|
|
Subject: Re: NAT on tun0 used with OpenVPN - msg#00013
List: security.firewalls.pfsense.user
Stefan Tunsch <stunsch <at> korrekto.com> writes:
>
>
> Hi!
>
>
> I need to set up
> outbound natting on tun0.
> tun0 is the virtual
> interface created and used by an OpenVPN client on my pfSense
> machine.
>
>
> This interface
> (tun0) isn't available for creating rules, NAT, etc on the web interface of
> pfSense.
>
>
> How can I set up
> outbound NATTING for this interface?
>
>
> Regards,
> Stefan
> --
> No virus found in this outgoing message.
> Checked by AVG Free Edition.
> Version: 7.1.409 / Virus Database: 268.14.3/530 - Release Date: 11/11/2006
>
Hi there,
i guess i understand the problem you have. i could not find the right button in
the web-gui for that setting, but i accomplished it through the cli. you need to
have some basic unix skills for that configuration!
just take a look at the nat-table with "pfctl -sn", you should see all
nat-rules. write them to a temp-file: "pfctl -sn > /var/tmp/tempfile"
then edit the temp-file: "vi /var/tmp/tempfile" and go to the last nat-item, for
example:
--snip--
nat on sis1 inet from 192.168.1.0/24 to any -> (sis1) round-robin
rdr-anchor "pftpx/*" all
--snip--
duplicate this line and change it to your needs (assume that 192.168.1.0 is your
local network and 192.168.2.0 is the ovpn remote network)
--snip--
nat on sis1 inet from 192.168.1.0/24 to any -> (sis1) round-robin
nat on tun0 inet from 192.168.1.0/24 to 192.168.2.0 -> (tun0) round-robin
rdr-anchor "pftpx/*" all
--snip--
do not touch the other lines! save the file and reread it with: "pfctl -Nf
/var/tmp/tempfile".
keep in mind that any change in firewall rules will delete this customization (a
reboot will also). try to automate it with grep and cron! if you need further
assistance just contact me.
if anyone has better solutions, let me know.
best regards
andreas
Was this page helpful?
Thread at a glance:
Previous Message by Date:
click to view message preview
Re: Problem with FW states
Thanks Espen, I already tried what you are suggesting, but it didn't solve my
problem.
I also tried to play with all the "Advanced" options reachable from the web
interface. It seems some of them (e.g. state: none, max states, states
timeout) don't work.
I've been able to analyze the network traffic: the situation is exactly the
one I described in the first post (reported below): when a state ESTABLISHED
exists because of a previous connection, the new SYN packet, coming from the
same IP:PORT of the previously established connection, is blocked by PF,
even if
- in FW rule page (firewall_rules_edit.php), ADVANCED options I set up PFSense
not to keep the state (working on the state number, timout)
- in the FW rule page (firewall_rules_edit.php) I set State Type to "none"
Furthermore, even if:
1.I set up the rule
pass DEV:high-ports=>SRV:23 State Type none
2. I don't set up any rule to enable the SRV's replies to pass through the
firewall
3. I reboot PFsense
I can establish new connections between DEV(:x) and SRV(:23).
I think this demonstrates that the "State Type none" option doesn't work. Does
it.
Any suggestion on how to solve my problem or fix PFSense State none issue?
Once again, thanks in advance and MC&HNY to everybody.
Odette
Alle 15:30, lunedì 18 dicembre 2006, Espen Johansen ha scritto:
> You can try to set "Firewall optimization options" in the Advanced
> page to "aggressive" and see if that helps.
>
> -lsf
>
> On 12/15/06, Odette <odette.nsaka@xxxxxxxxx> wrote:
> > Hi,
> >
> > I'm not able to find a solution to this problem:
> >
> > I've got some devices on the WAN net that need to open telnet connections
> > to a telnet server on the LAN net. (OK, don't tell me anything about
> > incoming telnet from WAN. At the moment I need to go on this way... )
> > LAN is bridged with WAN.
> > I've set up the FW rules and everything works fine.
> >
> > But...
> >
> > It often happens that the devices need to be resetted while a telnet
> > connection is estabilished. In this case, when the device "reboots" I
> > have to wait many minutes to estabilish a telnet connection.
> >
> > Looking at the FW state logs, I see that every "regular" telnet
> > connection is coming from the x port of the device where x is everytime
> > the same.
> >
> > Every time the device reboots, the new connection, estabilished just
> > waiting many minutes, comes from port x+1. On the FWStates log, I see
> > that the old state is still active.
> >
> > If I delete the FW states table before rebooting the device, the new
> > connection after reboot is estabilished immediately.
> > Furthermore, if I connect the device directly on the LAN switch,
> > (excluding PFSense filtering), I can reboot the device and have the new
> > connection immediately.
> >
> > I have not been able to analyze the network traffic, but I suppose that
> > the device tries everytime to estabilish the telnet connection form port
> > x and this is happening
> >
> > 1. A connection is estabilished
> > 2. PFSense keeps an active state DEV:x ==> SRV:23
> > 3. Devicereboot
> > 4. Device tries to estabilish a new connection (Syn from DEV:x to SRV:23)
> > 5. PFSense knows from it's states table that a connection DEV:x ==>
> > SRV:23 is already estabilished and drops the new DEV:x ==> SRV:23 Syn
> > packet 6. After some minutes the device reaches the time-out and tries a
> > new connection from port x+1. This new connection works fine.
> >
> > I've being trying to solve the problem by configuring PFSense
> > 1. inserting a new pass rule SRV:23 ==> DEV:(x...x+5)
> > 2. not to keep (Firewall: Rules: Edit: State Type: (Advanced) None) the
> > state for the "pass" rules
> > DEV:(x...x+5) ==> SRV:23
> > SRV:23 ==> DEV:(x...x+5)
> >
> > It doesn't work, even after rebooting PFSense. Furthermore, I can see the
> > the state in the States table. So I suppose that the advanced option
> > "State type: none" doesn't work.
> >
> > I also tried to set a state timeout to 10 seconds. The same effect: I can
> > see the connection state on the active state table for a long time.
> >
> > Any suggestion-info-idea?
> >
> > Thanks in advance to everybody
> >
> > Odette
Next Message by Date:
click to view message preview
2-node pfSense cluster failover
I really really like what I see in
http://pfsense.iserv.nl/tutorials/carp/carp-cluster-new.htm
and I happen to have a former HA NFS system based on mini-ITX
with two dead drives I would like to convert to a firewall.
This would work with the embedded image, right? I've flashed
a couple of CFDISK.1E (from PC Engines) with a 256 MByte
CF, which does seem to boot, and talk on serial. Unfortunately
I don't have access to these mini-ITX (VIA EPIA CL10000,
that's a C3 chip) to test. But these should work, right?
Unfortunately, these systems only have two NICs, and are in
a Travla C147 case http://206.14.132.88/products/Travla/c147/C147.html
so I'm not sure I can get more NICs in, even with a PCI riser.
My prior HA setup (drbd/heartbeat) used a crossover over serial
for signalling. No way to do that with pfsense, right?
--
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820 http://www.ativel.com
8B29F6BE: 099D 78BA 2FD3 B014 B08A 7779 75B0 2443 8B29 F6BE
signature.asc
Description: Digital signature
Previous Message by Thread:
click to view message preview
Problem with FW states
Hi,
I'm not able to find a solution to this problem:
I've got some devices on the WAN net that need to open telnet connections to a
telnet server on the LAN net. (OK, don't tell me anything about incoming
telnet from WAN. At the moment I need to go on this way... )
LAN is bridged with WAN.
I've set up the FW rules and everything works fine.
But...
It often happens that the devices need to be resetted while a telnet
connection is estabilished. In this case, when the device "reboots" I have to
wait many minutes to estabilish a telnet connection.
Looking at the FW state logs, I see that every "regular" telnet connection is
coming from the x port of the device where x is everytime the same.
Every time the device reboots, the new connection, estabilished just waiting
many minutes, comes from port x+1. On the FWStates log, I see that the old
state is still active.
If I delete the FW states table before rebooting the device, the new
connection after reboot is estabilished immediately.
Furthermore, if I connect the device directly on the LAN switch, (excluding
PFSense filtering), I can reboot the device and have the new connection
immediately.
I have not been able to analyze the network traffic, but I suppose that the
device tries everytime to estabilish the telnet connection form port x and
this is happening
1. A connection is estabilished
2. PFSense keeps an active state DEV:x ==> SRV:23
3. Devicereboot
4. Device tries to estabilish a new connection (Syn from DEV:x to SRV:23)
5. PFSense knows from it's states table that a connection DEV:x ==> SRV:23 is
already estabilished and drops the new DEV:x ==> SRV:23 Syn packet
6. After some minutes the device reaches the time-out and tries a new
connection from port x+1. This new connection works fine.
I've being trying to solve the problem by configuring PFSense
1. inserting a new pass rule SRV:23 ==> DEV:(x...x+5)
2. not to keep (Firewall: Rules: Edit: State Type: (Advanced) None) the state
for the "pass" rules
DEV:(x...x+5) ==> SRV:23
SRV:23 ==> DEV:(x...x+5)
It doesn't work, even after rebooting PFSense. Furthermore, I can see the the
state in the States table. So I suppose that the advanced option "State type:
none" doesn't work.
I also tried to set a state timeout to 10 seconds. The same effect: I can see
the connection state on the active state table for a long time.
Any suggestion-info-idea?
Thanks in advance to everybody
Odette
Next Message by Thread:
click to view message preview
2-node pfSense cluster failover
I really really like what I see in
http://pfsense.iserv.nl/tutorials/carp/carp-cluster-new.htm
and I happen to have a former HA NFS system based on mini-ITX
with two dead drives I would like to convert to a firewall.
This would work with the embedded image, right? I've flashed
a couple of CFDISK.1E (from PC Engines) with a 256 MByte
CF, which does seem to boot, and talk on serial. Unfortunately
I don't have access to these mini-ITX (VIA EPIA CL10000,
that's a C3 chip) to test. But these should work, right?
Unfortunately, these systems only have two NICs, and are in
a Travla C147 case http://206.14.132.88/products/Travla/c147/C147.html
so I'm not sure I can get more NICs in, even with a PCI riser.
My prior HA setup (drbd/heartbeat) used a crossover over serial
for signalling. No way to do that with pfsense, right?
--
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820 http://www.ativel.com
8B29F6BE: 099D 78BA 2FD3 B014 B08A 7779 75B0 2443 8B29 F6BE
signature.asc
Description: Digital signature
|
|