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

Re: [Openvpn-users] mssfix default (was: Incorrect tcp and udp checksums on packets from gateway to vpn clients (solved))


  • Subject: Re: [Openvpn-users] mssfix default (was: Incorrect tcp and udp checksums on packets from gateway to vpn clients (solved))
  • From: "Giampaolo Carmagnani" <giampaolo@xxxxxxxxx>
  • Date: Tue, 3 Apr 2007 23:34:06 -0300

Hi Peter

On 4/3/07, Peter Rabbitson <rabbit@xxxxxxxxx> wrote:
> Giampaolo Carmagnani wrote:
> > (snip)
> > Note that while mssfix only needs to be specified on one side of the
> > connection, fragment should be specified on both."
>
> Hi Giampolo,
>
> While my question does not relate directly to the original post, this
> statement of yours above contradicts with what I observed for mssfix. I
> even asked on this list a while ago but received no answer
> http://thread.gmane.org/gmane.network.openvpn.user/18652. Could you
> elaborate?
>
> Thanks
>
> Peter
>
The statement above is not mine but from the FAQ.
But I read your previous post and make a little research, I always
want to kow the openvpn's overhead.  Then I found this page:
http://mia.ece.uic.edu/~papers/volans/openvpn1.html

Quotting it:
"... Each layer of protocol adds some bytes of overhead. This is
illustrated in the above diagram. It is quite complicated to
pint-point the amount of overhead introduced by the openvpn later, as
it depends largely on thhe cipher + HMAC in use, and many other
factors....."

There are tables too, which have the experiment's results to find the overhead.
After reading  it, then we can conclude that the packet's data size
change based of a lot of variants.  And the fragment, mssfix and mtu
change in each implementation.

Best Regards

Giampaolo Carmagnani
______________________
OpenVPN mailing lists
https://lists.sourceforge.net/lists/listinfo/openvpn-users