When I rcp a file, the socket is never closed cleanly. But rsh itself works fine. I suspect this is a user level mistake as Solaris does the same thing. bash$ rcp /tmp/foo wumpus:/tmp/foo ; netstat -t Active Internet connections (w/o servers) Proto Recv-Q Send-Q Local Address Foreign Address State tcp 0 0 gleep.cs.Virginia.:1022 wumpus.cs.Virgini:shell FIN_WAIT2 bash$ rcp /tmp/foo wumpus:/tmp/foo ; netstat -t Active Internet connections (w/o servers) Proto Recv-Q Send-Q Local Address Foreign Address State tcp 1 0 gleep.cs.Virginia.:1020 wumpus.cs.Virgini:shell TIME_WAIT tcp 1 0 gleep.cs.Virginia.:1022 wumpus.cs.Virgini:shell TIME_WAIT The above is with redhat i386 5.2 and a 2.0.35 kernel. Below is 5.2 alpha with a 2.2.5 kernel: [ after several repeats... ] [lindahl@test00 /tmp]$ rcp /tmp/asdf test01:/tmp/asdf ; netstat -t Active Internet connections (w/o servers) Proto Recv-Q Send-Q Local Address Foreign Address State tcp 0 0 test00:1020 test01:shell TIME_WAIT tcp 0 0 test00:1021 test01:shell TIME_WAIT tcp 0 0 test00:1022 test01:shell TIME_WAIT tcp 0 0 test00:1023 test01:shell TIME_WAIT I only noticed because I repeatedly rcp to 320 machines, and this caused me to run out of low ports. Below is Solaris 2.5.something: adder:/tmp $ rcp /tmp/asdf mamba:/tmp/asdf ; netstat -f inet | grep mamba adder.1020 mamba.22 8760 0 8760 0 TIME_WAIT adder:/tmp $ rcp /tmp/asdf mamba:/tmp/asdf ; netstat -f inet | grep mamba adder.1020 mamba.22 8760 0 8760 0 TIME_WAIT adder.1019 mamba.22 8760 0 8760 0 TIME_WAIT -- g - To unsubscribe from this list: send the line "unsubscribe linux-net" in the body of a message to [EMAIL PROTECTED]
