----------  Forwarded Message  ----------

Subject: BOUNCE [EMAIL PROTECTED]:    Non-member submission from ["Brad 
Lilly" <[EMAIL PROTECTED]>]
Date: Sun, 17 Nov 2002 18:23:04 -0600
From: [EMAIL PROTECTED]
To: [EMAIL PROTECTED]

>From [EMAIL PROTECTED] Sun Nov 17 18:23:04 2002
Received: from pc-00050.gate1.shoal.net.au ([EMAIL PROTECTED]
 [203.202.164.4]) by server1.open.com.au (8.11.0/8.11.0) with ESMTP id
 gAI0N3C22833
        for <[EMAIL PROTECTED]>; Sun, 17 Nov 2002 18:23:03 -0600
Received: from localhost.localdomain ([127.0.0.1] helo=itbasecamp.com.au)
        by pc-00050.gate1.shoal.net.au with smtp (Exim 4.10)
        id 18DeF7-0006um-00
        for [EMAIL PROTECTED]; Mon, 18 Nov 2002 16:14:41 +1100
Received: from 192.168.0.12
        (SquirrelMail authenticated user bradley)
        by pc-00050.gate1.shoal.net.au with HTTP;
        Mon, 18 Nov 2002 16:14:41 +1100 (EST)
Message-ID:
 <[EMAIL PROTECTED]> Date:
 Mon, 18 Nov 2002 16:14:41 +1100 (EST)
Subject: accounting double up avoidance
From: "Brad Lilly" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
X-Priority: 3
Importance: Normal
X-MSMail-Priority: Normal
X-Mailer: SquirrelMail (version 1.2.7)
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
X-Spam-Score: 0.0 (/)
X-Scanner: exiscan for exim4 (I.T. Basecamp Pty Ltd)
 *18DeF7-0006um-00*ju3OGyhfG2Q*

Hello all,

Just wanted some opinions on accounting double up avoidance.

We are using mysql with radiator and currently we get double ups mainly
from VISP services (due to congestion) and sometimes from our local NAS's

We are using the bog standard insert statements. We have dupintervals at
10 secs for all clients

At first I thought it would be simple to have a unique index in the
accounting table based on the account session id but as we know our cisco
NAS's don't give a unique account session id. I then thought maybe a
concatenation of acct sessionid, nasport and curdate()(or something like
that) as a unique index in the table. Then this would leave the insert
queries failing on double ups

How would radiator handle this? Our double ups at best are sometimes ten
times in a row (the radius data comes across the wan for VISP) I know this
wouldn't be really great but when you look to other methods that I have
thought of below ....

I thought maybe a replace based on the same concatenated unique index
above in a where clause would work but then account start and stop times
shift right with every double up as they are based on timestamp -
acct-sessiontime(starttime) and timestamp(stop time)

I did notice in the doco the example of using the AcctInsertQuery with a
update or insert query. Again I can see that this will move the acctstart
& stop times right with every double up.

Anyway thats where I am at. Any feedback or comments would be appreciated

--
Brad Lilly
Ph +61244218555
Fax +61244218500

-------------------------------------------------------

-- 
Mike McCauley                               [EMAIL PROTECTED]
Open System Consultants Pty. Ltd            Unix, Perl, Motif, C++, WWW
24 Bateman St Hampton, VIC 3188 Australia   http://www.open.com.au
Phone +61 3 9598-0985                       Fax   +61 3 9598-0955

Radiator: the most portable, flexible and configurable RADIUS server 
anywhere. SQL, proxy, DBM, files, LDAP, NIS+, password, NT, Emerald, 
Platypus, Freeside, TACACS+, PAM, external, Active Directory, EAP, TLS, 
TTLS etc on Unix, Windows, MacOS etc.

===
Archive at http://www.open.com.au/archives/radiator/
Announcements on [EMAIL PROTECTED]
To unsubscribe, email '[EMAIL PROTECTED]' with
'unsubscribe radiator' in the body of the message.

Reply via email to