For all I know, the top of the output could be 10,000 (or more) lines up. Funny thing about endless loops, they tend to go on for quite a while. If you want, I'll post my conf files, which should be the same as the top of the output, no? The example.com realm should be in proxy.conf if you want any other confs just ask and I will post. $ grep -v -e \# proxy.conf proxy server { default_fallback = no } home_server localhost { type = auth ipaddr = 127.0.0.1 port = 1812 secret = testing123 require_message_authenticator = no response_window = 20 zombie_period = 40 revive_interval = 120 status_check = status-server check_interval = 30 num_answers_to_alive = 3 coa { irt = 2 mrt = 16 mrc = 5 mrd = 30 } } home_server virtual.example.com { virtual_server = virtual.example.com } home_server_pool my_auth_failover { type = fail-over home_server = localhost } realm example.com { auth_pool = my_auth_failover nostrip } realm LOCAL { } Like I said before though, I am running the default config (except for the nostrip line) so if authhost isn't set by default, I didn't add it. ~Huckle Berry On Tue, Jan 19, 2010 at 1:40 AM, Alan DeKok <aland@deployingradius.com>wrote:
Huckle Berry wrote:
Maybe proxy to itself was a bad way to describe it, you can interpret the output yourself if you'd like. I took the last 4096 lines of output
... from an endless loop which repeats the same thing.
Why not send the *top* of the output, before it starts to loop back to itself?
The debug output you posted does NOT match the other configs you sent. It clearly shows that the server is proxying to "example.com". This happens ONLY if you add "authhost" to the realm configuration for example.com.
The config you posted for example.com did *not* have an "authhost" entry.
And if you had posted the *top* of the debug output, it would have included the configuration for the "example.com" realm. Which would have showed *why* it was proxying
Alan DeKok. - List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html