/ INFO-VAX	Wed, 03 Jan 2001	Volume 2001 : Issue 5       Contents:- Re: Alpha nonstrategy, was Re: Happy Holidays - Re: Alpha nonstrategy, was Re: Happy Holidays 8 Applications using files by file ID instead of directory2 Re: Auditing VMS system for identifier references?6 RE: Bug - NLB0: crash, was: "NLA0: the null device..."6 RE: Bug - NLB0: crash, was: "NLA0: the null device..."+ Re: DECnet Error After Adding DE500 to 2100 + Re: DECnet Error After Adding DE500 to 2100 	 Re: echo? 	 eXcursion 
 Re: eXcursion $ File system performance and security Re: FTP is waiting to long!  Re: Giving up on VMS Re: Giving up on VMS Re: Giving up on VMS Re: Giving up on VMS Re: Giving up on VMS Re: Happy New Years  Re: Happy New Years  Re: Happy New Years  Re: Happy New Years  Re: Happy New Years 4 Re: Help for ORACLE v7.3 on VMS v7.1 and TCPIP v5.0A! Re: Problems with PATHWORKS V6.0C " Release date for 7.3 (and 7.3-1??)& Re: Release date for 7.3 (and 7.3-1??) Re: relocation Re: relocation Re: relocation Re: relocation Report Scheduler Re: Report Scheduler& Re: Sending messages to an application) Re: Source Code to Unsupported EDT Editor ) Re: Source Code to Unsupported EDT Editor  SSH2 under VMS?  Re: SSH2 under VMS?  Re: SSH2 under VMS? ' Suggestion for improvement for DECterm. + Re: Suggestion for improvement for DECterm. 5 TRAM and writebehind (was: Re: Giving up on the Best)  vms software administration  Re: vms software administration  Re: VR241 monitor and S-video  Re: VR241 monitor and S-video  Re: VR241 monitor and S-video   Re: XP1000 - which Graphics Card  F ----------------------------------------------------------------------   Date: 2 Jan 2001 19:13:31 GMT 1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon) 6 Subject: Re: Alpha nonstrategy, was Re: Happy Holidays+ Message-ID: <92t98r$i9p$1@info.cs.uofs.edu>   < In article <oo346.35612$1t.1793320@typhoon.ne.mediaone.net>,7  "Terry C. Shannon" <terryshannon@mediaone.net> writes:  |>  O |> Yep. And given the massive cost of porting the aforementioned apps to Alpha, , |> this restriction isn't likely to go away.  F I wondered how long before this came up again.  We had a rather livelyE discussion on this very subject not long ago and there were even some I rather interesting suggestions.  And while a number of people even agreed G that there was a potential solution, talking is as far as it ever gets.    bill     --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  # Date: Tue, 02 Jan 2001 20:18:34 GMT 4 From: "Terry C. Shannon" <terryshannon@mediaone.net>6 Subject: Re: Alpha nonstrategy, was Re: Happy Holidays= Message-ID: <uIq46.10667$BI2.2275080@typhoon.ne.mediaone.net>   > "Bill Gunshannon" <bill@triangle.cs.uofs.edu> wrote in message% news:92t98r$i9p$1@info.cs.uofs.edu... > > In article <oo346.35612$1t.1793320@typhoon.ne.mediaone.net>,9 >  "Terry C. Shannon" <terryshannon@mediaone.net> writes:  > |>J > |> Yep. And given the massive cost of porting the aforementioned apps to Alpha,. > |> this restriction isn't likely to go away. > H > I wondered how long before this came up again.  We had a rather livelyG > discussion on this very subject not long ago and there were even some K > rather interesting suggestions.  And while a number of people even agreed I > that there was a potential solution, talking is as far as it ever gets.  >  > bill  E Well, being as virtually none of the denizens of this noosegroup hold < influential positions at Compaq, there isn't much we can do.  K Of course, at $15 per share, it wouldn't be too difficult to pick up ten or J a hundred shares of CPQ, thereby gaining access to the Annual Meeting (notH too tough for the Texas constituency!) and a platform for voicing views.   ------------------------------  % Date: Tue, 02 Jan 2001 17:39:40 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca> A Subject: Applications using files by file ID instead of directory , Message-ID: <3A52589A.CFABC2DB@videotron.ca>  M Ok, you have an application that stores a gazillion small files and maintains H its own index of files (containing some attributes for each file such as sender, date etc).  G Would it be possible under VMS to have the application create the files F without having them entered in a VMS directory ? (i.e. only entered inF indexf.sys) and having the application use the files by file ID only ?  K Wouldn't this result in much faster file creation and especially deletion ? N Also, wouldn't file access (open) be faster since you're supplying the file ID3 right away and no need to search a directory file ?    ------------------------------  # Date: Tue, 02 Jan 2001 19:44:48 GMT 3 From: Tim Llewellyn <tim.llewellyn@cableinet.co.uk> ; Subject: Re: Auditing VMS system for identifier references? / Message-ID: <3A522F1E.E3BDA2F6@cableinet.co.uk>    JF Mezei wrote:    H > Also, is there a tool to ensure that SHOW.EXE is in fact the bona-fide" > SHOW.EXE and not some impostor ?  H I wrote a tool once to checksum all files in a directory tree, store theF results in a text files then compare this with the contents of another
 directory.@ I guess you'd call this "brute force" but you could then compare	 checksums C from an existing system with a freshly installed system of the same  version  and patch levels.   @ Unless there is some new "verification" feature in PCSI I havn't noticed,H there is no equivalent of the "setld -v" (I think thats what it is) from Tru64 for VMS.   Regards    ------------------------------  % Date: Tue, 02 Jan 2001 15:29:59 -0500 4 From: "Bochnik, William J" <BochnikWJ@bernstein.com>? Subject: RE: Bug - NLB0: crash, was: "NLA0: the null device..." J Message-ID: <2B37459189B0D211BE710000F8EF9D8508908AAC@nts0147.beehive.com>  J This message is in MIME format. Since your mail reader does not understand< this format, some or all of this message may not be legible.  ' ------_=_NextPart_001_01C074FA.D191CA14  Content-Type: text/plain;  	charset="iso-8859-1"   L how obvious does my tongue have to be planted in my cheek for it to be seen?   -----Original Message----- From: system@SendSpamHere.ORG  Sent: December 30, 2000 1:50 PM  To: Info-VAX@mvb.saic.com ? Subject: RE: Bug - NLB0: crash, was: "NLA0: the null device..."     J In article <2B37459189B0D211BE710000F8EF9D8508908AA1@nts0147.beehive.com>,6 "Bochnik, William J" <BochnikWJ@bernstein.com> writes:  K >It was my understanding that Windows NT has a hardcoded buffer on the null K >device of 32768 bytes requiring periodic reboots to empty the bucket.  VMS L >does not suffer from this constraint as the buffer is dynamically allocatedF >from non-paged pool, and a buffer relcaimation process (process id 0)' >emptied the bucked in a timely fasion.   J That's completely stupid and wrong -- VMS-wise that is.  I don't know what0 inane mechanism might be employed in Weendoze.     --2 VAXman- OpenVMS APE certification number: AAA-0001 VAXman(at)TMESIS(dot)COM             I city, n., 1. a place where trees are cut down and streets are named after  them.   ' ------_=_NextPart_001_01C074FA.D191CA14  Content-Type: text/html; 	charset="iso-8859-1" + Content-Transfer-Encoding: quoted-printable   1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">  <HTML> <HEAD>9 <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =  charset=3Diso-8859-1">@ <META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
 5.5.2651.65"> 9 <TITLE>RE: Bug - NLB0: crash, was: &quot;NLA0: the null =  device...&quot;</TITLE>  </HEAD>  <BODY>  G <P><FONT SIZE=3D2>how obvious does my tongue have to be planted in my =  cheek for it to be seen?</FONT>  </P>  3 <P><FONT SIZE=3D2>-----Original Message-----</FONT> 8 <BR><FONT SIZE=3D2>From: system@SendSpamHere.ORG </FONT>9 <BR><FONT SIZE=3D2>Sent: December 30, 2000 1:50 PM</FONT> 3 <BR><FONT SIZE=3D2>To: Info-VAX@mvb.saic.com</FONT> I <BR><FONT SIZE=3D2>Subject: RE: Bug - NLB0: crash, was: &quot;NLA0: the =  null device...&quot;</FONT>  </P> <BR>   <P><FONT SIZE=3D2>In article =G &lt;2B37459189B0D211BE710000F8EF9D8508908AA1@nts0147.beehive.com&gt;, = @ &quot;Bochnik, William J&quot; &lt;BochnikWJ@bernstein.com&gt; = writes:</FONT></P>  E <P><FONT SIZE=3D2>&gt;It was my understanding that Windows NT has a = # hardcoded buffer on the null</FONT> I <BR><FONT SIZE=3D2>&gt;device of 32768 bytes requiring periodic reboots = % to empty the bucket.&nbsp; VMS</FONT> D <BR><FONT SIZE=3D2>&gt;does not suffer from this constraint as the =& buffer is dynamically allocated</FONT>G <BR><FONT SIZE=3D2>&gt;from non-paged pool, and a buffer relcaimation =  process (process id 0)</FONT> D <BR><FONT SIZE=3D2>&gt;emptied the bucked in a timely fasion.</FONT> </P>  G <P><FONT SIZE=3D2>That's completely stupid and wrong -- VMS-wise that = " is.&nbsp; I don't know what</FONT>I <BR><FONT SIZE=3D2>inane mechanism might be employed in Weendoze.&nbsp; =  </FONT>  </P>   <P><FONT SIZE=3D2>--</FONT> > <BR><FONT SIZE=3D2>VAXman- OpenVMS APE certification number: =@ AAA-0001&nbsp;&nbsp;&nbsp;&nbsp; VAXman(at)TMESIS(dot)COM</FONT> <BR><FONT = G SIZE=3D2>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =  </FONT> F <BR><FONT SIZE=3D2>city, n., 1. a place where trees are cut down and =$ streets are named after them.</FONT> </P>   </BODY>  </HTML> ) ------_=_NextPart_001_01C074FA.D191CA14--    ------------------------------   Date: 2 Jan 2001 20:56:03 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)? Subject: RE: Bug - NLB0: crash, was: "NLA0: the null device..." 6 Message-ID: <92tf93$334$1@mailint03.im.hou.compaq.com>   In article <2B37459189B0D211BE710000F8EF9D8508908AAC@nts0147.beehive.com>, "Bochnik, William J" <BochnikWJ@bernstein.com> writes: K :This message is in MIME format. Since your mail reader does not understand = :this format, some or all of this message may not be legible.  ..M :how obvious does my tongue have to be planted in my cheek for it to be seen?   !   Rather further, obviously.  :-)   C   Humor is easily misinterpreted when in text form, and this is the B   basis for more than a few flame-wars, hurt feelings, and generalI   miscommunications, and this ease-of-miscommunications is also arguably  F   the basis for the existence of the "emoticons" -- or whatever those H   little text this-is-a-joke gizmo and friends are more properly called.=   (But you probably knew all that; no offense is intended...)   D   Please turn off MIME when posting to the newsgroup or (as happenedD   here, based on the headers) when sending messages to the INFO-VAX D   mailing list...  Thanks!  (Counting the quoted text, your one-lineD   response came through as a nice round 100 lines of MIME-encoding.)  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Tue, 02 Jan 2001 11:26:58 -0500 - From: "Richard D. Piccard" <piccard@ohio.edu> 4 Subject: Re: DECnet Error After Adding DE500 to 2100( Message-ID: <3A520150.3ED8879A@ohio.edu>  D Does the VMS 6.2 SPD list the DE500 as a supported device?  If it isF newer than the OS version, it should not be surprising that it doesn't work.   #                                 RDP      "Stuart, Ed" wrote:   3 > We're running VMS V6.2 on an AlphaServer 2100. We / > just added a DE500 to the system along with a 1 > number of ECOs. The system reports that the new 1 > device EWB-0 is active via LANCP, but if we try , > and add the device to DECnet (phase IV) we- > receive the following when we try and start 	 > DECnet.  > 3 > %RUN-S-PROC_ID, identification of created process 
 > is 29600171 0 > %DCL-I-SUPERSEDE, previous value of MOM$SYSTEM > has been superseded % > %DCL-I-SUPERSEDE, previous value of ) > MOM$SYSTEM_NOSOFTID has been superseded % > %DCL-I-SUPERSEDE, previous value of ' > MOM$SYSTEM_SOFTID has been superseded 0 > %NCP-W-INVIDE, Invalid identification format , > Line > Line = EWB-0) > %SYSTEM-F-IVDEVNAM, invalid device name 0 > %NCP-W-INVIDE, Invalid identification format ,	 > Circuit  > Circuit = EWB-0 ) > %SYSTEM-F-IVDEVNAM, invalid device name + > %NCP-I-NOINFO, No information in database 3 > %RUN-S-PROC_ID, identification of created process 
 > is 29600173  >  > Ed Stuart ' > Manager, Systems and Desktop Services ! > Information Technology Services  > City of Austin, Austin Energy  > Ed.Stuart@austinenergy.com > , > "Glittering prizes and endless compromises0 > shatter the illusion of integrity" - Neil Pert > D > *Please apply a generous amount of all the usual disclaimers here*   --B ==================================================================B Dick Piccard                           Academic Technology ManagerB piccard@ohio.edu                                 Computer ServicesB http://oak.cats.ohiou.edu/~piccard/                Ohio University   ------------------------------  # Date: Tue, 02 Jan 2001 19:35:36 GMT  From: goodman_j@my-deja.com 4 Subject: Re: DECnet Error After Adding DE500 to 2100) Message-ID: <92tai8$ddl$1@nnrp1.deja.com>   E DECnet line and circuit names are not the same as the VMS device name F displayed from the console, SHOW DEVICE or LANCP.  In this case deviceF EWB0: maps to line/circuit EWA-1.  In general after adding an ethernet> device run @SYS$MANAGER:NETCONFIG.COM, at least to examine theE recommended NCP commands.  You can then manually execute the ones you  require.  
 In articleG <6FACDDDFBD7BD411B38100D0B7B0CDCC40B5F7@ohms.electric.ci.austin.tx.us>, 2   "Stuart, Ed" <Ed.Stuart@austinenergy.com> wrote:3 > We're running VMS V6.2 on an AlphaServer 2100. We / > just added a DE500 to the system along with a 1 > number of ECOs. The system reports that the new 1 > device EWB-0 is active via LANCP, but if we try , > and add the device to DECnet (phase IV) we- > receive the following when we try and start 	 > DECnet.  >...0 > %NCP-W-INVIDE, Invalid identification format , > Line > Line = EWB-0) > %SYSTEM-F-IVDEVNAM, invalid device name 0 > %NCP-W-INVIDE, Invalid identification format ,	 > Circuit  > Circuit = EWB-0 ) > %SYSTEM-F-IVDEVNAM, invalid device name  --G "Any technology sufficiently advanced is indistinguishable from magic." 
 A.C.Clarke     Sent via Deja.com  http://www.deja.com/   ------------------------------   Date: 2 Jan 2001 16:30:27 GMT 2 From: mathog@seqaxp.bio.caltech.edu (David Mathog) Subject: Re: echo?, Message-ID: <92svn3$nkn@gap.cco.caltech.edu>  N In article <3a4bccd7.53294502@news1.telia.com>, asgd@sverige.nu (Asgd) writes:G >I trying to learn VMS (at the univerity) as a part in a bigger course. F >But now Im stuck and cant find the right command. Dont winkle your4 >nose to much...the command is at Unix and Dos: echo. >I want to print a line of text on the screen! >please help... B >If you know any god and easy-to-learn-VMS webbpage, that would be >great!   B Overview          http://seqaxp.bio.caltech.edu/www/vms_sheet.htmlJ Beginner's FAQ    http://seqaxp.bio.caltech.edu/www/vms_beginners_faq.html   Regards,   David Mathog mathog@seqaxp.bio.caltech.edu ? Manager, sequence analysis facility, biology division, Caltech     ------------------------------  % Date: Tue, 02 Jan 2001 17:12:45 -0300 ) From: fabio_compaq@ep-bc.petrobras.com.br  Subject: eXcursionL Message-ID: <OF66C23CFE.20DBFA71-ON032569C8.006EDD89@ep-bc.petrobras.com.br>  H It=B4s my first time configuring the eXcursion soft, and I am receiving=  F the following message when I try to start an application: in this case the Terminal from Dec Windows.  F TCP/IP is not running on your local system, please start it or  select another transport.  1 I am using Windows 95 and MS TCP/IP default stack    Regards    FC=    ------------------------------  $ Date: Tue, 2 Jan 2001 22:03:34 +0100. From: "Jesper Naur" <jesper.naur@post.tele.dk> Subject: Re: eXcursion, Message-ID: <92tfju$74o$1@news.inet.tele.dk>  6 <fabio_compaq@ep-bc.petrobras.com.br> wrote in messageF news:OF66C23CFE.20DBFA71-ON032569C8.006EDD89@ep-bc.petrobras.com.br...E Its my first time configuring the eXcursion soft, and I am receivingVF the following message when I try to start an application: in this case the Terminal from Dec Windows.  F TCP/IP is not running on your local system, please start it or  select another transport.  1 I am using Windows 95 and MS TCP/IP default stackr  K You don't exactly explain, what you are trying to do, but let me guess: You I are trying to use the X-server program "eXcursion" on a PC for displaying E the X-window belonging to a DECTerm running on a VMS-system (would be J helpful to know the machine type, VMS version and TCP/IP version). If this4 assumption is wrong, all bets below are perhaps off.  % The following questions come to mind:i  / 1. Exactly where does the error message appear?e6 2. Is TCP/IP configured and started on the VMS system?D 3. Is it possible to PING the VMS system from the PC and vice versa?I 4. Make sure, that the eXcursion server is started on the PC (can be done7 either explicitly, or from the     eXcursion Control Panel).eH 5. On the VMS system, in a locally started DECTerm, issue the following:  @         $ set display/create/trans=tcpip/node=<ip address of PC>         $ mc decw$puzzle  A     This should start the Puzzle program, with display on the PC.V  E The above mentioned steps should be successfully completed before youeC attempt to configure eXcursion Accounts and Applications, using thea eXcursion Control Panel.  J Have in mind, that the eXcursion distribution (the Host Support section, IJ believe), contains a file PCXSERVER.COM, which should be copied to the VMSL system under SYS$SYSTEM and renamed to PCX$SERVER.COM (the exact file namingC details may be slightly off, you should be able to find them in the  eXcursion description).r  I You can switch on Logging to a window from the eXcursion control panel one1 the PC - this may be helpful for troubleshooting.$  K TCP/IP on VMS must have the rexec service enabled, since eXcursion normallyn$ uses this for starting applications.  G Whenever you have defined an application under eXcursion, use the "Run"nD button to test it until it works. If you have to make changes to theJ application configuration AFTER you have created an icon, the icon must beK deleted and recreated, since the icon creation function takes a copy of thesL application configuration data, which will not be changed, if you change the application.         Best regards     Jesper Naurs   ------------------------------  % Date: Tue, 02 Jan 2001 15:18:29 -0500c- From: JF Mezei <jfmezei.spamnot@videotron.ca>c- Subject: File system performance and securityt, Message-ID: <3A52378E.9BE2BF42@videotron.ca>  K Recent throughs on file system performance has resulted in my asking myself- the following:  M If you give a process BYPASS and/or any/all all mighty privileges, would thatn( speed up file processing significantly ?  M For instance, supppose you run a news server which is the only app on a node.iN (no need for security). If a process has certain privileges, would any requestM for file access skip all the code to check for access, ACLs etc thus yieldingo performance improvements ?  L Or would VMS still go though the motions of checking the file protection andM ACLs and only at the end decide to grant access anyways due to the privilegesi% currently activated for the process ?    ------------------------------  $ Date: Tue, 2 Jan 2001 17:10:25 -0500. From: "M. T. Hollinger" <MyTH@ucx.lkg.dec.com>$ Subject: Re: FTP is waiting to long!6 Message-ID: <92tjs9$3ok$1@mailint03.im.hou.compaq.com>  8 "Roland Hauk" <roland.hauk@indramat.de> wrote in message' news:92hg2e$kvo1@sunny.mannesmann.de... L >     My problem is that FTP server on the other side is down, but the IP is > reachable.G >     In this case UCX is connecting to IP and waits to logon fore somel hours.8 >     I get no error, FTP is still waiting hour by hour!  J This delay suggests the FTP server host is responding in some way, perhapsL accepting the connection.  Since your other test, attempting a connection toJ a host which was entirely down, timed out in short order, the FTP client's" timeout does appear to be working.  $ What type of host is the FTP server?  F Can you get a protocol trace (such as TCPTRACE server-host-name) whileL attempting to connect?  That might help to determine what the server host isL sending back.  If it accepts the FTP connection (SYN ACK), but then sends noI further data, that would certainly explain your observed client behavior.                  - M    ------------------------------   Date: 2 Jan 2001 09:26:17 -0500a* From: young_r@eisner.decus.org (Rob Young) Subject: Re: Giving up on VMSt+ Message-ID: <gRz6Ntmkr7EQ@eisner.decus.org>d  _ In article <87vgrya6a1.fsf@k9.prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes:w   > - > BTW, requiring a UPS is admitting you lose.- >   8 	Not necessarily.  XIOtech's storage solution comes with; 	UPS builtin.  http://www.xiotech.com/  If you had said youc= 	need to bolt on a UPS to guarantee integrity, I would agree.-   				Rob1   ------------------------------  $ Date: Tue, 2 Jan 2001 10:58:37 -0500' From: "Bill Todd" <billtodd@foo.mv.com>e Subject: Re: Giving up on VMSs( Message-ID: <92stl9$dq3$1@pyrite.mv.net>  7 Paul Repacholi <prep@prep.synonet.com> wrote in messagen* news:87vgrya6a1.fsf@k9.prep.synonet.com...+ > "Bill Todd" <billtodd@foo.mv.com> writes:a >tE > > I think you may have misunderstood the context:  the 'poor' writeiJ > > performance is when compared to ext2fs on Linux, a very fast-and-looseK > > approach.  Those safer systems (that avoid the need for fsck) can still  look >  ^^^^D > > good compared with VMS, since with a well-designed log-protectedI > > implementation (I'm not privy to the internals of these products, but  IRIX'sF > > XFS seems very well-thought-of in this regard and indeed enjoys an enviableJ > > reputation for performance, including write performance) only a singleL > > synchronous log-disk write is required to attain all the safety of VMS'sL > > approach with in many cases significantly less operation latency (thoughL > > additional 'lazy' writes will eventually occur, with luck when the disksG > > aren't otherwise busy and with at least the hope of coalescing manyo updatesg) > > to the same pages into a single one).  >sJ > I think you should put the glossies where they belong, and sit down with > the code.0 >:J > Log based systems get their 'performance' by writeing a large contiguous( > chunk of dta and meta data in one hit.  B Hmmm.  Perhaps you're the one who needs to sit down and learn someL fundamental distinctions among different possible approaches to the problem.  H I have little use for *log-structured* storage, save for special-purposeI use.  That, however, is not what I was suggesting was superior:  it was a K *log-protected* approach, similar to the way a log protects a database, and K has none of the putative drawbacks you list below.  If you don't understandmL the difference, then there's little point in delving into greater detail; ifH you'd like to, start by re-reading what I wrote, and then study databaseK design for a bit (as I did, before I designed and implemented my first such   distributed file/object system).   - bill  !  However, to do this, you have toeG > have all reaads cached, the data in the cache, and the updates in thetH > cache, plus the extra overhead to the writelog, plus you must have theG > extra free space on the disk. If you run short, things can go to hell  > in a big way..." >tG > When you get to play catchup, then you must re-read, merge and write."C > The write phase is just what you had to do to start with. You canTB > cross your fingers and do bulk, non-attomic uppdates, but if youB > take a power or other failure then, you are back where you whereF > trying to get away from. Abiet, you can re-run the log if the designF > garentees an exact repeatable output for a given log. Note, exact...E > Otherwise you must do a roll-back, then an update. If you do thingsSF > like write coallesing, you loose the atomic, consistant state of the- > media, and can have to sort out the mess...t >EF > Now if you want to try to transfer the wonders of these systems into? > a cluster environment, best of british to you. And your data.0 >1@ > PAy now, or pay (more) later. But you do not get a free lunch.A > You can, however get most of the gain by either setting up your>F > app to 'do the right thing' with VMS, or setting the volume defaults% > or fstune'ing to get what you need.O > - > BTW, requiring a UPS is admitting you lose.r >i > --> > Paul Repacholi                               1 Crescent Rd.,9 > +61 (08) 9257-1001                           Kalamunda.tB >                                              West Australia 60760 > Raw, Cooked or Well-done, it's all half baked.   ------------------------------  $ Date: Tue, 2 Jan 2001 11:06:33 -0500' From: "Bill Todd" <billtodd@foo.mv.com>< Subject: Re: Giving up on VMS ( Message-ID: <92su45$e2g$1@pyrite.mv.net>  + Paul Sture <paul@sture.ch> wrote in message3% news:VA.00000201.25384c70@sture.ch...    ...e  F > What grabbed my attention was the problem of fsck-ing large volumes.  I Fsck is indeed an idea whose time has passed, if it was ever a reasonablenL solution in the first place:  it simply doesn't scale.  But modern Unix fileI systems (since about 1990, from Veritas, SGI, IBM, DEC - with the notable2L exception of the Linux toys:  even BSD has log-backed variants) use a log to avoid the need for fsck.    From myF > admittedly small amount of experience with fsck on my home Linux box (cheap, I > hardly state of the art), when it kicks in I might as well take a lunch  break.H > Extrapolate that to large systems and we are talking serious downtime. >DH > On the XFS front, I rather took the author's word for it, as I'm not a unixJ > expert. Re-reading in the light of your comments, I could deduce that he isB > speaking in terms of free / low cost unix when he says these are "definitelyn > not in a production state"  K I think he was referring to the state of these systems on Linux, where, foruG example, SGI's XFS port is still (last I knew, anyway) in an incomplete  beta.l   >- > > only a singlesL > > synchronous log-disk write is required to attain all the safety of VMS'sL > > approach with in many cases significantly less operation latency (thoughL > > additional 'lazy' writes will eventually occur, with luck when the disksG > > aren't otherwise busy and with at least the hope of coalescing manya updatess) > > to the same pages into a single one).a > >nH > Which description reminds me of NTFS.  I'm honestly not sure where the6 > performance of that lies with respect to the others.  I NTFS's performance is not stellar (though OK), mostly because it uses thekG log mostly for failure-protection and takes relatively little advantageaL (compared with my anecdotal understanding of what XFS does) of its potential to optimize performance.   ...-  ) > If you didn't read the complete articleMH > http://www.daemonnews.org/200012/tram.html , I think you might find itK > interesting. The author, Dru Nelson, goes on to propose Transactional-RAMc3 > (TRAM) as a method for addressing these problems.g  K Thanks - will try to get to it later.  I didn't know what it was about, andD it sounds interesting.   - bill   ------------------------------  % Date: Tue, 02 Jan 2001 15:03:22 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca>  Subject: Re: Giving up on VMSo, Message-ID: <3A523404.CC50AB28@videotron.ca>   Paul Repacholi wrote:d- > BTW, requiring a UPS is admitting you lose.r  E While I understand your argument,  I am not sure it is a valid one inLL industry. A corporation doesn't care how you maintain its data integrity, itI only cares that you do. Whether you do it through a UPS or through a lessnG efficient but safer file system, in the end, your data is just as safe.o  N UPS prevents failre, whereas the strong file systems minimises the impact of aG failure. Note that even with a strong file system, you still want a UPSvI anyways, so does the stronger file system really give you much additionalI value ?e   ------------------------------  % Date: Tue, 02 Jan 2001 15:07:45 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca>c Subject: Re: Giving up on VMSi, Message-ID: <3A52350B.A528D391@videotron.ca>   Rob Young wrote: > If you had said youoF >         need to bolt on a UPS to guarantee integrity, I would agree.  K But if bolting a UPS does garantee integrity, you've acheived the goal, andoC you probably end up with a much faster file system. You have betterS( performance with similar file integrity.   ------------------------------  # Date: Tue, 02 Jan 2001 13:07:24 GMTi% From: "Jim Tsetsos" <jim@tsetsos.net>a Subject: Re: Happy New Years? Message-ID: <gok46.41945$xW4.336321@news-server.bigpond.net.au>=   Hey,  E Here it don't snow and hardly even rains!  It reached about 37 Degreen3 Celcius today, (thats about 98.6 Degrees Fahrenheit L http://www.sanfranciscobay.com/live/temp.htm ) and haven't hit the middle of summer yet.-   Can you guess where I am?   % Thats right, Australia/Melbourne.....r    < "Hans Bachner" <Hans.Bachner@altavista.net> wrote in message" news:92si80.5l.1@hans.myfqdn.de...8 > David J. Dachtera (djesys.nospam@earthlink.net) wrote: >o > <snip>> > >> (remember, the internet is global in nature, SI rules...) >rL > May I help out with some numbers for the rest of us who are not really use > to inches and Fahrenheit?e >n> > 70 degrees Fahrenheit ~ 21 degrees Celsius (do you call them > 'centigrades'?)u >nL > Pretty warm for this part of the season - here in Austria we're around theH > freezing point. I'd love to be back to San Diego... only seen it once. >SK > Regarding the snow in Chicago - 20 inches are 50.8 cm, 50 inches are 1.270	 > meters.u >aL > Anyway, have a happy new millennium, regardless of units, scales, etc... -$ > as long as you're running OpenVMS! >v > Hans.    ------------------------------   Date: 2 Jan 2001 08:22:58 -0500v, From: koehler@eisner.decus.org (Bob Koehler) Subject: Re: Happy New Years+ Message-ID: <PxnOPBlw5mLq@eisner.decus.org>e  \ In article <3A50BFB8.77403E6C@videotron.ca>, JF Mezei <jfmezei.spamnot@videotron.ca> writes: > Koloth wrote: J >> The weather forecast is 70 degrees ABOVE zero.  Any thoughts of opening( >> a branch of OpenVMS development here. > : > Sorry, that is a bit hot.  30 more degrees and you boil. >   D Nope, it's quite cold.  About 230 more degrees and you thaw out.  8)  C IMHO Celcius is usefull for doing chemistry, Farenheit for weather.:H A simple concept known as using the right tool for the job:  Celcius wasE designed for dealing with the physical properties of water, Farenheity= was designed for measuring weather in Earth's temperate zone.e  D Of course the only really meaningfull measurement is -1/k where k is degrees Kelvin..  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporationl= NASA GSFC Flight Software       | Federal Sector, Civil Group E                                 | please remove ".aspm" when replyingp   ------------------------------  # Date: Tue, 02 Jan 2001 14:01:41 GMT / From: Mike Price <mike.price@littlewoods.co.uk>s Subject: Re: Happy New Years) Message-ID: <92sn02$s3o$1@nnrp1.deja.com>-  B I think you must have set up an auto-redirect for all your rain to Britain-9 If anyone wants some we seem to have more than we need!!!C   Mike --B All opinions are my own and do not necessarily reflect those of my employer     Sent via Deja.com, http://www.deja.com/   ------------------------------  % Date: Tue, 02 Jan 2001 16:51:35 +0100h  From: Paul Sture <paul@sture.ch> Subject: Re: Happy New Years+ Message-ID: <VA.00000202.26852e18@sture.ch>e  < In article <92si80.5l.1@hans.myfqdn.de>, Hans Bachner wrote: > 8 > David J. Dachtera (djesys.nospam@earthlink.net) wrote: >  > <snip>> > >> (remember, the internet is global in nature, SI rules...) > M > May I help out with some numbers for the rest of us who are not really use o > to inches and Fahrenheit?r > ? > 70 degrees Fahrenheit ~ 21 degrees Celsius (do you call them c > 'centigrades'?). > L Not 'centrigrades' but 'degrees Centigrade'. Celsius replaced Centigrade in J weather reports etc in the UK some years ago (15? 20? - I can't remember).  I Although in the UK it was promised that the Fahrenheit readings would be tK phased out at some point in the future, I don't think it has happened yet. a= AFAIK, UK weather forecasts give both Fahrenheit and Celsius.   I FWIW, the most confusing for me between F and C is when reading American "M literature. One quote went something like "It was 5 degrees above zero..." - sM "Huh, not very cold I thought", until I realised I should read it as degrees i Fahrenheit :-)  M > Pretty warm for this part of the season - here in Austria we're around the eH > freezing point. I'd love to be back to San Diego... only seen it once. > L Just above freezing here in the Zuerich area. A couple of inches of snow on 9 Friday at home, the last of which melted in today's rain.c  L > Regarding the snow in Chicago - 20 inches are 50.8 cm, 50 inches are 1.27 	 > meters.s > M > Anyway, have a happy new millennium, regardless of units, scales, etc... -  $ > as long as you're running OpenVMS! >  > Hans.p >n   ___ 
 Paul Sture Switzerland    ------------------------------  # Date: Tue, 02 Jan 2001 22:17:51 GMTo% From: "Jim Tsetsos" <jim@tsetsos.net>  Subject: Re: Happy New Years? Message-ID: <jss46.42284$xW4.340048@news-server.bigpond.net.au>c  K I must say though, our winter sees no snow, some rain but bearly gets belown
 10 C or 50 F.t      < "Mike Price" <mike.price@littlewoods.co.uk> wrote in message# news:92sn02$s3o$1@nnrp1.deja.com...yD > I think you must have set up an auto-redirect for all your rain to	 > Britainl; > If anyone wants some we seem to have more than we need!!!  >X > Mike > --D > All opinions are my own and do not necessarily reflect those of my
 > employer >e >h > Sent via Deja.coml > http://www.deja.com/   ------------------------------  % Date: Tue, 02 Jan 2001 16:50:20 -0000r- From: wspencer@ap.nospam.org (Warren Spencer) = Subject: Re: Help for ORACLE v7.3 on VMS v7.1 and TCPIP v5.0Ah/ Message-ID: <t541mcq68u9d03@news.supernews.com>u  A JBUTLER@utmem.edu wrote in <01JYE4RXC48290MW8K@UTMEM1.UTMEM.EDU>:e  H >     I have OpenVMS v7.1 Alpha and ORACLE v7.3.2 and was using UCX v4.2F >with ECOs.  I upgraded to TCPIP v5.0A ECO1 and the ORACLE Listener noG >longer works.  It seems to start up and you can successfully telnet tog< >the port being used, but the SQLNET connection gives error.C >     Does anyone have experience with this?  Do you have any hints@) >for why the version of TCP would matter?m$ >               Thanks, Jenny Butler! >               JBUTLER@UTMEM.EDUt >r  ; What is the exact error message and number you are getting?l   ws   -- w3 << What if there were no hypothetical questions? >>l   Warren Spencer Senior Software Engineer The Associated Press  ? ** My employer does not necessarily agree with my statements **h   ------------------------------  # Date: Tue, 02 Jan 2001 16:51:53 GMTp: From: brad.mccusker@compaq.remove_this.com (Brad McCusker)* Subject: Re: Problems with PATHWORKS V6.0C1 Message-ID: <3a5205ea.2031016758@news.compaq.com>i  , On Fri, 22 Dec 2000 05:21:11 GMT, "Alphaman"! <aaron@nospam-alphant.com> wrote:e     >We had in excess of aJ >quarter million files spread across close to 100 file shares and over 130F >GBytes.  Setting the security model to AS+VMS meant that for users inL >another domain to access a file also meant that we had to either set an ACLM >for that file or set it to W:R{W|E|D}.  Now multiply that across the 10's oflL >thousands of files that are in some of our larger shares, and that's eitherJ >a lot of VMS ACLs or a lot of World accessible files.  Or, as we wound upL >doing, a reliance on only Advanced Server security from the Advanced Server
 >perspective.  >r  F Hmmm... I think the lesson here is that we really need to work on the D documentation of hostmapping, and its relation to network users and > especially network users from trusted domains.  We are indeed  trying to address this.s  A Bottom line though is that adding OpenVMS security shouldn't makeo things less secure.  No way.  M >Oh, and our VMScluster is our PDC.  I'd be tempted to change the role of ourAM >BDC's to member servers if only MS allowed it...  (Can't down-grade a servere# >like that, at least not easily.)  k  @ Well, when we ship AS V7.3 (any day now...), you will be able toC change from *DC to MemberServer and Vice/versa without reisntallingrF the whole operating system.  Well, OK, you'll need to run config, but, thats still better than NT!   + >Who needs a PeeCee to act as a backup to aR >VMScluster??? 8^)  A Who needs a PeeCee to serve files at all!  At least not important6 files!       Regards,  
 Brad McCuskerl/ OpenVMS Advanced Server Engineering (PATHWORKS)r Compaq Computer Corporationo  3 All comments contained herein are my own and do notl* reflect those of anyone or anything else.    ------------------------------  # Date: Tue, 02 Jan 2001 12:46:34 GMTn/ From: Mike Price <mike.price@littlewoods.co.uk>-+ Subject: Release date for 7.3 (and 7.3-1??) ) Message-ID: <92sij9$ov8$1@nnrp1.deja.com>@  F Does anyone know the offical UK release date for VMS 7.3 (and 7.3-1 as& mentioned in a fibre channel thread)??  A The last I heard from Compaq was Feb 2001 but that was months ago.  H Maybe I can start planning the year out when I know what versions I need   Thanks   Mike   --B All opinions are my own and do not necessarily reflect those of my employer     Sent via Deja.como http://www.deja.com/   ------------------------------   Date: 2 Jan 2001 18:46:30 GMT 2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)/ Subject: Re: Release date for 7.3 (and 7.3-1??) 6 Message-ID: <92t7m6$1q4$2@mailint03.im.hou.compaq.com>  [ In article <92sij9$ov8$1@nnrp1.deja.com>, Mike Price <mike.price@littlewoods.co.uk> writes:aG :Does anyone know the offical UK release date for VMS 7.3 (and 7.3-1 asc' :mentioned in a fibre channel thread)??a  (   The OpenVMS roadmaps are available at:  F     http://www.openvms.compaq.com/openvms/roadmap/openvms_roadmaps.htm  K   I do not know that any more specific (nor any UK-specific) release dates h   have been announced. e  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Tue, 02 Jan 2001 18:26:55 GMTw= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)  Subject: Re: relocationu0 Message-ID: <009F5848.23970798@SendSpamHere.ORG>  k In article <OF98A77728.2496514E-ON802569C8.005FA7C4@qedi.quintiles.com>, steven.reece@quintiles.com writes:c >eH >I thought that addresses less than 200 were invalid anyway and would be >likely to cause ACCVIOs?  > @ >Bob Koehler (koehler at eisner dot decus dot org) wrote/quoted:6 >>>>Ashutosh Dhodapkar <dhodapka@ece.wisc.edu> writes: >>E >> I needed some help on relocation of binaries. basically, i want tolJ >> relocate a binary so that it occupies virtual address 0x00 however i amG >> unable to do this on the alpha system that we have. any help will beo >> appreciated.  >lE >You might be able to accomplish this with base=0 in the link optionsHC >file (or is that VAX only?).  You don't want to, I spent some timeiG >cleaning this out of an application a couple years ago, which would no 8 >longer build or run because someone used a base option. >hI >What are you trying to do?  There is probably a better way to do it than  >basing the image.<<<t >p   Simple example:r   File SIMPLE_BASE_TEST.MAR is:t            .PSECT  DATA,WRT,NOEXE,5 MSG:    .ASCID  "Hello World!"            .PSECT  CODE,NOWRT,EXE,5         .ENTRY  START_HERE,0         PUSHAB  MSGl#         CALLS   #1,G^LIB$PUT_OUTPUTe         RET          .END    START_HERE   $ MACRO SIMPLE_BASE_TEST7 $ LINK /BPAGE=13/SHAREABLE/NOTRACEBACK SIMPLE_BASE_TESTg $ RUN SIMPLE_BASE_TEST Hello World!    $ $ ANALYZE/IMAGE SIMPLE_BASE_TEST.EXE   .g   .    .w$         Image Activation Information  5                 first transfer address:  %X'00000000'u5                 second transfer address: %X'00000000'r5                 third transfer address:  %X'00000000'm   .b   .f   .i [QED]o --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMc            ,O city, n., 1. a place where trees are cut down and streets are named after them.t   ------------------------------   Date: 2 Jan 2001 18:37:21 GMTg2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) Subject: Re: relocation 6 Message-ID: <92t751$1q4$1@mailint03.im.hou.compaq.com>  k In article <OF98A77728.2496514E-ON802569C8.005FA7C4@qedi.quintiles.com>, steven.reece@quintiles.com writes:  : H :I thought that addresses less than 200 were invalid anyway and would be :likely to cause ACCVIOs?-  I   By default, OpenVMS VAX and OpenVMS Alpha LINKERS map a no-access page .I   at the bottom of virtual address space -- this in order to trap errant mK   pointer references and such.  This does not waste physical memory, as no 4I   physical page is mapped.  On OpenVMS VAX, this is the first 512/0x0200 fJ   bytes.  Given the variable page size, the inaccessable region of virtualI   address space on OpenVMS Alpha is 8192/0x02000 to 65536/0x010000 bytes.c  I   There are a few applications that expect references to virtual address ,H   zero to function, and there are a few programs that expect to overmap K   the lowest page.  This technique is comparatively rare, and dependencies dJ   on the application address space are not something I generally recommendI   incorporating and/or preserving within an application.  (One of the fewaM   cases where this level of control over the layout of image virtual address DM   space can be necessary involves working with embedded application code and xF   particularly with ROM and with RAM using RAM or NVRAM vector areas.)  L   The BASE directive is available within the OpenVMS VAX LINKER, while both L   OpenVMS VAX and OpenVMS Alpha users can utilize use the CLUSTER directive.  M   Given the "fun" I've had over the years maintaining applications that have tJ   (mis)used fixed address ranges and given the number of latent bugs that I   this no-access virtual address space has uncovered, I cannot generally  G   recommend the technique of basing images nor the technique of mapping C   accessable memory into the first page of virtual address space...e  N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 2 Jan 2001 14:25:54 -0500 , From: koehler@eisner.decus.org (Bob Koehler) Subject: Re: relocationt+ Message-ID: <OA6LFe$1wCKZ@eisner.decus.org>k  k In article <OF98A77728.2496514E-ON802569C8.005FA7C4@qedi.quintiles.com>, steven.reece@quintiles.com writes:a > I > I thought that addresses less than 200 were invalid anyway and would be  > likely to cause ACCVIOs? >   A On VAXen the guard page coveres addresses 0 - 1FF.  I don't thinkmD Alpha implements a guard page, but the linker may act as if it does.  H The LINKER manual states BASE is not supported for Alpha, but the effectB may be had by using the CLUSTER option if not creating a shareable image.  @ Furthermore, "The linker bases shareable images at address 0, byB default".  IIRC it still assumes they are relocateable, which BASE	 prevents.i  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences CorporationS= NASA GSFC Flight Software       | Federal Sector, Civil GroupIE                                 | please remove ".aspm" when replyinge   ------------------------------  # Date: Tue, 02 Jan 2001 22:46:45 GMT + From: John Santos <john.santos@verizon.net>f Subject: Re: relocations> Message-ID: <MPG.14bc24af9023ac129896ac@news.bellatlantic.net>  7 In article <92t751$1q4$1@mailint03.im.hou.compaq.com>, f% hoffman@xdelta.zko.dec.nospam says...n > m > In article <OF98A77728.2496514E-ON802569C8.005FA7C4@qedi.quintiles.com>, steven.reece@quintiles.com writes:a > :eJ > :I thought that addresses less than 200 were invalid anyway and would be > :likely to cause ACCVIOs?m > K >   By default, OpenVMS VAX and OpenVMS Alpha LINKERS map a no-access page 1K >   at the bottom of virtual address space -- this in order to trap errant  M >   pointer references and such.  This does not waste physical memory, as no eK >   physical page is mapped.  On OpenVMS VAX, this is the first 512/0x0200 hL >   bytes.  Given the variable page size, the inaccessable region of virtualK >   address space on OpenVMS Alpha is 8192/0x02000 to 65536/0x010000 bytes.   G When I first saw this on VMS, I thought the guard pages were a terrific E idea.  Of course, I had to find a way to bypass them in my first VMS ,
 program...  H This was ROSS/V, RSTS/E emulator for VMS.  The PDP-11 compatibility modeE on the VAX-11/* machines could only access the 1st 16 bits of virtualmG address space, so to execute PDP-11 images, you had to load them there.uD (There wasn't any sort of bias register or explicit emulation of theC PDP-11 memory management registers; the compatibility mode hardwarelD just relied on normal VMS virtual address mapping.  So I had to baseB the ROSS/V image at (or above) %X10000 and use CRMPSC or CRETVA to@ manage shared or unshared PDP-11 memory in the region of 0-FFFF.  D Fortunately the VAX linker had the BASE option to make this trivial.7 I think the RSX AME people at DEC had the same problem.h  F When the non-PDP-compatible VAXes came along, CEM$EMULATOR.EXE behaved
 the same way.e  F If one were writing a PDP-11 emulator from scratch (say for an Alpha),C one could incorporate a "bias" register to add or subtract from any0D PDP-11 virtual memory address to produce a real virtual address, andD thus map the PDP-11 code and data at any convenient virtual address.D This would cost at least one "ADD" on every memory reference though.9 (There is lots of room for optimization in this, though.)s  H One could also emulate the PDP-11 memory management registers.  I assume6 E-11 must do this, and it still gets good performance.  @ If one wanted to emulate I&D space or Supervisor mode, it would D definitely require at least a primitive memory mapping scheme, since? only one of these could be located at virtual addresses 0-FFFF.o  K >   There are a few applications that expect references to virtual address cJ >   zero to function, and there are a few programs that expect to overmap M >   the lowest page.  This technique is comparatively rare, and dependencies  L >   on the application address space are not something I generally recommendK >   incorporating and/or preserving within an application.  (One of the fewtO >   cases where this level of control over the layout of image virtual address  O >   space can be necessary involves working with embedded application code and -H >   particularly with ROM and with RAM using RAM or NVRAM vector areas.) > N >   The BASE directive is available within the OpenVMS VAX LINKER, while both N >   OpenVMS VAX and OpenVMS Alpha users can utilize use the CLUSTER directive. > O >   Given the "fun" I've had over the years maintaining applications that have iL >   (mis)used fixed address ranges and given the number of latent bugs that K >   this no-access virtual address space has uncovered, I cannot generally oI >   recommend the technique of basing images nor the technique of mappingdE >   accessable memory into the first page of virtual address space...t   Yup.     > P >  --------------------------- pure personal opinion ---------------------------N >    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com  @ I wonder if the original poster was attempting to write a virus?D He appeared to be asking about relocating code to low address space ; (where maybe he thought the boot sector or OS was located?)t   --   John Santos>   ------------------------------  $ Date: Tue, 2 Jan 2001 10:11:52 -06001 From: "Mark-Simon Pope" <mpope@NOSPAM@bristol.ca>  Subject: Report Schedulere0 Message-ID: <v9n46.5096$ok4.72336@news1.mts.net>  H I'm looking for a decent report scheduler, preferably written in DCL andJ Free.  Something that uses a .DAT file to define a schedule would be nice.  J It would have to be able to handle oddball rules like "Every Tuesday after the last Friday of each month."   J I have a long and complicated DCL that currently runs my  reporting and is becoming hard to manage.   Any suggestions?   Mark-Simon Popeb Bristol Aerospace Limited  Winnipeg, Manitoba.n   ------------------------------  $ Date: Tue, 2 Jan 2001 14:59:49 -0800- From: "Mark Rodgers" <mrodgers@earthlink.net>v Subject: Re: Report Schedulerc@ Message-ID: <%Ts46.90419$EM4.456863@e420r-sjo3.usenetserver.com>  L Take a look at ISE's EnterpriseSCHEDULE. You can schedule your reports basedF on named days, intervals and/or marked calendars. It uses two types ofL dependency and has advanced programming features based on variables. You canB use DCL commands or run jobs in a MOTIF interface. Check it out at www.i-s-e.com .     : Mark-Simon Pope <mpope@NOSPAM@bristol.ca> wrote in message* news:v9n46.5096$ok4.72336@news1.mts.net...J > I'm looking for a decent report scheduler, preferably written in DCL andL > Free.  Something that uses a .DAT file to define a schedule would be nice. >-L > It would have to be able to handle oddball rules like "Every Tuesday after! > the last Friday of each month."j >:L > I have a long and complicated DCL that currently runs my  reporting and is > becoming hard to manage. >o > Any suggestions? >  > Mark-Simon Popey > Bristol Aerospace Limited  > Winnipeg, Manitoba.e >i >l >l   ------------------------------  # Date: Tue, 02 Jan 2001 19:31:47 GMT 3 From: Tim Llewellyn <tim.llewellyn@cableinet.co.uk>o/ Subject: Re: Sending messages to an application / Message-ID: <3A522C12.2D665400@cableinet.co.uk>    Howard S Shubs wrote:  > G > In article <3a4f28ac$0$36815$272ea4a1@news.execpc.com>, Jack Patteeuwa! > <jjpatteeuw@voyager.net> wrote:  >  > >"Richard B. Gilbert" wrote: > >>M > >>         Offhand, I can think of a couple of ways to send a message to an O > >> application:  Common Event Flags, Mailboxes, Global sections!  Ok, I lied;4K > >> that's three.  Of course the application must be written to  Check its0F > >> Common Event Flags, Mailboxes, etc, and do something in response. > > G > >Don't forget Logical Names (obviously stored is a shared name table)h > ' > Four, four ways.  Then there's files.s   1 DECNET too, unless you count that as mailbox use.p   ------------------------------   Date: 2 Jan 2001 08:16:56 -0500y, From: koehler@eisner.decus.org (Bob Koehler)2 Subject: Re: Source Code to Unsupported EDT Editor+ Message-ID: <DgFfDxpRVNMS@eisner.decus.org>   v In article <EF2FC7AE84D716B6.F9FA0756E767E3A1.70BE0CA9C91EF815@lp.airnews.net>, "Headman" <headman@ticnet.com> writes:L > Can anyone direct me to the source code for the unsupported EDT editor for > AXP?  D The source code is probably on the source CD, but EDT is a supported editor.m  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporatione= NASA GSFC Flight Software       | Federal Sector, Civil Group5E                                 | please remove ".aspm" when replying    ------------------------------   Date: 2 Jan 2001 16:57:30 GMTs2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)2 Subject: Re: Source Code to Unsupported EDT Editor5 Message-ID: <92t19q$m1$1@mailint03.im.hou.compaq.com>o  v In article <EF2FC7AE84D716B6.F9FA0756E767E3A1.70BE0CA9C91EF815@lp.airnews.net>, "Headman" <headman@ticnet.com> writes:K :Can anyone direct me to the source code for the unsupported EDT editor fora :AXP?a  G   I'll assume you mean OpenVMS Alpha...  AFAIK, the EDT source code is e7   not particularly dependent on the OpenVMS platform...e  E   EDT was constrained by the available capabilities and compatibility ,   with the RSX operating system environment.  F   There are several other editors included on the Freeware, of course.E   (Your question does not include sufficient detail on your goals ford5   any particularly meaningful alternative responses.)   E   I'll see if I can clear the release of the EDT source code, and (if.E   release is permissible) include it onto some random future OpenVMS oH   Freeware distribution.  (At present, please acquire the CD-ROM OpenVMSH   source listings kit for OpenVMS Alpha, as others have suggested -- theF   part number of the CD-ROM media kit is included in the OpenVMS FAQ.)    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 2 Jan 2001 16:26:39 -0500 3 From: "Gareth V. Williams" <graff@cfa0.harvard.edu>6 Subject: SSH2 under VMS?, Message-ID: <3a52478f.0@cfanews.harvard.edu>  J Is anyone aware of a VMS port of ssh-2.4.0 or openssh-2.3.0p1?  Both these8 packages implement SSH2.  A web search turns up nothing.  H I am aware of the separate SSH1 client/server packages, but these do notJ include sftp.  Our Computation Facility is disabling Telnet and FTP at theF end of this month and requiring all connections to be made via ssh andG sftp.  Since I am unaware of a SSH2 implementation for VMS, that leaves-9 my group up a certain creek without a certain instrument.h  C I would prefer official support for SSH2 to be built into TCPIP.  ItA hear rumors that this is being considered/worked on.  Any idea ofs timescales?-   -- pH ------------------------------------------------------------------------H Gareth V. Williams, MS 18, 60 Garden Street, Cambridge, MA 02138, U.S.A.+ Associate Director, IAU Minor Planet CenteroH gwilliams@cfa.harvard.edu        http://cfa-www.harvard.edu/iau/mpc.html7 OpenVMS & RISC OS: refined choices in operating systemsm   ------------------------------   Date: 2 Jan 2001 14:04:00 PSTPT From: Fairfield@SLAC.Stanford.EDU (Ken Fairfield; SLAC: 650-926-2924; FAX: 926-3515) Subject: Re: SSH2 under VMS?3 Message-ID: <dNNxhia9+Mv$@mccdev.slac.stanford.edu>   - In article <3a52478f.0@cfanews.harvard.edu>, w:     	"Gareth V. Williams" <graff@cfa0.harvard.edu> writes:L > Is anyone aware of a VMS port of ssh-2.4.0 or openssh-2.3.0p1?  Both these: > packages implement SSH2.  A web search turns up nothing. > J > I am aware of the separate SSH1 client/server packages, but these do notL > include sftp.  Our Computation Facility is disabling Telnet and FTP at theH > end of this month and requiring all connections to be made via ssh andI > sftp.  Since I am unaware of a SSH2 implementation for VMS, that leaves9; > my group up a certain creek without a certain instrument.9 > E > I would prefer official support for SSH2 to be built into TCPIP.  IrC > hear rumors that this is being considered/worked on.  Any idea ofa
 > timescales?   H         You are correct.  At  present,  none  of the VMS implementationsH     support  SSH2.  Neither FISH (no further development, apparently, asH     Richard Levitte is busy with  OpenSSL)  nor  David  Jones'  OSU  SSHH     Server  support  SSH2,  and David has stated in the past that he wasH     not interested (at least at that time) in adding SSH2 support as the-     specifications were changing too rapidly.e  H         In addition, Process Software  has  ported  the Data Fellows SSHH     code   to   Multinet  4.3,  both  client  and  daemon,   but   firstH     implementation is SSH 1 only.  Process's stated roadmap  shows  themH     doing  SSH in TCPware next, probably including the SSH2 daemon, thenH     porting that  piece  back  to  Multinet.   Contact  Process for time     scales.   H         All in all, there's a paddle around  the next bend in the creek,,     but you'll need to be patient for now...  H         Also, it would behoove you to let your Computation Facility knowH     your concerns.  FWIW, our site went  to  SSH-only  a bit over a yearH     ago.   But  there  are so few clients (e.g., on PCs and  Macs)  thatH     support reasonable FTP over SSH, that we still allow FTP within  theH     site.   SSH2 was _not_ ready, apparently, when we switched over, andH     I've not heard of any big  push  to get there.  Not that things likeH     `scp'  aren't  tricky  in their own right.  In  particular,  scp  isH     implemented as sort of a script on unix, and is, according to  DavidH     Jones,  pretty ugly code.  One of the worst things about it is that,H     like cp, it makes assumptions  regarding text versus "binary" files.H     I  don't  know  anything  about sftp, but I do  hope  it  gets  that     distinction right.           -Ken -- aM  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:      650-926-3515N  -----------------------------------------------------------------------------B  These opinions are mine, not SLAC's, Stanford's, nor the DOE's...   ------------------------------  % Date: Tue, 02 Jan 2001 14:46:00 -0700-% From: Dan O'Reilly <dano@process.com>n Subject: Re: SSH2 under VMS?A Message-ID: <5.0.2.1.2.20010102144508.00ae2ca0@ntbsod.psccos.com>.  I SSH1 is currently available in Multinet 4.3.  It will be available in theeM next release of TCPware.  SSH2 will be available in both Multinet and TCPwarea shortly afterward.  / At 02:26 PM 1/2/2001, Gareth V. Williams wrote:tK >Is anyone aware of a VMS port of ssh-2.4.0 or openssh-2.3.0p1?  Both theser9 >packages implement SSH2.  A web search turns up nothing.p >.I >I am aware of the separate SSH1 client/server packages, but these do notaK >include sftp.  Our Computation Facility is disabling Telnet and FTP at thenG >end of this month and requiring all connections to be made via ssh andhH >sftp.  Since I am unaware of a SSH2 implementation for VMS, that leaves: >my group up a certain creek without a certain instrument. >aD >I would prefer official support for SSH2 to be built into TCPIP.  IB >hear rumors that this is being considered/worked on.  Any idea of >timescales? >  >--pI >------------------------------------------------------------------------ I >Gareth V. Williams, MS 18, 60 Garden Street, Cambridge, MA 02138, U.S.A.e, >Associate Director, IAU Minor Planet CenterI >gwilliams@cfa.harvard.edu        http://cfa-www.harvard.edu/iau/mpc.htmls8 >OpenVMS & RISC OS: refined choices in operating systems   ------I +-------------------------------+---------------------------------------+pI | Dan O'Reilly                  |                                       |nI | Principal Engineer            |  "Those are my principles. If you     | I | Process Software              |   don't like them I have others."     | I | http://www.process.com        |                    -- Groucho Marx    |hI +-------------------------------+---------------------------------------+k   ------------------------------  $ Date: Tue, 2 Jan 2001 13:48:24 +0100% From: "Fred Zwarts" <F.Zwarts@KVI.nl> 0 Subject: Suggestion for improvement for DECterm.- Message-ID: <92simp$ai$1@info.service.rug.nl>   E I would like an small improvement for DECterm, to make it easier to =' locate the user.E It would be a lot easier if the ACCPORNAM attribute of the FTA deviceK@ would be defined as the properties of the corresponding DISPLAY.@ E.g.:    "TCPIP/KVIX04.KVI.NL:0.0", or "DECNET/KVIX01:0.0", or = "LAT/KVIX02:0.0".mH In this way it would be much easier to locate a user, in particular if = he logged in=20t. with XDM from a (possibly distant) X-terminal.   ------------------------------   Date: 2 Jan 2001 17:14:15 GMT-2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)4 Subject: Re: Suggestion for improvement for DECterm.5 Message-ID: <92t297$sr$1@mailint03.im.hou.compaq.com>   U In article <92simp$ai$1@info.service.rug.nl>, "Fred Zwarts" <F.Zwarts@KVI.nl> writes:,F :I would like an small improvement for DECterm, to make it easier to = :locate the user.c :oF :It would be a lot easier if the ACCPORNAM attribute of the FTA deviceA :would be defined as the properties of the corresponding DISPLAY.l? :E.g.:    "TCPIP/KVIX04.KVI.NL:0.0", or "DECNET/KVIX01:0.0", or- :"LAT/KVIX02:0.0". :iG :In this way it would be much easier to locate a user, in particular ift< :he logged in with XDM from a (possibly distant) X-terminal.  M   This isn't specifically DECterm involved here, as ACCPORNAM is a construct IK   of the underlying terminal support and operates across most any terminal hK   device, and as the X Windows display specification is a construct of the  L   X Windows transport in use and operates across any DECwindows application.  H   If you have the DECwindows WSAn: workstation pseudo-device name -- theJ   device name can usually be acquired via the DECW$DISPLAY logical name --M   you can acquire the transport information.  SHOW DISPLAY [/SYMBOL][/FULL], hL   as well as the undocumented $qio interface -- examples of both are in the L   SRH_EXAMPLES area of the OpenVMS Freeware website.  The WSAu: workstation B   pseudo-device probably contains the information that you seek...  L   I'm somewhat hesitant to accept changes to the ACCPORNAM display, as thereJ   are folks that have already had application problems due to changes (or /   the errant omission of data) in that field...e  +   If I understand the question correctly...    	--l  J   I'd have to see how this ties in with XDM that is present in the TCP/IP K   Services field test, but I'd expect it to use the DECW$DISPLAY mechanism.     N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 2 Jan 2001 09:39:24 -0500E* From: young_r@eisner.decus.org (Rob Young)> Subject: TRAM and writebehind (was: Re: Giving up on the Best)+ Message-ID: <ofuh1HDS0MHe@eisner.decus.org>u  N In article <VA.00000201.25384c70@sture.ch>, Paul Sture <paul@sture.ch> writes:  * > If you didn't read the complete article I > http://www.daemonnews.org/200012/tram.html , I think you might find it eL > interesting. The author, Dru Nelson, goes on to propose Transactional-RAM 3 > (TRAM) as a method for addressing these problems.g > N > It made sense to me, although I haven't got my head around all the what-ifs  > yet. >   > 	I see it as a low-end solution also as the author talks about; 	32 MByte and $300 cost.  One thing to consider, what if itn$ 	blows out?  There goes your writes. 	aD 	I'm not sure if there are any way around this in a mission-criticalB 	Enterprise environment short of the writes hitting battery backed@ 	controller cache and even at that, working with shadowsets such: 	that shadowset members are on different controller pairs.  > 	Maybe.... what if you had a son-of-Wildfire box with multiple 	Galaxy instances such that:  5 			1)  Sucker has a nice sized battery for 10 minutes  			    of uptime v   			2) Locally attached storage a   	A write does a few things:   & 			1) Writes to writeback memory cache  4 			2) Writes to remote Galaxy member writeback cache  2 			3) Flushed to disk and coordination handled via
 				SCS/locksc  = 	Essentially, the write exists in two places at once and then A 	application program rolls on.  If one node crashes, writes stille@ 	exist in another Galaxy member.  Power lost, writes are flushed 	to locally attached storage.   	 	and on..e  A 	We know a VCC WRITEBEHIND sysgen parameter is on the horizon butn> 	one wonders how you use it with much confidence if you didn't0 	have a scheme similar or better than the above.   				Robn   ------------------------------   Date: 2 Jan 2001 15:32:43 GMTu9 From: Roar =?iso-8859-1?Q?Thron=E6s?= <roart@nvg.ntnu.no>-$ Subject: vms software administration3 Message-ID: <978449793.757827@tornado.itea.ntnu.no>p   Hi   Happy new year.a  A After a number of software installations (and especially the onesg= which is neither pcsi nor saveset based), the sylogin.com andoE systartup_vms.com files is starting to become a bit large and chaotic  with internal dependensies.S3 Also, keeping track of versions etc is also needed.,  G How is the availability of utilities capable of handling such software,aL including generating scripts to be run by sylogin.com and systartup_vms.com?   I have not found any such.  F I am in part underway of making a VMS port of a (perl) utility in Unix8 called Store. (Works good enough to put to ordinary use)7 (See http://www.pvv.ntnu.no/~arnej/store/storedoc.html)T VMS differences from that doc:  The use of .s and @s is changed.@ The linktree concept is not useful (even with DCL$PATH, I think)8 Generation of scripts for sylogin.com and systartup.com.   How does this utility sound?   --  
 -Roar Throns-   ------------------------------   Date: 2 Jan 2001 17:22:39 GMT02 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)( Subject: Re: vms software administration5 Message-ID: <92t2ov$sr$2@mailint03.im.hou.compaq.com>   o In article <978449793.757827@tornado.itea.ntnu.no>, Roar =?iso-8859-1?Q?Thron=E6s?= <roart@nvg.ntnu.no> writes:d  B :After a number of software installations (and especially the ones> :which is neither pcsi nor saveset based), the sylogin.com andF :systartup_vms.com files is starting to become a bit large and chaotic :with internal dependensies.4 :Also, keeping track of versions etc is also needed.  F   We tried that, and the interdependencies and interrelationships are ,   exceedingly complex problems to resolve...  G   You can find some of the underpinnings of the work that was performed ?   in this area in VMSINSTAL and in the SYSMAN startup database.t  H :How is the availability of utilities capable of handling such software,M :including generating scripts to be run by sylogin.com and systartup_vms.com?n  G   Generation of scripts is trivial for most cases, it's the sequencing oE   of the scripts that is the central problem...  (I've dealt with and E   developed a healthy respect for dependency analysis as part of the o9   OpenVMS Engineering VDE source code control package...)-   :I have not found any such.t  A   What would you have happen?  How would you expect this to work?5  G :I am in part underway of making a VMS port of a (perl) utility in Unixe: :called Store. :(Works good enough to put to ordinary use)  )   And therein lies the central problem...n  8 :(See http://www.pvv.ntnu.no/~arnej/store/storedoc.html) :VMS differences from that doc: ! :The use of .s and @s is changed.sA :The linktree concept is not useful (even with DCL$PATH, I think) 9 :Generation of scripts for sylogin.com and systartup.com.  :D :How does this utility sound?o  J   We have PCSI, VMSINSTAL, and VMSUPDATE...  As part of the DII COE work, K   we are  adding the COE segment installer...  I'd certainly be interested -J   in acquiring a port of this STORE tool for the Freeware CD-ROM, but I'd L   personally look to something like the COE segment installer for this task.    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------   Date: 2 Jan 2001 15:03:42 GMTDF From: lederman@star.enet.dec.DISABLE-JUNK-EMAIL.com (Bart Z. Lederman)& Subject: Re: VR241 monitor and S-video* Message-ID: <92sqke$b8d@usenet.pa.dec.com>  A S-Video is used on (and sometimes confused with) S-VHS recorders,n. but is now also used on other video equipment.  @ It uses a DIN connector that has separate channels for luminance= and chroma (or one cable for the "black and white" portion of-? the video signal and one for the "color" portion of the signal.oA The idea was that since some video equipment like video recorders > and DVD have the two signals recorded separately, there was no> reason to mix them together into a composite video signal, andA then have the TV set have to separate them back out into separatea; components again.  You generally get better picture qualityi= through S-video than through composite video, but not as goodA as component (3 cable) video.   ; I have only seen S-video used on "TV" type equipment.  I've ? never seen a computer monitor that will accept S-video as input A or output.  I suppose you could take the 'black and white' signal = from an S-video cable and feed it into the green channel of as= monitor, or into the only input on a black-and-white monitor,n4 or you could build something that would feed it into? all three channels of a color monitor, but I doubt very much if A it would be worth the effort.  And you would have to build or buyO> a color decoder to get a color signal from S-video into RGB to= get a color picture on a color monitor.  It would probably be = cheaper and would certainly be easier to buy a TV set with an < S-Video input than to set up a converter to a color monitor.   -- -(  B. Z. Lederman   Personal Opinions Only  8  Posting to a News group does NOT give anyone permission8  to send me advertising by E-mail or put me on a mailing  list of any kind.  5  Please remove the "DISABLE-JUNK-EMAIL" if you have a05  legitimate reason to E-mail a response to this post.3   ------------------------------  % Date: Tue, 02 Jan 2001 11:41:01 -06000* From: Brent Cook <busterb@mail.utexas.edu>& Subject: Re: VR241 monitor and S-video/ Message-ID: <3A5212AD.B399AD25@mail.utexas.edu>s   "Bart Z. Lederman" wrote:n  = > I have only seen S-video used on "TV" type equipment.  I'venA > never seen a computer monitor that will accept S-video as input  > or output.  G The Commodore C1702 Monitors had luminance and chroma inputs that could-E probably be adapted to svideo, but I guess that's not really what you  asked.   ------------------------------  % Date: Tue, 02 Jan 2001 15:11:51 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca>i& Subject: Re: VR241 monitor and S-video, Message-ID: <3A523601.DBA15E23@videotron.ca>   "Bart Z. Lederman" wrote:SB > It uses a DIN connector that has separate channels for luminance? > and chroma (or one cable for the "black and white" portion oftA > the video signal and one for the "color" portion of the signal.o  J Thanks (and to the others who contributed).  So it makes my S-video outputK rather useless for the RGB VR241 monitor. I was hoping it S-vieo would have-L had R B G and sync signals that would have been perfect for that monitor :-)    7 > I have only seen S-video used on "TV" type equipment.e  L Yes. And the video card I have on the mac is a a TV type card (with NTSC andK PAL input/output). Not designed to output the contents of the video screen. ) (for that there is a VGA and MAC outputs)r   ------------------------------  $ Date: Tue, 2 Jan 2001 14:00:26 -05005 From: "Fred Kleinsorge" <kleinsorge@star.zko.dec.com>F) Subject: Re: XP1000 - which Graphics Card , Message-ID: <92t8ig$7g2v$1@lead.zk3.dec.com>  J Hmmm.  That's news to me.  NVidea is/was taking over 3DFX (although I seemI to remember there being some problems there.  As far as I know, 3DLabs isnK alive and well, and not part of nVidea.  I even went to CNNFN.COM to see ifC I missed something.r  7 Maybe you are confusing 3DFX (TDFX) with 3DLabs (TDDD)?       G Paul Repacholi wrote in message <87ofy5jxqx.fsf@k9.prep.synonet.com>...e) >"Mike Burch" <mgburch@smart.net> writes:( >fI >> time of their release and I doubt that 3DLabs will stay in this marketS+ >> (which is dominated by Nvidia and 3DFX).  >p4 >As Nvidia have swallowed 3DLabs, that's a safe bet. >  >And a few more options gone.  >t >--)= >Paul Repacholi                               1 Crescent Rd.,g8 >+61 (08) 9257-1001                           Kalamunda.A >                                             West Australia 6076o/ >Raw, Cooked or Well-done, it's all half baked.m   ------------------------------   End of INFO-VAX 2001.005 ************************