The 'cluster' config file has two lines to allow N7TR to connect, the node
has always connected as N7TR-15, but the sysop that runs the N7TR node
wanted me to allow (no ssid)N7TR to connect too, I guess he might want to
connect as N7TR instead of N7TR-15. I only mention this in case it could be
part of the problem. My 'cluster' config file looks like this:
#####################
n7tr 2a-
n7tr-15 2a-
#####################
N7TR-15 initiates the connect, everything is working fine with his connect
when it comes in. I get the spots and everything works great. N7TR is
running the 5452 version of the Pavillion software, I'm reporting this same
version upon connect to him in the clx_par file, again, I don't know if this
matters, but I thought I would mention it.
My problem starts when there isn't activity on the link, CLX sends a
PC51(ping) to N7TR-15 if it doesn't hear anything for 5 minutes, this is
well documented. The problem is that the Pavillion node simply replies back
with a acknowledgement frame and never sends a ping back. Here are the live
monitored frames as it happens from a captured listen session:
NC7J sends a ping:
----------------------------
[Mon Apr 27 13:57:25 1998]
Port tnos: AX25: NC7J->N7TR-15 v NVIP N7RH-10 <I C S6 R4> pid=Text
PC51^N7TR-15^NC7J^1^
----------------------------
N7TR-15 sends the acknowledgement for this frame:
----------------------------
[Mon Apr 27 13:57:26 1998]
Port tnos: AX25: N7TR-15->NC7J v N7RH-10* NVIP* <RR R R7>
----------------------------
NC7J waits awhile and then sends a disconnect because it never gets a ping
back. I assume anyway, take a look at the time... after another 5 minutes of
not getting anything, CLX disconnects:
----------------------------
[Mon Apr 27 14:02:25 1998]
Port tnos: AX25: NC7J->N7TR-15 v NVIP N7RH-10 <D C P>
----------------------------
N7TR-15 acknowledges the disconnect frame:
----------------------------
[Mon Apr 27 14:02:25 1998]
Port tnos: AX25: N7TR-15->NC7J v N7RH-10* NVIP* <UA R F>
----------------------------
Now the link is lost, N7TR-15 attempts to reconnect every 15 minutes on it's
end, so then the link is down for usually about 12 minutes depending on the
5 minute offset. If spots are sent during this time, my clx doesn't get the
information, obviously. The link comes right backup as soon as N7TR-15
initiates the connection again.
The connect/disconnect happens over and over. The duration of the
connections from N7TR-15 depend on how active the link is. Here is a list
of connects and disconnects from N7TR (the active Pavillion node I'm getting
the activity from at the time I wrote this message):
-------------
28-Apr-1998 0019Z n7tr-15 connected.
28-Apr-1998 0007Z n7tr-15 disconnected.
27-Apr-1998 2355Z n7tr-15 connected.
27-Apr-1998 2347Z n7tr-15 disconnected.
27-Apr-1998 2329Z n7tr-15 connected.
27-Apr-1998 2317Z n7tr-15 disconnected.
27-Apr-1998 2304Z n7tr-15 connected.
27-Apr-1998 2252Z n7tr-15 disconnected.
27-Apr-1998 2239Z n7tr-15 connected.
27-Apr-1998 2227Z n7tr-15 disconnected.
27-Apr-1998 2213Z n7tr-15 connected.
27-Apr-1998 2212Z n7tr-15 disconnected.
27-Apr-1998 2149Z n7tr-15 connected.
27-Apr-1998 2137Z n7tr-15 disconnected.
27-Apr-1998 2123Z n7tr-15 connected.
--------------
I hope someone is still reading at this point! What can I tell the sysop of
the N7TR-15 node? Is there something special he needs to do to get his node
to respond to the PC51 Pings? I'm sure if he was responding correctly, my
CLX node would keep the link up.
Can someone tell me the correct way that a Pavillion node should initiate a
connection to a CLX node, could N7TR-15 be doing it wrong?
Any help would be appreciated, if you need more information, please let me
know.
Regards
mg NG7M
N7TR
-- Matthew George mailto:vnuwgb.drzoryzza@septentrio.com mailto:gsjzsws.voazamq@mail.dy.fi NG7M