This is a correct behaviour. We do auth first and then (if auth is successful) make a routing decision. So the destination IP is not known until the routing decision is made. That's also a good idea for some other reason - if you need to measure ASR for destination gateways, you don't want to count calls for which auth failed at the gatekeeper.
----- Original Message ----- From: "Aivis Olsteins" <aivis@xxxxxxxxxxxxxxxx>
Sent: Sunday, January 30, 2005 4:49 PM
you are right, it works, at least on authentication stage. I messed up with older log files, sorry
However, there is some problem (possibly related) with accounting module i guess:
In cases where call is denied due to access-reject from radius, destination ip address is not set (0.0.0.0:0) both in status port CDR and Radius Accounting "h323-remote-address". This does not affect call itself, but just creates improper CDR.
Example below: (call is rejected by radius by some valid reason):
2005/01/30 15:29:51.884 3 gkauth.cxx(1070) GKAUTH RadAliasAuth Setup check failed
2005/01/30 15:29:51.885 4 ProxyChannel.cxx(1446) Q931 Dropping call from XX.XX.XX.XX:42042 due to Setup authentication failure
2005/01/30 15:29:51.885 2 RasTbl.cxx(1969) CallTable::Insert(CALL) Call No. 119, tota
l sessions : 10
2005/01/30 15:29:51.885 3 radproto.cxx(2102) RADIUS
Sending PDU to RADIUS server XX.XX.XX.XY (XX.XX.XX.XY:1813) from port:18444[active requests: 0, ID space: 41-46], PDU: {
code = 4 (Accounting-Request)
id = 45
length = 408 octets
....................................
[15]= {
type = 26 (Vendor-Specific)
length = 35 octets
vendorId = 9
vendorType = 23
vendorLength = 29
vendorValue = 27 octets {
68 33 32 33 2d 72 65 6d 6f 74 65 2d 61 64 64 72 h323-remote-addr
65 73 73 3d 30 2e 30 2e 30 2e 30 ess=0.0.0.0
}
}
....................................
2005/01/30 15:29:51.890 3 gkacct.cxx(918) GKACCT RadAcct logged event 1 for call no
. 119
2005/01/30 15:29:51.890 2 gkacct.cxx(958) GKACCT Successfully logged event 1 for ca
ll no. 119
2005/01/30 15:29:51.890 1 RasTbl.cxx(2136) CDR|119|02 b2 1d d9 0d 9b 20 1e 0d fd 56 3
4 34 34 34 ef|0|unconnected|Sun, 30 Jan 2005 15:29:51
GMT|XX.XX.XX.XX:42042| |0.0.0.0:0| |70788:dialedDigits|id1:h323_ID|gk104;^M
2005/01/30 15:29:51.890 3 radproto.cxx(2102) RADIUS
Sending PDU to RADIUS server XX.XX.XX.XY (XX.XX.XX.XY:1813) from port:18444[active requests: 0, ID space: 41-47], PDU: {
code = 4 (Accounting-Request)
id = 46
length = 507 octets
.......................................
[18]= {
type = 26 (Vendor-Specific)
length = 35 octets
vendorId = 9
vendorType = 23
vendorLength = 29
vendorValue = 27 octets {
68 33 32 33 2d 72 65 6d 6f 74 65 2d 61 64 64 72 h323-remote-addr
65 73 73 3d 30 2e 30 2e 30 2e 30 ess=0.0.0.0
}
}
........................................
2005/01/30 15:29:51.901 3 gkacct.cxx(918) GKACCT RadAcct logged event 2 for call no
. 119
2005/01/30 15:29:51.901 2 gkacct.cxx(958) GKACCT Successfully logged event 2 for ca
ll no. 119
Best regards, Aivis mailto:aivis@xxxxxxxxxxxxxxxx
------------------------------------------------------- This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting Tool for open source databases. Create drag-&-drop reports. Save time by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc. Download a FREE copy at http://www.intelliview.com/go/osdn_nl
_______________________________________________________
List: Openh323gk-users@xxxxxxxxxxxxxxxxxxxxx Archive: http://sourceforge.net/mailarchive/forum.php?forum_id=8549 Homepage: http://www.gnugk.org/