Compaq management.)   C > What I like about VMS freeware (apart from the fact that here, in I > contrast to the Stallman crowd, the goal is not to undermine commercialCG > software) is that it is usually very much in the VMS mold.  By making ? > these true layered products, this would be just a bit better.d  I To be honest, I've used a few free products, and was happy that they wereu
 available.   Dave   --  4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596; 170 Grimplin Road               E-Mail: davef@tsoft-inc.come Vanderbilt, PA  15486i   ------------------------------  % Date: Sat, 15 Apr 2000 08:07:47 -0400n* From: David A Froble <davef@tsoft-inc.com>" Subject: Re: OpenVMS Certification- Message-ID: <38F85B93.651E8D15@tsoft-inc.com>e   Marty Kuhrt wrote: > _ > In article <38f61f21.116235507@news.wku.edu>, goathunter@PROCESS.COM (Hunter Goatley) writes:hC > > That was my goal, but Real Work always gets in the way of doingPJ > > everything I'd like to do.  If I could figure out how to make a living" > > doing freeware, I'd love that. > G > Change the name of MadGoat Software to MadGoat Linux.  IPO, using therJ > "we're making our money by supporting freeware" business plan that seemsA > to work for the RedHat, VA Linux, etc, crowd.  Get a market capoF > in the billions, and then announce that you're changing the spelling" > of your version of linux to vms. >  > Simple, really.  :)i  O I've a good friend with a vertical market application.  He doesn't sell it.  HeiO charges his customer's a flat monthly fee to use the application.  All upgradesl9 and enhancements are provided as part of the monthly fee.,  L While some might think this is an expensive proposition for the customer, itP doesn't work out that way.  There are new requirements coming out continually inK the vertical market in question, and he is very responsive to this.  As one I example, when EDI first became popular, his application was very quick in P providing this capability.  One reason he's so successful in the vertical marketB is that his customer's usually are the first with solutions to new- requirements.  In other words, a good vendor.n  N In some ways I envy him.  Every month there is a known income, and he can planN accordingly.  With the known income, he can plan and implement new features onP his schedule, not when a customer finally agrees that they need the enhancement,P and thus he is usually ahead of the competition, and that's why he dominates his vertical market.  I I guess that you could say he is in the service/support business, ie; thelO application is free, and the monthly fee is for service and support, and 'rightcP to new version'.  I'm not sure how often such an arrangement will work, but whenM it does I guess it works well.  My friend drives a new Mercedes and I drive ar beat up old Nissan truck. :-)m   Dave   -- n4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596; 170 Grimplin Road               E-Mail: davef@tsoft-inc.com  Vanderbilt, PA  15486e   ------------------------------  % Date: Sat, 15 Apr 2000 09:50:29 +0200y$ From: Paul Sture <sture.ch@sture.ch> Subject: Re: OpenVMS Times* Message-ID: <VA.00000003.00ac66e5@peacock>  ' > From: paddy.o'brien@zzz.tg.nsw.gov.auw > Newsgroups: comp.os.vmsm > Subject: Re: OpenVMS Times' > Date: Sat, 08 Apr 2000 18:47:05 +0010- >  a2 > missed Alan's post, but Hoff replied thusly .... >  sM > >In article <8chpkp$cn3$1@nnrp1.deja.com>, Alan Greig <agreig@my-deja.com> _
 > >writes:: > >:In article <8cg9qu$3vd$1@mailint03.im.hou.compaq.com>,G > >:A couple of UK Channel Partners I have shown this to are interested E > >:in printing and distributing this themselves to VMS customers. Isn$ > >:there any problem in doing this. > >oJ > >  Not that I am aware of.  Have the folks check with Sue directly, and M > >  she can discuss this with them -- we already make the masters available  M > >  for Compaq-internal folks to download and print...  Sue can also likely iJ > >  get the folks pointers to other OpenVMS available "collateral".  (eg:K > >  the new Compaq "OpenVMS and Windows NT Integration for Dummies" book.)d > >eK > >:Also there is a European print publication OpenVMS Now (Compaq Ireland)lG > >:very similar to OpenVMS Times which is not available on the web. ItnJ > >:strikes me that the two could be combined - perhaps retaining EuropeanE > >:and US editions. Just wondering if you were aware of OpenVMS Now.e > >:I > >  Yes, folks here in OpenVMS Engineering are aware of the OpenVMS Now.nI > >  (I don't know that are are (or are not) plans to combine the two...)  >  _M > You guys also have at least one bigot in the Antipodean region.  C'est moi.l >  tS > I have not been able to access any of these facilities.  At least I found no way 1* > to navigate the web sites to read these. >  6  : Try http://www.openvms.digital.com/openvmstimes/index.html  
 Paul StureR (Finally surfing from home, despite of micro$chlock and linux attempts to deny me 	 doing so)    ------------------------------  # Date: Sat, 15 Apr 2000 16:31:52 GMTn, From: kaplow_r@eisner.decus.org (Bob Kaplow)* Subject: Re: Scheduler & PSDC alternatives' Message-ID: <2000Apr15.123152.1@eisner>   ] In article <8d2t9m$qlm$1@lure.pipex.net>, "Graham HAZLEGREAVES" <qg38@dial.pipex.com> writes:-N > The company I'm currently working for (I've actually been contracting on andJ > off with them since '92) has 80+ Vax and Alpha systems dotted around UK.M > These run various products but the one's we need to find an alternative fori > are: > ' >     PCM  (Polycenter Console Manager) ( >     PSDC  (Performance Data Collector)" >     PSPA   (Performance Advisor) >     Schedulern > H > Due to a major change within this company, the supplier (no prizes forH > guessing who!!) want to charge us a huge (6 figure) sum to re-license,L > unlike other suppliers who've just said it doesn't change things (in termsM > of licensing costs etc.).  We did think about paying (I think it lasted for J > about .000000000001 of a second) before deciding - NO (I daren't put the7 > exact words that were said as it might be libelous!).f    Boy, does this sound familiar...  L > We've already decided on a replacement for PCM - ROBO Central from Heroix.D > We're already using it at one site plus we have plenty of in houseJ > experience of ROBOMON.  Another minor point is that myself and one of myD > colleagues know the developer from our Digital days (we all worked > together). > N > With regards to the other 3 we're having a bit of a problem in that we don't: > seem to be able to find anything that does what they do.  C Look in the OpenVMS roadmap. IIRC they have a slide that lists manye8 replacements for each of the former polycenter packages.  N > On a couple of our newer systems we've installed KRONOS as a replacement forF > Scheduler.  No disrespect to the developers, but it isn't as good asI > Scheduler and we've had to make extensive mods to some of our Schedulerg > jobs.a  H A sad tale here. Folks that are no longer with our company chose AutosysK from Platinum as our new scheduler, because it claimed support for Unisys AsK series, as well as VMS and Unix. Well, it never did work right on VMS, from.H day 1. And to make matters worse, guess who bought Platinum, less than aG month after we dumped big bux into Auosys? That's right, the same bloodsK suckers who bought Polycenter and sent us on this quest in the first place.r  I After struggling to get Autosys to work on VMS for a year (made harder byIK all the former Platinum folks taking a hike), CA just came back and offeredc+ us TNG and DECscheduler as a workaround :-)w  K [A funny sidebar: We lost a lot of colleagues to Platinum. I was talking toeK one about this after they got devoured, and my friend was looking for work.aK I jokingly suggested we should have a clause in our contracts that say thatnG if the product or company is sold, we have the option of asking for ourcL money back. he told me Platinum had LOTS of such contracts, IIRC to the tune. of over $1B. Most specifically referenced CA.]  M > As regards PSDC and PSPA, so far we haven't found anything that can providerJ > us with the same functionality.  We might not have looked hard enough, I. > admit, but we did only start this yesterday.  K We've gone with Best/1 on our Unix boxes, and have been about to buy it foryE VMS for at least 6 months now. The problem is they used the PSDC data G collector, so now all they have is something that parses binary MONITOR} data.#  L We also evaluated the new Compaq ECP tool, which is the old Capicity PlannerL with a new data collector. It's still a long way from what PSPA gave us, butJ it's gradually improving. The advisor isn't there, but some of the graphic displays are there.o  % Mostly, we're bak to using MONITOR...x   	Bob Kaplow	  E SPAM:	spamrecycle@ChooseYourmail.com	uce@ftc.gov	postmaster@127.0.0.1/   ------------------------------  % Date: Sat, 15 Apr 2000 08:22:18 -0400i* From: David A Froble <davef@tsoft-inc.com>. Subject: Re: Shock, horror! An OpenVMS advert!- Message-ID: <38F85EFA.D065025D@tsoft-inc.com>r   Christopher Smith wrote: > , > On Fri, 14 Apr 2000, Adrian Birkett wrote: > L > > Does anybody know if this is available in the UK? We could do with a bit > > more visual representation!i > J > I have no idea, but I'd like to hang one of those on the wall here, too.L > Unfortunately, I probably won't recieve one unless decus us starts mailing > them to people.e > 
 > Regards, >  > Chris   J Well, I'd suggest anyone who wants to receive such reply to the following:  
 ----------K This is Sue Skonetski from the OpenVMS Group, send me your name and addressi  and I will get a few out to you.  0 susan.skonetski@compaq.com or Star::S_Skonetski.  
 Best Regards,h   Suer
 ----------   Can't get any easier than that.e   Dave   -- r4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596; 170 Grimplin Road               E-Mail: davef@tsoft-inc.coms Vanderbilt, PA  15486v   ------------------------------  % Date: Sat, 15 Apr 2000 08:32:35 -0400 * From: David A Froble <davef@tsoft-inc.com>" Subject: Re: So who will buy VMS ?- Message-ID: <38F86163.D1A9CBAD@tsoft-inc.com>e   Nigel Arnot wrote: > 
 <big snip>  ! > or (b) if they have to go Unix,sF > they'll go Linux and never be held to ransom by an O/S vendor again,  P But then they'd be held ransom by whomever they depend upon for Linux support (aP very nebulous entity) and no, I won't accept the statement that it's open sourceP and they can 'do it themselves'.  Working on an operating system is not the sameM as working on an application system, and many shops cannot even do that well.   ) Dave, who has serious doubts about Linux..   -- u4 David Froble                       Tel: 724-529-04504 Dave Froble Enterprises, Inc.      Fax: 724-529-0596; 170 Grimplin Road               E-Mail: davef@tsoft-inc.coml Vanderbilt, PA  15486e   ------------------------------   End of INFO-VAX 2000.211 ************************