using ipfw seems to interfere with socket communication

Mikhail Teterin mi+mx at aldan.algebra.com
Tue Jun 27 22:13:44 UTC 2006


After I added the following rules to my ipfw configuration:

	ipfw pipe 1 config bw 2000000000Mbit/s
	ipfw add 200 pipe 1 tcp from any to any

The following test from Tcl's regression-test suite started to fail:

    set s [socket -server accept 0]
    set sock ""
    set s2 [socket 127.0.0.1 [lindex [fconfigure $s -sockname] 2]]
    vwait sock
    puts $s2 "one"
    flush $s2
    after 500
    fconfigure $sock -blocking 0
    set result a:[gets $sock]
    lappend result b:[gets $sock]  
    fconfigure $sock -blocking 1
    puts $s2 "two"
    flush $s2
    fconfigure $sock -blocking 0
    lappend result c:[gets $sock]
    fconfigure $sock -blocking 1
    close $s2
    close $s
    close $sock
    set result

Instead of the expected ``a:one b: c:two'', the result was consistently 
``a:one b: c:''. This means, the gets in the line:

	lappend result c:[gets $sock]

was always returning empty string, instead of the string "two", that was 
written into the socket and flushed.

Is the test wrong, and such result is possible, or is dummynet triggering a 
bug? Unloading the dummynet module allows the test to succeed...

Thanks!

	-mi


More information about the freebsd-net mailing list