tinc connection only usable after ping from other side

[uws]

På Fri, Jan 16, 2004 at 03:28:04PM +0100, Guus Sliepen skrev:
> On Fri, Jan 16, 2004 at 02:33:18PM +0100, uws wrote:
> > > It's caused by a masquerading firewall on your server's side.
> > > Fix your masquerading firewall (look at the examples on tinc's website)
> > > or use TCPOnly = yes (see the documentation).
> > I'm experiencing exactly the same problem as Moritz Maisel does. I run a
> > masquerading firewall on my server host. Even when using TCPOnly = yes
> > (which is what I use for all my connections) the problem persists. Any clue?
> Stateful firewall rules?
> The only other thing I can imagine is if you use the ethertap device
> instead of the tun/tap device and have set a different MAC address than
> fe:fd:0:0:0:0 and haven't disabled ARP.

You can view my firewall script at [1]. The tinc daemon runs on the same
machine as the firewall script, so I don't need portforwarding. This box has
(1) a pptp internet connection to the internet over a (2) dedicated cross
cable LAN to my ADSL modem and a (3) local area network connection to other
hosts. Everything works just fine, except for tinc.

Also, my 2.4 kernel's .config file contains

| # CONFIG_ETHERTAP is not set
| CONFIG_TUN=y

...so I suppose I'm using the correct device.


  mvrgr, Wouter


[1] http://www.xs4all.nl/~uws/firewall/
-- 
:wq                                                       mail uws at xs4all.nl

we'd drink and get high until late :: and now we're all alone     -- placebo

Tinc:         Discussion list about the tinc VPN daemon
Archive:      http://mail.nl.linux.org/lists/
Tinc site:    http://tinc.nl.linux.org/