1 INFO-VAX	Mon, 26 Apr 2004	Volume 2004 : Issue 230       Contents:% Re: ALPHAstation 255/233 and PBXGB-AA , Re: Billions wasted due to IT skills deficit, Re: Billions wasted due to IT skills deficit Re: DECW DECTerm startup errors + Re: errors on shadow sets and their members + Re: errors on shadow sets and their members + Re: errors on shadow sets and their members + Re: errors on shadow sets and their members + Re: errors on shadow sets and their members 4 Re: Linux on its way out - unless you are a geek ...4 Re: Linux on its way out - unless you are a geek ...& Re: OpenVMS newbie  - AlphaStation 233> Re: upgrading (or installing) VMS on a system with no CD drive/ Re: Using Windows XP hyperterminal with OpenVMS 0 VMS 7.3-2, TCPIP 5.4: which to upgrade to first?4 Re: VMS 7.3-2, TCPIP 5.4: which to upgrade to first?8 [Toulouse] vends MicroVAX 3400, Alphastation, CDs etc...  F ----------------------------------------------------------------------    Date: 25 Apr 2004 12:45:29 -0700$ From: n8wxs@arrl.net (Jeff Campbell). Subject: Re: ALPHAstation 255/233 and PBXGB-AA< Message-ID: <1a40a0b2.0404251145.4c6f2e7@posting.google.com>  | helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) wrote in message news:<c6euth$qga$1@online.de>...I > I have an ALPHAstation 255/233 which came with a "Number Nine Computer  I > Corporation" PCI graphics card.  The ALPHAstation is down now since it  G > is waiting to have a capacitor replaced.  I opened it up in order to  K > install an internal disk, and noticed a) that it has some free PCI slots  J > and b) that I have a "PBXGB-AA" graphics card lying around (one of many I > things I took "because I might need it sometime" when I collected some  , > used hardware being given away somewhere). > F > Will this card work with the ALPHAstation 255/233?  Is it "plug and J > play" or do I have to somehow configure it, get a driver for it, update K > my firmware or whatever?  What differences will I notice compared to the   > old graphics card?  N Yes, both the PowerStorm 3D30 and 4D20 (PBXGB-CA) are supported options on the AlphaStation 255. Check out   M  <http://h18002.www1.hp.com/alphaserver/workstations/retired/a255series.html>   H for the SOC (Systems and Options Catalog) or the Golden Eggs .PDF files.  J I dunno offhand if OpenVMS has native drivers for this card or if you willF need to find an OpenVMS OpenGL distribution kit from which to copy the
 driver(s).  N The default pixel clock frequency and monitor resolution are controlled by theJ rotary switch. For example, position 5 selects 1024 by 768 mode, 75.02 MHz6 pixel frequency with a vertical refresh rate of 70 Hz.  * There are two jumpers that need to be set:     1) VGAEN  VGA enable
   2) Alias  M For an EV4 processor, the Alias jumper must be open, not installed. The VGAEN K jumper is installed on the first PBXGB card, and not installed/removed from  any subsequent PBXGB cards.   ; For an EV5 processor, the Alias jumper is always installed.    HTH,  
 Jeff Campbell  n8wxs@arrl.net   ------------------------------  % Date: Sun, 25 Apr 2004 21:35:11 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>5 Subject: Re: Billions wasted due to IT skills deficit 6 Message-ID: <408C755F.69383CB9@NeOaSrPtAhMlNiOnWk.net>   John Smith wrote:  > [snip]H > Often what is required in those cases is a lesson from feudal JapaneseJ > society - ritual seppuku by those responsible. This allows for those not? > responsible the possibility of a 'clean slate' to start anew.   G Funny you should mention that. There was story recently of the Japanese D held hostage in Iraq who, upon returning home, had shame heaped uponH them for defying recommendations not to travel to Iraq and forcing theirF government to intervene. The article said that the Japanese gov't evenE intended to bill them for the expenses/airfare to bring them home (to ) the tune of the equivalent of $6,000 US).   E Perhaps homo sapiens is not so advanced as we like to think we are...   H Perhaps our genus should be renamed. "Homo sapiens" literally means "manE (who is) wise". Wisdom being in such short supply these days, perhaps 9 the Latin for "arrogant bastard" would be more apropos...   H Sorry, folks... I've been working too hard and the asshole manager (roleF model for Voldemort?) we had this time last year has left his scars on what is left of me...    Apologies...   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  % Date: Sun, 25 Apr 2004 21:25:33 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>5 Subject: Re: Billions wasted due to IT skills deficit 6 Message-ID: <408C731D.99323FBF@NeOaSrPtAhMlNiOnWk.net>   Jack Peacock wrote:  > M > "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> wrote in message 2 > news:408A8DDE.B9C11EA7@NeOaSrPtAhMlNiOnWk.net...H > > When I was young, I was always taught that "It takes a big person toK > > admit a mistake", yet in real life it now seems that admitting to being H > > wrong is virtually a cardinal transgression. People will go to theirK > > deaths, biological or professional, before admitting an error. (Optimal , > > case in point: OpenVMS and advertising.) > > N > When the mistake might bankrupt the company or force it into a unwanted saleL > (look at AT&T Wireless for a classic example), it's not enough to admit an > error in judgment.    F If "the mistake might bankrupt the company or force it into a unwantedH sale", what does it matter whether anyone admits the mistake or not? The% results are inevitable, are they not?   9 > Thousands of other people have to bear the consequences K > (lost jobs, lost stock value, wiped out pensions) of that mistake.  Think N > anyone will ever hire the CIO of AT&T Wireless after he wrecked the company?  F What does whether he admits it or not have to do with it in that case?F Everyone knows he screwed up. Everyone knows whether he admitted it or not.  E Would *YOU* hire someone who didn't have he balls to admit he screwed E up? What else would he try to shove under the carpet? (Say, "Enron".)   L > Would you invest in a company if you found out their chief computer expertH > had a history of massive incompetence and multi-billion dollar losses?  A Depends. Some of the geratest successes of our time were also the E greatest failures. (Say, "Thomas Alva Edison".) I seem to recall that : Henry Ford was considered an idiot by his contemporaries.   I > Admitting a mistake when it's too late doesn't help anyone, and is more F > likely to result in stockholder lawsuits than any productive change.  F Are stockholders less likely to sue in the absence of "owning up" than if someone does? (Say, "Tyco").    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  # Date: Sun, 25 Apr 2004 23:40:56 GMT ( From: Alder <PGDEHMKOKIMD@spammotel.com>( Subject: Re: DECW DECTerm startup errors+ Message-ID: <c0Yic.12102$en3.6550@edtnps89>    John Santos wrote:: # > On Sat, 24 Apr 2004, Alder wrote:  >  >  >>John Santos wrote::  >>$ >>>On Fri, 23 Apr 2004, Alder wrote: >>>  >>>  >>>  >>>>Bob Koehler wrote::  >>>> >>>> >>>>Z >>>>>In article <0tcic.7928$w63.7125@edtnps89>, Alder <PGDEHMKOKIMD@spammotel.com> writes: >>>>>  >>>>>  >>>>> J >>>>>>If I start a DECWindows (V1.2.6) session and then fire up a DECTerm I >>>>>>window, the DECTerm appears and functions, but an error message is   >>>>>>logged that reads: >>>>>>@ >>>>>>Fri Apr 23 10:24:39 2004 - output from process id 000001C2F >>>>>>-> RCV'D (pid 000001C3): %SET-W-NOTSET, error modifying MBA2501:W >>>>>>-> RCV'D (pid 000001C3): -SET-E-INVDEV, device is invalid for requested operation  >>>>>  >>>>> 3 >>>$ if .not. f$getdvi("sys$command:","trm") then -  >>>        exit  >>H >>I added a similar line to SYLOGIN.COM before my SET TERMINAL commands , >>and it eliminated the MBA errors.  Thanks. >> >>Terry  >  > C > Aha!  I should have read my own code!  I think that should do it. F > I'm a little confused about who posted what...  Are you the original1 > poster?  If not, did it solve the OP's problem?  >  > Glad it helped!  >  >   % Yup, I am, and it did.  Thanks again.    Terry    ------------------------------  % Date: Mon, 26 Apr 2004 02:20:58 +0800 , From: Paul Repacholi <prep@prep.synonet.com>4 Subject: Re: errors on shadow sets and their members- Message-ID: <87r7ubvrzp.fsf@prep.synonet.com>   R helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:  E > I've rearranged some hardware in my hobbyist cluster.  In one BA353 C > there is an RZ26 which I can't remove, so I'm "forced" to use it. @ > It is a member of a shadow set (the other member is on anotherD > node).  During SHADOW MERGE or COPY operations, it collects 10--15D > errors, but only on the node it has a direct connection to, not toE > all the other nodes, to which it is MSCP served.  The shadow set is + > mounted by all nodes and shows no errors.    > What does this mean?  F Your disk is about to shuffle off probably. The errors are only loggedC on the connected system, ie, with a controller. Only software error E codes are passed `up stream', and the whole point is to stop bad news  getting out :)  C Bricks can get pretty tight at times, but they soon come out with a F bit of moderate violence! Taking the front of the case off helps LOTS.   --  < Paul Repacholi                               1 Crescent Rd.,7 +61 (08) 9257-1001                           Kalamunda. @                                              West Australia 6076* comp.os.vms,- The Older, Grumpier Slashdot. Raw, Cooked or Well-done, it's all half baked.F EPIC, The Architecture of the future, always has been, always will be.   ------------------------------  % Date: Sun, 25 Apr 2004 13:18:47 -0700  From: Z <z@no.spam> 4 Subject: Re: errors on shadow sets and their members0 Message-ID: <108o789baebe16c@corp.supernews.com>  / Phillip Helbig---remove CLOTHES to reply wrote:   E > I've rearranged some hardware in my hobbyist cluster.  In one BA353 J > there is an RZ26 which I can't remove, so I'm "forced" to use it.  It isI > a member of a shadow set (the other member is on another node).  During I > SHADOW MERGE or COPY operations, it collects 10--15 errors, but only on H > the node it has a direct connection to, not to all the other nodes, toF > which it is MSCP served.  The shadow set is mounted by all nodes and > shows no errors.   >  > What does this mean?  8 Step 1: fire up $analyze/error or $diagnose and find out exactly WHAT the error is.   ------------------------------  # Date: Sun, 25 Apr 2004 21:25:05 GMT & From: Lee Mah <lytmah@telusplanet.net>4 Subject: Re: errors on shadow sets and their members) Message-ID: <R0Wic.4095$X52.885@clgrps12>   D  From my experience with shadowing, these have been my observations:  I Don't bother looking for errors on the shadow (DSA) devices.  They won't   log any.A The errors will show on the nodes which are doing, or have done,  $ MSCP-serving of the troubled device. For example,   $ mc sysman  SYSMAN> set env/cluster + %SYSMAN-I-ENV, current command environment: $         Clusterwide on local cluster<         Username Z99999       will be used on nonlocal nodes    SYSMAN> do show dev DSA123: - %SYSMAN-I-OUTPUT, command execution on node C G Device                  Device           Error    Volume         Free   	 Trans Mnt G  Name                   Status           Count     Label        Blocks  	 Count Cnt I DSA123:                 Mounted              0  SS123         12466386     256   4 C $2$DUA569:         (A)  ShadowSetMember      0  (member of DSA123:) D $2$DUA223:    (HSJM22)  ShadowSetMember     12   (member of DSA123:)  - %SYSMAN-I-OUTPUT, command execution on node A G Device                  Device           Error    Volume         Free   	 Trans Mnt G  Name                   Status           Count     Label        Blocks  	 Count Cnt I DSA123:                 Mounted              0  SS123         12466422     235   4 C $2$DUA569:    (HSJM11)  ShadowSetMember      0  (member of DSA123:) C $2$DUA223:         (C)  ShadowSetMember      0  (member of DSA123:)   - %SYSMAN-I-OUTPUT, command execution on node D G Device                  Device           Error    Volume         Free   	 Trans Mnt G  Name                   Status           Count     Label        Blocks  	 Count Cnt I DSA123:                 Mounted              0  SS123         12466422     193   4 C $2$DUA569:         (A)  ShadowSetMember      0  (member of DSA123:) C $2$DUA223:    (HSJM22)  ShadowSetMember      6  (member of DSA123:)   - %SYSMAN-I-OUTPUT, command execution on node B G Device                  Device           Error    Volume         Free   	 Trans Mnt G  Name                   Status           Count     Label        Blocks  	 Count Cnt I DSA123:                 Mounted              0  SS123         12466422     173   4 C $2$DUA123:    (HSJM11)  ShadowSetMember      0  (member of DSA123:) C $2$DUA223:         (D)  ShadowSetMember      0  (member of DSA123:)     / Phillip Helbig---remove CLOTHES to reply wrote:   D >I've rearranged some hardware in my hobbyist cluster.  In one BA353I >there is an RZ26 which I can't remove, so I'm "forced" to use it.  It is H >a member of a shadow set (the other member is on another node).  DuringH >SHADOW MERGE or COPY operations, it collects 10--15 errors, but only onG >the node it has a direct connection to, not to all the other nodes, to E >which it is MSCP served.  The shadow set is mounted by all nodes and  >shows no errors.  >  >What does this mean?  > I >Presumably, since the shadow set itself has no errors, and neither does  B >the other member, as long as the good member or both members are H >available, there will be no problems.  What happens if the good member J >goes, leaving only the member with the errors on the node it is directly = >connected to?  Will this result in errors on the shadow set?  > G >Are the errors likely to be physical problems with the disk, or could  H >they be due to a software problem or a configuration problem (say, too = >many packet collions on the LAN during the MERGE operation)?  >  >    >    ------------------------------  + Date: Sun, 25 Apr 2004 21:46:51 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney) 4 Subject: Re: errors on shadow sets and their members( Message-ID: <c6hbkb$ee2$1@pcls4.std.com>  R helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:  D >I've rearranged some hardware in my hobbyist cluster.  In one BA353I >there is an RZ26 which I can't remove, so I'm "forced" to use it.  It is H >a member of a shadow set (the other member is on another node).  DuringH >SHADOW MERGE or COPY operations, it collects 10--15 errors, but only onG >the node it has a direct connection to, not to all the other nodes, to E >which it is MSCP served.  The shadow set is mounted by all nodes and  >shows no errors.    >What does this mean?   B You have to look at the error log to see what the error really is.  H One possibility is a parity error on a disk block.  DKDRIVER will log an? error whenever it attempts to read a block with a parity error.   G How did the error get there?  It could mean bits flipped (mega badness) G but due to how shadowing works, it may have been deliberately put there D by shadowing (perhaps the corresponding block on the other drive wasE bad at one time, and this was "copied" during a copy operation in the  past).  F You may want to track down which files contain the bad blocks, if any.H $ ANALYZE/DISK/READ will read every block of every file but if you do itI to the shadowset it may go to the wrong member and miss the error blocks. E You can do it to the individual member if you are willing to break up D the shadowset and endure a copy when putting it back in (do you have  a version of VMS with minicopy?)     --   -Mike    ------------------------------  + Date: Sun, 25 Apr 2004 23:03:43 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)4 Subject: Re: errors on shadow sets and their members$ Message-ID: <c6hg4f$3fa$1@online.de>  H In article <c6hbkb$ee2$1@pcls4.std.com>, moroney@world.std.spaamtrap.com (Michael Moroney) writes:   H > You may want to track down which files contain the bad blocks, if any.J > $ ANALYZE/DISK/READ will read every block of every file but if you do itK > to the shadowset it may go to the wrong member and miss the error blocks. G > You can do it to the individual member if you are willing to break up F > the shadowset and endure a copy when putting it back in (do you have" > a version of VMS with minicopy?)  E The bad member is connected to a VAX with 7.3; the good member to an   ALPHA with 7.3-1.    ------------------------------  % Date: Sun, 25 Apr 2004 19:46:53 +0200   From: "Dr. Dweeb" <dr@dweeb.com>= Subject: Re: Linux on its way out - unless you are a geek ... , Message-ID: <c6gtih$979$1@news.cybercity.dk>   Karsten Nyblad wrote: 7 > "Bob Ceculski" <bob@instantwhip.com> wrote in message 8 > news:d7791aa1.0404241840.6cb5c69@posting.google.com... >>= >> I think VMS proves that you want security built in and not < >> an add on or you get what you have now with windoze/unix/; >> linux ... thousands of certs and becoming members of the  >> patch of the week club ...  > H > I think that this discussion has already demonstrated that many of theB > security features in VMS have been added after VMS was designed,E > e.g., APIs and ACLs.  Thus security may have to be built in, but an F > operating does not need to be born with all the security features inG > place.  Please note that the security features have been added to VMS / > without compromising backwards compatibility.  >   I I have not really seen that demonstrated in your posts.  That refinements D have been engineered into an already solid architecture is radicallyF differnet from the brown-bag kludge in evidence in the nix world whereH everyone just "rolled their own".  This has been demonstrated now for 30A years, and it still seems that there is a way to go.  There is an C engineering concept involved here, which I am not going to explain.   E > The security of VMS is also due to the low number of juvenile users G > on the platform.  The programmers on VMS have carriers and mortgages, B > and they do not want to get fired with no references and perhaps: > being send to the slammer and getting a criminal record. >   A The code kiddies attack Unix and Windoze because it is inherently G attackable.  Were "they" successful in attacking VMS, then "they" would G flock to it in droves, for a number of reasons which should be obvious.    I think your logic is inverted.   D > Please note  that many of the problems on the Windows platform are@ > due to the applications and because Microsoft does not use theF > security features of the operating system.  It is beyond me how theyG > could get the idea of letting E-mail clients execute code embedded in E > E-mails automatically and without the user doing anything.  Putting F > macro code in Word and Excel files has also proved a bad idea.  I doG > not understand why normal users have Administrator rights by default. F > Unfortunately, you can be sure that if you change the default setup,F > then some installations will fail, and you will not have a clue why.  H I can but agree on your points in this paragraph.  I suspect BillG mightJ even agree, given the current state of affairs, though I have seen nothing# to suggest this in print or action.   B > Thus even people knowing better run Windows without changing theG > default setup.  And then Microsoft does not use all the nice features  > for file protection. >  > Karsten Nyblad! > ibpit1202 at sneakemail dot com    med venlig hilsen 	 Dr. Dweeb    ------------------------------  + Date: Sun, 25 Apr 2004 18:31:08 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)= Subject: Re: Linux on its way out - unless you are a geek ... $ Message-ID: <c6h05b$gig$1@online.de>  G In article <c6gtih$979$1@news.cybercity.dk>, "Dr. Dweeb" <dr@dweeb.com>  writes:   F > > Please note  that many of the problems on the Windows platform areB > > due to the applications and because Microsoft does not use the0 > > security features of the operating system.    H Aren't you overlooking the important fact here that it is Microsoft who D aren't using the security features?  I doubt it matters to Joe User E whether the features are absent or just not used by the applications  I DEVELOPED BY THE SAME COMPANY WHO DEVELOPED THE OPERATING SYSTEM and who  > never miss an opportunity to point out how closely tied their G applications are to the operating system.  So, lack of knowledge can't   be the reason.  I I think most people who regularly upgrade Windows don't do so to get new  F features, but so that they can be sure that all the security holes in G the last release have been fixed.  If this weren't the case, Microsoft  E would probably sell a lot fewer upgrades (and make a lot less money).    ------------------------------  % Date: Sun, 25 Apr 2004 15:59:58 -0400 / From: "Rodney Kelp" <rodneykelp605@hotmail.com> / Subject: Re: OpenVMS newbie  - AlphaStation 233 1 Message-ID: <UaednSeQDawqhRHdRVn-jw@adelphia.com>    Why?  @ "Berdt van der Lingen" <berdt@n0spam-xs4all.nl> wrote in message0 news:408a2cd1$0$21804$e4fe514c@news.xs4all.nl...	 > Hi all,  > J > I'm a OpenVMS beginner and I want to learn the OpenVMS operating system.J > Is a AlphaStation 233 with 64 mb ram and a 2 gb SCSI disk suitable for a	 > newbie?  > Should I buy more ram? >  > Wish me luck!  > 
 > regards, >  > Berdt van der Lingen >  >      --- & Outgoing mail is certified Virus Free.: Checked by AVG anti-virus system (http://www.grisoft.com).@ Version: 6.0.663 / Virus Database: 426 - Release Date: 4/20/2004   ------------------------------  % Date: Sun, 25 Apr 2004 22:48:41 -0400   From: John Santos <JOHN@egh.com>G Subject: Re: upgrading (or installing) VMS on a system with no CD drive / Message-ID: <1040425214836.32308A@Ives.egh.com>   D On Sat, 24 Apr 2004, Phillip Helbig---remove CLOTHES to reply wrote:  G > Suppose one wants to upgrade (or install) VMS on a system with no CD  K > drive.  One method would be to copy the CD to a hard disk, then use that  H > in place of the CD.  (This might be a good idea anyway since it might K > speed things up quite a bit, perhaps even enough to make up for the time  F > spent copying the CD to disk, but even if not the actual upgrade or D > installation will be quicker (the copy to disk doesn't have to be H > watched over and hence can be done at leisure).)  However, this means 3 > one must have a spare disk and a place to put it.   C I've been thinking about the alternatives for installing V8.2 on my E microVAX (without CD drive) at home, when it is released.  (Currently C V7.3, which was on TK50, IIRC, but the cover letter said it was the + last version that would be issued on TK50.)   A One alternative I'm considering is to mount the VMS CD on a PC or @ Mac and copy it as an image to a spare RA72 (mounted foreign) onB the uVAX using FTP (or maybe NFS? or SAMBA?)  There are a lot moreC blocks on the RA72 than on a CD, so the cluster bit map will be too F small, but maybe VMS won't care, or maybe ANA/DISK/REPAIR will fix it.F Alternatively, I could copy it to a container file on an RA72, install? one of the virtual disk drivers on the mVAX, init another RA72, 3 mount the container file as VD0: (or whatever), and B backup/image VD0: DU2:/noinit.  Then boot DU2: and do the upgrade.  E (It might be easier to use OS X as the source, since I should be able C to run FTP on either the Mac or the VAX and upload/download the CD. E On the PC, I would have to either install an FTP server or figure out C how to make the FTP client be willing to deal with an image copy of  a CD...)   > G > Could one do the installation or upgrade on another system?  Suppose  K > node B needs a VMS upgrade.  Could I take the system disk out of node B,  J > put it in node A (which has a CD drive), do the upgrade there, then put G > it back in node B?  If the upgrade were only a matter of copying new  F > files to the system disk, then there would be no problem.  What I'm I > worried about, however, are differences between node A and node B: for  @ > example, the ethernet device might have a different name, the G > configuration (amount of RAM etc) might be different hence confusing  . > AUTOGEN if it is run during the upgrade etc.    C This should work just fine.  I've never done this, but I have moved C and copied system disks from on VAX to another or from one Alpha to A another many times and never had any problems using it on the new @ system (being sure that the new system is supported on both, butD that shouldn't be an issue since you're talking about upgrading from, one fairly recent VMS to a more recent one.)  B If you use the tailoring options (VAX only?) to remove unnecessaryD stuff, you'll want to make sure you don't tailor off stuff needed byB system B, but by default you should get a disk that is bootable on any VAX.  B On Alpha, the whole thing should be a piece of cake.  Just install@ the Node B system disk in Node A, boot the VMS CD on node A, andA tell it to upgrade the node B disk, instead of Node A's.  When it < finishes and tells you to reboot from the disk to finish the< upgrade process, just shutdown, move the disk back to Node B? and reboot there.  It should be done with the CD at that point,  so you won't need it any more.  D On VAX, you upgrade by booting minimum off the system disk and then B pointing VMSINSTAL at the CD.  You'll need to make sure the VAX isA set to reboot from the Node B disk, not the node A disk.  It will B reboot several times from the Node B disk as the upgrade proceeds.A One of the final reboots will run AUTOGEN, tuning for Node A, but @ using the feedback from node B, if you've saved feedback from it before starting.  @ After putting the system disk back in node B, you'll want to runB AUTOGEN if the hardware config is drastically different.  When the= upgrade runs AUTOGEN on Node A (VAX only, unless you actually @ reboot an Alpha disk while still on Node A), the tuning may be a= little off, unless you have a drastically different amount of A memory.  (Nothing else should have a major effect, since you will = be using the MODPARAMS.DAT and AUTOGEN FEEDBACK from node B.) B In in the worst case, the disk should still be bootable on Node B,% or any supported VAX for that matter.   C All the drivers should be there and should get loaded automatically < when you put the disk back into Node B.  Network stuff (both< TCP/IP and DECnet) probably won't work if you have differentA interfaces on Node A while Node A is booted from the Node B disk, ? but should be fine as soon as you move the disk back to Node B. A Don't reconfigure the networks until you move the disk back to B, . or you'll have to reconfigure them back again!  J > Since essentially all my disks are shadow sets, my long-term plan is to K > have one "master node" per architecture whose system disk I maintain.  I  K > can then propagate upgrades to other systems by breaking a shadow set on  K > the master node, removing a member and using that as a new shadow master  G > on the other nodes.  The problem is automating the process.  Since I  J > have stuff such as the queue database off the system disk, that's not a H > problem.  One worry is stuff such as different names for the ethernet H > device.  If one just sticks a system disk in another system, one will K > have to reconfigure TCPIP as a result.  Another worry is the danger that  I > some TCPIP stuff which is node specific might be in SYS$COMMON instead   > of SYS$SPECIFIC.  A You can install/upgrade TCPIP without reconfiguring it!  In fact, > with PCSI, the configuration step is totally separate from the@ install/upgrade step.  So if you don't mind the fact that TCP/IP= won't work until you move the disk back to node B, it doesn't B matter if the interfaces, IP addresses, routers, DNS servers, etc.F etc. are all wrong for site A...  They'll all be preserved (correctly) for site B.   > Generally for an upgrade, you don't need to reconfigure either> DECnet or TCP/IP, but you may want to, to enable new features,; or when they've fixed configuration bugs or default tunings @ but you don't get the fixes until you re-configure.  If so, wait# until you're back at site B/node B.   ? (You may want to try booting the disk at site A to make sure it ; works before taking the time to move it back to site B.  If < you do this, you may have to undo some configuration changes once you get back to site B.)   J > Even if one knew that all the node-specific stuff were in SYS$SPECIFIC, D > even copying it to somewhere else before the shadow copy and then I > copying it back wouldn't solve all problems, if the upgrade adds files   > to SYS$SPECIFIC. > ! > Has anyone solved this problem?   A I think once you make your Master disk (at site A) and first copy ? it to site B, you'll have to copy back any config changes (i.e. ? the whole [SYSn...] tree, except for [SYSn.SYSCOMMON...]), back ? to the master disk at site A.  Once you've done that, you would > only need to copy changed files back to Site A, to keep things? in synch.  (If you have unique roots already on all your system > disks and they are all at the same VMS version, you could copy: them to site A to initially create the master disk, before; cloning it for the 1st time.  If you are careful, you could < probably do this even if all the nodes are currently booting@ from [SYS0...]; just copy node1::[sys0...] to master::[sys1...],; node2::[sys0...] to master::[sys2...], and change node 1 to  boot from root 1, etc.)   > But I think you're worrying too much!  Just back everything up (in case :-( ), and try it!    --   John Santos  Evans Griffiths & Hart, Inc. 781-861-0670 ext 539   ------------------------------  % Date: Sun, 25 Apr 2004 21:14:56 -0500 @ From: "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net>8 Subject: Re: Using Windows XP hyperterminal with OpenVMS6 Message-ID: <408C70A0.C8DFF5E7@NeOaSrPtAhMlNiOnWk.net>   "Main, Kerry" wrote: >  > > -----Original Message-----7 > > From: Paul Repacholi [mailto:prep@prep.synonet.com]   > > Sent: April 24, 2004 1:57 PM > > To: Info-VAX@Mvb.Saic.Com < > > Subject: Re: Using Windows XP hyperterminal with OpenVMS > > F > > "David J. Dachtera" <djesys.nospam@NeOaSrPtAhMlNiOnWk.net> writes: > > G > > > The network is down - and I have an IP phone so I can't even call J > > > the helpdesk to tell them that I can't call them because the network > > > is down... > > ! > > > Isn't technology wonderful?  > > I > > Read the reports on the great net outage of IMP fame. They can't even  > > invent a NEW failure mode. > >  > > --@ > > Paul Repacholi                               1 Crescent Rd.,; > > +61 (08) 9257-1001                           Kalamunda. D > >                                              West Australia 6076. > > comp.os.vms,- The Older, Grumpier Slashdot2 > > Raw, Cooked or Well-done, it's all half baked.J > > EPIC, The Architecture of the future, always has been, always will be. > >  > 	 > David -  > F > Help desk response - "Well, why didn't you just send an email then?" >  > :-)   F Do you perchance recall from "The Day The Earth Stood Still" where theG lady tells the boss that the power is out, so he says to call the power E company. She responds that the phone doesn't work, so he says to call E the phone company then. She responds one last time, "...but the phone  doesn't work"...  # Somehow, "oy vay" doesn't cut it...    --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  ( Unofficial Affordable OpenVMS Home Page: http://www.djesys.com/vms/soho/    ------------------------------  + Date: Sun, 25 Apr 2004 21:00:15 +0000 (UTC) P From: helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply)9 Subject: VMS 7.3-2, TCPIP 5.4: which to upgrade to first? $ Message-ID: <c6h8sv$rbn$1@online.de>  E I am at VMS 7.3-1 and TCPIP 5.3.  I plan to upgrade to VMS 7.3-2 and  F TCPIP 5.4.  Normally, I upgrade the OS first, then layered products.  H Does it really make any difference?  Any advantages to doing one first,  or the other first?r  @ At the moment, I don't have a CD drive on the ALPHA (it's just aG substitute system until my 255/233 gets fixed), but have several in thelH cluster (connected to VAX nodes and MSCP served).  (Since it has a metalF case, the substitute ALPHA (a 3000/300 LX) is sitting beneath a coupleH of BA353s, a keyboard and a DECrepeater, all connected with rather shortG cables, so I can't just open up the case and connect up a CD drive, andeG I don't have any "external" CD drives to connect to the SCSI plug).  AsnH far as upgrading TCPIP goes, could I just do PRODUCT INSTALL, specifyingE the location of the saveset on CD (possibly copied to disk first), or G would I have to actually boot from the CD (or a copy of it on disk) andh upgrade TCPIP via the menu?    ------------------------------  + Date: Sun, 25 Apr 2004 21:53:29 +0000 (UTC) 7 From: moroney@world.std.spaamtrap.com (Michael Moroney)-= Subject: Re: VMS 7.3-2, TCPIP 5.4: which to upgrade to first?2( Message-ID: <c6hc0p$ee2$2@pcls4.std.com>  R helbig@astro.multiCLOTHESvax.de (Phillip Helbig---remove CLOTHES to reply) writes:  F >I am at VMS 7.3-1 and TCPIP 5.3.  I plan to upgrade to VMS 7.3-2 and G >TCPIP 5.4.  Normally, I upgrade the OS first, then layered products.  oI >Does it really make any difference?  Any advantages to doing one first, n >or the other first?  H TCPIP V5.4 is on the VMS 7.3-2 CD, so if you upgrade by booting from theJ CD or a copy, it'll ask you if you want to upgrade TCPIP at the same time.I Same for a couple of other products, such as Motif and DECnet (IV and V).) --   -Mikea   ------------------------------  % Date: Mon, 26 Apr 2004 06:17:50 +0200u" From: Didier Morandi <no@spam.com>A Subject: [Toulouse] vends MicroVAX 3400, Alphastation, CDs etc... . Message-ID: <c6i2l9$2otg$1@biggoron.nerim.net>  ( A vendre cause cessation d'hobbyistit :   2 MicroVAX 3400r 1 AlphaStation 250 4/2664 Tonnes de cables, CDs, distrib et autres docs utiles  / 500 euros chaque machine ou 1000 euros le tout.r   Enlvement  Toulouse.   D. -- n    Didier  Morandi Direction techniquen   Bureaumatique SA.5 Tl: 05 34 60 99 98p   ------------------------------   End of INFO-VAX 2004.230 ************************