Re: [nsp] CEF stability

From: Dave Bergum (bergum@MR.Net)
Date: Mon Jan 24 2000 - 15:19:59 EST


>>>>> On Mon, 24 Jan 2000 20:10:21 +0000, "Neil J. McRae" <neil@COLT.NET> said:

  Neil> On Mon, 24 Jan 2000 12:26:41 -0600 (CST)
  Neil> Dave Bergum <bergum@MR.Net> wrote:

>> Talk a bit about changing the mtu on a live box? I tried to up the mtu
>> from 1500 to 4470 on a live interface, and it did not seem to take effect.
>> (was having ospf db distribution issue with Juniper on the other end that
>> had default 4470 mtu, and ended up setting the Juniper down to 1500
>> instead.)

  Neil> I meant with a VIP2-50. If you change the MTU the cards do
  Neil> a brief reset. We've had no problems changing them if done
  Neil> at the pre-production stage

Ok, I had a case where changing the mtu from 1500 to 4470 was accepted, all
the show statements indicated that the interface was configured for 4470,
but it would not receive datagrams in excess of 1500...so I gave up.
Perhaps the reset is required to reallocate buffers and stuff.

Dave.



This archive was generated by hypermail 2b29 : Sun Aug 04 2002 - 04:12:09 EDT