[OpenVPN home] [Date Prev] [Date Index] [Date Next]
[OpenVPN mailing lists] [Thread Prev] [Thread Index] [Thread Next]
Google
 
Web openvpn.net

Re: [Openvpn-users] Problems With UDP and Server Mode


  • Subject: Re: [Openvpn-users] Problems With UDP and Server Mode
  • From: Leonard Isham <leonard.isham@xxxxxxxxx>
  • Date: Tue, 1 Nov 2005 18:41:58 -0500

On 11/1/05, Jez Rogers <jezndi@xxxxxxxxxxx> wrote:
> Giancarlo Razzolini wrote:
>
> > Where 300 is in seconds. This means that any peer has 5 minutes to
> > complete the handshake. I believe that this problem can happen to with
> > other conectivity problems, as mtu, etc. Try increasing the hand-window
> > and see what happens. If it doesn't work, check your connection for any
> > mtu problems (try to transfer a big file, and see if there are any
> > fragmentation, or lost packets).
> >
>
> Connection seems fine. Using ssh I transfered a 9 meg tar file in just
> over a minute. Same file gzipped to 1.9 meg took 5 seconds less. ( DSL
> lines at 25KB/S )

I have to point out that ssh is TCP  and you don't have problems when
you are using TCP.

> Upped the hand-window to 300 - no effect other than a time out after 300
> seconds.
>
> There does seem to be some sort of issue with firewalls going on.
>
> Client 1 is on the local lan to the server - this client is still
> connecting OK.
>
> Client 2 has a Zyxel router using NAT with a fixed IP. Can't connect.
>
> Client 3 has same router, no NAT, IP cop firewall fixed IP can't connect.
>
> Server has IP cop firewall, zyxel router, NAT, fixed IP, Internal red
> network.

Are you stating that there ia an IPCop firewall, and a Zylex router
between the server and Internet?  Is NAT done once or twice if this is
so?

> Client 2 sees UDP packets on the outgoing port coming from the server IP
> all over the firewall log.
>
> All had been working fine today, but this evening I plugged in another
> machine behind client2 firewall and initiated a connection to the server
> - at that point the rot set in. Having stopped both clients (behind
> client 2) and the server, waited 10 minutes are restarted it all,
> client1 is still connecting, 2 (only 1 off this time ) and 3 are not.
>
> Switching off client 3 and restarting server/client 2 seems to have
> restored the status quo. It now works again. Either that or having
> breifly switched over to TCP somehow forces something to come to it's
> senses somewhere.
>

I know it's a pain because the packets are encrypted (and UDP), but
have you tried a packet capture at each end to look for dropped
packets?

--
Leonard Isham, CISSP
Ostendo non ostento.

____________________________________________
Openvpn-users mailing list
Openvpn-users@xxxxxxxxxxxxxxxxxxxxx
https://lists.sourceforge.net/lists/listinfo/openvpn-users