xinetd entries

davec25

Does anyone know why on pfSense 2.4.2 I'd be seeing entries like this in xinetd.conf?

service 19004-tcp
{
type = unlisted
bind = 127.0.0.1
port = 19004
socket_type = stream
protocol = tcp
wait = no
user = nobody
server = /usr/bin/nc
server_args = -w 2000 172.16.0.119 80
}

It's leaving several hung netcat processes in the process stack.

johnpoz

@davec25 said in xinetd entries:

xinetd.conf

Where are you viewing this exactly.. What packages do you have installed?

Did you manually install something?

davec25

I can get the list of packages if we really need it, but after some research, it looks like this is how pfSense does NAT redirection. Maybe you can confirm that for me.

I'm seeing 111 hung processes for this particular rule, so I suspect there was a problem with host 127.16.0.119 for a while and these processes never self terminated.

If you would, please let me know if my analysis is correct.

Thanks

johnpoz

I have nat redirection setup.. And do not see that. Do you mean nat reflection?

What redirection do you have setup, and where exactly is this xinetd.conf file your seeing - what is the full path to the file?

And what exact redirection are you doing?

stephenw10

Those are almost certainly from using NAT reflection in NAT+proxy mode. You probably don't need that, Pure NAT mode works for almost all cases.

Steve

davec25

Thanks for the clarification!

johnpoz

to be honest any sort of nat "reflection" is just an abomination if you ask me.. Why not just have your local stuff resolve the local IP vs any sort of reflection off your public IP.. Simple host override is all it takes.

Only reason I can think of doing a reflection would be to work around the horrible coding of some app that uses a IP vs a fqdn as destination.