Subject: kern/20647: retry with old pppoe address first
To: None <gnats-bugs@gnats.netbsd.org>
From: None <perry@piermont.com>
List: netbsd-bugs
Date: 03/10/2003 09:16:50
>Number:         20647
>Category:       kern
>Synopsis:       retry with old pppoe address first
>Confidential:   no
>Severity:       non-critical
>Priority:       medium
>Responsible:    kern-bug-people
>State:          open
>Class:          change-request
>Submitter-Id:   net
>Arrival-Date:   Mon Mar 10 06:17:00 PST 2003
>Closed-Date:
>Last-Modified:
>Originator:     Perry E. Metzger
>Release:        NetBSD 1.6P
>Organization:
None
>Environment:
	
	
System: NetBSD alchemist 1.6P NetBSD 1.6P (ALCHEMIST) #0: Wed Mar 5 15:00:02 EST 2003 perry@alchemist:/usr/src/sys/arch/i386/compile/ALCHEMIST i386
Architecture: i386
Machine: i386
>Description:
	Often, a pppoe connection drops momentarily. Often, if a retry was
attempted with the same concentrator/address, it would succeed, leading
to no perceived loss of service on existing connections. However,
our pppoe does not attempt such a retry, leading to apparent disruption.
Having a mechanism that retried with the old address would be nice.
(It would also be nice to be able to clear the association if it turns
out that for whatever reason you really don't want the old address. For
example, you might suddenly get one someone was using for KaaZaa and
desperately want another address that isn't being bombarded with requests.)
>How-To-Repeat:
	
>Fix:
	none -- martin just asked me to send a PR to remind him.
>Release-Note:
>Audit-Trail:
>Unformatted: