| Message ID | 1467723643-19694-1-git-send-email-matthias.may@neratec.com |
|---|---|
| State | Changes Requested |
| Headers | show |
On Tue, Jul 05, 2016 at 03:00:43PM +0200, Matthias May wrote: > When a 4addr client suddenly looses its connection (no deauth/deassoc) > the AP still thinks it is connected. > If the client reconnects before the AP timeoutes the client, traffic > cannot flow. > > Fix this by making sure the WLAN_STA_WDS flag is unset in the sta->flags > when the client completes association. > diff --git a/src/ap/ieee802_11.c b/src/ap/ieee802_11.c > @@ -2829,6 +2829,7 @@ static void handle_assoc_cb(struct hostapd_data *hapd, > new_assoc = 0; > sta->flags |= WLAN_STA_ASSOC; > sta->flags &= ~WLAN_STA_WNM_SLEEP_MODE; > + sta->flags &= ~WLAN_STA_WDS; This looks a bit strange taken into account this same handle_assoc_cb() function is using the WLAN_STA_WDS flag just below this: if (sta->flags & WLAN_STA_WDS) { int ret; char ifname_wds[IFNAMSIZ + 1]; ret = hostapd_set_wds_sta(hapd, ifname_wds, sta->addr, sta->aid, 1); if (!ret) hostapd_set_wds_encryption(hapd, sta, ifname_wds); } All that would become dead code if this patch were applied. Is this code really supposed to be removed? If so, these lines should be deleted as well. If not, this issue would likely need to be fixed in some other manner.
On 07/23/2016 08:01 PM, Jouni Malinen wrote: > On Tue, Jul 05, 2016 at 03:00:43PM +0200, Matthias May wrote: >> When a 4addr client suddenly looses its connection (no deauth/deassoc) >> the AP still thinks it is connected. >> If the client reconnects before the AP timeoutes the client, traffic >> cannot flow. >> >> Fix this by making sure the WLAN_STA_WDS flag is unset in the sta->flags >> when the client completes association. > >> diff --git a/src/ap/ieee802_11.c b/src/ap/ieee802_11.c >> @@ -2829,6 +2829,7 @@ static void handle_assoc_cb(struct hostapd_data *hapd, >> new_assoc = 0; >> sta->flags |= WLAN_STA_ASSOC; >> sta->flags &= ~WLAN_STA_WNM_SLEEP_MODE; >> + sta->flags &= ~WLAN_STA_WDS; > > This looks a bit strange taken into account this same handle_assoc_cb() > function is using the WLAN_STA_WDS flag just below this: > > if (sta->flags & WLAN_STA_WDS) { > int ret; > char ifname_wds[IFNAMSIZ + 1]; > > ret = hostapd_set_wds_sta(hapd, ifname_wds, sta->addr, > sta->aid, 1); > if (!ret) > hostapd_set_wds_encryption(hapd, sta, ifname_wds); > } > > > All that would become dead code if this patch were applied. Is this code > really supposed to be removed? If so, these lines should be deleted as > well. If not, this issue would likely need to be fixed in some other > manner. > > Yes this looks a bit strange. ieee802_11_rx_from_unknown is the function which sets WLAN_STA_WDS bit once the WLAN_STA_ASSOC is set. Wouldn't this mean that this is already dead code now? BR Matthias
diff --git a/src/ap/ieee802_11.c b/src/ap/ieee802_11.c index 11f12f9..3262ae1 100644 --- a/src/ap/ieee802_11.c +++ b/src/ap/ieee802_11.c @@ -2829,6 +2829,7 @@ static void handle_assoc_cb(struct hostapd_data *hapd, new_assoc = 0; sta->flags |= WLAN_STA_ASSOC; sta->flags &= ~WLAN_STA_WNM_SLEEP_MODE; + sta->flags &= ~WLAN_STA_WDS; if ((!hapd->conf->ieee802_1x && !hapd->conf->wpa && !hapd->conf->osen) || sta->auth_alg == WLAN_AUTH_FT) { /*
When a 4addr client suddenly looses its connection (no deauth/deassoc) the AP still thinks it is connected. If the client reconnects before the AP timeoutes the client, traffic cannot flow. Fix this by making sure the WLAN_STA_WDS flag is unset in the sta->flags when the client completes association. Signed-off-by: Matthias May <matthias.may@neratec.com> --- src/ap/ieee802_11.c | 1 + 1 file changed, 1 insertion(+)