On Tue, 29 Mar 2005 14:36:42 -0500, Alan DeKok <[EMAIL PROTECTED]> wrote: > I believe it is. See src/modules/rlm_sql/rlm_sql.c, which calls > rad_check_ts().
Yup, I definitely see that.. And now that I'm digging deeper, I'm seeing the problem.. *sigh* So here's what I'm guessing is going on... We changed IP addresses a while back. The old IP's no longer exist, but there are apparently a number of radacct records that were never "stopped" correctly. So when the checkrad process runs, it sees these old records, can't identify the NAS, and reports that it's skipping them. Even if there are no records for a recognized NAS, the presence of "old" records there causes a reject. I tried looking through the source and I can see where this message is sent. It sends a return value of 1 if this happens, and it appears that a return of 1 indicates an MPP attempt... Is that about right? So I guess my best course of action right now is to clear out those old records. :) > Alan DeKok. -- Jason 'XenoPhage' Frisvold [EMAIL PROTECTED] - List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html
