Try adding 'radius-server timeout 10' to your Ciscos...
To check the SQL, start RadAdmin and enable all debug checkboxes, and set
Authenticating logfile to c:\radius\auth.log and Accounting to
c:\radius\acct.log, save and exit ... you can monitor those logs to see the
radius packets and the SQL code. (Don't forget about them after your
problem is solved, they can get real big real fast!)
Just above the ODBC 23000 errors, my guess is that you'll find 2 identical
radius packets (probably in acct.log) -- the second one creates the dup key
error. Caused by either too low a timeout setting, or maybe too slow a
response on your radius server (check resp times in the logfiles). We
solved a problem here by adding a faster SCSI drive controller.
HTH,
-Ted
At 03:48 PM 8/15/98 -0300, you wrote:
>Hello Everybody,
>
>I've recently installed Radius 2.5.
>I have 2 Cisco 2511 and 1 Ascend MAX 4000 running in my ISP (Cisco1, Cisco2
>and Waimea).
>The system works fine for them exept for Cisco2.
>From time to time I get an ODBC error and Calls and CallsOnLine tables are
>not updated.
>I've monitored radius using -x15 and the errors did not show up, I mean,
>they happenned but did not appear in the log).
>When I run Radius normally, the logfile in the data directory shows the
>following message:
>
>Sat Aug 15 14:47:24 1998: ODBC Error:23000:-1605:
> [Microsoft][ODBC Microsoft Access 97 Driver] The changes you requested to
>the table were not successful because they would create duplicate values in
>the index, primary key, or relationship. Change the data in the field or
>fields that contain duplicate data, remove the index, or redefine the index
>to permit duplicate entries and try again.
>
>My question is: How can I get the SQL statment that generated the error?
>
>Thanks.
>Best regards,
>
>Fernando
>MyWay Provedor Internet
>
>
>