Re: Root level prodir combined results

pays@faugeres.inria.fr Sat, 16 April 1994 12:47 UTC

Received: from ietf.nri.reston.va.us by IETF.CNRI.Reston.VA.US id aa01408; 16 Apr 94 8:47 EDT
Received: from CNRI.RESTON.VA.US by IETF.CNRI.Reston.VA.US id aa01404; 16 Apr 94 8:47 EDT
Received: from haig.cs.ucl.ac.uk by CNRI.Reston.VA.US id aa05266; 16 Apr 94 8:47 EDT
Received: from bells.cs.ucl.ac.uk by haig.cs.ucl.ac.uk with local SMTP id <g.01696-0@haig.cs.ucl.ac.uk>; Sat, 16 Apr 1994 12:52:55 +0100
Via: uk.ac.nsfnet-relay; Sat, 16 Apr 1994 12:52:46 +0100
Received: from faugeres.inria.fr by sun3.nsfnet-relay.ac.uk with Internet SMTP id <sg.14242-0@sun3.nsfnet-relay.ac.uk>; Sat, 16 Apr 1994 12:52:18 +0100
X400-Received: by /PRMD=inria/ADMD=atlas/C=fr/; Relayed; 16 Apr 94 13:52:11+0200
Date: Sat, 16 Apr 1994 13:52:11 +0200
Sender: ietf-archive-request@IETF.CNRI.Reston.VA.US
From: pays@faugeres.inria.fr
To: /DD.Common=tech-support/@nexor.co.uk, osi-ds@cs.ucl.ac.uk, panamas@paradise.ulcc.ac.uk, prodir-support@nexor.co.uk
Subject: Re: Root level prodir combined results
Message-ID: <766497131.29505.0-faugeres.inria.fr*@MHS>

Just one comment about the tool

Some of you may be surprised to have 2 results entries for
the french Master    C=fr, O=cnrs, CN=opaxdsa

This is certainly a result of the configuration that has been set up
here for some time now: a "distributed fault-tolerant" DSA

  In fact we have set up 2 physical different servers (synchronised
  by external means) both of which appears to be the official
  french master OPAXdsa.
  The trick to have one single fault-tolerant DSA was to
  register (in conformance with the X.500 model) one single
  server but which PSAP contains the NSAPs of both physical servers

  The result seems to be close to excellent, as it really
  provide a close to 100% level of availability

BUT, this unusual configuration seems to "bother" a bit the prodir tool,
as indicated by the results published

From my understanding the results contain
  1. a global availability result
  2.a second unidentified entry 
    (which may correspond to one of the physical servers)
    what is measured exactly?
    why do we have 2 entries and not either 1 or 3?

Could Nexor people have a look at the issue
  and if possible limit the results to ONE entry
  which correspond the the users perceived availability?

Thanks,

-- PAP