0 INFO-VAX	Wed, 11 Jan 2006	Volume 2006 : Issue 21      Contents:! Announcing the 2005 VMS SIG tapes # Re: Apache 2.1 running on OVMS IA64 ' Re: Beginner: nesting DCL command files ' Re: Beginner: nesting DCL command files ' Re: Beginner: nesting DCL command files ' Re: Beginner: nesting DCL command files / Re: foreign commands, .com files, and dcltables / Re: foreign commands, .com files, and dcltables / Re: foreign commands, .com files, and dcltables  Re: Honeypot stats Re: Honeypot stats, Re: How to prevent users from telnet-login ?, Re: How to prevent users from telnet-login ?, Re: How to prevent users from telnet-login ?@ Re: HP C - dup2(file_desc1,file_desc2) question: implicit CLOSE?7 Re: Licenses in a cluster (nodes know about each other) 7 Re: Licenses in a cluster (nodes know about each other)   Re: Licenses MOD_UNITS question.  OT: time to market with the 8086$ Re: OT: time to market with the 8086$ Re: OT: time to market with the 8086$ Re: OT: time to market with the 8086$ Re: OT: time to market with the 8086$ Re: OT: time to market with the 8086$ Re: OT: time to market with the 8086 Re: PDPs in the news!  Re: PDPs in the news! & Re: Samba v3 on VMS and HP VMS Roadmap Re: Strange SEARCH behaviour Re: Strange SEARCH behaviour Re: Strange SEARCH behaviour Re: Strange SEARCH behaviour1 Re: Suggestion for INSTALL PURGE and SHUTDOWN.COM 1 Re: Suggestion for INSTALL PURGE and SHUTDOWN.COM + RE: WVNETcluster uptime reaches 10 years... + RE: WVNETcluster uptime reaches 10 years...   F ----------------------------------------------------------------------  % Date: Tue, 10 Jan 2006 19:15:10 -0500 ' From: Glenn Everhart <Everhart@gce.com> * Subject: Announcing the 2005 VMS SIG tapes0 Message-ID: <11s8jp0bq97fsd7@corp.supernews.com>  L This is to announce the Spring 2005 sigtapes, which were mailed out a couple
 weeks ago.I The spring 2005 CD was rather late, so another for Fall 2005 was composed D with some updates of mainly similar material and sent along with it.  G Contents of the Spring 2005 CD are as follows (Fall 2005 has updates of  some of the same stuff):   Spring 2005 VMS / LT SIG Tapes [VMSLT05A...] Directory Tree ------------- --------- ----  G [EMULATORS...]	Alpha emulators (such as are available). None vms ready.   ? [GNU...]	Gnu C, Emacs, Gnumeric, Gnunet, etc. new free software   < [NET...]	This directory contains code which might be of some: 		interest, off the net. CDRTOOLS in particular is used to9 		write CD (and by rumor DVD) on other OSs. Some versions : 		work on VMS. UUCP is a copy of vms uucp, and new zip and4 		unzip sources here should build on VMS and add new; 		abilities and better compression. Botan is a crypto tool. = 		Bochs is an x86 emulator for generic machines and WINE is a ; 		windows emulator that runs on x86. Some versions of Bochs 7 		have been reported to work on VMS in the past, albeit   		slowly. XPDF is a .pdf viewer.  F [SAMBA...]	SAMBA is a system which allows systems to communicate using7 		SMB or CIFS protocols (used by MS Windows for network 6 		communication) thus permitting sharing either way of: 		printers and disks with such systems. (The VMS client is1 		however a utility, not a mountable filesystem.)   < [TK...]		Files recently added to Hunter Goatley's collection8 		CSVSEARCH - search comma separated variable data files: 		FLIST - File and directory manager (fullscreen oriented)5 		GBLSEC_SDA_VMS - global section info MAILSYM - mail 8 		symbiont for mailing files to yourself via PRINT MBU -7 		mailbox utility. Manipulate VMS system mailboxes from < 		command line RSYNC_2_5_6_PRE_1_SRC - synch class libs from 		CMS (VMS pieces)  < [VU...]		VMS things off the net This area has lots of useful= 		material that builds on VMS with sources and binaries where ; 		possible.  Among the items here:  Antiword - read MS Word = 		documents AUTOCONF - configuration generator like unix one, ; 		for VMS BLISS- partial Bliss compiler from FreeVMS effort 9 		BZIP2- better compress utility, compresses smaller than ; 		gzip CURL - get ftp or http served files off web, restart = 		partial transfers till done GNV - unit / gnu like tools for ; 		VMS Ghostscript - display Postscript on screen or printer ; 		for non postscript devices FreeVMS - VMS rebuilt for x86, : 		partial effort so far. Based on Linux sources but builds; 		VMS APIs.  Imagemagick - displays / manipulates images in 4 		many formats Mac64 - macro64 (assembler for alpha)= 		(freeware) MySQL - relational DBMS, built for VMS OpenSSL - < 		lots of cryoto routines and certificate generation ability= 		WGET - get http or ftp served data, remote mirror of sites, < 		will keep trying till it gets all of a file (picks up from9 		where a transfer failed).  Xscreensaver - lots of great ; 		display hacks and screen saver for Xwindows including VMS 7 		(obviously many others. Documents present generally.)   ? [WWW...]	WASD web server and utils This area contains a VMS web 7 		server and associated utilities, recent distribution.     L Thanks for all who wrote new material and released it, and anyone who cannotN find the tree please contact me for a copy, or for address updates. If you areK interested in being on the tree (and are willing to make copies for others)  please let me know.    Glenn C. Everhart  Everhart@gce.com   Tapecopy coordinator   ------------------------------    Date: 10 Jan 2006 12:03:44 -08005 From: "Tapani Rundgren" <Tapani.Rundgren@treville.se> , Subject: Re: Apache 2.1 running on OVMS IA64C Message-ID: <1136923424.642123.261480@f14g2000cwb.googlegroups.com>    Hi Ulf!   
 Yes we do! What are the problems You have?   
 Kind Regards,  Tapani Rundgren / web: http://www.treville.se for contact details    sysulfo wrote: > Hi ! > C > Does anyone have any experience regarding running CSWS/Apache 2.1 - > on Integrity, OpenVMS 8.2-1 AND Rdb 7.2 ???  > 3 > We are in great need of help to get this working.  > Config ... > Client(s) ... 	 > PHP ...  >  > Rgds.  > 
 > Ulf Ohlsson  >  >  > --  	 > sysulfo $ > posted via http://sysdminforum.com   ------------------------------    Date: 10 Jan 2006 11:17:16 -0800" From: dave.baxter@bannerhealth.com0 Subject: Re: Beginner: nesting DCL command filesC Message-ID: <1136920635.984424.190780@g43g2000cwa.googlegroups.com>   ; This is in response to DJD's question about TCPIP$SYSTARTUP    David J Dachtera wrote:  > F > TCPIP$DEFINE_COMMANDS seems to have an issue because it doesn't even, > work as intended on my site. Not sure why. >   F I also recieved a local tcpip parameter file (recommendations) from myA vendor, (probably same vendor as you) and I also decided that the ? appropriate way to implement this was by using the standard VMS 5 approach, i.e. create TCPIP$SYSTARTUP.COM and put the C @TCPIP$DEFINE_COMMANDS.COM  line in there, followed by the required A series of "sysconfig -r <sub-system> <attribute=value>" commands.   E I also had problems with this approach.    I don't remember whether I E was getting the same errors as you, however I ended up discarding the - approach.    I renamed TCPIP$SYSTARTUP.COM as 5 TCPIP$<VENDOR>_PARAMETERS.COM and then executed it in F SYSTARTUP_VMS.COM, right after the TCPIP$STARTUP.com and this seems toD work without a problem (I have no idea why, it is the identical fileB contents which fail as TCPIP$SYSTARTUP.COM).     {NOTE: I actuallyE wanted to do it "correctly", i.e. using the "sysconfigdb" command and C storing all of the values in SYSCONFIGTAB.DAT;  however my "Vendor" E wasn't comfortable with that so we end up having to redefine them all  at every boot}     Dave.    ------------------------------    Date: 10 Jan 2006 15:16:58 -0800 From: "R Boyd" <bob@hax.com>0 Subject: Re: Beginner: nesting DCL command filesC Message-ID: <1136935018.853643.216000@g14g2000cwa.googlegroups.com>   7 I've had the same issue with V7.3-2, TCPIP V5.4 eco 4/5 D I've also had some success with the SYSCONFIGTAB, but not with every value.  C I wound up modifying TCPIP$SYSTARTUP by using the translated values G that the DEFINE_COMMANDS procedure was supposed to setup and hardwiring G the actual image names into the commands instead.   I'm not sure it the G problem has to do with logical names not being created early enough, or F if there is some other issue, but the hardwiring of the commands seems to work ok.    E.G.  7 $ mcr tcpip$ifconfig we0 alias 'my_alias_ip_address'/24   3 I tried to do this with the command $ ifconfig ....  but it never seemed to work.   Robert  # dave.baxter@bannerhealth.com wrote: = > This is in response to DJD's question about TCPIP$SYSTARTUP  >  > David J Dachtera wrote:  > > H > > TCPIP$DEFINE_COMMANDS seems to have an issue because it doesn't even. > > work as intended on my site. Not sure why. > >  > H > I also recieved a local tcpip parameter file (recommendations) from myC > vendor, (probably same vendor as you) and I also decided that the A > appropriate way to implement this was by using the standard VMS 7 > approach, i.e. create TCPIP$SYSTARTUP.COM and put the E > @TCPIP$DEFINE_COMMANDS.COM  line in there, followed by the required C > series of "sysconfig -r <sub-system> <attribute=value>" commands.  > G > I also had problems with this approach.    I don't remember whether I G > was getting the same errors as you, however I ended up discarding the / > approach.    I renamed TCPIP$SYSTARTUP.COM as 7 > TCPIP$<VENDOR>_PARAMETERS.COM and then executed it in H > SYSTARTUP_VMS.COM, right after the TCPIP$STARTUP.com and this seems toF > work without a problem (I have no idea why, it is the identical fileD > contents which fail as TCPIP$SYSTARTUP.COM).     {NOTE: I actuallyG > wanted to do it "correctly", i.e. using the "sysconfigdb" command and E > storing all of the values in SYSCONFIGTAB.DAT;  however my "Vendor" G > wasn't comfortable with that so we end up having to redefine them all  > at every boot} >  >  > Dave.    ------------------------------  % Date: Wed, 11 Jan 2006 00:48:13 +0100 ( From: Paul Sture <paul.sture@bluewin.ch>0 Subject: Re: Beginner: nesting DCL command files, Message-ID: <42ivduF1jf80cU1@individual.net>  
 R Boyd wrote: 9 > I've had the same issue with V7.3-2, TCPIP V5.4 eco 4/5 F > I've also had some success with the SYSCONFIGTAB, but not with every > value. > E > I wound up modifying TCPIP$SYSTARTUP by using the translated values I > that the DEFINE_COMMANDS procedure was supposed to setup and hardwiring I > the actual image names into the commands instead.   I'm not sure it the I > problem has to do with logical names not being created early enough, or H > if there is some other issue, but the hardwiring of the commands seems
 > to work ok.  >  > E.G. > 9 > $ mcr tcpip$ifconfig we0 alias 'my_alias_ip_address'/24  > 5 > I tried to do this with the command $ ifconfig ....  > but it never seemed to work. >   H This reminds me of a problem I had back at the end of 2001. Someone had E been given a list of commands by an Oracle DBA (I assume he got them   from someone like Oracle)   I These didn't work, an din my research it appeared the sysconfig commands  E wouldn't tolerate anything other than a single space between command   elements. I ended up with:  D      $    @sys$manager:tcpip$define_commands  ! Define sysconfig etc6      $       sysconfig -r inet tcp_keepalive_default=1,      $       sysconfig -r inet tcp_keepcnt=2.      $       sysconfig -r inet tcp_keepidle=20.      $       sysconfig -r inet tcp_keepinit=60/      $       sysconfig -r inet tcp_keepintvl=20      I The manager for that system had taken the original instructions as given  I and formatted them with multiple spaces and maybe tabs, which broke them.    ------------------------------  % Date: Tue, 10 Jan 2006 19:13:14 -0600 2 From: David J Dachtera <djesys.nospam@comcast.net>0 Subject: Re: Beginner: nesting DCL command files+ Message-ID: <43C45BA9.29BC554B@comcast.net>    Paul Sture wrote:  >  > R Boyd wrote: ; > > I've had the same issue with V7.3-2, TCPIP V5.4 eco 4/5 H > > I've also had some success with the SYSCONFIGTAB, but not with every
 > > value. > > G > > I wound up modifying TCPIP$SYSTARTUP by using the translated values K > > that the DEFINE_COMMANDS procedure was supposed to setup and hardwiring K > > the actual image names into the commands instead.   I'm not sure it the K > > problem has to do with logical names not being created early enough, or J > > if there is some other issue, but the hardwiring of the commands seems > > to work ok.  > >  > > E.G. > > ; > > $ mcr tcpip$ifconfig we0 alias 'my_alias_ip_address'/24  > > 7 > > I tried to do this with the command $ ifconfig ....   > > but it never seemed to work. > >  > I > This reminds me of a problem I had back at the end of 2001. Someone had F > been given a list of commands by an Oracle DBA (I assume he got them > from someone like Oracle)  > J > These didn't work, an din my research it appeared the sysconfig commandsF > wouldn't tolerate anything other than a single space between command > elements. I ended up with: > F >      $    @sys$manager:tcpip$define_commands  ! Define sysconfig etc8 >      $       sysconfig -r inet tcp_keepalive_default=1. >      $       sysconfig -r inet tcp_keepcnt=20 >      $       sysconfig -r inet tcp_keepidle=200 >      $       sysconfig -r inet tcp_keepinit=601 >      $       sysconfig -r inet tcp_keepintvl=20  >  > J > The manager for that system had taken the original instructions as givenK > and formatted them with multiple spaces and maybe tabs, which broke them.   G Well, I never even got THAT far. "sysconfig" & Co. always return IVVERB  trying to do it "as designed".   --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  % Date: Tue, 10 Jan 2006 19:56:07 +0100 ( From: Paul Sture <paul.sture@bluewin.ch>8 Subject: Re: foreign commands, .com files, and dcltables, Message-ID: <42iea8F1ik74nU1@individual.net>   Christoph Gartmann wrote: v > In article <DTiotGxQ0bj6-pn2-dwcNPxFmRLgh@dave2_os2.home.ours>, "Dave Weatherall" <djw-nothere@nospam.nohow> writes: > F >>We use a project related xxxLOGIN.COM which is defined in the login F >>fields of the UAF. It enables us to define symbols, foreign and SET H >>COMMANDs specific to a project and not to clutter up SYLOGIN. Picking G >>up on DaveD's point - it also ensures that other users on the system  B >>do not have their vritual/process memory space polluted/used up  >>unnecessesarily (spellling?) >>H >>The alternative is have the user call the project-specific 'login.com'E >>as part of their own LOGIN.COM. I have installed my stuff at other    >>sites that prefer that method. >  > : > You may even modify your SYLOGIN.COM in a way like this:" >    $ grpnum = F$GETJPI("","GRP") >    $ IF grpnum .LE. 17 >    $    THEN >    ...O > or similar constructs. That's what we do here. So in SYLOGIN.COM is the stuff K > that is for all users and the stuff for specific groups or projects is in M > separate files that are called from SYLOGIN.COM . It is pretty ok to modify . > SYLOGIN, it won't be touched by OS upgrades. >     Expanding on that one could have  < $ group_login = "sys$manager:group_" + grpnum + "_login.com"7 $ if f$search(group_login) .nes. "" then @'group_login'   H (You could also put such files in group accessible dev+dirs to delegate ( maintenance to a project admin account.)  I Furthermore, if that command file has group logical name definitions, it  F can be executed at system startup via submit/user (user in that group E with GRPNAM privilege) to create and populate the group logical name   table with something like:   $ if f$setprv("GRPNAM")  $ then $     define /group ...  $ endif    ------------------------------  % Date: Tue, 10 Jan 2006 15:40:30 -0500  From: norm.raphael@metso.com8 Subject: Re: foreign commands, .com files, and dcltablesQ Message-ID: <OFCB5E6B29.1E48D421-ON852570F2.007131AF-852570F2.00718B12@metso.com>   C Paul Sture <paul.sture@bluewin.ch> wrote on 01/10/2006 01:56:07 PM:    > Christoph Gartmann wrote: C > > In article <DTiotGxQ0bj6-pn2-dwcNPxFmRLgh@dave2_os2.home.ours>, 6 > "Dave Weatherall" <djw-nothere@nospam.nohow> writes: > > G > >>We use a project related xxxLOGIN.COM which is defined in the login G > >>fields of the UAF. It enables us to define symbols, foreign and SET I > >>COMMANDs specific to a project and not to clutter up SYLOGIN. Picking H > >>up on DaveD's point - it also ensures that other users on the systemC > >>do not have their vritual/process memory space polluted/used up   > >>unnecessesarily (spellling?) > >>J > >>The alternative is have the user call the project-specific 'login.com'F > >>as part of their own LOGIN.COM. I have installed my stuff at other" > >>sites that prefer that method. > >  > > < > > You may even modify your SYLOGIN.COM in a way like this:$ > >    $ grpnum = F$GETJPI("","GRP") > >    $ IF grpnum .LE. 17 > >    $    THEN
 > >    ...D > > or similar constructs. That's what we do here. So in SYLOGIN.COM > is the stuffJ > > that is for all users and the stuff for specific groups or projects is inH > > separate files that are called from SYLOGIN.COM . It is pretty ok to modify0 > > SYLOGIN, it won't be touched by OS upgrades. > >  > " > Expanding on that one could have > > > $ group_login = "sys$manager:group_" + grpnum + "_login.com"9 > $ if f$search(group_login) .nes. "" then @'group_login'  > I > (You could also put such files in group accessible dev+dirs to delegate * > maintenance to a project admin account.) > J > Furthermore, if that command file has group logical name definitions, itG > can be executed at system startup via submit/user (user in that group F > with GRPNAM privilege) to create and populate the group logical name  K We do that with this kind of construct in the application startup procedure  invoked in SYSTARTUP_VMS.COM:    $   RUN/UIC=[306,000] - 9         /INPUT=SYS$SYSROOT:[SYSMGR.STARTUPS]306LOGS.COM - ;         /OUTPUT=SYS$SPECIFIC:[SYSMGR.STARTUPS]306LOGS.LOG -          SYS$SYSTEM:LOGINOUT = %RUN-S-PROC_ID, identification of created process is 20600209   4 where the ".COM" file defines the group logicals and purges back the ".LOG" files.    > table with something like: >  > $ if f$setprv("GRPNAM")  > $ then > $     define /group ... 	 > $ endif    ------------------------------  % Date: Tue, 10 Jan 2006 19:20:24 -0600 2 From: David J Dachtera <djesys.nospam@comcast.net>8 Subject: Re: foreign commands, .com files, and dcltables+ Message-ID: <43C45D58.7AF3985B@comcast.net>    Paul Sture wrote:  > [snip]" > Expanding on that one could have > > > $ group_login = "sys$manager:group_" + grpnum + "_login.com"9 > $ if f$search(group_login) .nes. "" then @'group_login'   ? You'd likely want to make that group number into something more  recognizable first, perhaps:  0 $ GRPNUM = F$FAO( "!6OL", F$GETJPI( 0, "GRP" ) )  A ...so it not only looks like what you see in AUTHORIZE and as the C LNM$GROUP_xxxxxx table name, but is always six digits, zero padded.    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  # Date: Wed, 11 Jan 2006 02:16:09 GMT 1 From: Keith Parris <keithparris_NOSPAM@yahoo.com>  Subject: Re: Honeypot stats 1 Message-ID: <JTZwf.1435$hJ1.924@news.cpqcorp.net>    JF Mezei wrote: A > VMS just hasn't had much bells and whistles added to it, so few  > vulnerabilities were added.   G On the contrary, VMS has had numerous and significant new features and  I capabilities added over the years, but new features are always carefully  F designed from the start so as not to add new security vulneratilities.   ------------------------------  % Date: Tue, 10 Jan 2006 23:50:39 -0500 ) From: "Neil Rieck" <n.rieck@sympatico.ca>  Subject: Re: Honeypot stats : Message-ID: <n80xf.16946$Pq4.184595@news20.bellglobal.com>  ? "Keith Parris" <keithparris_NOSPAM@yahoo.com> wrote in message  + news:JTZwf.1435$hJ1.924@news.cpqcorp.net...  > JF Mezei wrote: B >> VMS just hasn't had much bells and whistles added to it, so few >> vulnerabilities were added. > I > On the contrary, VMS has had numerous and significant new features and  K > capabilities added over the years, but new features are always carefully  H > designed from the start so as not to add new security vulneratilities. >   K I agree. In fact, the whole feel of OpenVMS seemed to change between 7.3-1  
 and 7.3-2.  
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.8 http://www3.sympatico.ca/n.rieck/links/cool_openvms.html   ------------------------------  + Date: Tue, 10 Jan 2006 13:27:29 -0600 (CST) * From: sms@antinode.org (Steven M. Schweda)5 Subject: Re: How to prevent users from telnet-login ? 2 Message-ID: <06011013272896_20331674@antinode.org>  ; From: koehler@eisner.nospam.encompasserve.org (Bob Koehler)   # > > UAF>  MODIFY user /REMOTE = 0-0 	 > > [...] 0 > > You are not authorized to login at this time1 > > %TELNET-S-REMCLOSED, Remote connection closed 4 > > -TELNET-I-SESSION, Session 01, host alp, port 23  G >    That's odd.  I thought REMOTE was for dial-up only.  But again, it # >    probably depends on the stack.   G    And I thought that it was normal, but I haven't dealt with an actual A dial-up connection for some years.  I should have added, however:    alp $ tcpip show version  ;   HP TCP/IP Services for OpenVMS Alpha Version V5.4 - ECO 5 D   on a COMPAQ Professional Workstation XP1000 running OpenVMS V7.3-2  D Of course, the original question included no such info, so, while it5 (demonstrably) worked for me, I did say, "might try".   H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode-org     Saint Paul  MN  55105-2547    ------------------------------    Date: 10 Jan 2006 16:16:39 -0600- From: Kilgallen@SpamCop.net (Larry Kilgallen) 5 Subject: Re: How to prevent users from telnet-login ? 3 Message-ID: <AZ0bqi+KcMHx@eisner.encompasserve.org>   q In article <nwTfEojRI85T@eisner.encompasserve.org>, koehler@eisner.nospam.encompasserve.org (Bob Koehler) writes:   G >    That's odd.  I thought REMOTE was for dial-up only.  But again, it # >    probably depends on the stack.   A The access types controlled by SYSUAF are Network, Batch, and the 3 three Interactive types - Local, Dialup and Remote.   > SET HOST is Remote, DECnet object access is Network, and otherD drivers are able to specify what type is ascribed to a given device.  7 $ search sys$Library:starlet.req jpi$_/match=and,values ; literal JPI$K_OTHER = 0;                !  JPI$_Mode values ; literal JPI$K_NETWORK = 1;              !  JPI$_Mode values ; literal JPI$K_BATCH = 2;                !  JPI$_Mode values ; literal JPI$K_INTERACTIVE = 3;          !  JPI$_Mode values > literal JPI$K_DETACHED = 0;             !  JPI$_JOBTYPE values> literal JPI$K_LOCAL = 3;                !  JPI$_JOBTYPE values> literal JPI$K_DIALUP = 4;               !  JPI$_JOBTYPE values> literal JPI$K_REMOTE = 5;               !  JPI$_JOBTYPE values   ------------------------------  % Date: Tue, 10 Jan 2006 19:34:16 -0600 2 From: David J Dachtera <djesys.nospam@comcast.net>5 Subject: Re: How to prevent users from telnet-login ? + Message-ID: <43C46098.4ADFBDF1@comcast.net>    Bob Koehler wrote: > a > In article <06010923511372_20331674@antinode.org>, sms@antinode.org (Steven M. Schweda) writes:  > > H > >    If you don't mind disabling DECnet access, too, you might try (in > > AUTHORIZE):  > > # > > UAF>  MODIFY user /REMOTE = 0-0  > >  > >  > > alp $ telnet alp) > > %TELNET-I-TRYING, Trying ... 10.0.0.9 4 > > %TELNET-I-SESSION, Session 01, host alp, port 23, > > -TELNET-I-ESCAPE, Escape character is ^] > >   > >       I'm ALP.  Who are you? > >  > > Username: sms 
 > > Password: 0 > > You are not authorized to login at this time1 > > %TELNET-S-REMCLOSED, Remote connection closed 4 > > -TELNET-I-SESSION, Session 01, host alp, port 23 > 9 >    That's odd.  I thought REMOTE was for dial-up only.    # I thought DIALUP was for dial-up...   " DJAS01::DDACHTERA$ uaf help modify   MODIFY  E      Changes values in a SYSUAF user record. Qualifiers not specified %      in the command remain unchanged.         Format   (        MODIFY  username /qualifier[,...]           #   Additional information available:      Parameter  Qualifiers C   /ACCESS    /ACCOUNT   /ALGORITHM /ASTLM     /BATCH     /BIOLM      /BYTLMC   /CLI       /CLITABLES /CPUTIME   /DEFPRIVILEGES        /DEVICE     /DIALUP C   /DIOLM     /DIRECTORY /ENQLM     /EXPIRATION           /FILLM      /FLAGSC   /GENERATE_PASSWORD    /INTERACTIVE          /JTQUOTA   /LGICMD     /LOCALC   /MAXACCTJOBS          /MAXDETACH /MAXJOBS   /MODIFY_IDENTIFIER     /NETWORKC   /OWNER     /PASSWORD  /PBYTLM    /PGFLQUOTA /PRCLM     /PRIMEDAYS 	 /PRIORITY :   /PRIVILEGES           /PWDEXPIRED           /PWDLIFETIMEH   /PWDMINIMUM           /QUEPRIO   /REMOTE    /SHRFILLM  /TQELM     /UIC    /WSDEFAULT /WSEXTENT  /WSQUOTA>   Examples   /IDENTIFIER           /PROXY     /SYSTEM_PASSWORD   MODIFY Subtopic?     > But again, it # >    probably depends on the stack.    Indeed.    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  % Date: Tue, 10 Jan 2006 22:19:29 -0600 6 From: "Craig A. Berry" <craigberry@mac.com.spamfooler>I Subject: Re: HP C - dup2(file_desc1,file_desc2) question: implicit CLOSE? D Message-ID: <craigberry-219976.22192910012006@news.isp.giganews.com>  B In article <1136885973.636663.93610@z14g2000cwz.googlegroups.com>,3  "Uli Bellgardt" <uli_bellgardt@hotmail.com> wrote:    > Hello, > E > I am using dup2 (file_desc, fileno(stdout)) to redirect a program's G > output to a file. After some time, or after some amount of output has E > been written (whatever comes first), the program opens a new output G > file, and repeats above dup2 (...) statement to route the output to a I > new file. The redirection works fine, all output appears in the correct  > files just as expected.  > F > But: After a while, the program is left with an increasing number ofH > open output files. The reason for this is clearly that I did not CLOSE+ > the output files, and the fix is trivial.  > I > The C documentation states "If file_desc2 is connected to an open file,  > that file is closed".  > C > >From this, I would expect that an previous output file is closed G > implicitely with the next dup2(...), and so on.  My program obviously > > does not behave like this. Am I wrong with this expectation? > , > Compaq C++ V6.5-046 for OpenVMS Alpha V8.2  B Without seeing your code, it's impossible to know exactly what is . happening, but there are some things to check.  H Are you checking return (and, if necessary, errno) values after calling 1 dup2?  If the call fails, the file is not closed.   H Do you know for certain, that file_desc1 and file_desc2 are not equal?  $ If they are, the file is not closed.   According to the standard:   -----------------  fid = dup2(fildes, fildes2);   shall be equivalent to:    close(fildes2); & fid = fcntl(fildes, F_DUPFD, fildes2); -----------------   B You could try the fcntl equivalent and see if the behavior is any 9 different.  The full description of the function is here:   @ http://www.opengroup.org/onlinepubs/009695399/functions/dup.html   ------------------------------  # Date: Tue, 10 Jan 2006 19:18:19 GMT ) From: Alfred Falk <falk@arc.REMOVE.ab.ca> @ Subject: Re: Licenses in a cluster (nodes know about each other)9 Message-ID: <Xns97477D2D54524falkarcabca@198.161.157.145>   0 JF Mezei <jfmezei.spamnot@teksavvy.com> wrote in% news:43C38BC0.15BF9C02@teksavvy.com:    D > I had been told that when you have 2 systems with different systemA > disks and different license databases, the licenses are totally D > independant. My experience seems to indicate this is not the case.  B I was told many years ago just the opposite.  If you have multipleG LMF$LICENSE.LDB files, each needs to contain all the licenses required  H on all nodes.  This is easier if all the files are actually identical.  E And its even easier just to have one file: In SYLOGICALS.COM add the   line@     	$ DEFINE/SYSTEM/EXEC LMF$LICENSE drive:[dir]LMF$LICENSE.LDB   C > My *guess* is that each node takes out clusterwide locks for each H > licence it has loaded indicating how many units it has consumed. (this/ > key is (or contains) the authorisation number    I think your guess is correct.     --  @ ----------------------------------------------------------------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: 10 Jan 2006 12:22:53 -0800$ From: "Ed Wilts" <ewilts@ewilts.org>@ Subject: Re: Licenses in a cluster (nodes know about each other)C Message-ID: <1136924573.646686.111080@g43g2000cwa.googlegroups.com>    JF Mezei wrote: J > I had been told that when you have 2 systems with different system disksH > and different license databases, the licenses are totally independant.  G This is not supported.  Although it may or may not work, you're on your F own.  Even if you can make it work now, it's not guaranteed to work in a future release.   E See the License Management Utility Guide for Database Location (in my G 7.3 manual, it's section 2.2.1).  "If you have multiple system disks in F an OpenVMS Cluster environment where some systems cannot access one ofF the system disks, you must keep separate identical License Databases."  G See also the section "Controlling Node Access to Licenses in Clusters": G  "In an OpenVMS Cluster environment, license units are available to all  nodes by default.".   * In short, whoever told you that was wrong.  	    .../Ed    ------------------------------  % Date: Tue, 10 Jan 2006 13:53:29 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> ) Subject: Re: Licenses MOD_UNITS question. , Message-ID: <43C402A9.1D3926AC@teksavvy.com>   Richard Brodie wrote: I > I don't think that is the correct explanation: it's more likely, if you  > clonedH > the system, that you have multiple, or the wrong interfaces in the NCPG > database. Make sure "list known lines" and "list known circuits" have  > the correct interface names   H I not only did that, but also dumped the decnet database files to see ifE there were any remnants of the previous interface and eventually just ? deleted the files and restarted the decnet config from scratch.   G The fact that the problem propagated to the Microvax II when I rebooted F it seesm to confirm to me that there is sharing of license information0 via the lock manager between nodes in a cluster.  F And this makes sense since if you have a common database with a singleC licence serving all nodes, then some mechanism must ensure that the / licence has sufficient units to feed all nodes.   G However, what is not clear to me is why a 0 units licence (which should H be able to feed all nodes in a cluster out of that single license) wouldE still not allow DECNET end-=node to activate if multiple nodes have a G single interface each. But it is the only explanation I can find of the  behaviour I have seen.   ------------------------------  % Date: Tue, 10 Jan 2006 16:31:20 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> ) Subject: OT: time to market with the 8086 , Message-ID: <43C4279D.C81E4812@teksavvy.com>  7 http://biz.yahoo.com/ap/060110/macworld_apple.html?.v=8   F Apple had announced in mid 2005  that it would begin to introduce 8086 based Macintoshes by mid 2006.  H Today, steve Jobs unveiled 2 8086 based macs, both using dual core intelE 8086s and said that the full product line will have been converted to H the 8086 by year end. That is partly because the 8086 is now faster thanB the custom G5 Apple chips  so there is no incentive to stick with ! Apple's custom Power based chips.   A Interesting comparison with the Alpha to IA64 transition which is   happening at a much slower pace.  G Imagine if, when VMS is ported to the 8086, they strike a deal with the E Charron people to integrate the VAX emulator into the 8086 version of F VMS so that VMS-8086 could run vax executables in emulation right from the $ sign.   H HP could produce a QBUS based 8086 and allow all remaining VAX customers+ to migrate to the new current VMS platform.    ------------------------------   Date: 10 Jan 2006 22:08:53 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)- Subject: Re: OT: time to market with the 8086 , Message-ID: <42ipjlF1ivhm2U1@individual.net>  , In article <43C4279D.C81E4812@teksavvy.com>,0 	JF Mezei <jfmezei.spamnot@teksavvy.com> writes:9 > http://biz.yahoo.com/ap/060110/macworld_apple.html?.v=8  > H > Apple had announced in mid 2005  that it would begin to introduce 8086  > based Macintoshes by mid 2006. > J > Today, steve Jobs unveiled 2 8086 based macs, both using dual core intelG > 8086s and said that the full product line will have been converted to J > the 8086 by year end. That is partly because the 8086 is now faster thanD > the custom G5 Apple chips  so there is no incentive to stick with # > Apple's custom Power based chips.   : Wow.  Is that a big step backwards from the M68K or what!!   > C > Interesting comparison with the Alpha to IA64 transition which is " > happening at a much slower pace. > I > Imagine if, when VMS is ported to the 8086, they strike a deal with the G > Charron people to integrate the VAX emulator into the 8086 version of H > VMS so that VMS-8086 could run vax executables in emulation right from
 > the $ sign.   4 Won't the segments cause a bit of a problem for VMS?   > J > HP could produce a QBUS based 8086 and allow all remaining VAX customers- > to migrate to the new current VMS platform.   @ I just wonder who is going to be manufacturing all these 8086's?   bill- (who still plays with CPM86 on occaision. :-)    --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  % Date: Tue, 10 Jan 2006 17:10:13 -0500 2 From: "Stanley F. Quayle" <squayle@insight.rr.com>- Subject: Re: OT: time to market with the 8086 / Message-ID: <43C3EA75.26846.19B7EC4D@localhost>   ( On 10 Jan 2006 at 16:31, JF Mezei wrote:E > Imagine if, when VMS is ported to the 8086, they strike a deal with H > the Charron people to integrate the VAX emulator into the 8086 versionF > of VMS so that VMS-8086 could run vax executables in emulation right > from the $ sign.  F Actually, I would expect HP to provide a free tool to migrate Itanium D and Alpha executables to run on the new platform, just as they have  done with VEST and AEST.  C AEST was written by "the CHARON people", and the latest version of  F VEST was updated by those same people.  They have a lot of experience ' writing those tools, as you've noticed.   8 [I'm not totally unbiased, being a CHARON-VAX reseller.]  
 --Stan Quayle  Quayle Consulting Inc.  
 ----------- Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363 3 8572 North Spring Ct., Pickerington, OH  43147  USA 0 stan-at-stanq-dot-com       http://www.stanq.com) "OpenVMS, when downtime is not an option"    ------------------------------  % Date: Tue, 10 Jan 2006 17:32:42 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> - Subject: Re: OT: time to market with the 8086 , Message-ID: <43C435FC.80407A3C@teksavvy.com>   "Stanley F. Quayle" wrote:G > Actually, I would expect HP to provide a free tool to migrate Itanium E > and Alpha executables to run on the new platform, just as they have  > done with VEST and AEST.    H But when you consider the number of VAX sites still out there, providingH a cheap and safe alternative would be very neat if the emulator were VMSH based because it would be a more politically correct way to upgrade from VAX to a more modern platform.  H Depending on when the announcement of the port to the 8086 is made, I am? not convinced that an IA64 emulator on the 8086 would really be E necessary. As long as Alpha remains a superset of available software, N emulating Alpha on the 8086 would be more important. (as would emulating VAX).  C Until there is a point where much software exists only on that IA64 3 thing, the need for the emulator won't be so great.    ------------------------------  % Date: Tue, 10 Jan 2006 19:10:59 -0500 ) From: "Neil Rieck" <n.rieck@sympatico.ca> - Subject: Re: OT: time to market with the 8086 9 Message-ID: <c2Ywf.7117$W03.396816@news20.bellglobal.com>   : "JF Mezei" <jfmezei.spamnot@teksavvy.com> wrote in message& news:43C435FC.80407A3C@teksavvy.com... [...snip...] >  > J > But when you consider the number of VAX sites still out there, providingJ > a cheap and safe alternative would be very neat if the emulator were VMSJ > based because it would be a more politically correct way to upgrade from  > VAX to a more modern platform. > J > Depending on when the announcement of the port to the 8086 is made, I amA > not convinced that an IA64 emulator on the 8086 would really be G > necessary. As long as Alpha remains a superset of available software, J > emulating Alpha on the 8086 would be more important. (as would emulating > VAX).  > E > Until there is a point where much software exists only on that IA64 5 > thing, the need for the emulator won't be so great.   G My apologies for this reply but your 8086 post provided me with a nice  I segue. This next paragraph came to me last week (Jan-2006) in a personal    email from an ex-Intel employee.   <quote>   L "Given your interest in VMS you might find this amusing. In the early 1990'sJ we visited Microsoft to try to ensure that their new OS "Windows NT" wouldK be available on IA32. We met with Dave Cutler, and he was adamant that IA32 I was doomed and would we please get lost so he could target Alpha and then I whatever 64-bit architecture was certain to replace IA32 by Intel. It was F not a polite disagreement; that guy HATED IA32 and wasn't reluctant toE transfer his displeasure to IA32's representatives (us). What an ugly % business meeting. Smart guy, though."    </quote>  H I asked for a clarification on the phrase "IA32" as it pertains to this J quote and was told that it refers to both 486 (which was in production in 1 1990) and Pentium which was still in development.   
 Neil Rieck Kitchener/Waterloo/Cambridge,  Ontario, Canada.8 http://www3.sympatico.ca/n.rieck/links/cool_openvms.html   ------------------------------  % Date: Tue, 10 Jan 2006 22:35:12 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> - Subject: Re: OT: time to market with the 8086 , Message-ID: <43C47CE4.A29D4B5B@teksavvy.com>   Found this neat tidbit today:  ##B The day Apple intros Intel-based Macs, its stock closes at $80.86.& That's a magic number for chip geeks.  ##   ------------------------------  % Date: Tue, 10 Jan 2006 23:27:48 -0500 2 From: "Stanley F. Quayle" <squayle@insight.rr.com>- Subject: Re: OT: time to market with the 8086 . Message-ID: <43C442F4.2240.1B1193BC@localhost>  ( On 10 Jan 2006 at 17:32, JF Mezei wrote:@ > But when you consider the number of VAX sites still out there,B > providing a cheap and safe alternative would be very neat if theH > emulator were VMS based because it would be a more politically correct5 > way to upgrade from VAX to a more modern platform.    C Gee, we've discussed this a couple of years ago.  About half of my  @ CHARON-VAX clients are using the Windows version *because* it's B Windows.  Looks good to management that it's on Windows.  Even if D it's really running VMS, "there's a desktop, mouse, and everything".  E Just goes to show that most managers should never be allowed to make   technical decisions.  
 --Stan Quayle  Quayle Consulting Inc.  
 ----------- Stanley F. Quayle, P.E. N8SQ  +1 614-868-1363 3 8572 North Spring Ct., Pickerington, OH  43147  USA 0 stan-at-stanq-dot-com       http://www.stanq.com) "OpenVMS, when downtime is not an option"    ------------------------------  % Date: Tue, 10 Jan 2006 14:00:50 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com>  Subject: Re: PDPs in the news!, Message-ID: <43C40462.99E85379@teksavvy.com>   Alan Greig wrote: , > PDP Planet is at http://www.pdpplanet.com/    C The links don't work because they make gratuitous use of javascript ( instead of a simple <a href   statement.  C And the front page has over 28 HTML errors. Obviously designed by a  microsoft person/software.   ------------------------------  % Date: Tue, 10 Jan 2006 19:43:27 -0800 2 From: Patricia Tressel <tressel@cs.washington.edu> Subject: Re: PDPs in the news!0 Message-ID: <43C47EDF.9020600@cs.washington.edu>   JF Mezei wrote:   E > The links don't work because they make gratuitous use of javascript * > instead of a simple <a href   statement.  P D'you mean they won't work without javascript enabled?  They work 4 me, and I'm   running mozilla on Linux...  ;-)  . > And the front page has over 28 HTML errors.   I Plus the mismatch in the character encoding -> 29 errors found by the W3  O validator (validator.w3.org -- very handy, 'specially if you're teaching HTML).   P But there may be other errors, and also some non-errors induced by the presence M of the /> end tag shortcuts, which are not valid 4.0 Transitional, and which  K weird the validator's parser out.  Only 4 of the errors might be due to />.   @ The missing alt tags and script types are all real errors.  Tsk.  4 > Obviously designed by a microsoft person/software.  M There's no "content generator" meta tag, and the page source has none of the  Q weird font cruft I associate with (e.g.) conversion from a Word doc, so this may   be hand-written.  	 -- Pat T.    ------------------------------   Date: 11 Jan 2006 01:51:58 GMT From: healyzh@aracnet.com / Subject: Re: Samba v3 on VMS and HP VMS Roadmap + Message-ID: <dq1obu0reh@enews2.newsguy.com>   8 John Malmberg <malmberg@dskwld.zko.hp.compaq.dec> wrote:F > I have forwarded the questions posted in this thread to him.  He is I > interested in feedback in what features are important to be in a SAMBA  / > based offering to help prioritize work on it.   I Something I'd like to see is it tested with Mac OS X.  Last I'd heard, no I one has gotten Mac OS X to access VMS via NFS, and I was rather disturbed I when I attempted to edit a VMS file via Samba on my Mac, only to have the @ attempts to save the changes I made delete versions of the file.  J Basically what I'm saying is that there is a need out there for a reliableJ method of accessing files on VMS systems from Mac's now that Pathworks forC Macs is no longer an option (not that it has been a real option for 	 *years*).    		Zane   ------------------------------    Date: 10 Jan 2006 17:30:12 -0800$ From: "AEF" <spamsink2001@yahoo.com>% Subject: Re: Strange SEARCH behaviour B Message-ID: <1136943012.884294.11300@g44g2000cwa.googlegroups.com>   comp.os.vms@hotmail.com wrote: > $ show log str$logG >    "STR$LOG" = "STR$LOCATION:[STR_LIVE.PLU.LOGS]" (LNM$PROCESS_TABLE)  > $ show symb str$warehouse  >   STR$WAREHOUSE = "PLU" , > $ dire/since=9-jan-2006/before=10-jan-2006  ! Can you please show the output of        $ SHOW SYMBOL DIRE  $ on your system? What version of VMS?  C Normally, DIRE would just list files without dates and sizes, so it  must be a symbol.   ( > str$log:strctg6000_'str$warehouse'.log > + > Directory LUXS01$DKA1:[STR_LIVE.PLU.LOGS]  > D > STRCTG6000_PLU.LOG;18890               34   9-JAN-2006 23:57:53.67 >  > ... clip ... > , > $ sear/since=9-jan-2006/before=10-jan-2006. > str$log:strctg6000_'str$warehouse'.log "-F-"! > %SEARCH-E-NOFILE, no file found F > $ sear/since=9-jan-2006 str$log:strctg6000_'str$warehouse'.log "-F-") > %SEARCH-I-NOMATCHES, no strings matched F > $ sear/befor=9-jan-2006 str$log:strctg6000_'str$warehouse'.log "-F-"! > %SEARCH-E-NOFILE, no file found  >  > ' > It seems that /before is not popular.  > G > I can not seem to generate this generally, just in this specific case  > with a 19000+ file count   ------------------------------    Date: 10 Jan 2006 17:49:03 -0800$ From: "AEF" <spamsink2001@yahoo.com>% Subject: Re: Strange SEARCH behaviour C Message-ID: <1136944142.983744.316770@z14g2000cwz.googlegroups.com>    comp.os.vms@hotmail.com wrote: > $ show log str$logG >    "STR$LOG" = "STR$LOCATION:[STR_LIVE.PLU.LOGS]" (LNM$PROCESS_TABLE)  > $ show symb str$warehouse  >   STR$WAREHOUSE = "PLU" , > $ dire/since=9-jan-2006/before=10-jan-2006  ! Can you please show the output of        $ SHOW SYMBOL DIRE  $ on your system? What version of VMS?  C Normally, DIRE would just list files without dates and sizes, so it  must be a symbol.   ( > str$log:strctg6000_'str$warehouse'.log > + > Directory LUXS01$DKA1:[STR_LIVE.PLU.LOGS]  > D > STRCTG6000_PLU.LOG;18890               34   9-JAN-2006 23:57:53.67 >  > ... clip ... > , > $ sear/since=9-jan-2006/before=10-jan-2006. > str$log:strctg6000_'str$warehouse'.log "-F-"! > %SEARCH-E-NOFILE, no file found F > $ sear/since=9-jan-2006 str$log:strctg6000_'str$warehouse'.log "-F-") > %SEARCH-I-NOMATCHES, no strings matched F > $ sear/befor=9-jan-2006 str$log:strctg6000_'str$warehouse'.log "-F-"! > %SEARCH-E-NOFILE, no file found  >  > ' > It seems that /before is not popular.  > G > I can not seem to generate this generally, just in this specific case  > with a 19000+ file count   ------------------------------  % Date: Tue, 10 Jan 2006 19:43:43 -0600 2 From: David J Dachtera <djesys.nospam@comcast.net>% Subject: Re: Strange SEARCH behaviour + Message-ID: <43C462CE.30C48686@comcast.net>    comp.os.vms@hotmail.com wrote: >  > $ show log str$logG >    "STR$LOG" = "STR$LOCATION:[STR_LIVE.PLU.LOGS]" (LNM$PROCESS_TABLE)  > $ show symb str$warehouse  >   STR$WAREHOUSE = "PLU" , > $ dire/since=9-jan-2006/before=10-jan-2006( > str$log:strctg6000_'str$warehouse'.log > + > Directory LUXS01$DKA1:[STR_LIVE.PLU.LOGS]  > D > STRCTG6000_PLU.LOG;18890               34   9-JAN-2006 23:57:53.67 >  > ... clip ... > , > $ sear/since=9-jan-2006/before=10-jan-2006. > str$log:strctg6000_'str$warehouse'.log "-F-"! > %SEARCH-E-NOFILE, no file found F > $ sear/since=9-jan-2006 str$log:strctg6000_'str$warehouse'.log "-F-") > %SEARCH-I-NOMATCHES, no strings matched F > $ sear/befor=9-jan-2006 str$log:strctg6000_'str$warehouse'.log "-F-"! > %SEARCH-E-NOFILE, no file found  > ' > It seems that /before is not popular.  > G > I can not seem to generate this generally, just in this specific case  > with a 19000+ file count   I think the answer is:  , $ sear/since=9-jan-2006/before=10-jan-2006 -/ 	str$log:strctg6000_'str$warehouse'.log;* "-F-"   = (Note the wildcarded version number. DIRECTORY uses a default G specification that includes "*.*;*" while SEARCH only uses ".*" for its 	 default.)    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------    Date: 10 Jan 2006 19:01:07 -0800$ From: "AEF" <spamsink2001@yahoo.com>% Subject: Re: Strange SEARCH behaviour C Message-ID: <1136948098.066483.219990@o13g2000cwo.googlegroups.com>    David J Dachtera wrote:   > comp.os.vms@hotmail.com wrote: > >  > > $ show log str$logI > >    "STR$LOG" = "STR$LOCATION:[STR_LIVE.PLU.LOGS]" (LNM$PROCESS_TABLE)  > > $ show symb str$warehouse  > >   STR$WAREHOUSE = "PLU" . > > $ dire/since=9-jan-2006/before=10-jan-2006* > > str$log:strctg6000_'str$warehouse'.log > > - > > Directory LUXS01$DKA1:[STR_LIVE.PLU.LOGS]  > > F > > STRCTG6000_PLU.LOG;18890               34   9-JAN-2006 23:57:53.67 > >  > > ... clip ... > > . > > $ sear/since=9-jan-2006/before=10-jan-20060 > > str$log:strctg6000_'str$warehouse'.log "-F-"# > > %SEARCH-E-NOFILE, no file found H > > $ sear/since=9-jan-2006 str$log:strctg6000_'str$warehouse'.log "-F-"+ > > %SEARCH-I-NOMATCHES, no strings matched H > > $ sear/befor=9-jan-2006 str$log:strctg6000_'str$warehouse'.log "-F-"# > > %SEARCH-E-NOFILE, no file found  > > ) > > It seems that /before is not popular.  > > I > > I can not seem to generate this generally, just in this specific case  > > with a 19000+ file count >  > I think the answer is: > . > $ sear/since=9-jan-2006/before=10-jan-2006 -1 > 	str$log:strctg6000_'str$warehouse'.log;* "-F-"  > ? > (Note the wildcarded version number. DIRECTORY uses a default I > specification that includes "*.*;*" while SEARCH only uses ".*" for its  > default.)   / Of course! It's either this, or he has DIRE :==  DIRECTORY/SIZE/DATE=EXPIRED !!!   D While the varying defaults for missing file-spec fields are sensibleD (at least from a certain point of view) for commands like DIRECTORY,? COPY, BACKUP, SEARCH, etc., it also means that you cannot use a E DIRECTORY command to see what files will be affected when you use the G same file-spec in other DCL commands! You have to know what the missing D file-spec fields default to for each command you use, or you have to% not leave any file-spec fields empty!    >  > -- > David J Dachtera > dba DJE Systems  > http://www.djesys.com/ > + > Unofficial OpenVMS Hobbyist Support Page: $ > http://www.djesys.com/vms/support/ > * > Unofficial Affordable OpenVMS Home Page:! > http://www.djesys.com/vms/soho/  > $ > Unofficial OpenVMS-IA32 Home Page:! > http://www.djesys.com/vms/ia32/  >  > Coming soon:( > Unofficial OpenVMS Marketing Home Page   ------------------------------  % Date: Tue, 10 Jan 2006 19:25:56 -0600 2 From: David J Dachtera <djesys.nospam@comcast.net>: Subject: Re: Suggestion for INSTALL PURGE and SHUTDOWN.COM+ Message-ID: <43C45EA4.9F717D87@comcast.net>    Bart Zorn wrote: >  > JF wrote:  > G > > Because P1 to P8 symbols are not available from a called procedure, E > > SYSHUTDWN has no way of knowing whether the user has requested an  > > automatic reboot or not. > @ > This is not true. If you do DELETE/SYMBOL/LOCAL P1 in a called@ > procedure, the instance of P1 of the calling procedure becomes$ > available in the called procedure!  G Someone posted that exact solution to a similar query not long ago. I'd D intended to include it in the Advanced portion of my DCL Programming" Hands-On, but I don't think I did.  G I didn't reasearch it, but I'm not at all certain the correspondents at  THAT time were not Bart and JF!    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  % Date: Tue, 10 Jan 2006 19:32:08 -0600 2 From: David J Dachtera <djesys.nospam@comcast.net>: Subject: Re: Suggestion for INSTALL PURGE and SHUTDOWN.COM+ Message-ID: <43C46017.15D8DFB4@comcast.net>   
 R Boyd wrote:  > [snip] > JF Mezei wrote: 
 > > [snip]3 > > Phillip Helbig---remove CLOTHES to reply wrote:  > > > [snip]  E What would be a large help is a sort of a DISMOUNT.COM procedure that H first checks for installed ("known") files on a volume before attemptingA a DISMOUNT, and optionally does INSTALL REMOVE on those images in ; response to being passed an appropriate command parameter.    G I know, that's rather a brash assumption to make at any time other than : system shutdown, and it would need to be invoked AFTER the. application-specific shutdowns have completed.   --   David J Dachtera dba DJE Systems  http://www.djesys.com/  ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/    Coming soon:& Unofficial OpenVMS Marketing Home Page   ------------------------------  % Date: Tue, 10 Jan 2006 19:21:26 -0500 ' From: "Main, Kerry" <Kerry.Main@hp.com> 4 Subject: RE: WVNETcluster uptime reaches 10 years...R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB773B6C@tayexc19.americas.cpqcorp.net>   > -----Original Message-----$ > From: bill@triangle.cs.uofs.edu=20A > [mailto:bill@triangle.cs.uofs.edu] On Behalf Of Bill Gunshannon  > Sent: January 8, 2006 5:46 PM  > To: Info-VAX@Mvb.Saic.Com 6 > Subject: Re: WVNETcluster uptime reaches 10 years... >=20 > In article=20 @ > <FD827B33AB0D9C4E92EACEEFEE2BA2FB773AC3@tayexc19.americas.cpqc > orp.net>, , > 	"Main, Kerry" <Kerry.Main@hp.com> writes: > >=20 > >> -----Original Message----- ) > >> From: bill@triangle.cs.uofs.edu=3D20 D > >> [mailto:bill@triangle.cs.uofs.edu] On Behalf Of Bill Gunshannon" > >> Sent: January 7, 2006 6:39 PM > >> To: Info-VAX@Mvb.Saic.Com9 > >> Subject: Re: WVNETcluster uptime reaches 10 years... 	 > >>=3D20 8 > >> In article <eWHVHo9INQti@eisner.encompasserve.org>,5 > >> 	Kilgallen@SpamCop.net (Larry Kilgallen) writes: A > >> > In article <43BF5649.51D600A6@teksavvy.com>, JF Mezei=3D20 + > >> <jfmezei.spamnot@teksavvy.com> writes:  > >> >> Kenneth Farmer wrote: ? > >> >>> I'm sure it's correct.  I'm wondering if other OS=3D20  > >> fanatics could argue it > >> >>> can be rigged.  > >> >>=3D204 > >> >> This rigging isn't really the question.=3D20 > >> >>=3D20F > >> >> Comparing apples to oranges is. Is it fair to compare the=3D20 > >> uptime of a: > >> >> VMS cluster  against that of individual machines ? > >> >=3D20 F > >> > The purpose of computers is not to keep a particular green=3D20 > >> light on the G > >> > machine lit, but rather to provide some service to humans. =3D20  > >> If VMS isF > >> > able to provide that service on a continuous basis that is=3D20 > >> what counts.  > >> >=3D20 C > >> > The involvement of multiple CPUs, threads of execution,=3D20  > >> power supplies A > >> > etc. is just so much geeky technical trivia not germane=20  > to the issue5 > >> > of whether the service was provided to humans.  > >> >=3D20 A > >> >> If you have 2 Solaris machines that provide HTTP/WEB=3D20  > >> servers and using aG > >> >> router to distribute traffic and stop sending traffic to a=3D20  > >> node that is F > >> >> down, the uptimes project won't report a "cluster" up time butD > >> >> individual nodes uptime even though functionally, those=3D20 > >> solaris boxes= > >> >> would offer about the same uptime than a VMS cluster.  > >> >=3D20 @ > >> > That service provided would not be the same if the web=20 > site involved > > >> > updating.  For read-only applications I have an even=20 > more reliable 1 > >> > technology pre-dating VMS called a "book". 	 > >>=3D20 < > >> And why could the above mentioned Solaris system not=20 > involve updating? @ > >> I have multiple servers with shared file systems so that=20 > any update on < > >> any system is universal.  I can (and do) do rolling =20 > updates so that > > >> system availability is continuous.  There are only two=20 > things missing. G > >> A "cluster uptime" value and thinking it matteried enough to care. 	 > >>=3D20  > >=20= > > And could you let us know what happens to the incoming=20  > writes when the B > > system hosting the writes for other systems via the network=20 > file sharingF > > you are talking about has to be rebooted or just plain halts or is > > powered off? >=20= > As long as we're not talking Linux they just wait til it=20  > comes back up.  H Obviously, if this is the case, then HA is not a requirement. If you areF running a student registration system on Linux in Sept, then a bad cpuH board (pick any HW failure) which needs to come from some place that may* take a day or two will likely be an issue.  G If HA is not a requirement, then clusters is clearly not a requirement.     ? > What happens to the incoming writes on my VAX Cluster when=20  > the HSJ serving A > all my disks dies?  There are failure modes for everything. =20   C Assuming typical cluster config's with volume shadowing across disk 4 controllers - failure is transparent to application.   > I have neverB > said VMS wasn't good at this, I just don't see the purpose of=20 > obsessing B > ove this uptime thing.  I remember when we were being fed the=20 > line thereB > was a "system" (a Vax no less) that had been up for 15 years.=20 >  Not beingA > possible, it eventually turned out to be a Cluster, which is=20 
 > a different  > animal entirely.=20   E If you don't need HA, then the uptime thing is obviously not an issue  for you.   >=20 > >=20< > > Or perhaps you could expand on how each system can do=20 > direct IO's to theH > > storage sub-system without the writes taking the long treck over the? > > network? Most folks think a DLM is required to do direct=20  > IO's from each > > system.  > >=20< > > Or perhaps you could expand on how you would shut one=20 > entire site downG > > without telling the end users in a multi-site config and not impact  > > application availability?  > >=20E > > Here is a pointer to a whitepaper that can refresh readers on the @ > > benefits of clustering and different UNIX implementations=20 > as compared to > > OpenVMS: > >=20@ > http://www.tru64unix.compaq.com/unix/illuminata_dt_unix_resear > ch_note.pd > > f  > >=20 > > Thanks,  >=20> > You just keep on obsessing. Eventually you may figure out=20 > that in the real/ > world most places just aren't that concerned.  >=20  H In a distributed world - perhaps. The impact of the distributed world isA much less than a centralized environment. Course, you pay for the / additional servers, management, monitoring etc.   G The reality in almost all large companies today is that server and data F center (DC) consolidation is one of the hottest topics on most C levelG types radars. They need to cut IT costs big time while at the same time 6 improve service levels and roll-out new stuff as well.  > Also, to be compliant with SOX and other regulatory complianceC requirements, many companies must ensure their IT environment meets  specific pre-determined levels.   E In a centralized world, HA (including server and DC's) is critical as H the costs of failure in terms of regulatory and legal compliance as well6 as applications not being available is extremely high.  A So, your comments about most people not being worried about HA is ; certainly not one shared by most med-large companies today.   = > Oh, by the way, how many VMS systems in New Orleans have=20  > Uptimes of more  > than a year? =20  D Not sure of any companies that had disaster tolerant clusters in NewD Orleans, but servers in clusters are supported up to 800km apart, soG that is all part of the HA planning. The New Orleans disaster is a good G reason why long distance clusters (greater than 100km) are being looked ! at by many large Customers today.   8 Every system in the world is subject to be taken down by@ > factors out of the control of the ones running the system. =20 > Keep tellingB > people about the advantages of VMS (God know's your employers=20
 > won't!) but @ > stop thinking that the results on some web site that tracks=20
 > "Uptime" is , > going to make people dance in the streets.  F I was talking about the technology - not about any news that would get people dancing in the street.   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------  % Date: Tue, 10 Jan 2006 20:09:45 -0500 ' From: "Main, Kerry" <Kerry.Main@hp.com> 4 Subject: RE: WVNETcluster uptime reaches 10 years...R Message-ID: <FD827B33AB0D9C4E92EACEEFEE2BA2FB773B6F@tayexc19.americas.cpqcorp.net>   > -----Original Message-----4 > From: Bill Todd [mailto:billtodd@metrocast.net]=20  > Sent: January 8, 2006 11:55 PM > To: Info-VAX@Mvb.Saic.Com 6 > Subject: Re: WVNETcluster uptime reaches 10 years... >=20 > Main, Kerry wrote: > >>-----Original Message-----& > >>From: bill@triangle.cs.uofs.edu=20 >=20 > ...  >=20? > >>I have multiple servers with shared file systems so that=20  > any update on H > >>any system is universal.  I can (and do) do rolling  updates so that= > >>system availability is continuous.  There are only two=20  > things missing. F > >>A "cluster uptime" value and thinking it matteried enough to care. > >> > >=20 > >=20= > > And could you let us know what happens to the incoming=20  > writes when the B > > system hosting the writes for other systems via the network=20 > file sharingF > > you are talking about has to be rebooted or just plain halts or is > > powered off? >=20A > Why should he bother?  I let you know multiple times several=20  > years ago=20J > back when NFS was the *only* cluster file system supported in Solaris=20= > (the client times out waiting for the server's response,=20  > resubmits the=20H > write which is of course idempotent, and hits the fail-over partner=20? > which has taken over the IP address previously used by the=20  > failed server=20D > - seeing no problem save for what appears to have been a random=20> > connection hiccup which the retry worked around), but you=20 > just sailed on=20 H > obliviously spewing the same incompetent drivel - and are obviously=20 > still doing so.  >=20  H Ok, so a backup server takes over the personality of the primary server.E Now, what happens to the stuff that was running on the backup server? G Since it has a new IP address, what happens to all those connections to # stuff running on the backup server?   H In addition, since one server is hosting all the writes, what happens ifD the primary server gets overloaded, but has not literally "failed"?.D Since that one server is the master for all the other system writes,C what is the impact on the other servers waiting for their writes to 	 complete?    > >=20< > > Or perhaps you could expand on how each system can do=20 > direct IO's to theH > > storage sub-system without the writes taking the long treck over the > > network? >=20? > Perhaps you should ask the author of the white paper which=20  > you yourself=20 @ > cite below:  it states clearly that by the time Sun Cluster=20 > 3.0 appeared=20 @ > they had "a cluster file system with shared-write access to=20 > volumes from=20 = > multiple nodes" (and to the best of my recollection that=20  > means exactly=20" > what one might suspect it does). >=20> > Not, of course, that such access is all that necessary to=20 > most uses of a=20 > > cluster - as demonstrated by the eminent viability of VMS=20 > clusters that=20I > don't have any directly-shared disks.  But it is kind of nice if you=20 1 > want to share a hardware RAID device or need=20  > low-latency/high-bandwidth=20 @ > access to some data from multiple nodes at once (more nodes=20 > than you'd=20 G > care to replicate it at, or for high-bandwidth update-intense loads).  >=20B >   Most folks think a DLM is required to do direct IO's from each > > system.  >=20A > Most folks would be wrong, then:  a central lock manager can=20  > handle that=20I > just fine (and does in quite a few commercially-successful systems -=20 / > SANergy having been one of the earlier ones).  >=20  B A distributed lock manager spreads the cluster traffic around eachE server such that a single node becoming very busy does not negatively , impact the performance on secondary systems.   > >=20< > > Or perhaps you could expand on how you would shut one=20 > entire site downG > > without telling the end users in a multi-site config and not impact  > > application availability?  >=20> > Applications (and application groups) in Solaris have for=20 > several years=20@ > been able to be bound to fail-overable IP addresses just as=20 > file systems=20 A > have - again something which I've explained to you in the past.  >=20  F Yes, fail-over which means you either have a pile of HW sitting at theF remote site either doing nothing waiting for a problem to occur on theD primary or running some secondary loads and/or connections which get) thrown out when the fail-over happens.=20   H In addition, how does one differentiate between a failed primary and oneE that is just extremely slow due to the loads on that primary system - < what process on the solution makes the call to fail-over the application?   > >=20E > > Here is a pointer to a whitepaper that can refresh readers on the @ > > benefits of clustering and different UNIX implementations=20 > as compared to > > OpenVMS: > >=20@ > http://www.tru64unix.compaq.com/unix/illuminata_dt_unix_resear > ch_note.pd > > f  >=202 > A paper almost entirely devoted to discussing=20 > disaster-tolerance rather=20G > than clustering per se and offering rather little insight into the=20 F > details of the latter on the various systems.  If you're going to=20< > encourage people to spend time looking at material, you=20 > should at least=20; > have made the effort to ensure that it has some actual=20  > relevance to the=20  > subject at hand. >=20  B Disaster tolerance is about using HA techniques like clustering toG improve the service levels offered to the end users i.e. the concept of  the IT utility "always on" IT.  F Active-active multi-site clustering such as what OpenVMS offers is notF the only way to improve the availability of an IT service offered on aA single system at a single site. There will always be cost-benefit 0 trade-off decisions that Mgmt will have to make.  E However, it is certainly popular when the following are requirements: @ - both availability and load sharing requirements for both sitesH (including greater than 100km distances) and servers are extremely high;B - requirements for RTO is extremely low (seconds) with RPO of zero committed data loss; - very high security;  - very high stability;     Regards   
 Kerry Main Senior Consultant  HP Services Canada Voice: 613-592-4660  Fax: 613-591-4477  kerryDOTmainAThpDOTcom (remove the DOT's and AT)=20  4 OpenVMS - the secure, multi-site OS that just works.   ------------------------------   End of INFO-VAX 2006.021 ************************