FR 2.1.3 and ASSERT FAILED event.c

Alan DeKok aland at deployingradius.com
Wed Feb 25 16:04:56 CET 2009


Chris Howley wrote:
> I encountered the following problem when the server received an Access-Challenge packet
> from a proxy server. Any help in fixing this problem would be appreciated.

  See doc/bugs for giving additional information, such as the rest of
the back trace.

  Also, a lot more of the debug log might help.

> Waking up in 0.9 seconds.
> rad_recv: Access-Challenge packet from host 194.82.174.185 port 1812, id=76, length=81
>         Tunnel-Type:0 = VLAN
>         Tunnel-Medium-Type:0 = IEEE-802
>         EAP-Message = 0x010300061920
>         Message-Authenticator = 0x193c8361dc660dd940460f693d6ebf9c
>         State = 0xad8b0646ad881f6aaefeee6ec7165a25
>         Proxy-State = 0x313730
> +- entering group post-proxy {...}
> [post_proxy_log]        expand: /usr/local/var/log/radius/radacct/%Y-%m-%d/post-proxy-detail-%H:00 -> /usr/local/var/log/radius/radacct/2009-02-24/post-proxy-detail-16:00
> [post_proxy_log] /usr/local/var/log/radius/radacct/%Y-%m-%d/post-proxy-detail-%H:00 expands to /usr/local/var/log/radius/radacct/2009-02-24/post-proxy-detail-16:00
> [post_proxy_log]        expand: %{Packet-Src-IP-Address} - %t -> 10.12.80.101 - Tue Feb 24 16:02:50 2009
> ++[post_proxy_log] returns ok
> [attr_filter.post-proxy]        expand: %{Realm} -> jrs
>  attr_filter: Matched entry DEFAULT at line 103
> ++[attr_filter.post-proxy] returns updated
> [eap] No pre-existing handler found
> ++[eap] returns noop
> ASSERT FAILED event.c[3593]: fun != NULL
> Abort (core dumped)

  This is a catastrophic error indicating that the server has a request
it doesn't know how to handle.

  The only way that this could happen is:

  a) buffer over-run somewhere
  b) source code modifications

  The code that receives a proxied response sets "fun", and doesn't do a
whole lot else before it hits that assertion.  If you're seeing this in
debugging mode (i.e. no threads), then there *very* few things that can
go wrong here.

  Alan DeKok.



More information about the Freeradius-Users mailing list