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

Conference star::wizards

Title: "ASK THE WIZARDS"
Moderator:QUARK::LIONEL
Created:Mon Oct 30 1995
Last Modified:Mon May 12 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:1857
Total number of notes:3728

1602.0. "Open: system configuration problem" by STAR::JKEENAN () Fri Feb 14 1997 08:14

Return-Path: "VMS001::WWW"@vms001.das-x.dec.com
Received: by vmsmkt.zko.dec.com (UCX V4.1-12, OpenVMS V6.2 VAX);
	Thu, 13 Feb 1997 12:24:24 -0500
Received: from vms001 by mail11.digital.com (8.7.5/UNX 1.5/1.0/WV)
	id MAA20576; Thu, 13 Feb 1997 12:16:49 -0500 (EST)
Date: Thu, 13 Feb 1997 12:17:34 -0500
Message-Id: <[email protected]>
From: "VMS001::WWW"@vms001.das-x.dec.com (13-Feb-1997 1217)
To: [email protected], [email protected], [email protected]
Subject: Ask the Wizard: '[email protected]'
X-VMS-To: [email protected]

Remote Host: (null)
Browser Type: Mozilla/3.0 (X11; I; SunOS 5.4 sun4m)
Remote Info: <null>
Name: Brian Cuttler
Email Address: [email protected]
CPU Architecture: VAX and Alpha
Version: v 6.2
Questions: 

Oops, sorry for the blank message, was looking for the
mailing address since mail from my web server fails (missing
a module) - usually fails.

Background:
I am running a mixed cluster, VAX6000-610 on CI, VAXserver 3100
on NI, Alpha 2100 on CI, VAX common system disk v6.1,
alpha system disk v6.2-1h3.

Since we will be phasing out the VAX, and the Alhpa is
not capable of running without the VAX (since we bought
the CI interface PCA-ci) I would like to enable the queue
manager to run on Alpha as well as VAX platforms.

I already had QMAN$MASTER pointing to the correct location
from both platforms, but starting the queue manager
$ init queue/manager/on=(alpha1,albny1) ! albny1 is the vax
resulted in a NOSUCHFILE, SYMDEL and INTERNALERROR errors.

Checking VMS cluster Systems manual p7-2, I see that the 
qman$master logical must be defined the "same" on all
nodes, I have corrected this but the error persists.

How do I correctly configure the systems so that the
queue manager can fail over across architectures?

					TIA,

					Brian Cuttler

University at Albany.
T.RTitleUserPersonal
Name
DateLines
1602.1Shared Queue Manager SetupXDELTA::HOFFMANSteve, OpenVMS EngineeringFri Feb 14 1997 17:0818
  Stop the queue manager on all nodes, check that the queue database
  device and directory is visible and mounted on all nodes, set up the
  logical names with the correct definitions, then restart the queue
  manager.

  Here is one local command used to start the queue manager:

  $ start /queue /manager /buffer='num_buffer' /extend=300 /restart -
	/on=(node,node,node,...) ddcu:[directory]

  The various queue-manager-related queue files (obviously) all reside
  in ddcu:[directory].  The system startup SYLOGICALS.COM procedure
  includes the line "$ DEFINE/SYS/EXEC QMAN$MASTER ddcu:[directory],
  and also makes sure that the target disk is mounted /SYSTEM.

  There are queue-manager-related patches available for various OpenVMS
  versions -- see http://www.service.digital.com/html/patch_public.html.