[quagga-dev 3772] Re: stub-router patch

Paul Jakma paul at clubi.ie
Wed Oct 26 14:13:04 BST 2005


On Mon, 24 Oct 2005, Paul Jakma wrote:

> So any opinions? :) Commit it? Let it rot? :)

..........*  <-- tumbleweed.

So shall i commit? The patch is fairly uncontraversial and shouldn't 
touch existing functionality if not enabled. Except:

1. In order to pull-off stub-router on-shutdown <seconds>, the 
shutdown path is pulled into pieces:

 	- shutdown of an ospf instance
 		- has to be via ospf_finish(), rather than
 	          ospf_terminate (which just deleted ospf instances).
 		- we don't actually support multiple instances yet,
 	          but be nice to be prepared for it
 		- ospf_finish has not had much use to date.
 	- ospf_finish does a double-take where it will end-up calling
 	  itself again, either delayed via the graceful-shutdown
 	  timer thread or directly from the graceful-shutdown thread.
 	  state is introduced so ospf_finish() can tell whether it is
 	  being called pre or post graceful shutdown.
 	- ospf_terminate now simply calls ospf_finish for each ospf
 	  instance, but does not exit. Instead, it just sets a
           flag to indicate shutdown and the last graceful shutdown
 	  function/thread will handle the exit.

2. It does not write out the command if stub router is 
administratively enabled. This might be a feature (you can't 'write 
file' and accidently leave your ospfd configured indefinitely as 
stub-router for next restart), but it probably should write out as 
otherwise there's no way to tell from 'show run' which command one 
must negate to clear it.

Well? Anyone? :)

regards,
-- 
Paul Jakma	paul at clubi.ie	paul at jakma.org	Key ID: 64A2FF6A
Fortune:
Consider well the proportions of things.  It is better to be a young June-bug
than an old bird of paradise.
 		-- Mark Twain, "Pudd'nhead Wilson's Calendar"



More information about the Quagga-dev mailing list