On Mon, Jul 30, 2012 at 11:47:03AM +0200, Marek Lindner wrote:
On Monday, July 30, 2012 11:23:43 Antonio Quartulli wrote:
On Mon, Jul 30, 2012 at 09:59:16AM +0200, Marek Lindner wrote:
On Monday, July 30, 2012 01:15:30 Antonio Quartulli wrote:
in case of client roaming a new global entry is added while a corresponding local one is still present. In this case the node can safely pass the WIFI flag from the local to the global entry
And why would we want that ?
To let the AP-isolation work. If a client is known to be WIFI and we do not "pass" this flag on roaming, we end up in a state where this client is not marked as WIFI anymore (even if it should be).
During this time period the client will be able to talk to the other WIFI clients even if the AP isolation is enabled
And why is this explanation not part of the commit message ?
Because it is left as exercise for the reader ;-) Actually I didn't mention it because the Ap-Isolation part is a "consequence". What this patch wants to fix is an issue in the client entry state consistency, regardless of the semantic of such state.
Anyhow, if you think it is worth adding it, I will do that.