Re: GRE tunnel - workaround found, possible bug in the kernel?

Linux Advanced Routing and Traffic Control

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



   The problem appears to be a bug in the kernel.  One thing I'd forgotten
to mention was that I regularly use VLAN interfaces on all of my machines.
The public exit interface on the CentOS boxes was a VLAN tagged interface
(eth0.2).  But when I did later tests on the Fedora boxes, I just did a test
on their native interfaces, which worked.  At first I thought perhaps I was
running into an MTU problem, but then I realized I was only sending small
ICMP packets to test.  And tcpdump isn't seeing any packets exit so I
*think* Linux is choking on the double tagging.

   If I exchange the addresses, moving the public address to the untagged
interface (eth0) and the private address to the tagged interface (eth0.2 for
example), GRE works fine since the default route is using the publicly
addressed interface and is much happier when it's untagged.

  Is this a bug or are 802.1q and GRE tagging just inherently incompatible?

Greg

_______________________________________________
LARTC mailing list
LARTC@xxxxxxxxxxxxxxx
http://mailman.ds9a.nl/cgi-bin/mailman/listinfo/lartc

[Index of Archives]     [LARTC Home Page]     [Netfilter]     [Netfilter Development]     [Network Development]     [Bugtraq]     [GCC Help]     [Yosemite News]     [Linux Kernel]     [Fedora Users]
  Powered by Linux