Hi Mohamed, On Thu, Nov 18, 2010 at 01:12:57PM -0800, Mohamed Abbas wrote: > This patch to solve bug #3604. We should not try to reconnect > to network that disappeared, this will cause having network > marked associating for ever. This scenario happen if the user > connected to AP then user walked away from AP reach this will > cause to disconnect then the autoconnect will try to connect > to the same AP which is out of reach. This patch will prevent > this and autoconnect will reconnect to this AP if user return > to the AP reach. Have you investigated why we are getting an unvailable network at that point ? After comming back from scanning, the unavailable networks should be removed. Is this a race between trying to join the AP while scanning is not done yet ?
Cheers, Samuel. > --- > src/service.c | 4 ++++ > 1 files changed, 4 insertions(+), 0 deletions(-) > > diff --git a/src/service.c b/src/service.c > index b1842d6..573207f 100644 > --- a/src/service.c > +++ b/src/service.c > @@ -2115,6 +2115,10 @@ static connman_bool_t is_ignore(struct connman_service > *service) > if (service->state == CONNMAN_SERVICE_STATE_FAILURE) > return TRUE; > > + if (service->network != NULL && > + connman_network_get_available(service->network) == FALSE) > + return TRUE; > + > return FALSE; > } > > -- > 1.7.2.3 > > _______________________________________________ > connman mailing list > [email protected] > http://lists.connman.net/listinfo/connman -- Intel Open Source Technology Centre http://oss.intel.com/ _______________________________________________ connman mailing list [email protected] http://lists.connman.net/listinfo/connman
