1 INFO-VAX	Tue, 21 Nov 2006	Volume 2006 : Issue 641       Contents: DLT-V4! Re: Educational VMS IA64 licenses ! Re: Educational VMS IA64 licenses ! Re: Educational VMS IA64 licenses C End of life question for VAX 3100-96 and Alphaserver 5/533 question G Re: End of life question for VAX 3100-96 and Alphaserver 5/533 question G Re: End of life question for VAX 3100-96 and Alphaserver 5/533 question < Re: First TCP/IP (UCX) Release, Oldest VMS with (limited) IP< Re: First TCP/IP (UCX) Release, Oldest VMS with (limited) IP  Re: Huge Bug in TPU (Capitalize) Re: OpenVMS Support Issues Re: OpenVMS Support Issues Re: OpenVMS Support Issues Re: OpenVMS Support Issues Re: OpenVMS Support IssuesC Re: Problem connecting to SSH, OpenVMS7.3-2, HP TCP/IP v5.4 patched C Re: Problem connecting to SSH, OpenVMS7.3-2, HP TCP/IP v5.4 patched * Re: Purpose of uppercasing a RUN command ?  Re: Simple coding example please  Re: Simple coding example please Re: Testing for mobile device  Re: Testing for mobile device # Trying to use ZLIB in a PERL script 2 Re: Using SYS$INPUT for input into an FTP command.2 Re: Using SYS$INPUT for input into an FTP command.2 Re: Using SYS$INPUT for input into an FTP command.2 Re: Using SYS$INPUT for input into an FTP command.2 Re: Using SYS$INPUT for input into an FTP command.2 Re: Using SYS$INPUT for input into an FTP command.  Re: volume shadowing over LAVC ?  Re: volume shadowing over LAVC ?( Western Dayligh Time. Starts in 12 days!, Re: Western Dayligh Time. Starts in 12 days!, Re: Western Dayligh Time. Starts in 12 days!, Re: Western Dayligh Time. Starts in 12 days!= Re: What layered products are important for a VAX VMS system? = Re: What layered products are important for a VAX VMS system? = Re: What layered products are important for a VAX VMS system?  Re: Who called me? Re: Who called me? Re: Who called me?  F ----------------------------------------------------------------------  # Date: Tue, 21 Nov 2006 18:24:30 GMT  From: dittman@dittman.net  Subject: DLT-V4 ) Message-ID: <yvH8h.5496$Uz.1153@trnddc05>   K Has anyone tried a Quantum DLT-V4 drive on a VMS system (Alpha or Itanium)?  If so, what were the results?   H I emailed Quantum tech support and they said I would have to contact the' vendor as they are unfamiliar with VMS.  --   Eric Dittman dittman@dittman.net    ------------------------------    Date: 21 Nov 2006 06:31:20 -0800/ From: "Volker Halle" <volker_halle@hotmail.com> * Subject: Re: Educational VMS IA64 licensesC Message-ID: <1164119479.771398.323370@h48g2000cwc.googlegroups.com>   G There is a VMS82I_LMF-V0100 patch around since 15-JUN-2005 (superseeded ? by VMS82I_LMF-V0300 ), which is supposed to enable 'Educational  Licenses' in OpenVMS I64 V8.2:  -  5.1  New functionality addressed in this kit   =           This  kit  adds  support  for  the  new  standalone 	 operating C           environment, The Education Operating Environment (EDUOE).  The C           EDUOE  contains  licenses  for  products  licensed  under  the D           educational   program.    The  LMF  PAK  used  by  EDU  OE is           OPENVMS-I64-EDUOE.  G Normally one would expect those changes to be automatically included in % the next higher OpenVMS I64 versions.    Volker.    ------------------------------  % Date: Tue, 21 Nov 2006 11:08:44 -0500 8 From: Stephen Hoffman <Hoff@HoffmanLabs-RemoveThis-.Org>* Subject: Re: Educational VMS IA64 licenses' Message-ID: <ejv8ad$40$1@pyrite.mv.net>    Volker Halle wrote: I > There is a VMS82I_LMF-V0100 patch around since 15-JUN-2005 (superseeded A > by VMS82I_LMF-V0300 ), which is supposed to enable 'Educational   > Licenses' in OpenVMS I64 V8.2: > / >  5.1  New functionality addressed in this kit  > ? >           This  kit  adds  support  for  the  new  standalone  > operating E >           environment, The Education Operating Environment (EDUOE).  > The E >           EDUOE  contains  licenses  for  products  licensed  under  > the F >           educational   program.    The  LMF  PAK  used  by  EDU  OE > is >           OPENVMS-I64-EDUOE.    F    IIRC, one reason to install the recent LMF updates for OpenVMS I64 + was support for the PCL-based license PAKs.   G    That written, the OPENVMS-I64-EDUOE is not the PAK that is arriving  B back from the hobbyist site, nor would that PAK be expected.  The B hobbyist license that is arriving back is currently OPENVMS-FOE64.    I > Normally one would expect those changes to be automatically included in ' > the next higher OpenVMS I64 versions.   1    The next higher version *after* the ECO ships.   G    I've seen cases where the ECO was not listed in the contents of the  H release (as it was not named when the release documentation closed) and G cases where a release was skipped and wasn't ECO'd, and cases where an   ECO was needed.   ?    I've contacted the maintainer of the hobbyist site off-line.    ------------------------------  % Date: Tue, 21 Nov 2006 13:11:43 -0500 8 From: Stephen Hoffman <Hoff@HoffmanLabs-RemoveThis-.Org>* Subject: Re: Educational VMS IA64 licenses( Message-ID: <ejvfgv$27s$1@pyrite.mv.net>      Heard back from the Deity...   G    The hobbyist OpenVMS I64 licenses being returned now appear to have  E the correct product name.  The OpenVMS I64 base license PAK that was  E just returned from the request was one of the expected license PAKs.  K Specifically, the OPENVMS-I64-FOE foundation operating environment license.   H    As for the educational license program, try that again (and load the 9 current LMF ECO, just for grins) and see if it works now.    ------------------------------  % Date: Tue, 21 Nov 2006 08:10:16 -0800  From: foo@bar.com L Subject: End of life question for VAX 3100-96 and Alphaserver 5/533 question8 Message-ID: <5296m2tog4v6ghnglpispvp9aqitkh2dkm@4ax.com>  5 Does anybody know when the official end of life (when F Digital/Compaq/HP stopped providing maintenance) to the above machinesF was? Also when was the end-of-life for the Vax architecture as a whole (hardware, not software)? * ------------------------------------------C NO, use John Mee3 at comcast dot net (remove the spaces and do the  # obvious with the "dot" and the "at"    ------------------------------    Date: 21 Nov 2006 09:24:41 -0800! From: "Ian Miller" <gxys@uk2.net> P Subject: Re: End of life question for VAX 3100-96 and Alphaserver 5/533 questionC Message-ID: <1164129881.146973.124390@f16g2000cwb.googlegroups.com>   F I don't think any alphaserver except parhaps very early ones have gone end of service life yet.   ------------------------------  % Date: Tue, 21 Nov 2006 13:02:36 -0500 8 From: Stephen Hoffman <Hoff@HoffmanLabs-RemoveThis-.Org>P Subject: Re: End of life question for VAX 3100-96 and Alphaserver 5/533 question( Message-ID: <ejvevs$22i$1@pyrite.mv.net>   foo@bar.com wrote:7 > Does anybody know when the official end of life (when H > Digital/Compaq/HP stopped providing maintenance) to the above machinesH > was? Also when was the end-of-life for the Vax architecture as a whole > (hardware, not software)?     I    Hardware service status like this traditionally depended on the input  E and participation of the local or regional HP service office, and on  * what is negotiated with the service folks.  9    The MicroVAX 3100 model 96 is a comparatively new VAX.   I    I'm not sure which AlphaServer is involved here, as there are several  F EV56-based systems at 533 MHz around -- AlphaServer 1200, AlphaServer G 4000, AlphaServer 4100, and maybe others?  These are mid-vintage Alpha   systems.  E    The general assumption I'd make is "at least five years" from the  C original purchase, but it could easily be longer.  There were some  I ancient PDP systems that were continuing with formal support for seeming  I eons...  Check with your service office, and with whomever is supporting   the particular box.   F    There are folks that may well be willing to assist with this other G than HP, as well -- folks that specialize in keeping older gear going,  F or providing a path to move older gear forward.  Whether to newer old E hardware, to CHARON-VAX or such, or to newer systems and/or to newer   architectures.   ------------------------------  % Date: Tue, 21 Nov 2006 19:00:36 +0800 3 From: "Richard Maher" <maher_rj@hotspamnotmail.com> E Subject: Re: First TCP/IP (UCX) Release, Oldest VMS with (limited) IP 1 Message-ID: <ejum0l$1qt$1@news-02.connect.com.au>    Hi Steve  ' > Deliveries begin in the fall of 1988.   J Yep, I have the V1.0 release notes in front of me that say that IPsec justD missed the cut but should definitely make it to the next release :-(   Cheers Richard Maher  E "Stephen Hoffman" <Hoff@HoffmanLabs-RemoveThis-.Org> wrote in message # news:ejtg8o$2j4m$1@pyrite.mv.net...  > F > There was a recent thread around here where a question on the oldestH > OpenVMS VAX (VAX/VMS or MicroVMS) release with IP support (optionally)I > available was raised.  Here's the related and oldest (and very limited) @ > UCX-related information; the oldest VMS version with UCX (IP). > I > As for CMU IP or other IP products that might be around, that's another 
 > discussion.  > I > [Found this announcement on the way to something else equally ancient.]  >  > -- >  > VMS/ULTRIX Connection V1.0 > K > This is a VMS layered software product that provides VMS services to UNIX B > clients.  This product includes the Internet (TCP/IP) networkingL > protocols running on VMS and the Network File System (NFS) server softwareG > running on VMS.  The importance of this product is that it allows any  singleL > VMS system or a large VAXcluster system to act as an NFS server to a largeI > number of UNIX-based workstations.  For the first time Digital provides C > UNIX-based workstations with access to the high availability data 
 management4 > features of a VAXcluster system such as shadowing. >  > Key Features: V1.0 support > ' > - Transmission Control Protocol (TCP)  > - Internet Protocol (IP)  > - User Datagram Protocol (UDP)% > - Address Resolution Protocol (ARP)   > - File Transfer Protocol (FTP) > D > Future versions will support mail (SMTP) and remote login (Telnet) > protocols. > ! > Prerequesite Software required:  > 9 > - For the VMS server, VMS 4.7 or V5.0, or MicroVMS V4.7 J > - For the UNIX client, any UNIX-based system which supports NFS V2.0 and the K >    networking protocols specified by the DDN (Defense Data Network) RFCs, J >    including ULTRIX-32 V2.2 or later and ULTRIX Worksystem Software V1.1 or >    later.  > ' > Deliveries begin in the fall of 1988.    ------------------------------  % Date: Tue, 21 Nov 2006 11:33:32 -0500 8 From: Stephen Hoffman <Hoff@HoffmanLabs-RemoveThis-.Org>E Subject: Re: First TCP/IP (UCX) Release, Oldest VMS with (limited) IP ' Message-ID: <ejv9ou$gp$1@pyrite.mv.net>    Richard B. Gilbert wrote:   K > UCX SMTP support did not work properly in any version from V2.0E to V3.3  H > ECO13.   V3.3 ECO14 fixed almost the last of dozens (or hundreds?) of K > bugs that made UCX "e-mail hell".  If Karol Zielonko is still around, he  G > should be able to give you chapter and verse.  (Tell him I said "Hi")   G    Um, if that request was aimed at me and if you happened to miss it,  G I'm "not still around."  You might see Karol before I do, depending on  G whether or not he shows up for the Partners' Roundhouse or some of the  4 sessions next month, or at some other similar event.  C    And as for the earlier versions of TCP/IP Services, I believe I  F commented on the relative robustness of the early product versions in  that other thread.  H    I'd personally tend to want to run V4.2 or later and maybe V4.1 in a A pinch, and with the most current ECO loaded.  As with most other  G non-trivial products of a similar scope and similar bent, the earliest  I releases were somewhat rocky, all things considered.  (The early Unix IP  G stacks were no joy to deal with, either.)  V4.2 is solid, as are other  F and more recent releases.  (If chasing down old bugs and locating old H ECOs and scrounging up old details and old documentation -- and getting E old software and hardware to work -- is something you happen to find  D interesting and rewarding, however, do have at.  Or should you have A business constraints that require such efforts, do call me. :-) )    ------------------------------    Date: 21 Nov 2006 03:36:53 -08003 From: "Big John" <john.powers@airwidesolutions.com> ) Subject: Re: Huge Bug in TPU (Capitalize) C Message-ID: <1164109013.482804.130710@h54g2000cwb.googlegroups.com>    JF Mezei wrote:    > Joseph Huber wrote: K > > If e.g. "-" would be a default word-separator, editing files containing L > > file-names,dates,... would become very ugly: "next word" would jump into- > > instead after those names containing "-".  > K > Good point. And that means I also don't want those commands in my default K > TPU commands. I just want that behaviour to apply to the uppercasing, not  > to other word operations.  >  > M > So I guess the solution would be to change the "capitalize" code to use its M > own private list of word separators which would not affect the rest of TPU.   F Fortunately, here you are lucky, since eve decided to call the commandG 'capitalize word', and nobody ever adds the word 'word', you can create D a separate command called 'eve_capitalize', which would override it.# And you can leave EVE's code alone.   D procedure eve_capitalize ! Capitalize word using eve_capitalize_word   local return_value;  eve$add_word_separators ("-");% return_status := eve_capitalize_word;  eve$restore_word_separators; return (return_status);   
 endprocedure;   D That way, you treat "-" as a word separator only for the duration of the capitalization.    - Cheers, John   ------------------------------    Date: 21 Nov 2006 00:06:55 -0800- From: "Beach Runner" <Bob4Health@hotmail.com> # Subject: Re: OpenVMS Support Issues C Message-ID: <1164096415.687454.206880@m73g2000cwd.googlegroups.com>    William Webb wrote: I >> > VMS is still great. And there still are lots of great VMS people and  > > applications.  > > D > > We can direct that our money be moved elsewhere. There are otherH > > organizations that will support VMS. Make sure that HP knows why theG > > big money is being spent elsewhere.  Just like a bad dog, rub it in I > > their noses.  And, please not at the engineer and support level.  Get I > > good examples of the types of support, sales and elevation issues you J > > run into. Don't blame it on the poor techie in India, it's not his/her
 > > fault. > > J > > The best thing would be if HP sells OpenVMS. They know of course, thatH > > allowed to compete in the real marketplace, it would be a gem again. > >  > >  > ) > As Bill Cosby used to say-  LookDammit!  > A > Sue asked for something specific and you just spit in her face.  > / > Thanks a lot.  Your post really helps things.  > E > Most of the time my issues are hardware-related.  I either send the D > onsite guy an email or I walk downstairs to his office and we talk% > about it.  Or I just fix it myself.  > G > I only call support on OS or "software" issues once or twice a month- G > it's usually when I find something isn't behaving the way TFM says it 7 > should, or something *seriously* obscure is involved.  > * > I reiterate what I've posted previously. >  > Write Sue.  Politely.  >  > Write Mr. Hurd.  Politely. >  > Write Ann McQuaid.  Politely.  > 3 > Write each and every member of the BOD, politely.  > E > Especially the outside directors- as the Corporate Rule is that the G > inside folks tend to listen when the outside directors are saying, "I 6 > got a bunch of letters about this- WTF is going on?" > ' > Don't just rant, you'll be dismissed.  > + > Sue, as usual, is right.  Give specifics.  > G > I have received follow-up calls which were clearly different that the G > typical "on a scale of 1 to 5, with 1 being highly dissatisfied and 5 C > being highly satisfied, please rate....---that we've all done for E > years-- on about 75 percent of the calls that I've placed that have  > been routed offshore.  > F > Remember, from the DECozoic Era through the Middle Compaq Period andE > even unto the present day, there is an extensive history of Surveys F > Being Important, and those of you who have been On Medallion Support( > for years know what I'm talking about. > R > If you bitch about offshoring per se, you will probably be dismissed as a bigot.H > And perhaps rightly so, although I don't know you personally (at least > I *think* I don't.)  > D > I have been completely honest in my discussions on these follow-upH > calls-  I don't give a rodent's posterior about where the support callD > goes- I've dealt with DECfolk all over the world in the past-- theE > pertinent issue isn't the location of the support person-- it's the 1 > level of training and resources that they have.  > E > Also, what I care about is that the service level has deteriorated-  > E > Where previously as a Gold Support customer, the call was taken and C > then forwarded to my TAM, who then decided upon either personally B > owning the call and researching the issue, or referring it to an > appropriate backline-- > E > Now after my call is taken, it is routed to a Tier 1 line where the = > level of knowledge is orders of magnitude less than my own, E > communication is sometimes okay and sometimes, well, challenging...  > G > But to be honest, they're where my VMS level of knowledge was in 1990 H > or so- and one of the things that burns me up is that it takes fifteen9 > or twenty minutes to even get a case number these days.  > G > The people who are fielding these calls have neither the training nor D > the expertise nor the access to certain in-house resources (STARS,A > anyone? or whatever it's called these days) to field calls from G > people, who, on average, have from fifteen to fifty times as much VMS  > experience as they do. > C > It's not fair to us and it's not fair to them, either.  It's like  > lambs to the slaughter.  > O > I know what's going on when I'm told that I'll get a callback in ten minutes.  > G > This is the support equivalent of reducing the pound bag of coffee to  > 11 1/2 ounces. > C > And once the experience and knowledge is lost, it can't be gotten ? > back.  That already happened with a StorageWorks design team.  > 6 > And a knowledgebase is no substitute for experience. > % > WWWebb, speaking purely for myself.    Mr Webb,  E I can understand your concept.  Use logic and reason, and that should  reach upper managementF of HP.  It's a great idea.  It's also a fairy tale.  It won't change a thing.  Not one thing.  C There is only one thing that will change things, and that's pulling  cash on the barrelhead.   D Perhaps Mr. Webb doesn't need VMS support, but many people have been? paying for remedial and advisory for many years, and others for : elevated support.  While support is never a substitute forD reading the documentation, it is reasonable for someone that takes a technical call to know VMS, not G simply have a very cursory understanding, some training, and limited if  any hands on experience at all.   D It is most clearly not a matter of prejudice as you wrote privately.G Many fine VMS people from around the world have lost their jobs. Either G they were RIF'd or left now to avoid the rush.  This is in support, VMS   engineering and layered product.  8 How are the few people left going to handle their calls.  B I can give examples of customers that complained and that got themC nowever.  They then held their check, cancelled their multi million A dollar orders, and THEN things changed.  THAT is what will make a  dramatic change.  D Do you not think all logic and reason have not been applied?  Do youG not think that the great people still doing support, some of the finest D engineers in the industry, did not know what would happen, saw would@ happen, and tried logic, reason and polite comments.    And that accomplished, what?   F I'm sorry if you think I spit in the face of Sue. I'm sure she is very@ well meaning and thinks that a few surverys and such will changeE things.   No logical or reasonable person would have been killing VMS 3 when there was a loyal and dedicated customer base.    So, you have my apologies.   ------------------------------    Date: 21 Nov 2006 08:25:35 -0800) From: "Sue" <susan_skonetski@hotmail.com> # Subject: Re: OpenVMS Support Issues A Message-ID: <1164126335.217008.47100@m7g2000cwm.googlegroups.com>   C I have requested information that I can take to management to prove A that there is an issue and as of yet I have not received one mail G message detaling the issues at hand.  Granted it is much easier to talk 9 about how bad the problem is but fixing it takes work.  I    sue    Dave Froble wrote: > Beach Runner wrote:  >  > > Management that B > > can replace experts with newbies in India does not value their0 > > employees, their customers or their product. > J > Now there is a good quote.  I'm thinking that at some time in the future > it will be 20/20 hindsight.  >  > --6 > David Froble                       Tel: 724-529-0450@ > Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com > DFE Ultralights, Inc.  > 170 Grimplin Road  > Vanderbilt, PA  15486    ------------------------------  % Date: Tue, 21 Nov 2006 12:28:52 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> # Subject: Re: OpenVMS Support Issues 8 Message-ID: <4d0eb$45633745$cef8887a$32530@TEKSAVVY.COM>   > Beach Runner wrote:  >  >> Management thatA >> can replace experts with newbies in India does not value their / >> employees, their customers or their product.     K When you announce to Wall Street that you will be firing 15,000 employees,  J you need to find ways to deliver on that promise. One of these ways is to L outsource as much as you can. This allows you to reduce employee count even D if in the end, you end up paying more for the outsourcing or losing - customers because of drop in service quality.   J When firing people becomes a higher priority than serving customers well,  then ...   ------------------------------  % Date: Tue, 21 Nov 2006 12:40:00 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> # Subject: Re: OpenVMS Support Issues 8 Message-ID: <8f597$456339e1$cef8887a$19747@TEKSAVVY.COM>   Dave Froble wrote:H > And then GM pulls their entire advertising budget from NBC.  (Anybody F > that still watches any TV will understand that automotive companies H > spend lots of money on advertising.)  After massive cardiac arrest in J > NBC corporate offices, a retraction was issued and I believe an apology.    L When Curly announced the euthanasia of Alpha, he bragged about Compaq sales G reps visiting their key customers on the very same day to make various  , offers. But the regular customers got squat.  J What if HP is giving its key VMS customers access to real support, whilst K the regular customers are sent to newbies in India ? What if HP only wants  - to retain a few key customers it can please ?   I When a single small customer doesn't renew a contract, they are probably  J not even asked why and nobody in upper manager hears about it.  And guess J what, if "GM" (aka: key customer) is poorly served, what if this is THEIR L argument to migrate off of VMS and by the time they tell HP, it is too late    for HP to save the account ?  H And since you brought up NBC. They recently announced major downsizing. K They saw that due to competition with cable, they will cut down their news  K division, and produce el-cheapo game shows for for the first hour of prime  J time. It won't be long before reruns of NBC's "Friends" on cable networks < will have higher ratings than NBC's new "downsized" network.  J Cutting down on service quality at a time when BCS is struggling is not a J good idea. It is like NBC cutting down on show quality at a time when its " ratings are down due to bad shows.   ------------------------------  % Date: Tue, 21 Nov 2006 13:04:20 -0500 * From: "Syltrem" <syltremzulu@videotron.ca># Subject: Re: OpenVMS Support Issues 0 Message-ID: <12m6ft6h9t0se84@corp.supernews.com>  5 "Sue" <susan_skonetski@hotmail.com> wrote in message  ; news:1164126335.217008.47100@m7g2000cwm.googlegroups.com... D >I have requested information that I can take to management to proveC > that there is an issue and as of yet I have not received one mail I > message detaling the issues at hand.  Granted it is much easier to talk ; > about how bad the problem is but fixing it takes work.  I  >  > sue  >  >     ' This morning I had a problem with TCPIP F I called HP, got in touch with someone who could not really help this L matter, and after 4-5 minutes another person was on the line with us and he ! could answer all of my questions.   C This is the same level of service that I had for the last 20 years.   L I had another call a couple of weeks back, for hardware, and again that was 
 excellent.   So it is NOT always bad.  L True, a couple of months ago it was a nightmare to open a new call even for D urgent issues, it could take hours before we could speak to someone K (hardware support) but it looks like things have impoved since, and I only  . spoke to people here in Canada, or in the USA.   My 2 cents..   Syltrem    ------------------------------    Date: 21 Nov 2006 07:22:01 -0600; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler) L Subject: Re: Problem connecting to SSH, OpenVMS7.3-2, HP TCP/IP v5.4 patched3 Message-ID: <IgKBZw2ZWL9r@eisner.encompasserve.org>   l In article <1164067407.240696.262600@f16g2000cwb.googlegroups.com>, "Milton" <milton.baar@gmail.com> writes:	 > Ah ha!!  >  > My logs show...  >  > $ Set NoOn1 > $ VERIFY = F$VERIFY(F$TRNLNM("SYLOGIN_VERIFY")) ; > %DCL-W-ACTIMAGE, error activating image DECW$SETSHODISSHR  > -CLI-E-IMGNAME, image file7 > VAX2$DKA0:[SYS0.SYSCOMMON.][SYSLIB]DECW$SETSHODISSHR.  > EXE;1 B > -SYSTEM-F-PRIVINSTALL, shareable images must be installed to run > privileged imag  > e : >   TCPIP$SSH    job terminated at 21-NOV-2006 09:54:57.40 > H > So, it looks like the install may have not *really* worked - it should3 > install the image as part of the install process.  > I > So, trash and rebuild VMS??  I have already disabled and re-enabled, it  > obviously didn't help....sigh   	    Never.   ?    In this case I think someone missed the required addition to H    SYSTARTUP_VMS.COM, or adjustments to GBLPAGES or GBLSECTIONS.  Those 4    are the usual reasons for INSTALLs to be missing.  3    And rebuilding VMS won't correct either of them.    ------------------------------  % Date: Tue, 21 Nov 2006 12:41:36 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> L Subject: Re: Problem connecting to SSH, OpenVMS7.3-2, HP TCP/IP v5.4 patched8 Message-ID: <989ad$45633a41$cef8887a$19747@TEKSAVVY.COM>   Dave Froble wrote:6 > Some parts of DECwindows are needed and used by VMS.    L This is a problem with AUTOGEN which generaes the VMSIMAGES.DAT file. I had K a similar issue with the CDA Converters requiring some DECWINDOWS files be  G installed.  It is just that AUTOGEN was never updated to include those   files "by default".    ------------------------------  % Date: Tue, 21 Nov 2006 10:40:40 -0500 8 From: Stephen Hoffman <Hoff@HoffmanLabs-RemoveThis-.Org>3 Subject: Re: Purpose of uppercasing a RUN command ? ) Message-ID: <ejv6m1$315d$1@pyrite.mv.net>    Robert Deininger wrote:   J > IIRC, the main reason VMSINSTAL deletes all global symbols is to make asI > much memory available as possible.  The procedure wants a lot of symbol K > space, and on small-memory systems quotas are low enough that it matters.   F    That, and to avoid problems where site-specific (process-specific) G symbols derail the VMSINSTAL.COM or the KITINSTAL.COM installation, or  I pick up left-overs from the last installation, or...  Problems triggered  F by unexpected local symbols can range from the obvious to the obscure.   ------------------------------    Date: 21 Nov 2006 07:07:44 -0600 From: briggs@encompasserve.org) Subject: Re: Simple coding example please 3 Message-ID: <v8zZf4MZ8+x2@eisner.encompasserve.org>   G In article <ejun5p$nns$1@lore.csc.com>, "Ade" <Ade@nowhere.com> writes: 	 > Hi all,  > O > Does anybody have a quick and simple example in C of how to use the SYS$FIND  K > call when reading sequential files rather than indexed files. The manual  . > says it's possible but I can't make it work.  : It is not possible to do keyed access on sequential files.    D It is possible to use _direct_ access on sequential files with fixedC length records.  The relevant key information is the record number. G RMS can multiply this by the fixed record size and divide by 512 to get A the relevant block number and use the remainder to get the offset  within the relevant block.  C Obviously that can't work on stream files or on files with variable  length records.     B It is possible to use _RFA_ access on all files.  The relevant keyB information is a Record's File Address which was harvested at some@ previous time.  The Record's File Address is essentially a blockA number and a record offset.  In indexed files, steps are taken so E that the RFA for a particular record is permanent, even if the record ? moves to a different block as the result of a bucket split.  In ? sequential and relative files, records never move and RFA's are  automatically permanent.    E If you want to find the record that starts with "DEW" in a sequential F file then you're stuck reading each record until you find the one that< starts with "DEW".  RMS has no provision to do this for you.   ------------------------------    Date: 21 Nov 2006 05:55:53 -0800" From: "kczwei" <kaycee@kaycee.net>) Subject: Re: Simple coding example please B Message-ID: <1164117353.318070.59780@b28g2000cwb.googlegroups.com>  I > > If you want to find the record that starts with "DEW" in a sequential J > > file then you're stuck reading each record until you find the one that@ > > starts with "DEW".  RMS has no provision to do this for you. > # > Ok, thanks, I was afraid of that.  >  > Ade   4 Without knowing what you are really trying to do....$ insert the following in your c code.  ( system("search/out=tmp.tmp aa.dat Dew"); fopen("tmp.tmp, "r"); > Then loop through only the text records you are interested in.   ------------------------------  % Date: Tue, 21 Nov 2006 11:54:33 -0500 8 From: Stephen Hoffman <Hoff@HoffmanLabs-RemoveThis-.Org>& Subject: Re: Testing for mobile device' Message-ID: <ejvb0c$rh$1@pyrite.mv.net>    David Gray wrote:   D > Looking for the smartest way of testing if a user logs in VMS on a: > hand held device or via a PC based terminal emulator.    > C > Users sat at workstations login using SmarTerm terminal emulation G > which is (usually)  VT400, mobile users login using a Symbol handheld B > using Windows CE terminal emulation which is VT100 or VT200.     > E > Some of our users would have a need to do both using the same login  > credentials.   > C > Using f$getdvi I can test DEVICE_TYPE_NAME &  TT_ACCPORNAM  which G > gives the VT version and remote port number.    This should be enough ? > but I wondered if there is another sure-fire way of telling.   > 3 > TT_ACCPORNAM = "Host: nnn.nnn.nnn.nnn Port: nnnn" # > DEVICE_TYPE_NAME = "VT400 Series"  >  > DEVICE_TYPE_NAME = "VT100"  F    What difference(s) do you seek to enforce here, between the remote I login from a Microsoft Windows terminal emulator and a remote login from   a Windows CE terminal emulator?   D    Do the users have a dedicated interface, or are they effectively I "loose" at the DCL prompt?  That the devices emulate a VT may or may not  , imply the users have access to the $ prompt.  I    If there is a dedicated user interface (other than the $ prompt), you  H can send over trinkets of information about the host platform (eg: http 6 has client information available to the web server)...  F    If the users are loose, you're somewhat stuck and somewhat exposed B -- a tracert or such could attempt to find the source (if they're I arriving via a different network path).  Users with DCL-level access can  H generally easily set the terminal device type themselves, so it's not a I rock solid protection -- though it might be sufficiently effective here,  I and might work for your needs.    If the mobile users all arrive at your  A application through a particular range of IP host addresses, for  H instance, you could use that to approximate the client type -- this can < be fetched via tracert or (usually) via the ACCPORNAM field.  D    You might want to encrypt it all, and treat the users similarly, H regardless of source.  If you need to differentiate sources and can set H keys on individual sources -- that's something that depends on features G within the particular terminal emulators, whether a USB or SD or other  G such dongle or a smart card or sim card or with digital certificate(s)  A purchased from a CA or even self-signed -- you could conceivably  G differentiate the source based on the particular device keys.  Combine  B that with the username and password...  (Tracking back through an I encrypted link might involve a little inner-mode poking, as -- depending  H on the protocol -- the ACCPORNAM doesn't get set the way you might want C in certain cases.  Getting from the FT to the BG (IIRC) requires a  @ simple peek at the UCB.  Some IP stacks might make this easier.)  G    There is also the terminal answerback buffer (which is available on  I most devices), and this can (usually) be set with user- or site-specific  H information.  That can be easily queried, and some terminal devices and 9 terminal emulators allow that information to be "locked".   I    IIRC, there was a discussion a while back about authenticated logins,  ! and that may be of interest here.    ------------------------------    Date: 21 Nov 2006 09:54:16 -0800- From: "Doug Phillips" <dphill46@netscape.net> & Subject: Re: Testing for mobile deviceB Message-ID: <1164131656.413520.190210@e3g2000cwe.googlegroups.com>   David Gray wrote: 	 > Hi all,  > D > Looking for the smartest way of testing if a user logs in VMS on a7 > hand held device or via a PC based terminal emulator.  > C > Users sat at workstations login using SmarTerm terminal emulation G > which is (usually)  VT400, mobile users login using a Symbol handheld > > using Windows CE terminal emulation which is VT100 or VT200. > E > Some of our users would have a need to do both using the same login  > credentials. > C > Using f$getdvi I can test DEVICE_TYPE_NAME &  TT_ACCPORNAM  which G > gives the VT version and remote port number.    This should be enough > > but I wondered if there is another sure-fire way of telling. > 3 > TT_ACCPORNAM = "Host: nnn.nnn.nnn.nnn Port: nnnn" # > DEVICE_TYPE_NAME = "VT400 Series"  >  > DEVICE_TYPE_NAME = "VT100" >    David,  D I'm making some wa presumptions based on having done similar things:D This question only makes sense in a captive or restricted (i.e. menuD driven, dedicated application, etc) production environment. It looksF like your access is via TELNET and you're not getting name translation from a host file or DNS.  C So, if you use static IP's on each handheld client you can identify A which client is logging in. Ignore the "Port: nnnn", use only the C "Host: IP".  (If you get name translation, then use that, but still G ignore the port). I've found that using other factors, such as terminal F type or answerback, is more difficult to maintain than keeping a table% of static IP's assigned to handhelds.   F So, login, check the IP table, then set a logical or token to indicateC client type (usually a handheld Yes or No indication is enough)  to G determine capability, screen size, location, functions, etc. Then check D the user (rights) to match authority / responsibility at that clientB type. Then, run the appropriate program, or allow and restrict the% appropriate logic within the program.   E This requires strict documentation and control of handheld IP address @ assignment (if DCHP is used for desktops, the hh's range must beB excluded) , and of user rights assignment (easier than managing by< individual users --- give the priv'ed people an identifier.)    Doug    ------------------------------    Date: 21 Nov 2006 04:59:23 -0800) From: "Scott Greig" <jsgreig@geminaq.com> , Subject: Trying to use ZLIB in a PERL scriptC Message-ID: <1164113963.224960.292930@j44g2000cwa.googlegroups.com>   G Actually, Im trying to make a Perl script called "TXT2PDF + crypt v9.0" 3 work on an Alpha VMS8.2 (+patches) with Perl 5.8.6.   G I know that Zlib is available on the freeware CD (I have it), and I can F find any number of "ZLIB.PM" files - I lack the basic understanding asA to how all of this stuff hangs together (particularly under VMS).   E Any pointers as to where best to install ZLIB (e.g. under the Perl...  tree, 9 or under the TXT2PDF... tree?) would be most appreciated.    TIA  Scott    ------------------------------    Date: 21 Nov 2006 05:48:44 -0800) From: "Scott Greig" <jsgreig@geminaq.com> ; Subject: Re: Using SYS$INPUT for input into an FTP command. C Message-ID: <1164116924.085186.139850@h48g2000cwc.googlegroups.com>   3 On Nov 21, 8:37 am, "Ade" <A...@nowhere.com> wrote:  > Hi,  >  > OpenVMS 7.1-1H1  > UCX V4.1 ECO 2 > + > Unfortunately, we can't change the above.  > , > Using this command in a command procedure: >     $FTP/INPUT=myfile.ftp  >  > where myfile.ftp contains  >     connect node >     username >     password >     ascii & >     put myfile.txt someotherfile.txt
 >     quit > N > works fine. I would like to combine the two files and direct FTP to take theN > input from the command file, ie DEFINE/USER SYS$INPUT SYS$COMMAND before theI > FTP/INPUT=SYS$INPUT command and the rest of myfile.ftp to follow in the / > procedure. This method however, doesn't work.   > You don't need to do all of that, just put the commands in the
 procedure as:    $ ftp  connect node username password  put myfile.txt someotherfile.txt quit  & Works for me on VMS8.2, TCPIP 5.5 eco1   > K > I've also tried this on an Itanium running OpenVMS 8.2, TCP/IP 5.5 and it  > still doesn't work > K > The reason I'm asking is we have to change a LOT of procedures due to the M > removal of decnet from a customers network meaning everything changes to an M > IP based solution and I was hoping to minimize the number of changes and/or # > extra files required for the job.  >  > Any suggestions? >  > Ade    ------------------------------  # Date: Tue, 21 Nov 2006 15:12:24 GMT + From: =?ISO-8859-1?Q?Jan-Erik_S=F6derholm?= ; Subject: Re: Using SYS$INPUT for input into an FTP command. 2 Message-ID: <sHE8h.23896$E02.9534@newsb.telia.net>  7 One problem with in-line FTP commands is that you can't 9 (or it's hard to) have variable commands through symbols.   6 Defining logical names for source and target filenames6 works just fine, but I'v had FTP scripts where storage3 parameters for an MVS dataset (LRECL and so on) was ! different between different runs.   4 By writing a temp ftp-command-file, there is no such problem, of course...     	 Jan-Erik.       
 Ade skrev:? > "Joseph Huber" <huber@NIRWANA-mppmu.mpg.de> wrote in message  ' > news:3Aj+yCNj8WQd@vms.mppmu.mpg.de... J >> In article <ejuvfv$ofe$1@lore.csc.com>, "Ade" <Ade@nowhere.com> writes: >>> OpenVMS 7.1-1H1  >>> UCX V4.1 ECO 2- >>> Unfortunately, we can't change the above. . >>> Using this command in a command procedure: >>>     $FTP/INPUT=myfile.ftp  >>> where myfile.ftp contains  >>>     connect node >>>     username >>>     password
 >>>     ascii ( >>>     put myfile.txt someotherfile.txt >>>     quitM >>> works fine. I would like to combine the two files and direct FTP to take   >>> the M >>> input from the command file, ie DEFINE/USER SYS$INPUT SYS$COMMAND before   >>> the K >>> FTP/INPUT=SYS$INPUT command and the rest of myfile.ftp to follow in the 1 >>> procedure. This method however, doesn't work. I >> Leave off the DEFINE/USER SYS$INPUT SYS$COMMAND, it redirects the FTP   >> input to & >> the terminal, not the command file. >> >> -- B >>   Joseph Huber , Muenchen,Germany:  http://www.huber-joseph.de/ > H > Thanks to you both for that. I also had to change the command line to K > include the /USER qualifier and the node name but it seems to be working   > correctly now. > 	 > Thanks,  >  > Ade  >  >    ------------------------------  % Date: Tue, 21 Nov 2006 10:45:13 -0500  From: norm.raphael@metso.com; Subject: Re: Using SYS$INPUT for input into an FTP command. Q Message-ID: <OFC881BD7E.CE33D01F-ON8525722D.005563CE-8525722D.005689D1@metso.com>   D "Scott Greig" <jsgreig@geminaq.com> wrote on 11/21/2006 08:48:44 AM:   >  > 5 > On Nov 21, 8:37 am, "Ade" <A...@nowhere.com> wrote:  > > Hi,  > >  > > OpenVMS 7.1-1H1  > > UCX V4.1 ECO 2 > > - > > Unfortunately, we can't change the above.  > > . > > Using this command in a command procedure: > >     $FTP/INPUT=myfile.ftp  > >  > > where myfile.ftp contains  > >     connect node > >     username > >     password
 > >     ascii ( > >     put myfile.txt someotherfile.txt > >     quit > > G > > works fine. I would like to combine the two files and direct FTP to  take theE > > input from the command file, ie DEFINE/USER SYS$INPUT SYS$COMMAND 
 before theK > > FTP/INPUT=SYS$INPUT command and the rest of myfile.ftp to follow in the 1 > > procedure. This method however, doesn't work.  > @ > You don't need to do all of that, just put the commands in the > procedure as:  >  > $ ftp  > connect node
 > username
 > password" > put myfile.txt someotherfile.txt > quit  F Well, yes, but to be formally correct, you should use a DECK/EOD pair:   $ ftp  $DECK  connect node username password  put myfile.txt someotherfile.txt quit $EOD $!  6 Also,  I didn't test this, but from an earier exchange   $pipe - & (write sys$input "connect ''node'" ; -! write sys$input "''username'" ; - ! write sys$input "''password'" ; - 4 write sys$input "put ''myfile' ''someotherfile'" ; - write sys$input "quit") | ftp  $!  1 should work with appropriate symbol substitution, 8 and eliminate the need for a temp file or logical names.   > ( > Works for me on VMS8.2, TCPIP 5.5 eco1 >  > > J > > I've also tried this on an Itanium running OpenVMS 8.2, TCP/IP 5.5 and it > > still doesn't work > > I > > The reason I'm asking is we have to change a LOT of procedures due to  the I > > removal of decnet from a customers network meaning everything changes  to an H > > IP based solution and I was hoping to minimize the number of changes and/or% > > extra files required for the job.  > >  > > Any suggestions? > >  > > Ade  >    ------------------------------  % Date: Tue, 21 Nov 2006 11:00:59 -0500 8 From: Stephen Hoffman <Hoff@HoffmanLabs-RemoveThis-.Org>; Subject: Re: Using SYS$INPUT for input into an FTP command. ) Message-ID: <ejv7rt$31gv$1@pyrite.mv.net>   
 Ade wrote:  , > Using this command in a command procedure: >     $FTP/INPUT=myfile.ftp  >  > where myfile.ftp contains  >     connect node >     username >     password >     ascii & >     put myfile.txt someotherfile.txt
 >     quit > O > works fine. I would like to combine the two files and direct FTP to take the  O > input from the command file, ie DEFINE/USER SYS$INPUT SYS$COMMAND before the  J > FTP/INPUT=SYS$INPUT command and the rest of myfile.ftp to follow in the / > procedure. This method however, doesn't work.  > L > I've also tried this on an Itanium running OpenVMS 8.2, TCP/IP 5.5 and it  > still doesn't work > L > The reason I'm asking is we have to change a LOT of procedures due to the N > removal of decnet from a customers network meaning everything changes to an N > IP based solution and I was hoping to minimize the number of changes and/or # > extra files required for the job.  >  > Any suggestions?      On V6.2 and later...       $ HELP COPY /FTP   /    As for your specific command requirements...       $ COPY/FTP/ASCII -       myfile.txt - 7     host.no.where"username password"::someotherfile.txt   ?    Or install DECnet-Plus over IP, and avoid changing anything.   H    The networking folks that are seeking to avoid non-IP protocols will H be happy with DECnet-Plus over IP (or at least, happier with not seeing C DDCMP wandering around the networks), and the folks that are using  H DECnet will be happy (having avoided changes), and there will be little G or no requirement to change existing procedures and processes that are  I using DECnet.  (Well, I've met some networking folks over the years that  I would be more happy if the OpenVMS hosts were off their network -- but I  I think these same folks might be really happy if all the pesky hosts were   off their IP network. :-) )   G    And as I've mentioned else-posting, Cisco and other modern gear can  F continue to operate non-IP networks.  Cisco can set up a private LAN, G and can bridge the SCS, MOP and DECnet protocols among multiple sites,  I for instance.  And can continue to route IP.   According to the roadmap,  ? clustering is moving onto IP, and DECnet is already there with  A DECnet-Plus and its IP transport support.  MOP service requests,  H however, are somewhat more of a problem as that's built into the system E consoles and the existing bootstrap pieces -- the project adding SCS  B over IP might or might not deal with this MOP stuff, I don't know.   ------------------------------  % Date: Tue, 21 Nov 2006 08:14:56 -0800 % From: DeanW <dean.woodward@gmail.com> ; Subject: Re: Using SYS$INPUT for input into an FTP command. I Message-ID: <3f119ada0611210814r141170fawd761248240ef737a@mail.gmail.com>   G On 21 Nov 2006 05:48:44 -0800, Scott Greig <jsgreig@geminaq.com> wrote:  > 5 > On Nov 21, 8:37 am, "Ade" <A...@nowhere.com> wrote:  > > Hi,  > >  > > OpenVMS 7.1-1H1  > > UCX V4.1 ECO 2 > > - > > Unfortunately, we can't change the above.  > > . > > Using this command in a command procedure: > >     $FTP/INPUT=myfile.ftp  > >  > > where myfile.ftp contains  > >     connect node > >     username > >     password
 > >     ascii ( > >     put myfile.txt someotherfile.txt > >     quit > > P > > works fine. I would like to combine the two files and direct FTP to take the  F I guess I'd want to ask what the percieved benefit is. with myfile.ftpE being external, you have a couple options- if it's relatively static, = you can change it easy enough; if it's not (file names change F constantly) you can create a new myfile.FTP on the fly quite easily. ID do a lot of this in batch; it's easier to create/change the external2 FTP command files than the one in the batch queue.  M > > The reason I'm asking is we have to change a LOT of procedures due to the O > > removal of decnet from a customers network meaning everything changes to an O > > IP based solution and I was hoping to minimize the number of changes and/or % > > extra files required for the job.   F Ah, now we have the root of the problem. Is the customer taking DECNetA off their VMS boxes, or just some bone-headed IT goons taking the = commands out of the router because they think it's a problem?   > DECNet over IP works quite well (unless *that* is what they're$ disabling.) There's also COPY/FTP...  G COPY /FTP /ASCII myfile node"username password"::dest:someotherfile.txt   , /ASCII is the default, included for clarity.   > > Any suggestions?  , Parting thought: Don't fix what ain't broke?   --   Dean Woodward    =o&o  dean.woodward@gmail.com    ------------------------------  % Date: Tue, 21 Nov 2006 08:33:53 -0800 % From: DeanW <dean.woodward@gmail.com> ; Subject: Re: Using SYS$INPUT for input into an FTP command. I Message-ID: <3f119ada0611210833j32ce71d3la11924411a7f1440@mail.gmail.com>   F On 11/21/06, Stephen Hoffman <Hoff@hoffmanlabs-removethis-.org> wrote:; > (Well, I've met some networking folks over the years that J > would be more happy if the OpenVMS hosts were off their network -- but IJ > think these same folks might be really happy if all the pesky hosts were > off their IP network. :-) )   = You have no idea how many times I wished I could just fire my ? customers and hire smarter ones. Usually this thought occurs at D oh-dark-early, when some external customer three time zones ahead ofA me forgot the new password they changed 30 minutes ago and locked 2 their entire site out trying to guess what it was.   ------------------------------  % Date: Tue, 21 Nov 2006 09:51:59 -0500 3 From: "Richard B. Gilbert" <rgilbert88@comcast.net> ) Subject: Re: volume shadowing over LAVC ? : Message-ID: <-92dnYd4wKMaj_7YnZ2dnUVZ_tqdnZ2d@comcast.com>  
 Pierre wrote:    > hi,  > G > is it possible to do volume shadowing over LAVC without HSx controler * > (let say with local disk on each node) ? >  > TIA,	 > Pierre.  >   F IFF you can afford the volume shadowing licenses required.  ISTR that F you need a license for each machine in the cluster for each disk.  If F you have a three node cluster with three disks that's nine licenses!! D I've managed clusters with HBVS but I never had to buy licenses for H them.  The rules may have changed since then; my experience with it was 
 1997-1999.  < Check out what it costs before you get all excited about it!  I Also, since volume shadowing is software based, you place additional CPU  6 load on your cluster.  Then there is the network load.  C When a cluster member crashes, the shadowset has to be checked for  I consistancy, block by block.  It takes a while.  And your cluster is dog   slow while it's going on.   C I've run both host based volume shadowing and HSZ40 or HSZ50 based  D hardware RAID and believe me, hardware is better!!!  I wouldn't use I HSZ40/50 today of course; I believe that HSZ70 was the last of that line  G and that would be the way to go for wired SCSI.  The really modern way  K to go is fiber optic and the HSG contollers but that gets pretty expensive.   9 Shop around on the used market and see what you can find.    ------------------------------  % Date: Tue, 21 Nov 2006 12:55:18 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> ) Subject: Re: volume shadowing over LAVC ? 8 Message-ID: <5cc99$45633d77$cef8887a$21218@TEKSAVVY.COM>   Richard B. Gilbert wrote: > > Check out what it costs before you get all excited about it!  I There are licences that cover shadowing, as opposed to just disk by disk.   K > Also, since volume shadowing is software based, you place additional CPU  8 > load on your cluster.  Then there is the network load.  I Yes. But sometimes, the ethernet is all you have and you can't afford to  K have other links between the 2 nodes=, at which point host based shadowing   becomes a good solution.  E > When a cluster member crashes, the shadowset has to be checked for   > consistancy, block by block.    F Nop. It is when the crashed node that comes back on line and wants to L re-integrate its disks that the rebuilding has to take place. And there are D now parameters to control how much cpu/bandwidth is used to do this.      E > I've run both host based volume shadowing and HSZ40 or HSZ50 based  6 > hardware RAID and believe me, hardware is better!!!   D Hardware shadowing that is totally separate of VMS can be dangerous K depending on configuration with regards to cluster partitioning. With host  K based shadowing, your disks are "part of the cluster" and benefit from the  L cluster partitioning safeguards. And VMS has flags in each disk to indicate J which disk is the most recent. So, if you remount yor shadowset, it knows K which disk in teh shadow set has the most recent data to be used as source  " of data to update the other disks.  K With hardware systems, you don't have that extra logic that is part of VMS.    ------------------------------  % Date: Tue, 21 Nov 2006 17:09:46 +0800  From: prep@prep.synonet.com 1 Subject: Western Dayligh Time. Starts in 12 days! 0 Message-ID: <87ejrxyxlx.fsf@k9.prep.synonet.com>  @ The excuses for leaders we have in the local parlement have just> passed a 3 year trial of daylight saving time. Starting on the 3rd of December.  > You all WILL have your systems updated for this, won't you....  E We appologise for the idiocy. I said that they needed 2 years notice, 
 preferably 3.    --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------    Date: 21 Nov 2006 06:01:29 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) 5 Subject: Re: Western Dayligh Time. Starts in 12 days! 3 Message-ID: <YcBT11Esx$fi@eisner.encompasserve.org>   N In article <87ejrxyxlx.fsf@k9.prep.synonet.com>, prep@prep.synonet.com writes:B > The excuses for leaders we have in the local parlement have just@ > passed a 3 year trial of daylight saving time. Starting on the > 3rd of December. > @ > You all WILL have your systems updated for this, won't you.... > G > We appologise for the idiocy. I said that they needed 2 years notice,  > preferably 3.   D We don't blame you personally.  If you were of the legislative class$ you would not know how to post here.   ------------------------------  % Date: Tue, 21 Nov 2006 12:26:03 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> 5 Subject: Re: Western Dayligh Time. Starts in 12 days! 7 Message-ID: <f2da$4563369c$cef8887a$32530@TEKSAVVY.COM>    prep@prep.synonet.com wrote:B > The excuses for leaders we have in the local parlement have just@ > passed a 3 year trial of daylight saving time. Starting on the > 3rd of December.  G Is that for Western Australia ?  Did they have any reasoning for this ?    ------------------------------  % Date: Tue, 21 Nov 2006 11:23:02 -0700 6 From: "Michael D. Ober" <obermd.@.alum.mit.edu.nospam>5 Subject: Re: Western Dayligh Time. Starts in 12 days! 6 Message-ID: <45634418$0$10301$815e3792@news.qwest.net>  G On Dec 3rd, turn off all Daylight Savings time support and change your  " system time to the "new" timezone.  
 Mike Ober.  ) <prep@prep.synonet.com> wrote in message  * news:87ejrxyxlx.fsf@k9.prep.synonet.com...B > The excuses for leaders we have in the local parlement have just@ > passed a 3 year trial of daylight saving time. Starting on the > 3rd of December. > @ > You all WILL have your systems updated for this, won't you.... > G > We appologise for the idiocy. I said that they needed 2 years notice,  > preferably 3.  >  > --  > > Paul Repacholi                               1 Crescent Rd.,9 > +61 (08) 9257-1001                           Kalamunda. A >                                             West Australia 6076 , > comp.os.vms,- The Older, Grumpier Slashdot0 > Raw, Cooked or Well-done, it's all half baked.I > EPIC, The Architecture of the future, always has been, always will be.     ------------------------------  # Date: Tue, 21 Nov 2006 09:49:00 GMT L From: winston@SSRL.SLAC.STANFORD.EDU (Alan Winston - SSRL Central Computing)F Subject: Re: What layered products are important for a VAX VMS system?6 Message-ID: <00A5F097.ADD82B8C@SSRL.SLAC.STANFORD.EDU>  q In article <NbJkdr5beWmL@eisner.encompasserve.org>, koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes: d >In article <qsGdnS4XJN_VKsHYnZ2dnUVZ_sednZ2d@libcom.com>, Dave Froble <davef@tsoft-inc.com> writes: >>  F >> Being a bit alarmist here, since I've seen multiple posts from you N >> mentioned embedded real-time systems, and doubt that Microsoft is a player. > H >   I've been recently working on an embedded real-time system for whichH >   the initials MS would have been very convenient.  I started to writeH >   that down until someone misread it as "Microsoft" outloud.  Horrors! > E >   Its bad enough MS Word keeps crashing while I document the system J >   (required to be delivered in "Word format").  I've got to get handier  >   with OpenOffice.  L I have a DSR to RTF converter, if you want to write your docs in the subset K of RUNOFF that I use myself (and if you want to debug the occasional ACCVIO  crash).    -- Alan    ------------------------------  % Date: Tue, 21 Nov 2006 05:18:16 -0500 ' From: Dave Froble <davef@tsoft-inc.com> F Subject: Re: What layered products are important for a VAX VMS system?9 Message-ID: <Bp-dnatZ35TtT__YnZ2dnUVZ_h6dnZ2d@libcom.com>    Beach Runner wrote:  > Bob Koehler wrote: >> In article <paul.sture.nospam-B52EF9.18334617112006@mac.sture.homeip.net>, Paul Sture <paul.sture.nospam@hispeed.ch> writes: I >>> You could bash in the documentation with something more reliable, and , >>> sort out the "Word format" bit later on.G >>    Exactly the point of using OpenOffice (more than just word, there 0 >>    is a standard for which fonts, ... I use). > H > As someone that loved VAX 11-780s, many, many, many years ago, I can'tH > help really wondering why start it up. The power alone is absurd.  ForH > the cost one could get something much smaller and faster, and still be > a VAX. > G > If it is a public university taking public funds, paying for licenses & > just would be real tough to justify. > F > The only possible justification is to run something that can only be: > used on that specific hardware platform, as some ancientH > legacy application.  In which, buying some type of compiler just can't
 > make sense.  > G > Let's face it, this was the first generation VMS system. A classic. I F > would LOVE the faceplate.   It wasn't even the "Performance Plus" of
 > the 785. >   C As Rich has already stated, this is for a museum.  That is a total   justification.   --  4 David Froble                       Tel: 724-529-0450> Dave Froble Enterprises, Inc.      E-Mail: davef@tsoft-inc.com DFE Ultralights, Inc.  170 Grimplin Road  Vanderbilt, PA  15486    ------------------------------   Date: 21 Nov 2006 12:29:16 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)F Subject: Re: What layered products are important for a VAX VMS system?/ Message-ID: <4sg9osFuol69U1@mid.individual.net>   9 In article <Bp-dnatZ35TtT__YnZ2dnUVZ_h6dnZ2d@libcom.com>, * 	Dave Froble <davef@tsoft-inc.com> writes: > E > As Rich has already stated, this is for a museum.  That is a total   > justification.   C Speaking of mueseums.  Is there anyone here interested in three (3) D VAX 7000-640's?  I would really hate to see these loaded in the back? of a dump truck with a forklift and hauled off to the landfill.    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: 21 Nov 2006 08:21:35 -0800 From: bryan.s.a@gmail.com  Subject: Re: Who called me? A Message-ID: <1164126095.232435.32920@m7g2000cwm.googlegroups.com>    Stephen Hoffman wrote: > bryan.s.a@gmail.com wrote: > B > > $ link/trace/debug get_invo_info,SYS$SHARE:TRACE.EXE/shareable > H >    I'd use a linker options file, and put the TRACE.EXE command there. > > >    LINK/TRACE/DEBUG get_invo_info,ddcu:[dir]trace.opt/option > - >    And over in trace.opt, there's the line:  >  >    ddcu:[dir]trace.exe/share > I >    The /SHAREABLE on the command line doesn't do what you want here, as D > it creates a shareable image as its output.  The /SHAREABLE in theH > options file references a shareable image for inclusion in the linking2 > operation.  [Confusing and inconsistent, yes...] > 3 > > I'm wondering if the trace.exe file is corrupt.  > E >    The file is not likely corrupt, but you could verify that with a I > DIFFERENCES command against a known good copy; against the TRACE.EXE on M > a distribution kit, or the version from the most recent associated ECO kit.  > ( >    -- another approach, off the top -- >  > #include <lib$routines.h>  > #include <ssdef.h> > #include <string.h>  >  > main() >    {$ >    char ascic_debug_commands[128]; >    char *dbgcmd = F > "*set log ddcu:[dir]crash.dat;set output verify;show calls;go;exit"; > , >    strcpy( ascic_debug_commands, dbgcmd );; >    ascic_debug_commands[0] = (char) strlen( dbgcmd ) - 1;  > 2 >    lib$signal(SS$_DEBUG,1,ascic_debug_commands); >  >    return 1; >    }  
 Thank You.  F I put the sys$share:trace.exe/share in the .OPT file and no longer getC the illegal record errors, however I still get the undefined symbol C errors so it doesn't appear to be pulling in the traceback routines  into the image.   B We are unable to use the method of output debug info to a file andF reading because we are wanting to use this to obtain audit informationG to store with each of the hundreds of thousands of writes/updates we do B each day and using such an output file would increase our overhead# several-fold for each write/update.   A It seems like this should be a simple thing to do, but so far the  solution escapes me.  C I really appreciate any help towards resolution.  Each post gets me  closer.   # Any additional help is appreciated.   
 Thank You, -Bryan   ------------------------------    Date: 21 Nov 2006 09:02:00 -0800/ From: "Volker Halle" <volker_halle@hotmail.com>  Subject: Re: Who called me? C Message-ID: <1164128520.122610.163070@h54g2000cwb.googlegroups.com>    Bryan,  B prior to OpenVMS V8.3, there is the TBK$SYMBOLIZE internal routine8 available, but this interface/API is not documented (see [TRACE]TBKSYM).    Volker.    ------------------------------  % Date: Tue, 21 Nov 2006 12:52:18 -0500 8 From: Stephen Hoffman <Hoff@HoffmanLabs-RemoveThis-.Org> Subject: Re: Who called me? ( Message-ID: <ejveck$1sg$1@pyrite.mv.net>  E    Please consider posting the specific OpenVMS platform and version  D involved here, and please also consider posting the specific linker I error messages you mention in your reply.  These details can help tailor  G the answers to your needs and your environment; these details can help  - answer your current questions in this thread.     H > I put the sys$share:trace.exe/share in the .OPT file and no longer getE > the illegal record errors, however I still get the undefined symbol E > errors so it doesn't appear to be pulling in the traceback routines  > into the image.   D    I'd initially assume that you're not running the minimal OpenVMS G version needed for whatever feature you're calling, but -- without the  H command and the errors and the version info -- that can only be a guess G on my part.  If you're using some of the TBK calls discussed, you need  H OpenVMS Alpha V8.3 or OpenVMS I64 V8.3.  Older releases and OpenVMS VAX < releases do not have various of the routines discussed here.  G    nb: system and library and shareable image routines don't generally  G get pulled into the image being built by the linker, a reference to an  I external image where the code resides or a reference into OpenVMS itself  D gets added into the image being built.  During the image activation G process, the image activator finds the external reference, and locates  F and ensures the necessary executable code is present and/or is merged B into the run-time environment, and ensures that the code can then H correctly access the virtual address where the target routine exists in I process virtual address space.  This linker processing as differentiated  G from how the linker and the image activator process object code.  With  I object code, yes, a copy of the object code tends to get merged into and  G built into the particular image being built.  For details on shareable  F images and linking and such, you might want o see the Shareable Image H Cookbook in the left navigation of the old OpenVMS Ask The Wizard area.    But I digress... :-)  H    I'm making a number of assumptions here around the application scale H and the environment based on unfortunately rather information, which is H a situation I find rather frustrating -- it means I can potentially and G unintentionally provide you with bad or ill-suited information.  Which  B is unfortunate.  (I'm not purposely trying to be, um, dense. :-) )  G    One assumption: that you're close enough to your performance limits  E that activating the debugger to get this information is problematic.  H (It might well be, or it might not -- that really depends on the system I and the current system load and the frequency of activating the debugger  H image, for instance.)  If you're not close to the performance limits, a F comparatively inefficient solution -- in terms of processor cycles -- I might be the most appropriate -- in terms of the engineering and testing  - time involved, and the speed of the solution.   G    Another potential option here is to look to security auditing here,  I as write auditing will produce some of what is required in the auditing,  E and -- in conjunction with the linker maps and compiler listings and  H maybe even a few creative uses of identifiers to mark where in the code H the application is executing from -- might well suffice for your needs.    But again, that's a guess.  F    If you're in a relative hurry here, I'd tend to suggest hiring the G formal assistance of someone with skills in this area, and the ability  E to con-call or "whiteboard" this effort with and for you.  This will  C likely expedite whatever  compliance-, auditing-, security- and/or  @ accountability-related solution is required for your particular H environment.  There are various vendors -- and in the interests of full D disclosure, HoffmanLabs is one -- that provide this sort of service.  G    But in the interim and assuming you are not using OpenVMS V8.3 here  I (and assuming you're not bringing somebody in to help with this), locate  9 the undocumented routine I mentioned a few postings back.       TBK$SHOW_TRACEBACK   3    <http://h71000.www7.hp.com/wizard/wiz_5175.html>   I    That (undocumented? and unsupported? and subject to change?) call may  F or may not work for you here (and over future upgrades), if something D here or elsewhere within this thread in general has not piqued your E interest and/or resolved your auditing particular requirement.  (And  C assuming you can't compartment the calling and called code in some  B fashion...  Whether a set of "where-am-I" identifiers or a shared 0 "where-am-I" text global symbol or otherwise...)   ------------------------------   End of INFO-VAX 2006.641 ************************