Hello, I'd like to ask for help to the programmers of ax25.
I am ananlyzing the behaviour of ax25 protocol in Linux, and I am seeing things
that I am not able to understand, even looking the sources:
Here an example, maybe related with ax25 timers. The remote station
(ea4rct, running Linux) is sending me RRs acks of packets that have been
already been ackownedged via the piggy-back procedure, wich is in theory
impossible because the flag of "ack pending" is unset
(ax25->condition &=~AX25_COND_ACK_PENDING) in ax25_kick() function, wich is
called always just before t2 timer and this flag are tested to send a RR.
If the t2 timer expired before sending the I packet, the RR would be before it,
not after.
I don't find any explanation. This happens sometimes. t2 is 15 (1,5 seg), at
1200 bauds.
That is the trace from the sending station. The receiving is similar.
[Tue Feb 2 15:18:49 1999]
Port 1200: AX25: EA4ACL-3->EA4RCT-5 <I C S3 R6> pid=IP
IP: len 40 44.133.228.17->44.133.228.65 ihl 20 ttl 64 tos 16 DF prot TCP
TCP: 1104->telnet Seq xef971775 Ack x6603247b ACK Wnd 4320
[Tue Feb 2 15:18:49 1999]
Port 1200: AX25: EA4RCT-5->EA4ACL-3 <I C S6 R4> First seg; remain 1 pid=IP
IP: len 132 44.133.228.65->44.133.228.17 ihl 20 ttl 64 tos 16 DF prot TCP
TCP: telnet->1104 Seq x6603247b Ack xef971775 ACK PSH Wnd 4320 Data 88
0000 s..login.gz.. moni_rct..moni_2_rct.. textos_periodicos_listas.
0040 .[julian@gateradio julia
[Tue Feb 2 15:18:49 1999]
Port 1200: AX25: EA4RCT-5->EA4ACL-3 <I C S7 R4> remain 0
0000 n]$
[Tue Feb 2 15:18:50 1999]
Port 1200: AX25: EA4RCT-5->EA4ACL-3 <RR R R4>
--
Saludos de Juli�n
-.-
>> Powered by KDE / Linux <<
