From:	CRDGW2::CRDGW2::MRGATE::"SMTP::CRVAX.SRI.COM::RELAY-INFO-VAX" 30-JUL-1990 20:46:39.98
To:	MRGATE::"ARISIA::EVERHART"
CC:	
Subj:	Booting VAXstations remotely

Received:  by crdgw1.ge.com (5.57/GE 1.70)
	 id AA27339; Mon, 30 Jul 90 20:28:37 EDT
Received: From OPERATIONS.DCCS.UPENN.EDU by CRVAX.SRI.COM with TCP; Mon, 30 JUL 90 16:43:39 PDT
Message-Id: <9007302340.AA14095@operations.dccs.upenn.edu>
Date: Mon, 30 Jul 90 19:40 EDT
From: MARTIN@wharton.upenn.edu
Subject: Booting VAXstations remotely
To: info-vax@sri.com
X-Vms-To: IN%"info-vax@sri.com"


 # Russell D Wilton                 Internet: WILTON@HG.ULeth.CA #
 # Systems and Comm Manager         BITNET:    WILTON%HG@UNCANET #
 # Computing Services               Voice:        (403) 329-2525 #
 # University of Lethbridge         FAX:          (403) 329-2022 #
 # 4401 University Drive   Lethbridge, Alberta, CANADA   T1K 3M4 #
 #===============================================================#
  
	Writes..

>Is there some way to boot a VAXstation remotely?  I was hoping the NCP TRIGGER
>command would do it, but no such luck.  Am I missing something obvious?  


	Hi:

	There are lots of ways to do this.

	My favorite lets you do just about anything with the boot node.

	Kind of sneaky but effective...

	First go into the NCP on the node (OR NODES) that respond
	to the boot request for each of the VAXstations.
	The command will be
	NCP>show char node NAME_OF_THE_SATELLITE

	You can tell if the boot node will respond if the response to
	the above command looks something like this.

Node Volatile Characteristics as of 30-JUL-1990 18:18:10

Remote node =   1.1023 (NAME_OF_NODE)

Hardware address         = 08-00-2B-FF-EE-AA
Tertiary loader          = SYS$SYSTEM:TERTIARY_VMB.EXE
Load Assist Agent        = SYS$SHARE:NISCS_LAA.EXE
Load Assist Parameter    = DISK$VAXVMSRL053:<SYSFF.>
   
	You can then FOOL the boot node into not responding the satellite.
	just use the ncp command 
	NCP>DEFINE node NAME_OF_THE_SATELLITE hardware FF-00-2B-FF-EE-AA
	NCP>SET node NAME_OF_THE_SATELLITE hardware FF-00-2B-FF-EE-AA	

	Notice that I only flipped the first 2 bytes of the 48 bit Ethernet
	address.. Thats because DEC addresses always start out 08-00-2B-.
	Sort of a trick so you dont have to write anything down.
	 
	Then shut down the satellites with auto reboot requested.

	Then shut down the "boot" node, reboot (standalone if you like).
	Do backups or whatever. Bring up the node completely if you wish.
	When you are done mucking about reset the ethernet addresses in the
	Decnet database (perm and volatile). You can even add the nodes in one
	at a time to improve boot performance.

	Ignore any of the OPCOM /DECNET messages about LINE OPEN ERROR
	because this is just DECNET saying it does not know what to do with
	load requests for the node at FF-00-2B-FF-EE-AA.

	Actually you can also accomplish this by commenting out the
	@sys$manager:startnet.com from systartup_v5.com but then you
	loose all decnet functionality.

	Have fun

................................................................................
Christopher A. Martin			 The Wharton School of
manager, System managment/programming	 The University of Pennsylvania
					 Philadelphia, PA 19019-6301
martin@wharton.upenn.edu		 pstn 215.898.5392

Treat the EARTH well!  It was not given to you by your parents...
                       It was loaned to you by your children.
-Kenyan proverb