Re: A router bug?

David (david@killerlabs.com)
Sun, 31 Jan 1999 06:03:04 -0800 (PST)


On Sun, 31 Jan 1999, Alan Cox wrote:
> Thats what I would expect. That means the MTU size is the answer

changing the MTU sizes wasn't fixing it. i had them all set to 576.

> > in the above situation, one can "fix" NT sites by forcing my ppp0 to an MTU
> > of 576 and they'll work, but this isn't the answer to both situations.
>
> See ?

yes, i saw. on the -router- that fixed it. setting the MTU on the eth0 of
the LAN machines did -not- fix it. as noted in the documentation, setting
the MTU only changes for new connections, not established. i had set my
eth0 init scripts to use the MTU of 576.

again. this did -not- fix it.

> I've yet to see a non masquerade trace set where this is the case.

this isn't masq, it's GRE tunneling. no masq options are enabled in the
kernel's config.

the solution as provided and explained by AK was setting the mss on the
routes for the machines on the LAN. this -did- fix it.

(for hj) route add default gw <gw> mss 576

the reason why mtu didn't work and mss did is for idle speculation i
would expect.

-d

-- 
  Look, Windows 98  Buy, lemmings, buy!  MCSE, Must Consult Someone Experienced
__ (c) 1998 David Ford.  Redistribution via the Microsoft Network is prohibited
\/  for linux-kernel: please read linux/Documentation/* before posting problems

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/