0 INFO-VAX	Wed, 15 Feb 2006	Volume 2006 : Issue 91      Contents:) 2006 OpenVMS Advanced Technical Boot Camp - Re: 2006 OpenVMS Advanced Technical Boot Camp  Re: 4000 vlc motherboard  Re: Alpha CPU testing equipment? Re: Boy, do I like VMS humor!  Re: Boy, do I like VMS humor!  Re: Boy, do I like VMS humor!  Re: Boy, do I like VMS humor!  Re: Boy, do I like VMS humor! % Re: cluster-wide logicals and startup % Re: Excessive paging problem - SOLVED = Re: Gartner wakes up company executives to X86-64 scalability 7 Re: How does a node know it needs to rebuild a volume ? 7 Re: How does a node know it needs to rebuild a volume ? 7 Re: How does a node know it needs to rebuild a volume ? 7 Re: How does a node know it needs to rebuild a volume ? ; Re: how to check for the existence of a logical-name table? ; Re: how to check for the existence of a logical-name table? ; Re: how to check for the existence of a logical-name table? ( Mtools on VMS v. USB flash storage gizmo Re: null terminated strings  safe signals (signal / sigset) Re: Samba info request Re: Samba info request$ TCPIP SET CONF PROTOCOL TCP /FORWARD  F ----------------------------------------------------------------------    Date: 14 Feb 2006 19:09:07 -0800) From: "Sue" <susan_skonetski@hotmail.com> 2 Subject: 2006 OpenVMS Advanced Technical Boot CampB Message-ID: <1139972947.197322.43970@z14g2000cwz.googlegroups.com>   Dear News Group,  F Just giving you an advanced notice that next week I will be opening upE registration for the boot camp.  The following is the list of topics, D the reason you see the same topic listed more than once is that eachD session (other than the general sessions) are offered at least twice and hands on 3-4 times.    I will post the url next week,  * miss you folks, sorry I have been so busy.  
 Warm regards,  Sue      Session Name  C Programming and OpenVMS  C Programming and OpenVMS&  Hyperthreads on Montecito and OpenVMS&  Hyperthreads on Montecito and OpenVMS)  OpenVMS Calling Standard for Programmers )  OpenVMS Calling Standard for Programmers $  OpenVMS Debugging and Tracing Tools$  OpenVMS Debugging and Tracing Tools2 Hands-on overview with OpenVMS Performance Tools "2 Hands-on overview with OpenVMS Performance Tools "4 "Alpha Hardware Virtualization & CockpitMgr Hands-On% A Picture of Performance on Integrity % A Picture of Performance on Integrity " A Survey of Processor Technologies" A Survey of Processor Technologies' A Survey of Virtualization Technologies ' A Survey of Virtualization Technologies  Advanced RMS tuning  Advanced RMS tuning 3 Alpha Hardware Virtualization & CockpitMgr Hands-On C An in-depth look at the ELF object and image file  format and DWARF  debug formatC An in-depth look at the ELF object and image file  format and DWARF  debug format5 An Introduction to SAMBA on  OpenVMS and OpenVMS CIFS 5 An Introduction to SAMBA on  OpenVMS and OpenVMS CIFS # Analyzing objects and images on I64 # Analyzing objects and images on I64 . Backup and Archive automation on OpenVMS - ABS. Backup and Archive automation on OpenVMS - ABS+ Cookbook approach to porting code using GNV + Cookbook approach to porting code using GNV  Detailed OpenVMS Hardware Plans  Detailed OpenVMS Roadmaps ( Developing for OpenVMS from a Windows PC( Developing for OpenVMS from a Windows PC( Developing for OpenVMS from a Windows PC( Developing for OpenVMS from a Windows PC9 Developing highly interactive Web applications on OpenVMS 9 Developing highly interactive Web applications on OpenVMS , Disk partitioning on VMS: LDdriver's secrets, Disk partitioning on VMS: LDdriver's secrets e-Biz and Integration Update e-Biz and Integration UpdateC Finding and analyzing performance problems with OpenVMS Performance  Tools C Finding and analyzing performance problems with OpenVMS Performance  Tools : Hands on Lab - Java wrappers for your OpenVMS Applications: Hands on Lab - Java wrappers for your OpenVMS Applications: Hands on Lab - Java wrappers for your OpenVMS Applications: Hands on Lab - Java wrappers for your OpenVMS Applications Hoff's Hints Hoff's Hints+ Host Based Volume Shadowing Features (HBVS) + Host Based Volume Shadowing Features (HBVS)  How to Spell WSIT  How to Spell WSIT  HP Secure Server HP Secure Server= I64 Alternate Booting - InfoServer, Memory Disk and Sattelite = I64 Alternate Booting - InfoServer, Memory Disk and Sattelite  I64 Cluster Satellite Boot I64 Cluster Satellite Boot Image activation on I64  Image activation on I64 8 Intro to Collaborative Performance Management on OpenVMS8 Intro to Collaborative Performance Management on OpenVMSE Introduction to VMS Engineering Tool "IOHammer Load Generator" / Test  ToolE Introduction to VMS Engineering Tool "IOHammer Load Generator" / Test  Tool Itanium Administrators class Itanium Administrators class Itanium Administrators class Itanium Administrators class0 Java on OpenVMS : Ugly Duckling or Racing Pigeon0 Java on OpenVMS : Ugly Duckling or Racing Pigeon LAN Update & Troubleshooting LAN Update & Troubleshooting Linking applications on I64  Linking applications on I64 % Making the most out of a system crash % Making the most out of a system crash F Maneuvering Through the Integrity Server MP (Management Processor) and) the BMC (Baseboard Management Controller) F Maneuvering Through the Integrity Server MP (Management Processor) and) the BMC (Baseboard Management Controller) , New I/O features in OpenVMS  V8.3 and beyond, New I/O features in OpenVMS  V8.3 and beyond New Improved OpenVMS Security  New Improved OpenVMS Security  OpenVMS Compiler Update  OpenVMS Compiler Update  OpenVMS Directions. OpenVMS Disaster Recovery & Disaster Tolerance. OpenVMS Disaster Recovery & Disaster Tolerance% OpenVMS Information desk - the sequel - OpenVMS Internals Changes V7.3-1 through V8.2 - OpenVMS Internals Changes V7.3-1 through V8.2  OpenVMS Lock Manager Update  OpenVMS Lock Manager Update - OpenVMS on HP Integrity servers -- The Basics - OpenVMS on HP Integrity servers -- The Basics  OpenVMS Process Internals  OpenVMS Process Internals  OpenVMS System Management  OpenVMS System Management / OpenVMS Tranistion Modules - Overview and Usage / OpenVMS Tranistion Modules - Overview and Usage  OpenVMS Utilities update OpenVMS Utilities update; OpenVMS V8.3 Alpha & Integrity CI Measument and Performance ; OpenVMS V8.3 Alpha & Integrity CI Measument and Performance ( OpenVMS Virtualization becomes a reality( OpenVMS Virtualization becomes a reality6 Oracle 9iRAC with OpenVMS HBVS over extended distances6 Oracle 9iRAC with OpenVMS HBVS over extended distances7 Oracle Rdb on I64, how you can get the best performance 7 Oracle Rdb on I64, how you can get the best performance 
 Oracle Update " Overview of Performance Management" Overview of Performance Management Printing from OpenVMS with DCPS  Printing from OpenVMS with DCPS ; SOA and OpenVMS: Principles and Practices with Web Services ; SOA and OpenVMS: Principles and Practices with Web Services  Storage Hands On Storage Hands On Storage Hands On Storage Hands On  Storage/Cluster Technical Update  Storage/Cluster Technical Update' T4 & Friends Advanced Hands-on Workshop ' T4 & Friends Advanced Hands-on Workshop ( TCP/IP in High Availability Environments( TCP/IP in High Availability Environments TCP/IP Networking  TCP/IP Networking , The OpenVMS UNIX Portability (UP) Initiative, The OpenVMS UNIX Portability (UP) Initiative The Performance Cookbook The Performance CookbookB The Performance Data Collector Application Integration for OpenVMSB The Performance Data Collector Application Integration for OpenVMS( Troubleshooting Hung & Looping Processes( Troubleshooting Hung & Looping Processes, Understanding the Intel Itanium Architecture, Understanding the Intel Itanium Architecture Use of System Service Logging  Use of System Service Logging A Using LAMP (Linux/Apache/MySQL/Perl-PHP) to  Re-VAMP your OpenVMS  SystemA Using LAMP (Linux/Apache/MySQL/Perl-PHP) to  Re-VAMP your OpenVMS  System4 Using SDA to analyze an Alpha Crash , A Crash Course4 Using SDA to analyze an Alpha Crash , A Crash Course0 Using SDA To Analyze OpenVMS Alpha Process Dumps0 Using SDA To Analyze OpenVMS Alpha Process DumpsC Using state of the art tools to develop Software for OpenVMS in the  21st CenturyC Using state of the art tools to develop Software for OpenVMS in the  21st Century' Using Vevamon to manage EVA performance ' Using Vevamon to manage EVA performance ) Why Integrity is the right choice for VMS  Introduction to OpenVMS  Introduction to OpenVMS / WASD: why is it chosen when there's VMS Apache? / WASD: why is it chosen when there's VMS Apache?    ------------------------------    Date: 14 Feb 2006 19:18:38 -0800) From: "Sue" <susan_skonetski@hotmail.com> 6 Subject: Re: 2006 OpenVMS Advanced Technical Boot CampB Message-ID: <1139973518.665292.39030@g44g2000cwa.googlegroups.com>  A Registration will be limited and the tuition is $1495 we start on E Sunday May 21 and end on Friday May 26th and the boot camp will be at ! the Sheraton hotel in Nashua, NH.   E If you have never been before we start at 7:45 AM and end around 9 PM 4 Monday, Tuesday and Thursday and possibly Wednesday.  C Breakfast, Lunch and breaks everyday are included as well as dinner  Tuesday and Thursday.    Sue    ------------------------------    Date: 14 Feb 2006 18:19:15 -0800" From: chris_doran@postmaster.co.uk! Subject: Re: 4000 vlc motherboard C Message-ID: <1139969955.551770.175310@z14g2000cwz.googlegroups.com>    Cliff Miller wrote: G > I've a 4000 vlc motherboard on the way.  I hope to decipher the power @ > connection and provide it from a non-traditional source.  What$ > challenges do I face?  Any advice? > H > There is no graphics card - will the MB work without one?  I believe I? > have memory that will work, already.  Any suggestions will be   > appreciated!  ---thanks, Cliff  D The power connections I've just measured, viewed from the top of the board are:-    |  |                     +-------+  |                     |       | & |              unused | o   o | unused |                     |       | & |                +5.2 | o   o | unused |                     |       | ! |                +5.2 | o   o | 0   |                     |       ||& |                   0 | o   o ||unused  |                     |       ||& |                   0 | o   o | unused |                     |       | # |                   0 | o   o | +12  |                     |       | $ |                 -12 | o   o | +5.2 |                     |       |  |                     +-------+  |  |  | 1 +------------------------------------------------   G +5.2V is what I measured; I'd think 5.0 will be OK. There don't seem to F be any special sense lines to support unlike other DEC models. The PSU> is rated 105W, so I guess a similar PC supply can be used. I'dE recommend a higher wattage if possible, as the VLC's normal one seems C to be only just up to the job and gives up on the more power-hungry  disks.  G The memory is 72-pin as used in early PCs (I'm not sure which speed) to  a maximum of 24Mb.  F I can't say what happens without the graphics board, but I guess I can try it if no one else knows.   Chris    ------------------------------  % Date: Tue, 14 Feb 2006 12:47:05 -0600 / From: Chris Scheers <chris@applied-synergy.com> ) Subject: Re: Alpha CPU testing equipment? 2 Message-ID: <43F225A9.6030705@applied-synergy.com>  
 cmk wrote: > Hi group!  >  > - > I've found something strange at scrap-shop.  > ( > It is an unit size of an ossciloscope,0 > which is claimed to be an "ALPHA CPU TEST SET") >    No sign of an indentification plate.  >    Manufacturer unknown. > A > I'm unable to verify that such devices were ever sold publicly. ; > Google is mum about actual equipment - seems to talk only @ > in general about alpha's functional testing - mostly performed > on complete systems. > B > I doubt this is a real thing, and don't believe it could be usedE > (or usefull) in the wild. Previous owner is said to be Symbol Tech. 7 > the peolpe who make bar-code, RFID scanners and such.  >  > < > Has anyone seen anything alike? How should I check if it's? > the real thing? (nice to have - from enthusiast perspective).  > F > Maybe someone can tell a bit about Alpha's testing at manufacturing?C > (or Itanic - it is of interest how these 3-D colour X-ray imaging  > technologiesD > which Intel bought along witch NewFocus Inc. are actually applied)   Is there a date on it?  @ Data General had a single chip CPU that was sort of an enhanced F Nova/stripped Eclipse that they referred to as the "Alpha".  This was 2 used extensively in their intelligent controllers.  B If this is for the Data General chip, I know someone who might be  interested in it.    --  G ----------------------------------------------------------------------- $ Chris Scheers, Applied Synergy, Inc.  B Voice: 817-237-3360            Internet: chris@applied-synergy.com    Fax: 817-237-3074   ------------------------------    Date: 14 Feb 2006 11:29:07 -0600 From: gleason@encompasserve.org & Subject: Re: Boy, do I like VMS humor!3 Message-ID: <3Zw$a+Lpl8fA@eisner.encompasserve.org>   i In article <6.1.2.0.2.20060214070152.027819c0@raptor.psccos.com>, Dan O'Reilly <dano@process.com> writes: D > Of course, there's always the well-reviewed humor of Datatrieve & K > wombats.  In DTR-11 (I don't know about DTR-32), if you tried to use, in  I > the report generator, "AT MIDDLE OF PAGE", it would come back with "AT  3 > MIDDLE OF PAGE is innovative, but alas, illegal".  > M > But one of my favorites was the RSX error code that you got from DECnet if   > a link dropped: IE.NFW.  > *   And the numeric value of IE.NFW was -69.   Lee K. Gleason N5ZMR Control-G Consultants  lgleason@houston.rr.com    ------------------------------  + Date: Tue, 14 Feb 2006 23:08:08 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)& Subject: Re: Boy, do I like VMS humor!$ Message-ID: <dstnso$i5i$4@online.de>  F In article <43f1dded$0$1231$91cee783@newsreader01.highway.telekom.at>,) "Walter Kuhn" <w.kuhn@ksg.co.at> writes:    % > $ WRITE SYS$OUTPUT F$MESSAGE (2928)   > I knew that some things were not implemented on VAX, but I am 6 disappointed that amphibious vehicles are among these.   ------------------------------  + Date: Tue, 14 Feb 2006 23:08:35 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)& Subject: Re: Boy, do I like VMS humor!$ Message-ID: <dstntj$i5i$5@online.de>  B In article <45e5svF67gfrU1@individual.net>, bill@cs.uofs.edu (Bill Gunshannon) writes:   H > In article <43f1dded$0$1231$91cee783@newsreader01.highway.telekom.at>,+ > 	"Walter Kuhn" <w.kuhn@ksg.co.at> writes:  > > Or, how about  > > ' > > $ WRITE SYS$OUTPUT F$MESSAGE (2928)  > >  > * > %SYSTEM-S-NOMSG, Message number 00000B71  & You just outed yourself as a VAX user!   ------------------------------  % Date: Tue, 14 Feb 2006 18:54:37 -0500 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> & Subject: Re: Boy, do I like VMS humor!, Message-ID: <43F26DB1.3E67288D@teksavvy.com>  / Phillip Helbig---remove CLOTHES to reply wrote: , > > %SYSTEM-S-NOMSG, Message number 00000B71 > ( > You just outed yourself as a VAX user!    E When roughly 1/3 of the remaining installed base is VAX based, nobody 2 should be ashamed to declare themselves VAX users.  E And consider that Charon VAX will allow VAX-VMS to run on modern 8086 1 hardware while the other platforms are idle/late.    ------------------------------   Date: 15 Feb 2006 00:15:13 GMT( From: bill@cs.uofs.edu (Bill Gunshannon)& Subject: Re: Boy, do I like VMS humor!+ Message-ID: <45fa4hF6duh0U1@individual.net>   $ In article <dstntj$i5i$5@online.de>,S 	helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes: D > In article <45e5svF67gfrU1@individual.net>, bill@cs.uofs.edu (Bill > Gunshannon) writes:  > I >> In article <43f1dded$0$1231$91cee783@newsreader01.highway.telekom.at>, , >> 	"Walter Kuhn" <w.kuhn@ksg.co.at> writes: >> > Or, how about >> >  ( >> > $ WRITE SYS$OUTPUT F$MESSAGE (2928) >> >   >>  + >> %SYSTEM-S-NOMSG, Message number 00000B71  > ( > You just outed yourself as a VAX user!  A Outed myself???  I have always been proud of my VAXen!!  The only ' systems I like better are my PDP-11's!!    bill      --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------  + Date: Tue, 14 Feb 2006 22:51:55 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply). Subject: Re: cluster-wide logicals and startup$ Message-ID: <dstmub$i5i$1@online.de>  D In article <dsr2f9$u2p$1@online.de>, helbig@astro.multiCLOTHESvax.de3 (Phillip Helbig---remove CLOTHES to reply) writes:    6 > In article <43F0FD4C.74AC02D@teksavvy.com>, JF Mezei) > <jfmezei.spamnot@teksavvy.com> writes:   > 3 > > Phillip Helbig---remove CLOTHES to reply wrote: F > > > in SYLOGICALS.COM.  I need the cluster-wide logicals so that, ifE > > > necessary, I can use them to stall the startup while a MINICOPY * > > > initiated on another node completes. > > L > > Is that just so that the booting node doesn't become responsible for theB > > minicopy or whatever shadow operation it feels is neceessary ? > G > Yes, since VAX can't do MINICOPY.  When a MOUNT is done and a COPY is K > needed, then ANY node can handle the COPY; it doesn't have to be the one  I > from which the MOUNT command came and it doesn't have to have a direct  I > connection to any of the members of the shadow set.  For a MINICOPY to  F > happen, the MOUNT command cannot come from a VAX and the node which K > handles the COPY cannot be a VAX (and, of course, the member needing the  K > copy has to have been DISMOUNTed with /POLICY=MINICOPY, though this will  I > be relaxed in a future version of VMS).  Thus, I define a cluster-wide  F > logical before I do the dismount.  At boot, I check this.  If it is G > defined, the startup waits after setting SHADOW_MAX_COPY to 0 on the  I > booting node.  I then set SHADOW_MAX_COPY to 0 on all other VAXes (the  J > booting node has to be done separately from within SYLOGICALS.COM since K > SYSMAN isn't running yet) and issue a MOUNT command.  After the MINICOPY  : > stars, I deassign the logical and the startup continues.  G Sounds complicated, but I am running quite old hardware.  My main user  H disk is a pair of 9-GB SEAGATE disks hosted by a VAXStation 4000/90 and I a VAX 105A (used to be a 4000/60 and a 100/A).  A full copy took about 7  G hours and choked up the network.  In a typical case (rebooting the VAX  G after applying some patches, configuring new hardware, changing "hard"  I parameters etc---downtime of half an hour at most), the minicopy takes a  + few seconds.  Definitely worth the trouble.    ------------------------------  % Date: Tue, 14 Feb 2006 11:38:31 -0500 , From: "Richard Tomkins" <tomkinsr@istop.com>. Subject: Re: Excessive paging problem - SOLVED< Message-ID: <43f2086b$0$8056$6d36acad@titian.nntpserver.com>  ; Something to do with Dynamic Garbage Collection of Classes.   H -X are non-standard Java VM options and are subject to arbitrary change.  I This one may be specific to OpenVMS only, as it does not show up in my PC % version. It could also be deprecated.   F Execute a command line, java -h or java -X and you can see the various  options available to run the VM.    ) <norm.raphael@metso.com> wrote in message K news:OF65AE4BF8.1BB85835-ON85257115.004FAA12-85257115.004FC2B3@metso.com...  >  > E > Paul Sture <paul.sture@bluewin.ch> wrote on 02/13/2006 10:25:29 AM:  >  > > Bernhard Dorninger wrote:  > > > Hello again! > > >  > > > , > > > For those, who might be interested.... > > > C > > > After experimenting with the Java cmd line args and long time . > > > observation, we finally found a solution > > > L > > > Activating the switch -Xdynclassgc did the trick. The Java VM now runs  > > > with a stable working set. > > >  > > ) > > Thanks for letting us know, Bernhard.  > 3 > ..and for the uninitiated, that switch does what?  >     . *** Free account sponsored by SecureIX.com ***X *** Encrypt your Internet usage with a free VPN account from http://www.SecureIX.com ***   ------------------------------  % Date: Tue, 14 Feb 2006 23:05:22 +0100 = From: =?ISO-8859-1?Q?Markus_D=F6hr?= <wantbottom@t-online.de> F Subject: Re: Gartner wakes up company executives to X86-64 scalability/ Message-ID: <dstk7c$sdh$02$1@news.t-online.com>    Alan Greig wrote:  >  >  > Main, Kerry wrote: >  >>D >> As an example - Oracle charges $40K/CPU list (Linux, Windows, anyJ >> platform). Add $50% /CPU for RAC clustering sw. So, assuming one gets aI >> corp discount to $30K/cpu, it does not take to much math to figure out F >> that a Linux DB x86 cluster with individual servers running at onlyH >> 10-20% busy is not a good investment. And then there is the annual SW9 >> maint costs that can run 10%-25% of the SW list costs.  > F > Danger here is this argument can turned on its head and be taken as K > implying that X64 based systems are far more powerful then what they are  E > typically used for. In other words why should firms consolidate on  7 > Itanium when X64 based hardware has so much headroom.  >  >>I >> Its kind of like buying a brand new expensive car, but only driving it ' >> for a few hours on Sunday each week.  > G > If that's how you describe a "Linux DB x86 cluster" I'd hate to hear  B > your description of a Superdome/HP-UX/SAP system running at 10% + > utilization :-) And I bet there are some!  > G > And without getting too far into Oracle pricing black-magic - last I  K > checked Oracle licenses were cheaper for multi-core X64 than the cost to  L > achieve a similar Oracle benchmark on currently available Itanium systems. >   A The year before last we heavily discussed with HP sales reps, HP  I technicans and SAP what road to choose. HP said: "Go for Itanium, it's a  G future investment, it's fast and scalable". SAP said "we will postpone  * the x86_64 port for at least 6 months...".  I A couple of weeks later on the CeBIT fair we got asked if we can "return  G the IA64 boxes and replace by x86_64" - question from the SAP people -  G since they saw, that the market is questioning the availability of SAP  I on x86_64 (Linux). They are even thinking of dropping support for SAP on  6 Itanium, Windows and Linux and just go on with x86_64.  E As of now our production is running on an 4-way IA64 system. We have  H several multi-core systems as well - and they are performing exceedinly < better than the IA64 ones; especially when it comes to java.  E We can buy two 64Gig 4way dualcore AMDs for the same price as of one  H IA64 with two CPUs and 32GB - and that's what we are doing. It's easier H to have a spare box around to just plug the cables in case something is H going wrong. It's MUCH cheaper then using an IA64 box with HPVM, WLM or D anything like that, in purchasing, maintenance and support contract.     [...]    --   Markus  D (who's not a VMS user but an active reader of this very interesting 
 newsgroup)   ------------------------------  % Date: Tue, 14 Feb 2006 16:11:22 -0500 . From: JF Mezei <jfmezei.spamnot@vaxination.ca>@ Subject: Re: How does a node know it needs to rebuild a volume ?- Message-ID: <43F24778.99CA1813@vaxination.ca>    Michael Moroney wrote:I > These counts normally match.  If a node crashes without dismounting the I > disk, the lock count decrements but SCB$W_WRITECNT doesn't.  If $ MOUNT H > sees a mismatch it either rebuilds the volume or prints a message that > the rebuild is needed.     Ok, so in a 4 node cluster:   H node 1 is brought down gracefully. The SCB$W_WRITECNT is decremented and4 still matches the cluster-wide lock for mount count.   node 2 crashes.   G If you then reboot node 1 , it will then do the /REBUILD even though it " wasn't the one to cause the mess ?    6 During a rebuild, does activity in other nodes stall ?   ------------------------------    Date: 14 Feb 2006 13:53:24 -0800) From: "Ken Robinson" <kenrbnsn@gmail.com> @ Subject: Re: How does a node know it needs to rebuild a volume ?B Message-ID: <1139954004.837944.88680@z14g2000cwz.googlegroups.com>   JF Mezei wrote (in part): I > If you then reboot node 1 , it will then do the /REBUILD even though it $ > wasn't the one to cause the mess ?  D When a node goes to mount a disk, it doesn't know or care which node: caused the disk to need a rebuild, just that it needs one.  C You can always specify "/norebuild" on the mount and do the rebuild 7 when you know there won't be much activity on the disk.    The    $ sho dev/rebuild d   6 command will tell you which disks  need to be rebuilt.   Ken    ------------------------------    Date: 14 Feb 2006 14:27:02 -0800 From: davidc@montagar.com @ Subject: Re: How does a node know it needs to rebuild a volume ?C Message-ID: <1139956022.184814.188550@g14g2000cwa.googlegroups.com>    JF Mezei wrote:  > Michael Moroney wrote:K > > These counts normally match.  If a node crashes without dismounting the K > > disk, the lock count decrements but SCB$W_WRITECNT doesn't.  If $ MOUNT J > > sees a mismatch it either rebuilds the volume or prints a message that > > the rebuild is needed. >  >  > Ok, so in a 4 node cluster:  > J > node 1 is brought down gracefully. The SCB$W_WRITECNT is decremented and6 > still matches the cluster-wide lock for mount count.   Yes.   > node 2 crashes.  > I > If you then reboot node 1 , it will then do the /REBUILD even though it $ > wasn't the one to cause the mess ?  G Yes.  It doesn't matter who crashes, the WRITECNT's don't match up with G expectations, and the next node to mount the disk will rebuild it (even * if it's currently mounted by other nodes).  8 > During a rebuild, does activity in other nodes stall ?  @ Mostly.  Read-only processes will not be impacted, but processesC writing to the volume will likely always stall until the rebuild it 	 complete.   G This is why some people use MOUNT/NOREBUILD at system startup to insure G that stalls during a rebuild can be done at a convenient time - not the  next node reboot.    ------------------------------  + Date: Tue, 14 Feb 2006 22:52:47 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney) @ Subject: Re: How does a node know it needs to rebuild a volume ?( Message-ID: <dstmvv$lij$1@pcls4.std.com>  0 JF Mezei <jfmezei.spamnot@vaxination.ca> writes:   >Michael Moroney wrote: J >> These counts normally match.  If a node crashes without dismounting theJ >> disk, the lock count decrements but SCB$W_WRITECNT doesn't.  If $ MOUNTI >> sees a mismatch it either rebuilds the volume or prints a message that  >> the rebuild is needed.      >Ok, so in a 4 node cluster:  I >node 1 is brought down gracefully. The SCB$W_WRITECNT is decremented and 5 >still matches the cluster-wide lock for mount count.    Yes.   >node 2 crashes.  H >If you then reboot node 1 , it will then do the /REBUILD even though it# >wasn't the one to cause the mess ?   E Yes, assuming the mount wasn't specified as /NOREBUILD.  There really H isn't anything node-specific about the "mess" as you put it.  The "mess"E consists of file headers and disk blocks marked as allocated in their I respective bitmaps when they, in fact, are free.  They were pre-allocated J by Node 2 so that Node 2 didn't have to go to the bitmap to allocate stuffE every time it wanted to create or extend a file. There are some other F things that may need to be cleaned, but other than disk quota, I don't remember what they are.   7 >During a rebuild, does activity in other nodes stall ?   H File creation and extension will stall, at least when the relevant stuffG is being rebuilt.  These may not stall (I don't know) if the local node J can satisfy them from its own cache, but I think all the caches are wiped. Other I/O shouldn't stall.   ------------------------------  + Date: Tue, 14 Feb 2006 18:47:03 +0000 (UTC) - From: klewis@OMEGA.MITRE.ORG (Keith A. Lewis) D Subject: Re: how to check for the existence of a logical-name table?. Message-ID: <dst8j7$r89$1@newslocal.mitre.org>   Ryan Moore <rmoore@rmoore.dyndns.org> writes in article <Pine.LNX.4.64.0602120916470.18998@jaipur.local> dated Sun, 12 Feb 2006 17:19:14 GMT: L >F$TRNLNM will tell you if a table exists or not.  Look it up in the system  >directory:  > M >$ write sys$output f$trnlnm("table_name", "lnm$system_directory",,,,"table")  > E >This will return "TRUE" when the table exists and "" when the table   >doesn't exist.   H And "FALSE" if it exists but it's not a table, just a name.  This can be somewhat misleading.  I $ write sys$output f$trnlnm("NOTHING", "lnm$system_directory",,,,"table")   L $ write sys$output f$trnlnm("LNM$SYSTEM", "lnm$system_directory",,,,"table") FALSE R $ write sys$output f$trnlnm("LNM$SYSTEM_TABLE", "lnm$system_directory",,,,"table") TRUE  0 --Keith Lewis              klewis {at} mitre.org> The above may not (yet) represent the opinions of my employer.   ------------------------------    Date: 14 Feb 2006 18:12:51 -0800$ From: "AEF" <spamsink2001@yahoo.com>D Subject: Re: how to check for the existence of a logical-name table?C Message-ID: <1139969571.077054.282450@g43g2000cwa.googlegroups.com>    Keith A. Lewis wrote:  > Ryan Moore <rmoore@rmoore.dyndns.org> writes in article <Pine.LNX.4.64.0602120916470.18998@jaipur.local> dated Sun, 12 Feb 2006 17:19:14 GMT: M > >F$TRNLNM will tell you if a table exists or not.  Look it up in the system 
 > >directory:  > > O > >$ write sys$output f$trnlnm("table_name", "lnm$system_directory",,,,"table")  > > F > >This will return "TRUE" when the table exists and "" when the table > >doesn't exist.  > J > And "FALSE" if it exists but it's not a table, just a name.  This can be > somewhat misleading. > K > $ write sys$output f$trnlnm("NOTHING", "lnm$system_directory",,,,"table")  > N > $ write sys$output f$trnlnm("LNM$SYSTEM", "lnm$system_directory",,,,"table") > FALSE T > $ write sys$output f$trnlnm("LNM$SYSTEM_TABLE", "lnm$system_directory",,,,"table") > TRUE  : So to use this method you would need the equivalent of theF TRANSLATE_QUEUE function of F$GETQUI(), but for table names instead ofA queue names. Other than that, you'd have to do your own iterative F logical-name translation. (Unless something new's been added since VMS v6.2.)   > 2 > --Keith Lewis              klewis {at} mitre.org@ > The above may not (yet) represent the opinions of my employer.   AEF    ------------------------------  % Date: Tue, 14 Feb 2006 20:27:22 -0600 2 From: David J Dachtera <djesys.nospam@comcast.net>D Subject: Re: how to check for the existence of a logical-name table?+ Message-ID: <43F2918A.D81C5081@comcast.net>    "Keith A. Lewis" wrote:  >  > Ryan Moore <rmoore@rmoore.dyndns.org> writes in article <Pine.LNX.4.64.0602120916470.18998@jaipur.local> dated Sun, 12 Feb 2006 17:19:14 GMT: M > >F$TRNLNM will tell you if a table exists or not.  Look it up in the system 
 > >directory:  > > O > >$ write sys$output f$trnlnm("table_name", "lnm$system_directory",,,,"table")  > > F > >This will return "TRUE" when the table exists and "" when the table > >doesn't exist.  > J > And "FALSE" if it exists but it's not a table, just a name.  This can be > somewhat misleading. > K > $ write sys$output f$trnlnm("NOTHING", "lnm$system_directory",,,,"table")  > N > $ write sys$output f$trnlnm("LNM$SYSTEM", "lnm$system_directory",,,,"table") > FALSE T > $ write sys$output f$trnlnm("LNM$SYSTEM_TABLE", "lnm$system_directory",,,,"table") > TRUE   So, you could do this:   $ got_it = -B   f$trnlnm("SYSMAN$NODE_TABLE", "lnm$system_directory",,,,"table")8 $ if (got_it .nes. "") .and. got_it then ...	! It exists   - or -A $ if (got_it .eqs. "") .or. (.not. got_it) then ... ! Not a table    --   David J Dachtera dba DJE Systems  http://www.djesys.com/  & Unofficial OpenVMS Marketing Home Page! http://www.djesys.com/vms/market/   ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/   " Unofficial OpenVMS-IA32 Home Page: http://www.djesys.com/vms/ia32/   ) Unofficial OpenVMS Hobbyist Support Page: " http://www.djesys.com/vms/support/   ------------------------------  + Date: Tue, 14 Feb 2006 18:42:04 -0600 (CST) * From: sms@antinode.org (Steven M. Schweda)1 Subject: Mtools on VMS v. USB flash storage gizmo 2 Message-ID: <06021418420405_20331674@antinode.org>  G    The local computer store recently advertised a semi-bargain price on F a Kingston USB flash memory gizmo (512MB), so I finally broke down andF invested in one.  After only a bit of fooling around (sucking down andH installing DNDRIVER), I plugged it into the cheap ($10) Chinese USB cardB (with the NEC PD720100AGM chip) in my (spare/experimental) XP1000F running VMS V8.2, and it seemed to work about as expected (some of the> time, anyway).  I'm not an actual paying customer, but it I doF appreciate the effort that went into making this stuff work.  (But see below for some whining.)  D    After a bit of fiddling, I also got a recent version of Mtools toF work a little (within limits), and it was apparently able to cope withA some data written using a Windows 98 SE system.  For the morbidly  curious:  ! ALP2 $ show device /full MTOOLS_U   O Disk ALP2$DNA0:, device type KingstonDataTraveler 2.0, is online, file-oriented F     device, shareable, available to cluster, error logging is enabled.  O     Error count                    0    Operations completed               3136 O     Owner process                 ""    Owner UIC                      [SYSTEM] O     Owner process ID        00000000    Dev Prot            S:RWPL,O:RWPL,G:R,W O     Reference count                0    Default buffer size                 512 O     Total blocks             1001471    Sectors per track                    32 O     Total cylinders             1010    Tracks per cylinder                  31      ALP2 $ MTOOLS -c mdir  u:    MTOOLS_U (_ALP2$DNA0:) mounted.  Volume in drive U has no label "  Volume Serial Number is 1B00-1A18 Directory for U:/   ( DTI      EXE   1236215 2006-02-12  19:20D KINGST~1     <DIR>     2006-02-12  19:53  KINGSTON TECHNOLOGY CO INC)         2 files           1 236 215 bytes .                         508 944 384 bytes free    _ALP2$DNA0: dismounted.   ALP2 $ mtools -V. MTOOLS version 3.9.10a0, dated March 2nd, 2005P configured with the following options: disable-xdf disable-vold disable-new-vold  disable-debug disable-raw-term     G    This Mtools stuff is still pretty ragged.  Wildcard file names don't A work, for example, and I'd be willing to bet that there are other H problems lurking near the surface (if not actually sticking out of it).   H    In the VMS complaint department, I hit an annoyance in the UCM, where> SHOW EVENTS /BEFORE or /SINCE demands a time value, instead of0 defaulting to TODAY (which is common elsewhere):  
 ALP2 $ ucm8 Universal Serial Bus Configuration Manager, Version V1.0   UCM> show events /since N %CLI-W-VALREQ, missing qualifier or keyword value - supply all required values  \SINCE\  F I assume that a minor adjustment to the command definition should cure" that one. (Well, it _was_ "V1.0".)  F    Also, I haven't figured out how to reproduce it (or stop it when itG happens), but I have seen situations where an INITIALIZE or MOUNT would E hang (RWAST), and then yanking the USB gizmo would crash the system.  F And yanking the gizmo and then plugging it into a different port seems5 to cause some confusion.  But it's fun when it works.   H    Just curious: Does mounting one of these things with /NOCACHE make it  safe(r) to yank before DISMOUNT?  H ------------------------------------------------------------------------  4    Steven M. Schweda               (+1) 651-699-98183    382 South Warwick Street        sms@antinode-org     Saint Paul  MN  55105-2547    ------------------------------  + Date: Tue, 14 Feb 2006 18:15:41 +0000 (UTC) - From: klewis@OMEGA.MITRE.ORG (Keith A. Lewis) $ Subject: Re: null terminated strings. Message-ID: <dst6od$pgd$2@newslocal.mitre.org>  t bill@cs.uofs.edu (Bill Gunshannon) writes in article <453cckF4ov0hU2@individual.net> dated 10 Feb 2006 11:40:04 GMT:/ >In article <dsgvtp$3j5$1@newslocal.mitre.org>, 1 >	klewis@OMEGA.MITRE.ORG (Keith A. Lewis) writes:  >> Dave Froble <davef@tsoft-inc.com> writes in article <GdednZ5VLKVpb3beRVn-iQ@libcom.com> dated Thu, 09 Feb 2006 20:57:47 -0500:  >>> 0 >>>The little I looked at Java reminded me of C. >>  O >> Java's primitive statements and blocs mimic those of C, but that's where the N >> similarity ends.  Java is a strongly typed language.  You can typecast, butL >> typecasts are checked at runtime.  There is no such thing as a pointer.   >>   > E >Yes, Java became our starting language for the first two programming I >courses of our CS and CIS program (replacing Ada, which replaced Pascal) K >a few years ago.  Just the other day one of the professors was complaining E >about the lack of pointers because there just are times when using a H >pointer is the best way to do something.  And this particular professorJ >has probably never written a C program more complicated tha "hello world"& >so we are not talking a C bigot here.  J Just recently I was using a package that had been ported from C to Java.  G Where the C code (this is a guess, as I never saw the C code) had added L something to a byte-array pointer and passed the new value as a pointer, theK guy doing the port copied the array.  This was inefficient but it worked... C until I used it on a set of input data which re-used the header and L re-loaded only the data part of the record from the file.  The copied arrays. were still loaded with old data, imagine that!  I Of course I'm not stuck in the pointer rut.  I created a new class called J ByteArrayReference, which keeps track of the original array object and theI "offset" into it.  Small overhead, better than pointer arithmetic because K the checks are more robust.  No possibility of buffer overrun exploits in a  JVM!  0 --Keith Lewis              klewis {at} mitre.org> The above may not (yet) represent the opinions of my employer.   ------------------------------    Date: 14 Feb 2006 12:16:20 -05007 From: colonel@monmouth.com (World's Largest Leprechaun) ' Subject: safe signals (signal / sigset) - Message-ID: <dst394$a53$1@shell.monmouth.com>   E We have a threaded VMS application that tries to catch floating-point ? exceptions and jump back in.  In a stress test where 50 threads ? divide by zero at once, only a few threads catch the trap.  The @ others seem to be proceeding as if the division were legal.  TheB jump buffers are unique to their threads, unless there's something+ wrong with thread-specific keys in OpenVMS.   B The code is ported from Linux to GNV.  We want the signal handlersG to be on at all times, so we would use sigset(), but it's not available F with GNV.  We use sigaction() instead, which should leave the handlersE on when they trigger.  I would hope that when a thread uses a handler B it masks the signal only in itself, not in other threads.  I wouldB also hope that sigaction() without RESETHANDLER leaves the handlerH in place, but I can't find any documentation about it.  As I understand,H W*****s always clears the handler on invoking it, so I have doubts about& VMS.  Does anybody here know for sure?   -:- ; 	"Oh, you don't need to have a moment's concern about that, : 	 Grampa.  Tippy's always at his best in time of disaster.9 	 He'll be here with the lumber any minute now.  He's the 0 	 finest, smartest dog in the whole wide world!"  " 				--Bob Elliott and Ray Goulding --   Col. G. L. Sicherman home: colonel@mail.monmouth.com  work: sicherman@att.com ( web: <http://www.monmouth.com/~colonel/>   ------------------------------  % Date: Tue, 14 Feb 2006 13:28:19 -0500  From: "Jilly" <jilly@hp.com> Subject: Re: Samba info request , Message-ID: <43f22144$1@usenet01.boi.hp.com>  E <abirkett@csc.com> wrote in message news:dst2lq$frd$1@lore.csc.com... J > Just in case you've seen this elesewhere, I'll post the CLUE CRASH info. > K > Unfortunately, an upgrade isn't possible at this point as the base system G > mimics a customers of theirs but my client requires Samba to transfer * > files. Thanks for the suggestion anyway. >  >  >   > Crashdump Summary Information:  > ------------------------------, > Crash Time:        10-FEB-2006 14:18:00.51= > Bugcheck Type:     INVEXCEPTN, Exception while above ASTDEL 1 > Node:              XXX-removed-XXX  (Clustered) * > CPU Type:          AlphaServer 800 5/400 > VMS Version:       V7.1-1H2  > Current Process:   SMBD_BG522 : > Current Image:     $15$DKA200:[SAMBA228.][BIN]SMBD.EXE;2< > Failing PC:        FFFFFFFF.8AC35C38    UCX$BGDRIVER+41C38& > Failing PS:        00000000.00000803! > Module:            UCX$BGDRIVER  > Offset:            00041C38   K I do not find any record of this crash profile having been seen by the CSC.    Jilly    ------------------------------  + Date: Tue, 14 Feb 2006 12:06:03 -0500 (EST)  From: abirkett@csc.com Subject: Re: Samba info request ' Message-ID: <dst2lq$frd$1@lore.csc.com>   H Just in case you've seen this elesewhere, I'll post the CLUE CRASH info.  J Unfortunately, an upgrade isn't possible at this point as the base system F mimics a customers of theirs but my client requires Samba to transfer ( files. Thanks for the suggestion anyway.       Crashdump Summary Information:  -----------------------------* Crash Time:        10-FEB-2006 14:18:00.51; Bugcheck Type:     INVEXCEPTN, Exception while above ASTDEL / Node:              XXX-removed-XXX  (Clustered) ( CPU Type:          AlphaServer 800 5/400 VMS Version:       V7.1-1H2  Current Process:   SMBD_BG522 8 Current Image:     $15$DKA200:[SAMBA228.][BIN]SMBD.EXE;2: Failing PC:        FFFFFFFF.8AC35C38    UCX$BGDRIVER+41C38$ Failing PS:        00000000.00000803 Module:            UCX$BGDRIVER  Offset:            00041C38   * Boot Time:         10-FEB-2006 11:05:03.00* System Uptime:               0 03:12:57.51 Crash/Primary CPU: 00/00 System/CPU Type:   1B05  Saved Processes:   44 ' Pagesize:          8 KByte (8192 bytes) < Physical Memory:   512 MByte (65536 PFNs, contiguous memory)  Dumpfile Pagelets: 119858 blocks: Dump Flags:        olddump,writecomp,errlogcomp,dump_style' Dump Type:         compressed,selective ) EXE$GL_FLAGS:      poolpging,init,bugdump 7 Paging Files:      3 Pagefiles and 1 Swapfile installed    Stack Pointers: : KSP = 00000000.7FFA1B28   ESP = 00000000.7FFA6000   SSP =  00000000.7FFAC100  USP = 00000000.7AF90170    General Registers:: R0  = 00000000.00000000   R1  = 00000000.0000000C   R2  =  00000000.7FFA1D10 : R3  = FFFFFFFF.84CBCE00   R4  = 00000000.7FFA1B80   R5  =  00000000.7FFA1CF8 : R6  = 00000000.7FFA1D40   R7  = 00000000.00000803   R8  =  FFFFFFFF.8AC4D3B0 : R9  = 00000000.0000000C   R10 = 00000000.7AF9021C   R11 =  00000000.0C1C2001 : R12 = 00000000.00008001   R13 = FFFFFFFF.8AC4C300   R14 =  FFFFFFFF.81603A40 : R15 = FFFFFFFF.84CC1F88   R16 = 00000000.000001CC   R17 =  00000000.7FFA1B80 : R18 = 00000000.7FFA1D40   R19 = 00000000.00000001   R20 =  FFFFFFFF.8101E374 : R21 = 00000000.00000000   R22 = 00000000.7AF9021E   R23 =  FFFFFFFF.8AC4A490 : R24 = 00000000.00000001   AI  = 00000000.00000005   RA  =  FFFFFFFF.8AC0CD80 : PV  = FFFFFFFF.8AC4A490   R28 = FFFFFFFF.8AC35BF8   FP  =  00000000.7FFA1D80 1 PC  = FFFFFFFF.80065160   PS  = 28000000.00000800    Exception Frame:: R2  = FFFFFFFF.8AC4A490   R3  = 00000000.00000001   R4  =  FFFFFFFF.8101E374 : R5  = FFFFFFFF.81559440   R6  = 00000000.7FF721C0   R7  =  00000000.00000006    Crashdump Summary Information:  -----------------------------1 PC  = FFFFFFFF.8AC35C38   PS  = 00000000.00000803   = Signal Array:                            64-bit Signal Array: B Arg Count    = 00000005                  Arg Count      = 00000005: Condition    = 0000000C                  Condition      =  00000000.0000000C : Argument #2  = 00010000                  Argument #2    =  00000000.00010000 : Argument #3  = 0000001C                  Argument #3    =  00000000.0000001C : Argument #4  = 8AC35C38                  Argument #4    =  FFFFFFFF.8AC35C38 : Argument #5  = 00000803                  Argument #5    =  00000000.00000803    Mechanism Array:: Arguments    = 0000002C                  Establisher FP =  00000000.7FFA1D80 : Flags        = 00000000                  Exception FP   =  00000000.7FFA1D40 : Depth        = FFFFFFFD                  Signal Array   =  00000000.7FFA1CF8 : Handler Data = 00000000.0000000C         Signal64 Array =  00000000.7FFA1D10 : R0  = 00000000.00000000   R1  = 00000000.00000000   R16 =  00000000.00000000 : R17 = FFFFFFFF.8130E9E0   R18 = 00000000.00000006   R19 =  00000000.00000001 : R20 = FFFFFFFF.8101E374   R21 = 00000000.00000000   R22 =  00000000.7AF9021E : R23 = FFFFFFFF.8AC4A490   R24 = 00000000.00000001   R25 =  00000000.00000005 : R26 = FFFFFFFF.8AC0CD80   R27 = FFFFFFFF.8AC4A490   R28 =  FFFFFFFF.8AC35BF8    System Registers: 1 Page Table Base Register (PTBR) 00000000.000097A1 0 Processor Base Register (PRBR) FFFFFFFF.8100E0006 Privileged Context Block Base (PCBB) 00000000.12F400802 System Control Block Base (SCBB) 00000000.000001A0< Software Interrupt Summary Register (SISR) 00000000.00000000, Address Space Number (ASN) 00000000.0000004D8 AST Summary / AST Enable (ASTSR_ASTEN) 00000000.0000000F- Floating-Point Enable (FEN) 00000000.00000001 0 Interrupt Priority Level (IPL) 00000000.000000084 Machine Check Error Summary (MCES) 00000000.000000009 Virtual Page Table Base Register (VPTB) FFFFFFFC.00000000    Crashdump Summary Information:  ----------------------------- Failing Instruction:5 UCX$BGDRIVER+41C38:     LDL             R0,#X001C(R0)   * Instruction Stream (last 20 instructions):4 UCX$BGDRIVER+41BE8:     BIC             R26,#XFF,R263 UCX$BGDRIVER+41BEC:     BIS             R26,R27,R26 7 UCX$BGDRIVER+41BF0:     STL             R26,#X0004(R25) 4 UCX$BGDRIVER+41BF4:     BR              R31,#XFFFF525 UCX$BGDRIVER+41BF8:     LDA             SP,#XFFC0(SP) 3 UCX$BGDRIVER+41BFC:     XOR             R18,#X06,R1 0 UCX$BGDRIVER+41C00:     STQ             R27,(SP)4 UCX$BGDRIVER+41C04:     BIS             R31,#X05,R256 UCX$BGDRIVER+41C08:     STQ             R26,#X0008(SP)5 UCX$BGDRIVER+41C0C:     STQ             R2,#X0010(SP) 5 UCX$BGDRIVER+41C10:     STQ             R3,#X0018(SP) 5 UCX$BGDRIVER+41C14:     STQ             R4,#X0020(SP) 5 UCX$BGDRIVER+41C18:     STQ             R5,#X0028(SP) 5 UCX$BGDRIVER+41C1C:     STQ             R6,#X0030(SP) 5 UCX$BGDRIVER+41C20:     STQ             FP,#X0038(SP) 1 UCX$BGDRIVER+41C24:     BIS             R31,SP,FP 6 UCX$BGDRIVER+41C28:     LDL             R0,#X000C(R17)2 UCX$BGDRIVER+41C2C:     BIS             R31,R27,R22 UCX$BGDRIVER+41C30:     BIS             R31,R19,R32 UCX$BGDRIVER+41C34:     BIS             R31,R20,R45 UCX$BGDRIVER+41C38:     LDL             R0,#X001C(R0) 2 UCX$BGDRIVER+41C3C:     BIS             R31,R31,R63 UCX$BGDRIVER+41C40:     BIS             R31,R31,R31 2 UCX$BGDRIVER+41C44:     BIS             R31,R3,R196 UCX$BGDRIVER+41C48:     LDA             R27,#XEE88(R2)   ------------------------------  % Date: Tue, 14 Feb 2006 16:48:51 -0500 . From: JF Mezei <jfmezei.spamnot@vaxination.ca>- Subject: TCPIP SET CONF PROTOCOL TCP /FORWARD - Message-ID: <43F2503E.70BEB4AB@vaxination.ca>   H I tried to re-setup the slip interface with my PSION yesterday, now that. I know that SLIP still works on VAX TCPIP 5.3.  F The SLIP interface is in a different subnet hosted by VELO.  And I hadH setup the routes to point to VELO to access that subnet, but that didn'tD work. VELO didn't respond in a traceroute (which it should have been' first hop before getting to the PSION).   % After stumbling in the dark, I found:   ) on velo: sysconfig -r inet ipforwarding=1   > But that wasn't enough, it also needs sysconfig -r ipgateway=1  E This made it work on the live system. I had set this up years ago and  never worried since.  Q The only relevelant TCPIP command I found was TCPIP SET CONF PROTOCOL IP /FORWARD   F Unfortunatly, the documentation is rather terse on this subject. CouldE someone confirm that SET CONF PROTOCOL IP /FORWARD result in both the C ipforward and ipgateway being set to 1 when TCPIP services starts ?     G And is there a good explanation of the difference between ipforward and  ipgateway ?    ------------------------------   End of INFO-VAX 2006.091 ************************                                                                                                      