I've got a NetBSD 1.6 firewall sitting behind a Telstra
Bigpond aDSL link, using IP Filter: v3.4.27 (336) (vanilla
netBSD 1.6).

We're running a small ruleset, and trying to NAT an
internal box for backups etc over rsync. The
ipnat entry is :

rdr pppoe0 0.0.0.0/0 port 22 -> 192.168.0.3 port 22 tcp

ipf.conf has :

pass in quick on pppoe0 proto tcp from xxxx to yyyy port = 22 flags S/SA keep state

This seems to work just fine, I can ssh in to the box
from the outside and all seems well.

_but_

When I run some larger rsyncs from the outside
to the inside box I see this :

Read from remote host foo.bar: Connection timed out
rsync: connection unexpectedly closed (465051 bytes read so far)
rsync error: error in rsync protocol data stream (code 12) at io.c(150)
rsync: connection unexpectedly closed (465051 bytes read so far)
rsync error: error in rsync protocol data stream (code 12) at io.c(150)

I don't think this is an MTU issue, although I don't really know
enough about it to be sure. We're clamping the DSL link
(RP pppoe). The rsyncs are using ssh, and have worked
previously when we had a RH ip-drains firewall in place, but
no redirect (the firewall was also the rsync host). Now we've
got a proper seperate firewall I need to get it working again :)

The two boxes on either side are Linux boxes (yeah ... I know ..
but I can't convert them to BSD yet, I'm working on it!), and
running reasonably current versions of OpenSSH and rsync. Before
I go into upgrading everything etc, has anyone seen this before
and maybe has a quick suggestion to make? Kernel settings etc?

thanks

Carl

--
=======================
Vivitec Pty. Ltd.
Level 2, 2-4 Bridge Rd.
Richmond, 3121.
Ph. +61 3 9429 3555
Fax +61 3 9421 8969
=======================

Reply via email to