That seems fair enough for the debuggee process, but the problem is that
in this case, jps doesn't report other java processes either.


Here is a script that demonstrates this.
#!/bin/sh
if [ $# = 0 ] ; then
    java=java
    jps=jps
    javac=javac
else
    java=$1/bin/java
    jps=$1/bin/jps
    javac=$1/bin/javac
fi

cat <<EOF > /tmp/hello.java


public class hello {
    public static void main(String args[]) {
        try {
            Thread.sleep(50000);
        }
        catch (Exception ee) {
        }
    }
}
EOF

$javac -d . /tmp/hello.java

echo "jps without a waiting debuggee running:"
$java hello &
pid1=$!
$jps

echo
echo "run a debuggee that waits for a connection:"
$java -Xrunjdwp:transport=dt_socket,address=46225,server=y,suspend=y hello &
pid=$!

echo
sleep 5
echo "jps again"
$jps

kill $pid
kill $pid1
This failure mode has been seen in nightly testing. Here is
my analysis report entry:

New MISC_REGRESSION failures (from 2007.11.22)
    sun/tools/jps/jps-lm.sh
    sun/tools/jps/jps-l_2.sh
    sun/tools/jps/jps-m_2.sh
    sun/tools/jps/jps-Vm_2.sh
    sun/tools/jps/jps-Vvml.sh
    sun/tools/jps/jps-Vvml_2.sh
    sun/tools/jps/jps-V_2.sh
        These tests failed due to "Could not synchronize with target" on
        Win32 Client VM (machine vm-8450r-3).

        Update: Swamy believes these failures are due to problems with
            the test machine.

        Update: Swamy found the following existing bug:
            6602355 4/4 If a debuggee is waiting for a connection, jps
                        fails with msg: Could not synchronize with target

        Update: Swamy verified that the failures on vm-8450r-3 are due
            to 6602355.
Here is another analysis report entry:

New MISC_REGRESSION failures (from 2007.11.28)
    sun/tools/jps/jps-Defaults.sh
    sun/tools/jps/jps-l_1.sh
* sun/tools/jps/jps-m.sh
    sun/tools/jps/jps-Vm_2.sh
        These tests failed due to "Could not synchronize with target"
        on Win-AMD64 Server VM (machine vm-x2100-01). These failures
        are an occurrence of the following bug:

            6602355 4/4 If a debuggee is waiting for a connection, jps
                        fails with msg: Could not synchronize with target
Here is another analysis report entry:

New MISC_REGRESSION failures (from 2007.12.06)
* sun/tools/jps/jps-Vvm.sh
* sun/tools/jps/jps-v_1.sh
* sun/tools/jps/jps-vm_1.sh
        These tests failed due to "Could not synchronize with target"
        on Win32 Server VM (machine colfax005). These failures are an
        occurrence of the following bug:

            6602355 4/4 If a debuggee is waiting for a connection, jps
                        fails with msg: Could not synchronize with target