Re: Problems with counter module

2004-10-12 Thread Kostas Kalevras
On Tue, 12 Oct 2004 [EMAIL PROTECTED] wrote: > Hi all, > > > rlm_counter: Could not find Service-Type attribute in the request. > > Returning NOOP. > > So fix that. See allowed-servicetype configuration directive ( i thought > it > would be rather obvious). > > In radiusd.conf: > > count

Re: Problems with counter module

2004-10-12 Thread macleod
Hi all, > rlm_counter: Could not find Service-Type attribute in the request. > Returning NOOP. So fix that. See allowed-servicetype configuration directive ( i thought it would be rather obvious). In radiusd.conf: counter daily { filename = ${raddbdir}/db.daily

Re: Problems with counter module

2004-10-11 Thread Kostas Kalevras
On Mon, 11 Oct 2004 [EMAIL PROTECTED] wrote: > Hi all, > maybe I found the reason... but I don't know how to fix it. Thanks in advance > > > rlm_counter: Could not find Service-Type attribute in the request. > Returning NOOP. So fix that. See allowed-servicetype configuration directive

Re: Problems with counter module

2004-10-11 Thread macleod
Hi all, something more: if i put in users file time limits on login-time, that works! but if i use a time quota of 1 hour every day, after that hour i can connect again for 1 hour and again... Example (with login time limits, it works!): Pablo Auth-Type := Local, User-Password == "Pablo", Logi

Re: Problems with counter module

2004-10-11 Thread macleod
Hi all, maybe I found the reason... but I don't know how to fix it. Thanks in advance rlm_counter: Could not find Service-Type attribute in the request. Returning NOOP. modcall[accounting]: module "daily" returns noop for request 2 modcall[accounting]: module "unix" returns o

Re: Re: Problems with counter module

2004-10-11 Thread Kostas Kalevras
On Sun, 10 Oct 2004 [EMAIL PROTECTED] wrote: > Thanks a lot for your answers. > > You said: > "So check if the user sessions are recorded. Maybe the accounting stop do > not contain a session-time attribute. Post an accounting-stop debug output > for the user." > > I understand what you mean, but