1 INFO-VAX	Fri, 26 Oct 2001	Volume 2001 : Issue 595       Contents:0 Re: Another DSSI VAX cluster connection question/ Anyone know of an ASCII-based mailmerge program 3 Re: Anyone know of an ASCII-based mailmerge program 3 Re: Anyone know of an ASCII-based mailmerge program 3 Re: Anyone know of an ASCII-based mailmerge program 3 Re: assaults on web servers - multinet IP filtering / Re: assaults on web servers from the billyworld / Re: assaults on web servers from the billyworld / Re: assaults on web servers from the billyworld / Re: assaults on web servers from the billyworld / Re: assaults on web servers from the billyworld / Re: assaults on web servers from the billyworld / Re: assaults on web servers from the billyworld  Re: BMQ on Itanium Re: BMQ on Itanium1 Re: CDD RDB + Datatrieve - what license do I need  Re: Compaq financial humour ! $ Re: Compaq: VMS is alive and kicking% Re: Controller-based Mirror Set Usage ! Re: Delta time between two date's  Re: DSSI VAX cluster questions Re: DSSI VAX cluster questions2 Re: Errors on RTA-ports in OpenVms version 5.5-2 ?2 Re: Errors on RTA-ports in OpenVms version 5.5-2 ?6 Re: Finding GNU "find" that builds from source for VMS7 Re: Future Programming Platforms - Your Opinions Wanted 7 RE: Future Programming Platforms - Your Opinions Wanted 7 Re: Future Programming Platforms - Your Opinions Wanted  How to configure decnet? Re: I am changing jobs3 Re: KNOW WHEN A SUBMIT JOB IS FINISHED RETURN VALUE 3 Re: KNOW WHEN A SUBMIT JOB IS FINISHED RETURN VALUE 3 Re: KNOW WHEN A SUBMIT JOB IS FINISHED RETURN VALUE 3 Re: KNOW WHEN A SUBMIT JOB IS FINISHED RETURN VALUE ! Looking for a bit of ACL/ACE help % Re: Looking for a bit of ACL/ACE help % Re: Looking for a bit of ACL/ACE help   Re: MIME Compliant Mail from VMS  Re: MIME Compliant Mail from VMS  Re: MIME Compliant Mail from VMS* Re: More official info on Compaq/HP merger On hanging processes On hanging processes, Re: Optimizing indexed file while adding key, Re: Optimizing indexed file while adding key, Re: Optimizing indexed file while adding key- Re: Spiralog (was Re: BCC support in VMSMAIL)  Re: SSH for Alpha VMS % Re: SYS$EXAMPLES:DAYLIGHT_SAVINGS.COM % Re: SYS$EXAMPLES:DAYLIGHT_SAVINGS.COM 1 Where to find the DCL command procedure document? 5 Re: Where to find the DCL command procedure document? # Re: Windows XP reality check please   F ----------------------------------------------------------------------  # Date: Thu, 25 Oct 2001 18:20:27 GMT " From: Alfred Falk <falk@arc.ab.ca>9 Subject: Re: Another DSSI VAX cluster connection question 9 Message-ID: <Xns91457D88B78E9falkarcabca@205.233.108.180>   > "Brian Tillman" <tillman_brian@notnoone.notnohow.com> wrote in news:3bd83d07$1@news.si.com:    F > I have a cluster whose two boot nodes are VAX 4000 Model 600As.  TheF > disks are attached via a pair of HSD30 controllers, not in redundantF > configuration.  I have two shelves of RZ28/RZ29s each controller and. > the third SCSI port is reserved for a TSZ07. > I > I have inherited another shelf of RZ29s that I'd like to use, but since A > both HSDs are tapped out, I believe I need to buy another.  The I > question I have is: can I put another DSSI node in this configuration?  & > How may HSD controllers are allowed?  M A single DSSI bus can have 8 nodes.  You have at least 4.  More if there are  = internal DSSI disks in the VAXes.  Simple check on this rule.   H A DSSI bus can be something like 85' long.  Unfortunately standard DSSI M cables are fairly long, and "internal" cabling counts for quite a bit, too.   I Typically you are limited to 5 "cabinets", so you're likely okay on this  
 rule, too.  ! I have a configuration like this:      	AlphaServer 800     	VAX 4000/100A
     	HSD05
     	HSD05     	AlphaServer DS20 L This works perfectly.  I once tried adding another HSD05. "Almost" worked.  H The AlphaServers at each end of the bus could "see" everything but each ; other at the >>> prompt.  Horrible results on boot attempt. @ ----------------------------------------------------------------A   A L B E R T A         Alfred Falk               falk@arc.ab.ca  @ R E S E A R C H         Information Systems Dept   (780)450-5185+   C O U N C I L         250 Karl Clark Road 1                         Edmonton, Alberta, Canada  http://www.arc.ab.ca/   T6N 1E4   http://www.arc.ab.ca/staff/falk/   ------------------------------  # Date: Thu, 25 Oct 2001 22:58:51 GMT L From: winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")8 Subject: Anyone know of an ASCII-based mailmerge program8 Message-ID: <00A040F6.672C95DB@SSRL04.SLAC.STANFORD.EDU>  	 VMSers --   ) Trying to save myself a little work here.   P I have an Rdb database residing on VMS (7.2-1, though it doesn't matter  much). E We've done mailmerge using Word-11 list processing; basically I  have M Datatrieve select the specific data of interest and then write a file full of O logical records ("logical" to  distinguish it from the RMS sense, in which each " line is a record)  that looks like  
 <startrecord>  <fieldname1>value  <fieldname2>value  <fieldname3>value  <endrecord>   I and I can then import this into Word-11 and do a mailmerge against a file  that looks like  ---  Dear <fieldname1>,  * Your <fieldname2> is full of <fieldname3>.  
 Sincerely,   My Name  ---   M and produce a bunch of pages with customized bodies.  The virtue of this was, K back when our users still remembered how to edit documents in Word-11, that N the users controlled the text and I didn't have to recode everything when they wanted to change a line.    I Now instead of paper we'd like to produce email.  I have in the past made ( Datatrieve write a file that looked like ----J $ mail sys$input actual-email-address /SUBJECT="some kind of subject line"   Dear the-actual-user-name,  A text and here's a value substituted in and oh here's another one.    Yours    My Name J $ mail sys$input actual-email-address /SUBJECT="some kind of subject line"   Dear the-actual-user-name,  A text and here's a value substituted in and oh here's another one.    Yours    My Name  $  -----   M Which is easy enough to do, but tends to bury the message text in the script, M and the users are extremely likely to break the Datatrieve script if they try G to edit it to change the text, and any major changes really mess it up.     M So what I'd really like is some kind of macro processor that can do tag-based M substitution into a template file from another file, in a loop where it reads L completely through the second file, working entirely in ASCII and running on VMS.  For free, please.   O Users seem to be able to understand tag-based substitution (thus the popularity  of PHP).  O I can readily change my input file to it to look more like SGML, if that helps;    <record> <field1> value </field1> <field2> value </field2> <field3> value </field3>
 </record>   L This seems like somewhere between one and five day's work in PERL or Python,L but it also seemed like the sort of thing that somebody would probably have M come up with already.  (And of course it would be easy to get distracted into L tweaking it forever, like being able to do arithmetic in tags or callouts toL application programs or whatever, and then you could play with it for a longN time.  At present I really just need to substitute field values and allow text wrapping.)    J Has anybody out there got one of these to share?  I've got Perl 5.6.1, canO install Python 2, have C, C++, Fortran, Pascal (I'm on CSLG and have large disk A drives, so could even install Cobol if needed to compile a tool).   3 Or is there some way to coerce PHP into doing this?    Thanks,    -- Alan     O =============================================================================== 0  Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDUM  Disclaimer: I speak only for myself, not SLAC or SSRL   Phone:  650/926-3056 M  Physical mail to: SSRL -- SLAC BIN 69, PO BOX 4349, STANFORD, CA  94309-0210 O ===============================================================================    ------------------------------  % Date: Thu, 25 Oct 2001 22:26:20 -0400 - From: JF Mezei <jfmezei.spamnot@videotron.ca> < Subject: Re: Anyone know of an ASCII-based mailmerge program, Message-ID: <3BD8C9BB.F25BD739@videotron.ca>  * Alan Winston - SSRL Admin Cmptg Mgr wrote:K > Now instead of paper we'd like to produce email.  I have in the past made * > Datatrieve write a file that looked like > ----L > $ mail sys$input actual-email-address /SUBJECT="some kind of subject line"  N You may wish to look into ALL-IN-1 (office server).  It has powerful scriptingI and text processing capabilities for text files and can produce emails or / generate a file and send that file via VMSmail.   J Not sure i they still sell it, but it used to be available as a single useJ licence that was very affordable and works fine for such purposes. It alsoM gives you access to SYSUAF.DAT to merge information into your emails etc etc.    ------------------------------  % Date: Thu, 25 Oct 2001 22:41:22 -0500 C From: "Craig A. Berry" <craig.berry@nospam.SignalTreeSolutions.com> < Subject: Re: Anyone know of an ASCII-based mailmerge programI Message-ID: <craig.berry-3DF838.22412225102001@newsrump.sjc.telocity.net>   8 In article <00A040F6.672C95DB@SSRL04.SLAC.STANFORD.EDU>,H  winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr")   wrote:   O > So what I'd really like is some kind of macro processor that can do tag-based O > substitution into a template file from another file, in a loop where it reads N > completely through the second file, working entirely in ASCII and running on > VMS.  For free, please.   N > This seems like somewhere between one and five day's work in PERL or Python,N > but it also seemed like the sort of thing that somebody would probably have  > come up with already.     C It may not be that much work if you use existing extensions.  I've  D never used these myself, but the Perl modules Mail::Bulkmail and/or  Text::Merge may be of use:  5 <http://search.cpan.org/search?module=Mail::Bulkmail> / <http://search.cpan.org/search?dist=Text-Merge>    ------------------------------   Date: 26 Oct 2001 03:46:01 GMT5 From: ccburgess@uqstu.jdstory.uq.edu.au (Ian Burgess) < Subject: Re: Anyone know of an ASCII-based mailmerge program. Message-ID: <9ram9p$1ft$1@bunyip.cc.uq.edu.au>   In article <00A040F6.672C95DB@SSRL04.SLAC.STANFORD.EDU>, winston@SSRL.SLAC.STANFORD.EDU ("Alan Winston - SSRL Admin Cmptg Mgr") writes: 
 >VMSers -- > * >Trying to save myself a little work here. > Q >I have an Rdb database residing on VMS (7.2-1, though it doesn't matter  much).  F >We've done mailmerge using Word-11 list processing; basically I  haveN >Datatrieve select the specific data of interest and then write a file full ofP >logical records ("logical" to  distinguish it from the RMS sense, in which each# >line is a record)  that looks like  >  ><startrecord> ><fieldname1>value ><fieldname2>value ><fieldname3>value ><endrecord> > J >and I can then import this into Word-11 and do a mailmerge against a file >that looks like >--- >Dear <fieldname1>,  > + >Your <fieldname2> is full of <fieldname3>.  >  >Sincerely,  >  >My Name >--- .  .  .    >Thanks, >-- AlanP >===============================================================================1 > Alan Winston --- WINSTON@SSRL.SLAC.STANFORD.EDU P >===============================================================================  L Just remove <startrecord> and replace <endrecord> with a Form-Feed character: and you can use this routine I wrote in Teco 10 years ago.   DCL procedure:& ______________________________________' $ !   Define template template_filespec : $ !   define data data_filespec  (pages separated by <ff>) $ ! $ $      on error then goto teco_errorO $      edit/teco/nocom/nomem/execute=former.tec   ! a fancy form of MUNG FORMER  $      print full.txt  $ exit   $Teco_error:/ $      write sys$output "Something went wrong!"  $ exit& ______________________________________  3 TECO program:  former.tec -- no special characters. & ______________________________________ @er|TEMPLATE|<:a;>hx1hk  @ew|FULL.TXT|@er|DATA|ey/ !loop:!j<@s|<|;r@i|j<@fs\|@s|>|@i|\|l2r@i|\;>|> 	 hxmhkg1mm  :p"SOloop:' ex& ______________________________________     TEMPLATE file:& ______________________________________ Dear <title> <surname>,   ? Your <folder> folder is full, <nfiles> files, <nblocks> blocks. & ______________________________________  
 DATA file:& ______________________________________ <title>Miss  <surname>Doolittle <folder>wastebasket  <nfiles>1234 <nblocks>121334   ! [insert form-feed character here]   	 <title>Mr  <surname>Bloggs  <folder>big rubbish bin 
 <nfiles>12 <nblocks>12134& ______________________________________
            FULL.TXT output file: & ______________________________________ Dear Miss Doolittle,  ; Your wastebasket folder is full, 1234 files, 121334 blocks.      <FF> Dear Mr Bloggs,   < Your big rubbish bin folder is full, 12 files, 12134 blocks.    & ______________________________________   Explanation of TECO program & ______________________________________ !start:!, !2eh?<esc><esc> at this point for debugging!C ! Try adding 4ed to suppress memory expansion if that is a problem!  !Read in template!= @er|TEMPLATE|<:a;>         !append until the whole file read! 0 hx1  hk                    !put it into q-reg 1!1 @ew|FULL.TXT|              !open the output file! : @er|DATA|ey                !read first data set from file!  !loop:! M j<@s|<|;r@i|j<@fs\|@s|>|@i|\|l2r@i|\;>|>  ! make, e.g., j<@fs\<TITLE>\Mr\;> ! M @i||@i|| hxm hk          !put the generated macro into q-reg m, wipe slate! - g1                         !get the template! ; mm                         !do all parameter substitutions! J :p"SOloop:'               !write page, read new details, loop if success! ex7 ! <parameter> text is obtained from the details file. ! > !TEMPLATE contains any number of parameters, as "<parameter>"! !excluding the quotes.  e.g., !  !  Dear <TITLE> <SURNAME>,    ! 1 !DATA has a page per letter (ends with a <FF>), ! L !each line of which replaces a <parameter> parameter whereever it occurs in!J !the template. <TITLE> (for example) may occur any number of times in the! !template.   e.g.      ! !<TITLE>Miss                  !  !<FIRSTNAME>Sally             !  !<SURNAME>Brown               !  ! !  !  This algorithm is:!% ! read template, store it in q-reg 1!  ! open the output file! G ! open the input file and read the first page (details for one letter)!  !loop to here!H ! make the page into a TECO macro to edit the template for this letter.! !   find end-of-line,!L !   generate a "replace <parameter> by ... for all occurrences in template"!( !      where ... is this line contents.!/ !   repeat for each line in the details page. ! 6 ! put the macro into q-reg m, wipe the working buffer! ! get the template!  ! perform the macro in q-reg m! + ! output the letter and append a form-feed! > ! if there are more details to read, read and loop, else exit!& ______________________________________    B This bit of Teco has produced thousands of letters over the years.   Enjoy.     Ian Burgess  University of Queensland I.Burgess[at]its.uq.edu.au www.its.uq.edu.au    ------------------------------  % Date: Fri, 26 Oct 2001 00:30:46 -0400 0 From: "Jeff Morgan" <vmswiz@geonospamcities.com>< Subject: Re: assaults on web servers - multinet IP filtering- Message-ID: <9rap0m$26pj$1@news3.infoave.net>   I FYI, if you are running Multinet, you can create an IP filter list on the F network interface that will block all connections from the IP address.  E I just figured out how to do it today and it may be useful to some...    Create a file:   MULTINET:FILTER-SE1.DAT & drop ip 208.165.181.11 255.255.255.255& drop ip 208.165.228.69 255.255.255.255& drop ip 64.210.39.253  255.255.255.255& drop ip 208.180.241.43 255.255.255.255 permit ip 0.0.0.0 0.0.0.0    And activate it with:   9 MULTINET SET/INTERFACE/FILTER=MULTINET:FILTER-SE1.DAT SE1     Your interface may be different.  I This is documented in the administrator guide, but it wasn't intuitive on  first or second read to me.   I My OSU access log is clean now. Of course you have to manually add all of D the attacking IP addresses to the list. And watch for new addresses.  . If my filter is incorrect, please let me know.   ------------------------------   Date: 25 Oct 2001 13:15:45 CDT= From: wayne@tachysoft.xxx.062469.killspam.00be (Wayne Sewell)o8 Subject: Re: assaults on web servers from the billyworld. Message-ID: <c34HlZVW4u3h@tachxxsoftxxconsult>  ` In article <6kWB7.958$RL6.10455@news.cpqcorp.net>, "Rick Barry" <barry@star.zko.dec.com> writes:7 >> 'GET' '/c/winnt/system32/cmd.exe?/c+dir', (HTTP/1.0)  > etc. > L > As mentioned by previous posters, these entries indicate attempted attacks? > on port 80, looking for a vulnerable IIS server to infect andrM > self-propogate. For OpenVMS-based web servers, the results of these attacksaN > range from annoynance to denial-of-service to legitimate clients. The worms'N > payload are inaffective on OpenVMS, but they consume connections which are aI > limited resource and also consume disk space as a result of hundreds orq1 > thousands of entries in access/error log files.n > N > It does more harm than good to configure the server to return data for theseN > "requests" because the worms do nothing with it and it only consumes network > bandwidth.    M If they don't do anything with it, why do they ask for it?  Does the 'get' infJ itself infect the billy system?  In order to infect something, you have toM modify it in some way.  Does the 'get' do this?  I thought the purpose of theTM 'get' was to determine if infection was possible, i.e. this is a billy systemAI with a particular directory structure.  I had assumed that if the get wasiO successful, the rogue software would then do something else to cause the actual 
 infection.  G My goal in responding was to make the infecting software *think* it hadeL succeeded in finding a target so maybe it wouldn't try all the *other* 'get'O commands. If I knew what the get was supposed to return, I would send that backg instead of the Lumberjack Song.          -- kO ===============================================================================rM Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxx : http://www.tachysoft.xxx/www/tachyon.html and wayne.html  K change .xxx to .com in addresses above, assuming you are not a spambot  :-)oO ===============================================================================-H Randolph Duke (in Trading Places): "Mother always said you were greedy.". Mortimer Duke: "She meant it as a compliment!"   ------------------------------  % Date: Thu, 25 Oct 2001 15:49:39 -0400D+ From: "Rick Barry" <barry@star.zko.dec.com>n8 Subject: Re: assaults on web servers from the billyworld2 Message-ID: <X0_B7.975$RL6.10553@news.cpqcorp.net>  J "Wayne Sewell" <wayne@tachysoft.xxx.062469.killspam.00be> wrote in message( news:c34HlZVW4u3h@tachxxsoftxxconsult...A > In article <6kWB7.958$RL6.10455@news.cpqcorp.net>, "Rick Barry"o  <barry@star.zko.dec.com> writes:9 > >> 'GET' '/c/winnt/system32/cmd.exe?/c+dir', (HTTP/1.0)f > > etc. > >IF > > As mentioned by previous posters, these entries indicate attempted attacksnA > > on port 80, looking for a vulnerable IIS server to infect andtG > > self-propogate. For OpenVMS-based web servers, the results of theseD attacksaI > > range from annoynance to denial-of-service to legitimate clients. Thei worms'J > > payload are inaffective on OpenVMS, but they consume connections which are a.K > > limited resource and also consume disk space as a result of hundreds org3 > > thousands of entries in access/error log files.i > > J > > It does more harm than good to configure the server to return data for theseiH > > "requests" because the worms do nothing with it and it only consumes network  > > bandwidth. >  >hL > If they don't do anything with it, why do they ask for it?  Does the 'get' inL > itself infect the billy system?  In order to infect something, you have toK > modify it in some way.  Does the 'get' do this?  I thought the purpose of- the-H > 'get' was to determine if infection was possible, i.e. this is a billy systemK > with a particular directory structure.  I had assumed that if the get was5J > successful, the rogue software would then do something else to cause the actual > infection. >dI > My goal in responding was to make the infecting software *think* it had-H > succeeded in finding a target so maybe it wouldn't try all the *other* 'get'oL > commands. If I knew what the get was supposed to return, I would send that back! > instead of the Lumberjack Song.r >   K In the case of CODE RED II, the HTTP GET request payload takes advantage ofgJ an IIS indexing service buffer overflow flaw to execute code on the target9 (see http://www.cert.org/incident_notes/IN-2001-09.html).i  E Nimda propogates itself to IIS servers using tftp on UDP port 69 (seecI http://www.cert.org/advisories/CA-2001-26.html). It also tries to exploito the flaw used by CODE RED II.i  
 Rick Barry) Compaq Secure Web Server Development Teamt OpenVMS Engineeringa Compaq Computer Corporation    ------------------------------  % Date: Thu, 25 Oct 2001 17:48:41 -0400p- From: JF Mezei <jfmezei.spamnot@videotron.ca>t8 Subject: Re: assaults on web servers from the billyworld, Message-ID: <3BD888B8.FD12580C@videotron.ca>   Jim Becker wrote: E > My firewall drops approximately 60,000 of these connection attemptseH > per day, mostly because they're attempting to connect to systems where" > I'm not publishing web services.  I When A wants to send an HTTP request to B, doesn't it first establish thenK TCPIP connection to port 80 of B, and then send a packet of data containingh& the HTTP "GET" (or HEAD etc) request ?  N How does the firewall function to be able to prevent such a virus request from reaching your web server ?  L Does the firewall actually have a proxy web server that actually accepts theM connection requests, the HTTP request and then it does the connect request too< the actual web server to transmit the data it has received ?  M What happens when you have a POST request that contains multipart data with amI very large file ? Does the proxy/firewall store all the data into its own K buffers and connects to the real web server only once all the data has beene
 received ?   ------------------------------   Date: 25 Oct 2001 17:16:17 CDT= From: wayne@tachysoft.xxx.062469.killspam.00be (Wayne Sewell)a8 Subject: Re: assaults on web servers from the billyworld. Message-ID: <AYdMQd$Qwzjy@tachxxsoftxxconsult>  ` In article <X0_B7.975$RL6.10553@news.cpqcorp.net>, "Rick Barry" <barry@star.zko.dec.com> writes:L > "Wayne Sewell" <wayne@tachysoft.xxx.062469.killspam.00be> wrote in message* > news:c34HlZVW4u3h@tachxxsoftxxconsult... >> >>M >> If they don't do anything with it, why do they ask for it?  Does the 'get'e > inM >> itself infect the billy system?  In order to infect something, you have to L >> modify it in some way.  Does the 'get' do this?  I thought the purpose of > thetI >> 'get' was to determine if infection was possible, i.e. this is a billye > systemL >> with a particular directory structure.  I had assumed that if the get wasK >> successful, the rogue software would then do something else to cause the  > actual
 >> infection.  >>J >> My goal in responding was to make the infecting software *think* it hadI >> succeeded in finding a target so maybe it wouldn't try all the *other*r > 'get' M >> commands. If I knew what the get was supposed to return, I would send thatr > back" >> instead of the Lumberjack Song. >> > M > In the case of CODE RED II, the HTTP GET request payload takes advantage ofoL > an IIS indexing service buffer overflow flaw to execute code on the target; > (see http://www.cert.org/incident_notes/IN-2001-09.html).= > G > Nimda propogates itself to IIS servers using tftp on UDP port 69 (seeeK > http://www.cert.org/advisories/CA-2001-26.html). It also tries to exploitt > the flaw used by CODE RED II.4    O I see.  So the get *is* the actual attack then.  Maybe I will alter my redirect=L to run a command procedure instead of returning an html page.  The dcl couldN tell the tcp stack to disable access from that ip address to port 80.  My goalO all along has been to abort the *repeated* hits from the same site.   I realize70 that not much can be done about the initial hit.     -- tO =============================================================================== M Wayne Sewell, Tachyon Software Consulting  (281)812-0738  wayne@tachysoft.xxxm: http://www.tachysoft.xxx/www/tachyon.html and wayne.html  K change .xxx to .com in addresses above, assuming you are not a spambot  :-)eO ===============================================================================fH Randolph Duke (in Trading Places): "Mother always said you were greedy.". Mortimer Duke: "She meant it as a compliment!"   ------------------------------  # Date: Thu, 25 Oct 2001 23:27:02 GMTe= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)s8 Subject: Re: assaults on web servers from the billyworld0 Message-ID: <00A04113.7CCBFB16@SendSpamHere.ORG>  n In article <AYdMQd$Qwzjy@tachxxsoftxxconsult>, wayne@tachysoft.xxx.062469.killspam.00be (Wayne Sewell) writes:
 >{...snip...}g >> yH >> Nimda propogates itself to IIS servers using tftp on UDP port 69 (seeL >> http://www.cert.org/advisories/CA-2001-26.html). It also tries to exploit  >> the flaw used by CODE RED II. >o > P >I see.  So the get *is* the actual attack then.  Maybe I will alter my redirectM >to run a command procedure instead of returning an html page.  The dcl couldeO >tell the tcp stack to disable access from that ip address to port 80.  My goal P >all along has been to abort the *repeated* hits from the same site.   I realize1 >that not much can be done about the initial hit.d  J Be sure to follow-up with an invoice to billywarez, Inc for your time and I effort to eradicate yourself from problems caused by his substandard bits'2 of shit and, in particular, the turd known as IIS.   --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMd            kJ   "And of course, I'm a genius, so people are naturally drawn to my fiery I   intellect.  Their admiration overwhelms their envy!" -- Calvin & Hobbest   ------------------------------    Date: 25 Oct 2001 20:49:02 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)@8 Subject: Re: assaults on web servers from the billyworld3 Message-ID: <Zb8ci+m5Kvjy@eisner.encompasserve.org>   n In article <AYdMQd$Qwzjy@tachxxsoftxxconsult>, wayne@tachysoft.xxx.062469.killspam.00be (Wayne Sewell) writes:  Q > I see.  So the get *is* the actual attack then.  Maybe I will alter my redirectWN > to run a command procedure instead of returning an html page.  The dcl couldP > tell the tcp stack to disable access from that ip address to port 80.  My goalQ > all along has been to abort the *repeated* hits from the same site.   I realizec2 > that not much can be done about the initial hit.  D Instead of disabling access, how about programming a DECtalk to callF the phone number for the originating domain (nslookup/whois/whatever).E Have it give them a good long lecture about Microsoft.  Don't disablei. access -- do it every time an attempt is made.   ------------------------------  % Date: Thu, 25 Oct 2001 22:32:45 -0400=' From: Jim Becker <jbecker@ui.urban.org>=8 Subject: Re: assaults on web servers from the billyworld, Message-ID: <3BD8CB4D.1C8D34CD@ui.urban.org>   JF Mezei wrote:= >  > Jim Becker wrote: G > > My firewall drops approximately 60,000 of these connection attemptstJ > > per day, mostly because they're attempting to connect to systems where$ > > I'm not publishing web services. > K > When A wants to send an HTTP request to B, doesn't it first establish the M > TCPIP connection to port 80 of B, and then send a packet of data containinge( > the HTTP "GET" (or HEAD etc) request ?  E We're getting a bit OT (given that I have a non-VMS firewall), but inrA my particular firewall's toolbox I can take two basic approaches._  F The first approach -- the one that I use -- basically acts as a packetC filter. The firewall nabs the initial attempt without caring that aEC GET or whatever is to come. If the attempted connection is going to>F the correct port on one of my published web servers (and if the sourceF is not on my blacklist), the firewall allows the session. It's then upF to the web server to handle any GETs that may arrive. If the attemptedE connection is going to one of my addresses where I'm NOT publishing aAC web service, the attempt goes into the bit bucket. It's this latteraF case that occurs roughly 60,000 times per day. Corollary: My publishedB web servers are getting bunches of these attempted attacks. It's aB nuisance, but at least they're safe from these particular attacks.  P > How does the firewall function to be able to prevent such a virus request from > reaching your web server ? > N > Does the firewall actually have a proxy web server that actually accepts theO > connection requests, the HTTP request and then it does the connect request to > > the actual web server to transmit the data it has received ?  A The second approach is that I can implement a smart (rule-based),>E http-aware "security server" (basically a transparent proxy). In this F approach, the initial connection still has to be allowed or dropped asF above, but then a GET (for example) has to pass the proxy's additionalD tests. The proxy inspects the GET request for improprieties, and canC take action accordingly. This was the scenario that overburdened my C firewall, which normally doesn't break into a sweat. Inspecting allf? the HTTP flying back and forth turned out to be a CPU-intensive  problem.  O > What happens when you have a POST request that contains multipart data with a K > very large file ? Does the proxy/firewall store all the data into its own!M > buffers and connects to the real web server only once all the data has beena > received ?  ; I haven't tested this particular case, and I don't have theaF documentation ready to hand. In a similar case, the firewall can shuntE the file off to a virus scanner, so in that situation it would indeed E have to grab the whole file, send it off, wait for the thumbs up/down 3 from the virus scanner, then proceed appropriately.m   --
 Jim Becker+ The Urban Institute (http://www.urban.org/) ' Encompass (http://www.encompassus.org/) . ESILUG (http://encompasserve.org/lugs/esilug/)   ------------------------------  % Date: Thu, 25 Oct 2001 21:40:51 -0500e1 From: "David J. Dachtera" <djesys.nospam@fsi.net>f Subject: Re: BMQ on Itaniumt' Message-ID: <3BD8CD33.AF7C7E12@fsi.net>n   Bruce Stephens wrote:h >  > Hi folks,i > A > Anyone in this forum still using BEA's MessageQ product on VMS?  > , > Would like to see it supported on Itanium? > 6 > Now is the time to hold up your hand and be counted. > = > You can reply here, to me directly, or to the person below.i > 	 > Thanks,e > Brucep >  > Cindy McDonnell Feinberg > BEA Systems, Inc.e
 > www.bea.come > Global Account Manager > Compaq Computer Corporationa > Phone: 908-580-3214  > Fax: 908-580-3050_ > Mobile: 610-360-9733 > email: cindy.feinberg@bea.coma   Hello, Bruce and Co.,e  D If you follow comp.os.vms, you know that a large part of the OpenVMS= battle cry includes, "We need app.'s!" and, "We need tools!".-  A So, the answer to your question is both obvious and simple: if itvE currently runs on either OpenVMS-VAX or OpenVMS-Alpha, your customersi# will need it to run on OpenVMS-IPF.$  G Don't ask, just port - everything! Your stockholders will thank you fora: having the vision to plan ahead and take reasonable risks.  H We can't expect Compaq or HP to have faith in OpenVMS if the ISVs don't.  G No app.'s + no tools = no ISVs = no customers = no VMS. In spite of the-G way everyone wants to overcomplicate things (market cap.'s, ROI, etc.),2 it really is that simple.2   -- 5 David J. Dachtera  dba DJE Systems4 http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/F   ------------------------------    Date: 25 Oct 2001 22:25:54 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)d Subject: Re: BMQ on Itanium 3 Message-ID: <q$vfZDzXmXpB@eisner.encompasserve.org>s  [ In article <3BD8CD33.AF7C7E12@fsi.net>, "David J. Dachtera" <djesys.nospam@fsi.net> writes:   F > If you follow comp.os.vms, you know that a large part of the OpenVMS? > battle cry includes, "We need app.'s!" and, "We need tools!".: > C > So, the answer to your question is both obvious and simple: if ituG > currently runs on either OpenVMS-VAX or OpenVMS-Alpha, your customersl% > will need it to run on OpenVMS-IPF.c > I > Don't ask, just port - everything! Your stockholders will thank you for>< > having the vision to plan ahead and take reasonable risks.  7 My question to the originator is what risk do you see ?   E Even if BEA code runs in inner mode, a port from Alpha to IA64 shouldeF be straightforward.  Getting involved early as a major ISV will enableF you to get prompt assistance from VMS Development if you find a manner: in which the behaviour on the two platforms _does_ differ.  A Depending on the nature of your code, VAX may be similar to AlphaiA or different, but you already have that one solved.  IA64 will beq like Alpha.    ------------------------------  % Date: Thu, 25 Oct 2001 20:28:59 +0200o7 From: Lothar Geyer <Lothar.Geyer@EDV-Berater-Online.de>t: Subject: Re: CDD RDB + Datatrieve - what license do I need5 Message-ID: <3BD859EB.B33F1098@EDV-Berater-Online.de>i   Brian Tillman schrieb: > I > >If you only use Rdb for storing CDD information then you do not need ao > >separate license for it.i > C > If I recall, the current version of CDD doesn't need Rdb to storegE > definitions, either.  I believe it can also use non-database files.s > --  / That's the next step to be able to to kill Rdb.e   Lothar Geyer   ------------------------------  % Date: Thu, 25 Oct 2001 17:59:48 -0400e- From: JF Mezei <jfmezei.spamnot@videotron.ca>n& Subject: Re: Compaq financial humour !, Message-ID: <3BD88B52.479E67E7@videotron.ca>   Paul Repacholi wrote:l@ > > So payment for the murder of Alpha may start only next year. > = > Do you mean after HP take over? If so, the Q's shareholders " > would have a field day in court!   Why ?=  K 1- If Alpha was portrayed as a potential money loser, then the shareholdersaK would have enthousiastically approved of its murder in onrder to focus on awK single platform. Remember that shareholders don't knwo VMS exists and don'tu7 know that Alpha was actually keeping the company alive.=  F And consider that the expertise those Wildfires will now be focused on4 building fancy wintel servers with an HP brand name.  I It is a given that Compaq's murder of IA64's competitor was structured to-M benefit both HP and Compaq. Since mergers are often very bad for the company,rN waiting until the merger commences to start cashing in the Intel money for the: ALpha murder would help hide the bad news from the merger.   ------------------------------  % Date: Thu, 25 Oct 2001 21:22:37 -0500V1 From: "David J. Dachtera" <djesys.nospam@fsi.net>5- Subject: Re: Compaq: VMS is alive and kickingb' Message-ID: <3BD8C8ED.230B0C35@fsi.net>'   John McLean wrote: > [snip]C > Okay so I can say this with hindsight.  I'm not a CEO.  CEO's ares& > supposed to have business foresight.  $ Operative key-phrase: "supposed to".   -- I David J. Dachtera  dba DJE Systemso http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/d   ------------------------------  % Date: Thu, 25 Oct 2001 21:50:24 -0500e1 From: "David J. Dachtera" <djesys.nospam@fsi.net>h. Subject: Re: Controller-based Mirror Set Usage' Message-ID: <3BD8CF70.ABF56EAA@fsi.net>    David Harrold wrote: > G > We are trying to come up with a way to create a reporting copy of oureE > database.  The application code does not allow us to use a stand-bys? > instance of oracle, opened read-only, to run reports against.. > E > One of the options were are looking at is to use a controller-basedsH > mirror set.  We are using HSJ50s with V5.7J-1 firmware and HSG80s withE > V8.6F-1 firmware.  The idea is to use a controller based mirror setIC > for the database volume, split the mirror, create a unit for that E > split member, mount it on a node, start oracle and run the reports.r > F > So, has anyone done this?  Can it be done programmatically (Assuming/ > we'll need the SAN command scripter product)?   A Short of Oracle, yes, you can do this programmatically, if you'ree@ willing to use unsupported software and write some DCL yourself.  & For talking to HSJs, use HSDSA-SCRIPT:, http://www.djesys.com/freeware/vms/hsdsa.zip  H I think that may be on the current OpenVMS Freeware (V5) also. It's fromC DEC(now Compaq) originally. The .ZIP file on my web site includes ak( "read me" file that tells how to use it.  F For talking to HSGs, you can use the old "HSZTERM" (SET HOST/SCSI) andB the device name presented by the HSGs (DGAxxxx:). HSZTERM has beenF retired, but may be found on some older ConDists. Test this THOROUGHLY8 before attempting to use it in a production environment.   -- C David J. Dachterak dba DJE Systemsc http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/-   ------------------------------   Date: 25 Oct 2001 18:04 CDT ' From: carl@gerg.tamu.edu (Carl Perkins) * Subject: Re: Delta time between two date's- Message-ID: <25OCT200118043377@gerg.tamu.edu>d  A Jan-Erik =?iso-8859-1?Q?S=F6derholm?= <noone@dummy.com> writes...0 }OK. }Thanks all for all examples !( }Shouldn't there be a lexical for this ? } ; }I just tried the COM file below (timediff.com) and got the  }following result :d }  }$ timediff "boottime" }0273-00:14:07 }$ sh sys/noprocE }OpenVMS V7.2-1  on node xxxxxx  25-OCT-2001 09:11:53.06  Uptime  272 	 }22:59:04E }$ } # }Anyone know why this defference ??r }  }Jan-Erik Sderholm.  = A lexical for what? Getting the difference between two times?   F I don't know why the specific interval, but the uptime in the SHOW SYSD appears to always be several minutes after the actual boot time. TheG value it starts from is aparently set at the end of the system startup,dH possibly after some delay after that even - all of the processes startedA at system startup appear to have been running for longer than the  SHOW SYSTEM uptime:u   $ sh proc/id=202000af/accl [...]rN                           Node: GERGX            Process name: "DECW$SERVER_0" [...]a)  Connect time:             25 01:52:04.02a $ sh sys/noproc9K OpenVMS V7.2-1  on node GERGX  25-OCT-2001 17:52:03.45  Uptime  25 01:48:04   G So the DECwindows server, which is pretty much the last process started0E during system startup, has been running for 4 minutes longer than theoJ SHOW SYSTEM indicates as the uptime. You can see an even bigger differenceF in the creation date of the OPERATOR.LOG file or the earlier processes started during boot.  J The "BOOTTIME" time is, on the other hand, a few seconds after the initial< entry in the error log for the most recent boot's timestamp.   --- Carl   ------------------------------  # Date: Thu, 25 Oct 2001 18:27:49 GMT>" From: Alfred Falk <falk@arc.ab.ca>' Subject: Re: DSSI VAX cluster questionsy9 Message-ID: <Xns91457EC8DB1F9falkarcabca@205.233.108.180>m  4 Robert DiRosario <rdirosario@starpower.net> wrote in% news:3BD75DA9.E8ABEC9@starpower.net:    ! >>I never saw a model 105 or 106.>I > The 105 and 106 both have a SCSI bus and two DSSI busses.  The SCSI bus>A > and one of the DSSI busses each have three internal connectors.h >  >>4000/100'sH > On Sunday a friend just gave me two 4000/100's for free! :-)  It looks1 > like they have a one SCSI bus and one DSSI bus.   I That is correct.  The 4000/10xA's have two DSSI busses, but non-A models e have just one DSSI bus.e  @ ----------------------------------------------------------------A   A L B E R T A         Alfred Falk               falk@arc.ab.ca p@ R E S E A R C H         Information Systems Dept   (780)450-5185+   C O U N C I L         250 Karl Clark Road.1                         Edmonton, Alberta, Canada  http://www.arc.ab.ca/   T6N 1E4*  http://www.arc.ab.ca/staff/falk/   ------------------------------  % Date: Fri, 26 Oct 2001 00:02:02 +0000 1 From: Robert DiRosario <rdirosario@starpower.net> ' Subject: Re: DSSI VAX cluster questionsr- Message-ID: <3BD8A7FA.5D8DB13B@starpower.net>   I Can you put two HSD10's on a single SCSI bus to create two different DSSIb2 paths to a drive, making it more like a CI system?  L I opened up my new 4000/100's and I now have two RF31T and four RF35 drives. :-)oM On the web I found average access times of 7 ms for the 31T and 10 ms for thex0 35.  (On www.aquascape.com, not a DEC web site.)  H Can VMS be "split" across two or more system disks,  like a Unix system?   dittman@dittman.net wrote:  6 > Tim Llewellyn <tim.llewellyn@cableinet.co.uk> wrote:L > :> For the cluster system disk would it be better to use a DSSI drive or aG > :> SCSI drive on a HSD10?  The only DSSI drives I have that are large2I > :> enough to hold VMS are 5.25" RF7x drives.   I have fast DEC (SeagateeK > :> w/ DEC microcode) SCSI drives, but I don't know if it's good or bad togL > :> have a HSD10 in-between the system disk and the DSSI bus.  (The HSD10's= > :> were inside the VAXEN, which I got from a surplus sale.)S > :>H > :> For SCSI things like CDROM and tape drives is there much differenceM > :> between putting them on a SCSI bus on one of the systems vs putting themm= > :> on a HSD10 on the second DSSI bus on one of the systems?  > H > : my experience with 4000/100's and HSD05 (not quite a HSD10, possibly > : they areI > : different) is that direct DSSI is better than HSD05 and if you have ae > : SCSIJ > : port on the back as well as a DSSI port then use that in preference to5 > : the others. Sorry I never saw a model 105 or 106.  >eK > I did some tests on my 4000/500 with an RF72, RF73, and HSD05/RZ28 combo.sJ > The HSD05/RZ28 combo was faster than the RF72, and equal in speed to the > RF73.d > -- > Eric Dittman > dittman@dittman.netb? > Check out the DEC Enthusiasts Club at http://www.dittman.net/.   ------------------------------    Date: 25 Oct 2001 13:28:34 -0700@ From: Dirk.Van.Bouchaute@Kender-Thijssen.be (Dirk Van Bouchaute); Subject: Re: Errors on RTA-ports in OpenVms version 5.5-2 ?i= Message-ID: <383301ac.0110251228.4d2b4644@posting.google.com>-   Dirk.Van.Bouchaute@Kender-Thijssen.be (Dirk Van Bouchaute) wrote in message news:<383301ac.0110250158.714fd80e@posting.google.com>...w > Hello, > G > Some VMS systems (I suppose the problem is related to the vms-version  > (5.5-2?))YG > log errors on the RTAx (RTA1,RTA2,RTA3,..) port (which is mapped when G > a decnet 'set host' is done towards that system). It does not seem ton+ > be a serious problem but some people hereaE > are curious to know the reason of these errors. Does anyone know if @ > this is a 'known problem' and how it could possibly be cured ? >  > Many thanks in advance ! >  > Dirk Van Bouchaute' > Dirk.Van.Bouchaute@Kender-Thijssen.be    .n   ------------------------------  # Date: Thu, 25 Oct 2001 20:54:22 GMT,2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman); Subject: Re: Errors on RTA-ports in OpenVms version 5.5-2 ? 2 Message-ID: <2__B7.980$RL6.10713@news.cpqcorp.net>   In article <383301ac.0110251228.4d2b4644@posting.google.com>, Dirk.Van.Bouchaute@Kender-Thijssen.be (Dirk Van Bouchaute) writes: :Dirk.Van.Bouchaute@Kender-Thijssen.be (Dirk Van Bouchaute) wrote in message news:<383301ac.0110250158.714fd80e@posting.google.com>...  H :> Some VMS systems (I suppose the problem is related to the vms-version :> (5.5-2?))H :> log errors on the RTAx (RTA1,RTA2,RTA3,..) port (which is mapped whenH :> a decnet 'set host' is done towards that system). It does not seem to, :> be a serious problem but some people hereF :> are curious to know the reason of these errors. Does anyone know if :> this is a 'known problem'...   F   Yes, this is known.  This bug was introduced in V5.5 and was removed   in V6.0.    * :> ...and how it could possibly be cured ?  C   Yes.  Upgrade to an OpenVMS VAX release from circa 1994 or newer.h  J   As the error count is simply incorrect and there is no underlying error,H   the fix was incorporated into OpenVMS VAX V6.0 and later releases, and   no ECO kit was released.  K   While zdec (a recently posted tool) or the sequence in the FAQ that zaps  ;   the error count would clear this, the error will reoccur.f  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Wed, 24 Oct 2001 10:17:12 -0500 * From: Patrick Spinler <pspinler@yahoo.com>? Subject: Re: Finding GNU "find" that builds from source for VMSa) Message-ID: <3BD6DB78.8FFF8092@yahoo.com>    John:y  H I just wanted to extend a vote of thanks for the GNV package !  I use it quite a lot here.  t  F I find myself frequently dropping into bash to do large scale multipleH file operations that I would otherwise have to have written a DCL script for.   Thanks ! -- Pat   John Reagan wrote: > H > There is a backing CVS library to many of the GNV binaries, but you'llH > find (pardon the pun) that the sources don't quite match the binaries. > I > Much of what I've been doing over the past 10 months is to get matchingiG > sources for the binaries, port more of the GNU components, and updatec > the CVS library. > G > The good news is that the library will be updated sometime this week.y > G > The bad news is that GNV still doesn't provide the autotools.  I handtI > crafted config.h files and makefile.com files for each GNU component (IaG > did findutils, diffutils, textutils, fileutils, sh-utils, grep, less,oE > and a few others).  Also, the "bad" news is that the config.h filesoG > match the COE version of OpenVMS.  A future task should be to add thedI > appropriate #if on _CRTL_VER in the config.h files.  I should have donetG > that as I went along, but I didn't.  Sorry.  Also, there is some worko9 > underway to try to provide the autotools in the future.  > J > As for the find in findutils, I started with findutils 4.1 from gnu.org.G >   I had to make changes to find/pred.c to change its use of fork() toaJ > vfork() (used with find -exec) and additional changes with restoring theG > current working directory since vfork() isn't quite the same.  I also-" > fixed one typo in find/parser.c. > C > So if you can wait until next week, the GNV CVS library should bej8 > refreshed with files that might be more useful to you. > 
 > John Reagant) > Compaq Pascal/{A|I}MACRO Project Leadero   ------------------------------  % Date: Thu, 25 Oct 2001 14:40:49 -0400i% From: "Lou Solomon" <lny98@yahoo.com> @ Subject: Re: Future Programming Platforms - Your Opinions Wanted/ Message-ID: <ttgng66q1nrn26@corp.supernews.com>b   Christopher:  6 Thank you for your thoughtful reply.  It really helps.  J First off,  I agree with you about the web client issues.  We publish someL of our VMS programs via telnet over the internet.  They are light years moreJ usable than 90% of comparable E-commerce sites, yet, of course, we have toI update them to be web compatable, basically, because our marketing peopleeK believe clients want web-like programs (color, graphics, fonts, etc).  SureoL the basic web paradigm is just about on par with the block send terminals ofJ 40 years ago (fill out fields, press a submit key, get back errors, HAH!),/ but even I admit, that's what people are using.w  I To be a little more specific... our programs are your basic business type.K programs. Screens to enter data, reading and writing from some kind of data L store, viewing the results on screen, writing reports.  Etc.  Nothing fancy,. no real need for multidemensional arrays, etc.  C To me, the relevation of Visual Basic after using VAX basic was thecG libraries out their available to do things.  Want to display stuff in a L grid, buy a grid control.  Want to let a user "browse" to a location to openJ a file, use the appropriate active X control.  (I use both those examples,J cause I had to write my own grid, and my own 'file/directory browswer' forI my VMS programs).  To me, a language is really the sum of the "libraries"JD available to it, and one can't deny that for VB (and other microsoftI languages) there are a lot of very nice libraries out there, that do reale2 useful work, and save LOTS of time in programming.  K I do disagree with one of your thoughts on Java, that there is not a lot ofoK 'extra' stuff to it.  From what I see, java has lots of good stuff built in J with the basic distribution, and lots of 3rd party controls, add ins, etc,K available to it, many/most written in Java to allow them to be added to anyi> Java platform (just look in any programmers paradise catalog).   Once again, thanks.n   Lou Solomon,  8 "Christopher Smith" <csmith@amdocs.com> wrote in messageF news:3B55D7F383B0D31197D9009027541CBF1170DDC6@cmiexch1.cmi.itds.com... > > -----Original Message-----. > > From: Lou Solomon [mailto:lny98@yahoo.com] >lJ > > programming in Visual Basic on Windows, and DEC Basic on VMS.  Data is@ > > stored on either SQL 7.0 DB or RMS files (accessed from PC's > > via CONNX odbc
 > > drivers).  >-K > I know which of the two I'd choose, and I've only ever used the other. ;)o >rA > > He is an open source advocate, and would love a solution that+ > > is reasonablyl> > > priced, multiplatform (or at least not locking us into one > > OS, or one clienta) > > platform), powerful (of course), etc.s >iF > Wouldn't we all?  All solutions will have their problems, of course. > A > > For example: should we go towards a solution where the clientr > > is always a,I > > web browser?  (sure makes it easy to have workers @ home, and to porte" > > programs over to our clients). > I > I think that anyone would need much more information about your current1 (andD > future) needs to make that call.  My first instinctual response is	 somethinga > like "NO!" :)e >gH > Generally the web-browser-as-client-for-real-application thing doesn't workE > well.  The interface ends up being terribly counter-intuitive.  ThefJ > limitations of a web-browser are difficult to overcome.  Once you try toD > overcome them you get pulled into the trap of using javascript (or worse...K > plugins, vbscript, etc), and later find out that the implementation isn'teH > consistent across browsers.  Don't think of even using CSS to get your page  > to render more nicely, either! >c> > > What about those open source languages like Python or TCL, > > etc.?  And ofe >iI > I use them all the time.  I'm confident that any one of them could be a  goodJ > replacement for Viral Basic.  Personally, I like TCL for nearly anythingL > that other people would write in visual basic.  I have used it somewhat inJ > the Real World, to develop small custom applications, and I can tell youL > that the applications are instantly portable across platforms.  (Note thatD > by instantly, I mean "change the directory separation character.") >JG > TCL has specific weakness in its theoretical lack of multidimensionalnD > arrays, and its relatively (compared to C, Modula) unwieldy way of	 producingi > binary output. >aJ > You can work around the multidimensional array problem relatively easily byH > using numbered one-dimensional arrays.  There is also a list construct whichbA > may contain nested lists, so if you don't mind addressing a twon dimensional:K > array in the form of "lindex(lindex($List, $Elem),$Elem)" then you can do  > that as well.2 >0I > When you get to about four dimensions of the above, it becomes somewhat: > confusing, though. >SE > Perl and Python are nice.  I haven't done much with Python, myself,t though, K > so I'll limit my comments to Perl.  I'd say that Perl has nearly the sameaF > benefits of TCL, except that I'm not sure whether you can get a PerlL > interpreter with Tk on windows or MacOS.  (That would mean that you eitherI > use text-based interfaces, or re-write the GUI -- on the other hand, it  maysK > be available...)  The strength of Perl over TCL is in text processing ande inD > the availability of several add-on modules that do everything from	 calculateo > Pi to butter your toast. >pL > There is also a relative newcomer called Ruby.  It most likely has all the= > problems of any relative newcomer, but may be worth a look.g >  > > course, what about Java? >eJ > Ahh, yes... what about Java? :)  Java is a good idea, in theory.  It mayH > eventually be a decent language.  The pseudo-code format is useful forI > portability.  On the other hand, it will put lots of stress on most anysH > computer.  There are also some things that, IMO, they should have done very > differently with Java. >nG > But, in trying to be objective, I'll say that if your Java program is > > well-constructed, it may serve your needs.  It has some nice object-orientedSL > features.  You should expect long development times, though, especially inG > the initial stages.  If you write modularly and in an object-orientedoJ > fashion, you can decrease the times later.  There aren't a lot of extrasJ > added in to Java.  No very good way to do GUI stuff, etc.  I'd also noteA > that the Sun JDK is very spartan.  Some people don't like that.- >-E > If you're considering Java, I'd highly recommend that you look intoa InfernosF > and Oberon.  They are two very different (from Java, and each other)J > languages which offer a lot of the strengths of Java without some of the
 > weaknesses.0 >OF > I'm not sure about source-code availability for Inferno.  It used to belongJ > to Bell Labs, now Lucent.  I believe there is another company selling itD > now, but there may be an "open source" type license available too. >OJ > All implementations of Oberon I've seen are completely free with source.L > Grab it and try it out.  I'm not sure whether either Inferno or Oberon runK > on VMS (I think that Oberon does, at least...), but then, Java doesn't don$ > very well either from what I hear. >i7 > > I guess I'd love any real world experiences or evena > > philisophical musings.@ > > (We are still in the philosophical stage... we know it isn't > > going to beb > > easy to move over).  >tH > Feel free to quiz me about my experiences with any of these languages.E > They're mixed, of course, and I have more experience with some than2 others.l >.
 > Regards, >  > Chris  > # > Christopher Smith, Perl Developer: > Amdocs - Champaign, IL >n > /usr/bin/perl -e 'A > print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");2 > '- >1 >:   ------------------------------  % Date: Thu, 25 Oct 2001 16:03:34 -0500 + From: Christopher Smith <csmith@amdocs.com> @ Subject: RE: Future Programming Platforms - Your Opinions WantedL Message-ID: <3B55D7F383B0D31197D9009027541CBF1170DDCA@cmiexch1.cmi.itds.com>   > -----Original Message-----, > From: Lou Solomon [mailto:lny98@yahoo.com]  @ > First off,  I agree with you about the web client issues.  We  > publish some> > of our VMS programs via telnet over the internet.  They are  > light years more: > usable than 90% of comparable E-commerce sites, yet, of  > course, we have to; > update them to be web compatable, basically, because our e > marketing people; > believe clients want web-like programs (color, graphics, r > fonts, etc).  Sure= > the basic web paradigm is just about on par with the block a > send terminals of>> > 40 years ago (fill out fields, press a submit key, get back  > errors, HAH!),1 > but even I admit, that's what people are using.o  I Yes, perhaps we should start writing "command-line" driven web apps.  OnesG window to type a command, one window (or just the rest of the page) fors. output, and a button to submit the command. :)  H In keeping with the web-based theme here for a second, I'll also mentionK that there's a TCL web browser plugin that will work on (at least) NetscapelK under Unix, Netscape/Internet Exploder under windows, and Internet Exploder G under MacOS.  I have used this plugin, and it does, indeed run TCL apps.6 "inline" in web pages, with little or no modification.  E > To me, the relevation of Visual Basic after using VAX basic was the-? > libraries out their available to do things.  Want to display   > stuff in a> > grid, buy a grid control.  Want to let a user "browse" to a  > location to open= > a file, use the appropriate active X control.  (I use both   > those examples, ? > cause I had to write my own grid, and my own 'file/directory i > browswer' fors@ > my VMS programs).  To me, a language is really the sum of the 
 > "libraries"iF > available to it, and one can't deny that for VB (and other microsoft? > languages) there are a lot of very nice libraries out there, e > that do real4 > useful work, and save LOTS of time in programming.  I In that case, Oberon and Inferno -- again -- might be interesting to you.e  > > I do disagree with one of your thoughts on Java, that there  > is not a lot of ? > 'extra' stuff to it.  From what I see, java has lots of good a > stuff built in? > with the basic distribution, and lots of 3rd party controls, C > add ins, etc,w> > available to it, many/most written in Java to allow them to  > be added to any @ > Java platform (just look in any programmers paradise catalog).  J Well, I should clarify some.  The language has several nice features whichH may be considered "extra" in comparison to other languages.  Notably theH arbitrary precision arithmetic, for instance.  My compliant lies more inI that -- in my admittedly limited experience -- SWING, for instance, isn'tpJ all it's cracked up to be, and AWT, while being all it's cracked up to be,J isn't cracked up for much. :)  My complaint about the lack of "extras" wasH more with the writing of user-interface components in Java than anything else.      Regards,   Chriso    ! Christopher Smith, Perl Developere Amdocs - Champaign, IL   /usr/bin/perl -e '? print((~"\x95\xc4\xe3"^"Just Another Perl Hacker.")."\x08!\n");e 'h  t   ------------------------------  % Date: Thu, 25 Oct 2001 21:54:11 -0700a6 From: "Howard Taylor" <Howard.Taylor@pacificcoast.net>@ Subject: Re: Future Programming Platforms - Your Opinions Wanted  Message-ID: <3bd8f196$1@nubby2.>  0 "Lou Solomon" <lny98@yahoo.com> wrote in message) news:ttgcgta0m9sb07@corp.supernews.com...i > All: >eL > My (new) manager has given me the task of deciding what is going to be ourH > future programming platform.  Currently, we are both a Windows and VMS shop,eH > programming in Visual Basic on Windows, and DEC Basic on VMS.  Data isH > stored on either SQL 7.0 DB or RMS files (accessed from PC's via CONNX odbc > drivers).r >sB > He is an open source advocate, and would love a solution that is
 reasonablyG > priced, multiplatform (or at least not locking us into one OS, or oneo client' > platform), powerful (of course), etc.  > K > For example: should we go towards a solution where the client is always a G > web browser?  (sure makes it easy to have workers @ home, and to porti  > programs over to our clients).J > What about those open source languages like Python or TCL, etc.?  And of > course, what about Java? >rL > I guess I'd love any real world experiences or even philisophical musings.J > (We are still in the philosophical stage... we know it isn't going to be > easy to move over).< >3  K My advice is to continue doing what you're doing - use the appropriate toolnJ for each platform. Using BASIC means it's not too hard to move from one toJ the other, and you end up with supported, compiled code at the end, ratherJ than interpreted code (which forces you to install and maintain a run-timeL system on each box). Different platforms do things differently, and althoughJ certain tools may claim to behave exactly the same on each platform, there will always be "gotchas".>  
 Howard Taylorb   ------------------------------   Date: 25 Oct 2001 18:09:42 GMT- From: forkosh@panix1.panix.com (John Forkosh)w! Subject: How to configure decnet?o) Message-ID: <9r9kh6$jic$1@news.panix.com>s  > I have a couple of VAXstation 4000/60's, and a couple of VLC's@ (and transceivers for the latter, with, as best I understand it,A heartbeat off) wired together with T's, bnc cables, and a pair ofo terminators.?      I've also installed VMS 7.2 with hoobyist licenses for vmsr= and layered products (including decnetend pak, but apparentlye? not decnetrtg).  While vms runs okay, and I can compile and runr; C programs, I haven't been able to net$configure and/or ncp-1 decnet so that I can set host, etc, between them.-<      I tried following the "basic" installation instructions; at http://www.openvms.compaq.com:8000/ and also looked over-> the other decnet-plus documents there, but I haven't been able to get it working.<      I was then hoping to find some simplified "quick-start"< instructions for installing and configuring decnet, like the: vax-kit.pdf document at montagar, but haven't been able to7 locate anything.  (The faq doesn't seem to discuss it.)s;      Is there some document like that somewheres?  I'm justn6 looking for "intuitively basic" network functionality,7 but apparently wasn't able to figure out how to do thatd) from the documents at openvms.compaq.com.e9      (By the way, I'd eventually like to add a linux nodee7 runnig decnet for linux, which, I believe, means I haver0 to configure Phase IV support into decnet-plus.) Thanks for any suggestions,' John (forkosh@panix.com)   ------------------------------  % Date: Thu, 25 Oct 2001 21:29:18 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>n Subject: Re: I am changing jobso' Message-ID: <3BD8CA7E.F18CD10B@fsi.net>c   John McLean wrote: >  > "David J. Dachtera" wrote: > >  > > Hamlyn Mootoo wrote: > > > [snip] I honestly didn't< > > > know there was a group specifically for VMS marketing. > > H > > Neither did I. When do they start work? That is to say, when do theyJ > > come forth with their first effort? So, far we've seen nothing I wouldI > > associate with a marketing group, rather perhaps a freshman marketingO > > student. > >S > > -- > > David J. Dachterad > > dba DJE Systems  > > http://www.djesys.com/ > >f > ? > Must be a Virtual Marketing System.  You know, one that neverr > *physically* does anything.   E I thought "Virtual" meant it's "imaginary" - it doesn't really exist,t1 it's just something else trying to be (whatever).f  ! ...which would explain a lot, no?e   -- n David J. Dachterai dba DJE Systemst http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/e   ------------------------------  + Date: Thu, 25 Oct 2001 10:53:52 -0700 (PDT)e. From: Fabio Cardoso <fabiopenvms@yahoo.com.br>< Subject: Re: KNOW WHEN A SUBMIT JOB IS FINISHED RETURN VALUE@ Message-ID: <20011025175352.93692.qmail@web20202.mail.yahoo.com>   Sandra   Why do you want this ?=20-4 You should define a logical name, changing its value to$ record the end of a job for example.     (...)c .R! $ DEFINE/SYSTEM logical-name texto $ EXIT  / I am not sure if you can check the $status of ae
 finished job.i   Regards    F=E1bio Cardosot    , --- SANDRA <SROMERO@CAIXAPENEDES.COM> wrote: > Hi,  >=204 > I would like to know if there is some way to get a > return value when an' > submitted job is finished and use it.r# > I work with VMS and DCL language.  > Thanks a Lot,=20 >=20  > petonets a tots=A1=A1=A1=A1=A1 >=20 > Sandriqui.     =3D=3D=3D=3D=3DrL =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D  F=E1bio dos Santos Cardoso OpenVMS System Manager Rio de Janeiro - Brazil  fabiopenvms@yahoo.com.brL =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3Dh  2 __________________________________________________ Do You Yahoo!?, Make a great connection at Yahoo! Personals. http://personals.yahoo.com   ------------------------------    Date: 25 Oct 2001 13:54:27 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen),< Subject: Re: KNOW WHEN A SUBMIT JOB IS FINISHED RETURN VALUE3 Message-ID: <wOl8wI2EE36T@eisner.encompasserve.org>   g In article <2efbec60.0110250946.54578391@posting.google.com>, SROMERO@CAIXAPENEDES.COM (SANDRA) writes:W  H > I would like to know if there is some way to get a return value when a' > submitted job is finished and use it.a  ? Read the Description section for the Synchronize command in theyA DCL Dictionary.  The URL on the VMS V7.3 Documentation CD-ROM is:s  6 	file:///VMSDOC073/v73/9996/9996pro_081.html#brass_156   ------------------------------  # Date: Thu, 25 Oct 2001 19:02:03 GMTc2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)< Subject: Re: KNOW WHEN A SUBMIT JOB IS FINISHED RETURN VALUE2 Message-ID: <LkZB7.971$RL6.10572@news.cpqcorp.net>  g In article <2efbec60.0110250946.54578391@posting.google.com>, SROMERO@CAIXAPENEDES.COM (SANDRA) writes:o  G :I would like to know if there is some way to get a return value when at& :submitted job is finished and use it." :I work with VMS and DCL language.  ?   You could use SPAWN/WAIT, and capture the operation directly,o  F   More commonly, the batch job can write the necessary information to    a file somewhere...0  E   Or probably the easiest approach here, you can use the DCL command -E   SYNCHRONIZE -- the exit status from the SYNCHRONIZE command is the     status from the batch job.  B   I'm assuming a recent version of OpenVMS -- when posting, please8   include the version and platform information.  Thanks!  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Thu, 25 Oct 2001 17:57:37 -0400o2 From: norm lastovica <norman.lastovica@oracle.com>< Subject: Re: KNOW WHEN A SUBMIT JOB IS FINISHED RETURN VALUE* Message-ID: <3BD88AD1.EE2EA7B9@oracle.com>  3 use the synchronize command.  HELP SYNCH to get youy started.  
 SANDRA wrote:f >  > Hi,o > H > I would like to know if there is some way to get a return value when a' > submitted job is finished and use it.:# > I work with VMS and DCL language.s > Thanks a Lot,a >  > petonets a tots >  > Sandriqui.   -- e> norman lastovica / oracle rdb engineering / usa / 610.696.4685   ------------------------------  # Date: Fri, 26 Oct 2001 00:48:26 GMTt" From: Don <donald@nospam.invalid!>* Subject: Looking for a bit of ACL/ACE help8 Message-ID: <pubhtt8mipje8opbsfpi7qv9rghf9e3fa7@4ax.com>   VMS 7.2o  C Can I completely control file access via an ACE even if the file isa( owned by the user attempting the access?  D Let's say user JONES is trying to access file JUNK.TXT and that file has an owner UIC of: [TEST,JONES]   File protection of:h  (S:RWED, O:RWED, G:RWED, W:RWED)   And an acl of:( (IDENT=[TEST,JONES],ACCESS=READ+EXECUTE)  D With this, shouldn't the flow first start at the ACE?  In this case,> the identifier is a match, so it should only allow R+E, right?  A Preliminary testing on my part shows that the flow seems to firsttF check ownership of the file, which is a match, then it uses the O:RWEDB file protection to determine access; it will allow a delete of the file in this case.    E If I remove O: completely, leaving W:RWED, the ACE then seems to take / over thereby denying delete access to the file.o  D So, again, can I provide full file control strictly with an ACE evenC on a file owned by the user performing the operation - even if thats' file has the protection set to O:RWED ?t    I hope I made that clear..... ;)  A Thanks for any help.  Please post any/all responses to the group.n   ------------------------------  # Date: Fri, 26 Oct 2001 01:42:06 GMTy2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman). Subject: Re: Looking for a bit of ACL/ACE help2 Message-ID: <Ob3C7.985$RL6.10752@news.cpqcorp.net>  ] In article <pubhtt8mipje8opbsfpi7qv9rghf9e3fa7@4ax.com>, Don <donald@nospam.invalid!> writes:m  D :Can I completely control file access via an ACE even if the file is) :owned by the user attempting the access?O  J   No, you can't.  If you own the object, you have control access.  If you C   have control access, then you can (eventually) get at the object.e  I   If you don't want the user to have control access, the user cannot own eH   the object.  Look around for the discussions of subsystem identifiers I   and particularly of resource identifiers, and also look at the creator iI   ACE entry mechanism, discussions both here in the newsgroup and in the oC   OpenVMS documentation set -- particularly in the security manual.J    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------    Date: 25 Oct 2001 20:53:29 -0500- From: Kilgallen@SpamCop.net (Larry Kilgallen)m. Subject: Re: Looking for a bit of ACL/ACE help3 Message-ID: <xJq8K65Apllo@eisner.encompasserve.org>n  g In article <Ob3C7.985$RL6.10752@news.cpqcorp.net>, hoffman@xdelta.zko.dec.nospam (Hoff Hoffman) writes: _ > In article <pubhtt8mipje8opbsfpi7qv9rghf9e3fa7@4ax.com>, Don <donald@nospam.invalid!> writes:, > F > :Can I completely control file access via an ACE even if the file is+ > :owned by the user attempting the access?a > L >   No, you can't.  If you own the object, you have control access.  If you E >   have control access, then you can (eventually) get at the object.o  B The fact that the owner can always control access is a fundamental@ of Discretionary Access Control -- the mechanism built into VMS.  C To restrict the freedom of the file owner you need Mandatory AccesspE Control.  That is available in SEVMS, but the latest version is V6.2.v  A If you want mandatory access control for hardware currently beinge@ built, you will have to switch to something like MVS or Solaris.   ------------------------------  # Date: Thu, 25 Oct 2001 18:52:25 GMTt2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)) Subject: Re: MIME Compliant Mail from VMS'2 Message-ID: <JbZB7.969$RL6.10589@news.cpqcorp.net>  g In article <LOWB7.458$Mw.95212@news.uswest.net>, "Michael D. Ober" <mdo.@.wakeassoc.com.nospam> writes: H :How can I attach a file to a VMS generated SMTP Mail Message?  The file1 :needs to appear as an attachement in MS-Outlook.e  J   The OpenVMS FAQ section "How do I send or read attachments in VMS MAIL?"J   and the Ask The Wizard topics (4492) and (5580) might be a good start...    N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  % Date: Thu, 25 Oct 2001 13:55:07 -0600g4 From: "Michael D. Ober" <mdo.@.wakeassoc.com.nospam>) Subject: Re: MIME Compliant Mail from VMSw1 Message-ID: <v6_B7.910$Mw.153192@news.uswest.net>S  K Thanks, the MIME utility on VMS is new for me.  Now does anyone know how toeJ script this in DCL, given a message body in one file and an attachement in
 another file?b -- Thanks,b
 Mike Ober.  7 "Peter LANGSTOEGER" <eplan@kapsch.net> wrote in messaget" news:3bd83b7d@news.kapsch.co.at...D > In article <LOWB7.458$Mw.95212@news.uswest.net>, "Michael D. Ober"$ <mdo.@.wakeassoc.com.nospam> writes:J > >How can I attach a file to a VMS generated SMTP Mail Message?  The file3 > >needs to appear as an attachement in MS-Outlook.u >gA > With MX you can do MAIL>SEND/FOREIGN/NOEDIT file-to-send.binarybA > Without MX, check the MIME utility of more recent VMS versions.t >  > --> > Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651= > Network and OpenVMS system manager  Fax.    +43 1 81111-888n> > <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netJ > A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  % Date: Thu, 25 Oct 2001 23:26:27 +0200n+ From: Jimmi Aakjaer <aakjaer@post7.tele.dk>t) Subject: Re: MIME Compliant Mail from VMSe8 Message-ID: <fq0httog2ph346cgndimt6pusj5aqpj1l4@4ax.com>  5 On Thu, 25 Oct 2001 10:09:15 -0600, "Michael D. Ober"d# <mdo.@.wakeassoc.com.nospam> wrote:   H >How can I attach a file to a VMS generated SMTP Mail Message?  The file1 >needs to appear as an attachement in MS-Outlook.o   Try with mailbox for vms a  2 http://www.jelparis.org/plequere/mailbox/index.htm  	 :-) jimmiw   ------------------------------    Date: 25 Oct 2001 19:47:55 -0400. From: Christopher K Davis <ckd-post@ckdhr.com>3 Subject: Re: More official info on Compaq/HP mergerg4 Message-ID: <w4u1wn7044.fsf@kline-station.ckdhr.com>  2 Monty Brandenberg <mcbinc@ne.mediaone.net> writes:   > JF Mezei wrote: H >> What has surprised me was no visible action by Microsoft to help with4 >> the disasters (donations to those funds etc etc).  I > In the interest of fair reporting (not that that's an RFC or anything),jD > Gates coughed up $10M in the first day or two which would be a few% > days' interest on MS' cash-on-hand.   E I heard it as $5M cash, $5M "in-kind" donations (software which costsm pennies to duplicate).   -- aI Christopher Davis * <ckd-sig@ckdhr.com> * <URL:http://www.ckdhr.com/ckd/>d?          If you want to give blood, try to make an appointment. E   Call 1-800-GIVE LIFE (in USA) or your local blood center/Red Cross.r?          Blood will be needed for weeks to come and won't keep.r   ------------------------------    Date: 25 Oct 2001 22:57:32 -0700' From: roose_chua@yahoo.com (Roose Chua)d Subject: On hanging processesk= Message-ID: <127ddcff.0110252157.17d5d05c@posting.google.com>   D HI! I would just like to ask some help regarding hanging processes.   B 1. When using $ Monitor utility or any other monitoring utilities,C what should you observe to verify that a process is indeed hanging? F 2. When a process has been determined to be hanging due to waiting for: a resourse, how would I know what resource it was waiting?B 3. Is there a documentation available in the internet that discuss hanging processes?   Regards,
 Roose Chua	 Accenture   F *The posted message is purely my personal opinion and does not reflect@ the view of the organization that I am currently connected with.   ------------------------------    Date: 25 Oct 2001 22:58:23 -0700' From: roose_chua@yahoo.com (Roose Chua)p Subject: On hanging processes,= Message-ID: <127ddcff.0110252158.48c6d8fa@posting.google.com>   D Hi! I would just like to ask some help regarding hanging processes.   B 1. When using $ Monitor utility or any other monitoring utilities,C what should you observe to verify that a process is indeed hanging? F 2. When a process has been determined to be hanging due to waiting for: a resourse, how would I know what resource it was waiting?B 3. Is there a documentation available in the internet that discuss hanging processes?   Regards,
 Roose Chua	 Accentures  F *The posted message is purely my personal opinion and does not reflect@ the view of the organization that I am currently connected with.   ------------------------------  % Date: Fri, 26 Oct 2001 14:22:42 +1000 / From: "Phil Howell" <phowell@snowyhydro.com.au> 5 Subject: Re: Optimizing indexed file while adding keyt1 Message-ID: <Rz5C7.1563$Fi4.89672@ozemail.com.au>c  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3BD79244.1EA3B5F7@videotron.ca...K > I have a large indexed file (60,000 records) with a single key. I want top addd > 2 more keys to it. >dJ > The file is stable (no changes, no additions). But access time is fairly
 important.: So you are changing to the alternate keys for access then?H If there are more than about 3 concurrent users then investigate using a& global_buffer_count of > 0 on the fileB (rms tries to use global buffers for the primary key first though)J > I know how I can use ANA/RMS/STAT and then used edit/FDL to optimize the fileJ > and then convert it to the better FDL. And I know how to use EDIT/FDL to addcE > keys and then use convert to create the indices for those new keys. ; You may be better off using ANAL/RMS/FDL to generate an fdl)E then use your favourite text editor to add the new keys (cut & paste) J as there are no changes or additions you can leave all %fill parameters at 100s5 you should also remove any superfluous fdl parameterstL I also remove the AREA directives as I believe convert does a reasonable job anyway   leaving something like TITLE   <whatever>   FILEK         ALLOCATION              <current size + an estimate for 2 new keys>n#         BEST_TRY_CONTIGUOUS     yesp!         BUCKET_SIZE             0 !         EXTENSION               0r'         ORGANIZATION            indexedk   RECORD         FORMAT          fixed          SIZE             800   KEY 0f"         SEG0_LENGTH             18!         SEG0_POSITION           4 &         TYPE                    string   KEY 1n"         CHANGES                 no#         DUPLICATES              yesi"         SEG0_LENGTH             25"         SEG0_POSITION           22&         TYPE                    string   KEY 2u"         CHANGES                 no#         DUPLICATES              yes("         SEG0_LENGTH             17"         SEG0_POSITION           72&         TYPE                    string  K If there is a "null key" value that you don't want to index then specify it ; If there are no duplicate secondary keys then DUPLICATES no  >eJ > However, considering the size of the file, I am curious whether it would beL > possible to do both the optimisation and the adding of the new keys with a > single convert ?@ You could but the optimisation will be be without the extra keys (but it will work) >hK > Does the code in edit/fdl use the output of ANA/RMS/STAT  when I add keys  ?  > K > Or is the only way to first add the keys, convert the file, then ana/rms,o) > optimize the fdl and covernt it again ?e@ Use /nosort when you convert it as it is already in key sequence Phil   ------------------------------  % Date: Fri, 26 Oct 2001 01:10:52 -0400t- From: JF Mezei <jfmezei.spamnot@videotron.ca>y5 Subject: Re: Optimizing indexed file while adding key , Message-ID: <3BD8F040.2389711B@videotron.ca>   Phil Howell wrote:B > Use /nosort when you convert it as it is already in key sequence  N Since I am adding secondary keys to the file, is /NOSORT still applicable ? Or, does /NOSORT apply only to the primary key ?   ------------------------------  % Date: Fri, 26 Oct 2001 07:47:38 +0200t7 From: Lothar Geyer <Lothar.Geyer@EDV-Berater-Online.de>-5 Subject: Re: Optimizing indexed file while adding key 5 Message-ID: <3BD8F8FA.4F88C154@EDV-Berater-Online.de>:   Phil Howell schrieb: >  ....N > I also remove the AREA directives as I believe convert does a reasonable job > anyway ....  D You should never remove the AREA directives. Use different areas for" each key and for the data records.F The edi/fdl/nointer does optimization. However, it generates a maximumG of two areas. So for files with more than one key you should always use ) edi/fdl/inter and add the areas manually.D   Lothar Geyer   ------------------------------  % Date: Thu, 25 Oct 2001 21:25:57 -0500 1 From: "David J. Dachtera" <djesys.nospam@fsi.net>o6 Subject: Re: Spiralog (was Re: BCC support in VMSMAIL)' Message-ID: <3BD8C9B5.4B559505@fsi.net>o   Phillip Helbig wrote:1 > J > > remember "SPIRALOG", the 64 bit file system to replace RMS ... why wasJ > > that killed?  i asked several people in software support and they saidK > > because of poor performance ... i don't buy that, if that was true they3L > > would have never finished it for release ... i think oracle had a lot toI > > do with killing it as spiralog would have decreased sales for oracle,a' > > rdb, and other high priced db's .... >  > Please explain.  :-|  F Curious. I thought Spiralog was abandoned because it didn't work, just like host-based snapshots.   -- l David J. DachteraR dba DJE SystemsB http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/<   ------------------------------  # Date: Thu, 25 Oct 2001 18:24:14 GMTM( From: "Amy Lewis" <amylewis@pacbell.net> Subject: Re: SSH for Alpha VMSB Message-ID: <iNYB7.3138$Ou1.1034800247@newssvr21.news.prodigy.com>  + "L Arut" <larut@yahoo.com> wrote in messagec7 news:73f00c8f.0105282321.13854644@posting.google.com...D; > Is there anyway of getting sn SSH server pre-compiled forrK > an Alpha VMS system since I don't have access to DEC C having surrendered0$ > our VAX C license many years ago ?   Run Multinet 4.3    	 Amy Lewisj   ------------------------------  # Date: Thu, 25 Oct 2001 18:13:39 GMTZ8 From: hammond@not@peek.ppb.cpqcorp.net (Charlie Hammond). Subject: Re: SYS$EXAMPLES:DAYLIGHT_SAVINGS.COM2 Message-ID: <nDYB7.968$RL6.10436@news.cpqcorp.net>  0 The official patch for this is VMS73_DTSS-V0100.3 As stated below, this applies ONLY to OpenVMS V7.3.w   The kit is available at:  J http://ftp1.support.compaq.com/public/vms/axp/v7.3/vms73_dtss-v0100.README  G If possible, please obtain and use the official patch rather than usingeN the work-around explained below.  If you do use, or have used the work-around,K please obtain and apply the patch as soon as practical.  The patch containse8 other minor fixes to SYS$EXAMPLES:DAYLIGHT_SAVINGS.COM.   @         ********************************************************    m In article <%xUB7.951$RL6.10319@news.cpqcorp.net>, hammond@not@peek.ppb.cpqcorp.net (Charlie Hammond) writes:- >  > 3 >    This information applies ONLY to OpenVMS V7.3.n) >    It applies to OpenVMS Alpha and VAX.-= >    It does not apply to any V7.2, V7.2-* or prior versions. 0 >    Versions beyond V7.3 will contain this fix. >l5 >    ------------------------------------------------e< >N.B. This information is provided on a "best effort" basis.@ >It has not completed engineering review in OpenVMS engineering.> >You are solely responsible for determining the applicability  >to your system.5 >    ------------------------------------------------D >w >eC >A while back somebody posted that they had noticed a sign error intA >SYS$EXAMPLES:DAYLIGHT_SAVINGS.COM.  I acknowledged this and said E >that it had been fixed, and that the fix was in the remedial stream.tB >This was correct, but today (25-OCt-2001) it came to my attention< >that this fix has never been made available to customers.  4 >The work to make this available is now in progress. >    @C >Unfortunately, in the USA we change from Daylight to Standard time-C >this coming weekend; it is unlikely that the fix will be published * >in time.  I apologize for this situation. >i >WHAT TO DO? >wE >If you are not running OpenVMS V7.3, you do not need to do anything.g > @ >If you do not use SYS$EXAMPLES:DAYLIGHT_SAVINGS.COM, you do not >need to do anything.  >hD >If you do use SYS$EXAMPLES:DAYLIGHT_SAVINGS.COM, modify it as shown >below before using it.  >iA >Look for these lines, starting at approximately line number 260, + >and change the sign for TIME$ADJ as shown.a >w >    $!h0 >    $!  Set the new TDF and the time adjustment >    $   if dst$std .eqs. "D"w
 >    $   thenf& >    $       new$tdf = new$tdf + 60   8 >    $       time$adj = -60  <---------- make this "+60" >    $   endif >    $   if dst$std .eqs. "S"r
 >    $   thent# >    $       new$tdf = new$tdf - 60m7 >    $       time$adj = 60 <----------- Make this "-60"e >    $   endif >    $!o >eD >If you modify SYS$EXAMPLES:DAYLIGHT_SAVINGS.COM take care to ensureE >that the modified version is placed in SYS$COMMON, not SYS$SPECIFIC.eD >This can be done by using an EDIT command similar to the following: >.K >    $  EDIT SYS$SYSDEVICE:[VMS$COMMON.SYSHLP.EXAMPLES]DAYLIGHT_SAVINGS.COM- >  >aD >NOTE: SYS$EXAMPLES:DAYLIGHT_SAVINGS.COM was completely rewritten inF >OpenVMS V7.3.  The V7.3 version of SYS$EXAMPLES:DAYLIGHT_SAVINGS.COM E >is not compatible with prior versions of OpenVMS.  DO NOT attempt toeI >make the changes shown above for any version of OpenVMS other than V7.3.C >I >-- L >    Charlie Hammond -- Compaq Computer Corporation -- Pompano Beach  FL USAI >       (hammond@not@peek.ppb.cpqcorp.net -- remove "@not" when replying) K >      All opinions expressed are my own and not necessarily my employer's.t >  >    -- eK     Charlie Hammond -- Compaq Computer Corporation -- Pompano Beach  FL USAeH        (hammond@not@peek.ppb.cpqcorp.net -- remove "@not" when replying)J       All opinions expressed are my own and not necessarily my employer's.   ------------------------------    Date: 25 Oct 2001 23:06:41 +0200* From: eplan@kapsch.net (Peter LANGSTOEGER). Subject: Re: SYS$EXAMPLES:DAYLIGHT_SAVINGS.COM* Message-ID: <3bd87ee1$1@news.kapsch.co.at>  m In article <nDYB7.968$RL6.10436@news.cpqcorp.net>, hammond@not@peek.ppb.cpqcorp.net (Charlie Hammond) writes:-1 >The official patch for this is VMS73_DTSS-V0100.-4 >As stated below, this applies ONLY to OpenVMS V7.3. >n >The kit is available at:z >dK >http://ftp1.support.compaq.com/public/vms/axp/v7.3/vms73_dtss-v0100.README   & Does the problem only exist on Alpha ?   -- u< Peter "EPLAN" LANGSTOEGER           Tel.    +43 1 81111-2651; Network and OpenVMS system manager  Fax.    +43 1 81111-888$< <<< KAPSCH AG  Wagenseilgasse 1     E-mail  eplan@kapsch.netH A-1121 VIENNA  AUSTRIA              "I'm not a pessimist, I'm a realist"   ------------------------------  % Date: Thu, 25 Oct 2001 14:54:01 -0700o; From: Cheng-Liang Shen <Cheng-Liang.Shen@cas.honeywell.com> : Subject: Where to find the DCL command procedure document?1 Message-ID: <3BD889F9.35CF85D2@cas.honeywell.com>   G Can anyone tell me where I can find the DCL command procedure tutorial?h4 It had better have many examples to imitate. Thanks.  
 Chengliang   ------------------------------  # Date: Thu, 25 Oct 2001 22:35:35 GMTr2 From: hoffman@xdelta.zko.dec.nospam (Hoff Hoffman)> Subject: Re: Where to find the DCL command procedure document?2 Message-ID: <Xs0C7.984$RL6.10411@news.cpqcorp.net>  o In article <3BD889F9.35CF85D2@cas.honeywell.com>, Cheng-Liang Shen <Cheng-Liang.Shen@cas.honeywell.com> writes:-H :Can anyone tell me where I can find the DCL command procedure tutorial?5 :It had better have many examples to imitate. Thanks.   F   Please acquire and read the OpenVMS Frequently Asked Questions (FAQ)H   document.  You will find pointers to books, to manuals, to tutorials, >   and to the answers to a variety of common OpenVMS questions.  G   Shameless plug: Writing Real Programs in DCL is a book available fromcG   Digital Press.  The book provides an introduction to DCL programming dE   up through writing webserver CGI scripts and such using DCL, and it(   provides many examples..  N  ---------------------------- #include <rtfaq.h> -----------------------------N       For additional, please see the OpenVMS FAQ -- www.openvms.compaq.com    N  --------------------------- pure personal opinion ---------------------------L    Hoff (Stephen) Hoffman   OpenVMS Engineering   hoffman#xdelta.zko.dec.com   ------------------------------  # Date: Fri, 26 Oct 2001 04:23:34 GMT . From: "mulp" <michaelpettengill@earthlink.net>, Subject: Re: Windows XP reality check pleaseD Message-ID: <az5C7.2400$2K5.231131@newsread2.prod.itd.earthlink.net>  J I'm watching Mr Bill on Charlie Rose right now and Mr Bill has just statedG that they have calculated that Windows XP will save you a week per yearnC because of not having to reboot, faster reboots and shutdowns, etc.k  J Mr Bill also says that Windows XP Home is now compatible enough to replace" the DOS based versions of Windows.  I I've spent nearly 3 hours downloading the update advisor and according tocG its report, I will probably need a week to get my system working at itseK current level with all the updates I need to find for my slightly more thanIG 6 month old PC (AMD Athlon, 1.2 ghz, 384 meg, 60 gig, DVD, CD-RW, etc.)o   Some of the issues reported:  Handspring Visor (USB)T  Nikon CoolPix (USB)  Visioneer 4400 (USB)e  Lexar Jumpshot (USB)m  Epson Photostylus 1280 (USB)8  Lexar Digital Film Reader
  Wacom Tablet-/  Autodesk Plot Style manager (Autocad LT 2000i)   Autodesk Plotter manageri  ConfluentLM  Digital Dashboard
  HSP56 WDM  Sensiva  Speechh  Drive Image 4.0  DirectCD 5   Easy CD Creator 5  DVD Movie player.  3 A number of programs are know to require reinstall:g
  Outlook 20002  DirectCD 5p  Easy CD creator 5  J Nearly all the software and hardware is new since I bought the PC (ok, theK Nikon CoolPix 990, Epson Printer, and scanner, I bought 2 months before thee PC).  F I'd rather move a VMS development environment from VAX to an IA64 betaL release than deal with moving from my current Win ME system, which does suckL from a reliability standpoint compared to Win 98, Win NT, and Win 2k, to WinJ XP.  And while I've worked with people who claim that Win NT or Win 2k wasK rock solid, I've found the way that I worked and the apps that I use, wouldo3 cause them to behave bizzarely several times a day.e  H I'd rather be using VMS, and I tried, but Compaq couldn't figure out theJ salvage value of a DEC 3000 Model 300 because the people who knew how wereK laid off, so I had to give back my VMS system so that it could be scrapped.jI But in the end, Compaq may well have driven me from the computer businessaH all together so all I need is a PC to do research on wood working tools, paints, amazon, and ebay.   I Over the past several months I've come to the conclusion that M$ has mademE computers too painful to deal with.  Give me dry rot, carpenter ants,dI incorrectly built foundations, etc. over Windows and PC problems any day;aH they're much easier to deal with and there is real satisfaction once youJ have fixed the problems - and you can spend an extra $20 and know that theE problem will be fixed for the next decade minimum and with luck a fewg decades.  E Oh, yeah, another thing that Mr Bill said was that "we have a problemIF communicating".  We have phone numbers and email addresses that are soK complicated to manage and what is needed is software to manage them for us.   L Of course, from Mr Bill's point of view the communcation problem is this: "II want you to send me money every month" - we're not listening.  So this isnL where Win XP comes in - it incorporates nagware which starts off nagging youI to get a passport account and then it will check for updates and I'm surerJ that M$ will nag you about getting a Hotmail account for $9.95 a month, anG MSN account for $19.95 a month, priority interrup IM for $6.95 a month,-/ wireless IM to nag you for $29.95 a month, ....2  : "JF Mezei" <jfmezei.spamnot@videotron.ca> wrote in message& news:3BD32238.55FFAAF3@videotron.ca...I > OK, Microsoft is attempting to trumpet its new revolutionary Windows XPo > operating system.i >hI > What is radically different in XP than in previous versions ? Are theren arelA > serious imporvements in terms of an "enterprise" side of thingsh (clusteringW > etc etc) ? > = > What about on desktops, are there any significant changes ?o >o   ------------------------------   End of INFO-VAX 2001.595 ************************