[quagga-dev 11336] Re: BGP debug CLI

Daniel Walton dwalton at cumulusnetworks.com
Fri Jun 20 13:40:48 BST 2014

I do like the time machine idea :)

We have some scale tests that I will run to sanity check the performance impact of the slightly more complicated "is debug X for peer/prefix Y enabled" checks.  If the debugs are off I don't think there will be a noticeable difference.  If they are on it should be minimal unless the user enables debugs for 100s of prefixes...but that seems unlikely.


----- Original Message -----
From: "Alexis Rosen" <quagga-users at alexis.users.panix.com>
To: "Daniel Walton" <dwalton at cumulusnetworks.com>
Cc: quagga-dev at lists.quagga.net, quagga-users at lists.quagga.net
Sent: Friday, June 20, 2014 1:05:41 AM
Subject: Re: [quagga-dev 11334] BGP debug CLI

On Jun 19, 2014, at 3:53 PM, Daniel Walton <dwalton at cumulusnetworks.com> wrote:
> I am working on a patch to enhance the debug cli for bgp so the user can:
> - enable keepalive debugs for a specific peer
> - enable fsm debugs for a specific peer
> - enable inbound and/or outbound updates debugs for a specific peer
> - enable update debugs for a specific prefix
> - enable zebra debugs for a specific prefix

That's really great!

> Thoughts/feedback?

Yes, once you're done, please build a time machine and submit the patches in 2009.

Also, any info you may have on resource consumption would be very helpful. That is, it would be good to have confidence that doing some debugging won't blow up a production system.


More information about the Quagga-dev mailing list