[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference vaxaxp::vmsnotes

Title:	VAX and Alpha VMS
Notice:	This is a new VMSnotes, please read note 2.1
Moderator:	VAXAXP::BERNARDO

Created:	Wed Jan 22 1997
Last Modified:	Fri Jun 06 1997
Last Successful Update:	Fri Jun 06 1997
Number of topics:	703
Total number of notes:	3722

668.0. "Vax Satellite randomly booting or not (lancp)" by PRSSOS::MENICACCI () Mon Jun 02 1997 05:41

			Hi,

several customers reported us the following problem with satellites defined
in LANCP.

Vaxstations previously down line loaded by a VAX 6410.

Then, the customer bought a boot node AXP V6.2-1h3 and used the cross boot
architecture facility.

NCP service circuit was disabled.

First VAX (VAX1) satellite was defined in LANCP and booted OK.
Second VAX (VAX2) satellite was defined in LANCP and booted OK.

...

Some time later third VAX (VAX3) was defined in LANCP and didn't boot.

The mop (lancp) down line load was received, but this demand loops without
success nor error.

VAX2 was also rebooted without success.

The customer doesn't know what lead to this. 

Satellites "seems" to be well defined in LANCP.

LANCP> sh dev ewa0/mop

Device Listing, volatile database:
              --- MOP Downline Load Service Characteristics ---
   Device     State   Access Mode      Clients           Data Size
   ------     -----   -----------      -------           ---------
   EWA0      Enabled NoExclusive    KnownClientsOnly     246 bytes
LANCP> sh node mm 

Node Listing, volatile database:

MM (08-00-2B-12-34-56):
  MOP DLL:  Load file:   NISCS_LOAD.EXE
            Load root:   DKA0:[SYS10.]
            Boot type:   VAX satellite


I tried lancp> set node vax2 /perm ==> nothing changed.
I tried lancp> set node/v3         ==> nothing changed.


Then, I wanted to create a logfile (lancp>set acc/echo/full).
In order to read it, I had to stop LANCP and then I restarted it and OH!
the satellites booted !!!

I need to know if this is a known problem, I would like also to know if a lancp
TIMA kit exists. I didn't find any. Some months ago, I had a new LANCP image :


        Image Identification Information

                image name: "LANCP"
                image file identification: "X-19A1"
                image file build identification: "X61Q-SSB-IH00"
                link date/time:  9-DEC-1996 11:20:48.61
                linker identification: "A11-12"


Could this image solve my problem ? Can we give it to customers ?

Thanks for any input,

Maria MENICACCI.

T.R	Title	User	Personal Name	Date	Lines
668.1		STAR::JFRAZIER	What color is a chameleon on a mirror?	`Mon Jun 02 1997 12:19`	7
	Re: .0 By any chance, are there also FDDI adaptors on the boot node? If so, are they mop enabled? James :-)
668.2		STAR::STOCKDALE		`Mon Jun 02 1997 17:40`	16
	As .-1 mentionned, if FDDI was involved they you might have a similar situation which failed in .-1's configurations. But it was only necessary to have multiple nodes responding to the boot requests. The booting node would send a multicast 'request memory load', then a would respond with an 'assistance volunteer' messages, then the booting node send a unicast 'request memory load' to the node it selected, then the booting node would receive another 'assistance volunteer' message which it should have ignored. At this point the booting node decided to continue the boot with the new volunteer and this immediately fails (on the booting node) so it starts the boot process over again. The problem was a console problem in the VAX satellite, and the way to resolve was to allow only one node to respond to that satellite's boot requests. - Dick
668.3	No FDDI interface	PRSSOS::MENICACCI		`Tue Jun 03 1997 04:53`	15
	Hi, thanks for your inputs. In the previous VAX boot node, these satellites weren't known. (ncp clear/purge sat). The AXP (only one boot node) has no FDDI interface. Any other idea ? How can we explain that stopping/restarting the lanacp process solved the problem ? Maria.
668.4	Need more info	STAR::STOCKDALE		`Thu Jun 05 1997 08:49`	28
	>>How can we explain that stopping/restarting the lanacp process solved the >>problem ? This would imply that there was a problem with enabling MOP in non-exclusive mode, where starting/restarting LANACP would cure it. Actually, disabling/ re-enabling MOP would accomplish the same thing. The fix for this problem is in the LAN common routines, in V6.2-1H3. If you do an ANA/SYS and SHOW LAN/DEV=EWA, the LAN version should appear as LAN version 00A30002.06020089 or later. If it is earlier than A3 then you need to update SYS$EWDRIVER.EXE. If it is up to date, could send me output from SDA and LANCP for the following (after boot failure happens)? In SDA: SHOW LAN/FULL E EWA;2000 In LANCP: SHOW MOP Thanks. - Dick
668.5	The LAN version is OK	PRSSOS::MENICACCI		`Fri Jun 06 1997 10:08`	12
	Hi all and Dick, I verified the version and it's 00A30002.06020089. So I gave the commands you asked for to my customer. As this problem happens oftenly, as soon as I receive the output, I will post it here. Fortunately, stop/restart is still an effective workaround. Maria.