[Dxspider-support] Long mail/bulletin messages

[Charlie Carroll]

I'm using the latest version of the Windows version and noticed a potential
problem with long mail & bulletin messages.  I sent a message that contained
about 32/33 PC29s from an AK1A node.  To another AK1A node this works fine
as it has for years.  When I attempt to send that same message to my
DXSpider node (W1HR), the message never arrives at W1HR.  Since the AK1A
node is only about 5 feet away from W1HR, I can see the packets on displays
at both ends.  It appears that the complete message is leaving the AK1A, but
not ending up in DXSpider.  As a result, the AK1A node eventually times out
on that message.  One other thing.  There is an intermediate gateway between
the AK1A node and the W1HR node.  This gateway is my access to the internet
and handles the interface between the COM port from the AK1A node to the
gateway and the 10BaseT lan from the gateway to the W1HR node.  That being
said, here's what's ending up at the W1HR gateway.

<- I K1XX PC29^W1HR^K1XX^25^desires. The same goes for sh/dx. You can do a
whole lot of filtering here^~

<- I K1XX PC29^W1HR^K1XX^25^too. Note that the spots from GB7DXA are not
being forwarded to K1XX, and^~

<- I K1XX PC29^W1HR^K1XX^25^vice versa. There's a lot of power in this
software and so far I really^~t.^~

Here's what's on the ethernet feeding the W1HR node.
0100  a whole lot of filtering here^~..PC29^W1HR^K1XX^20^too.  Note th
0140  at the spots from GB7DXA are not being forwarded to K1XX, and^~.
0180  .PC29^W1HR^K1XX^20^vice versa.  There's a lot of power in this s
01c0  oftware and so far I really^~..PC29^W1HR^K1XX^20^like i

lan send: Tue Apr 03 20:26:55 2001
ETHER: len 60 00:40:05:2b:f1:6fÍ00:e0:29:6f:55:a5 type IP
IP: len 46 192.168.1.1Í192.168.1.3 ihl 20 ttl 254 prot TCP
TCP: 23Í2256 Seq x476a88fe Ack x5e1ca2f PSH ACK Wnd 2048 Data 6
0000  t.^~..

Most of the last PC29 seems to have gone astray.  If you look closely, you
will see that the last PC29 shown in the partial first packet (01c0) is
missing.  However, the "t.^~"  in the second packet (0000) is there.  Even
though there are another 3 or 4 PC29s coming from the AK1A node, they never
show up at all at W1HR.  Hence, the message timeout at K1XX.

Any one have any thoughts?

73 Charlie, k1xx