lagg trouble

Leif Walsh leif.walsh at gmail.com
Fri Sep 24 13:54:59 UTC 2010


I have the following in /boot/loader.conf:

if_iwn_load="YES"
iwn5000fw_load="YES"
wlan_wep_load="YES"
wlan_ccmp_load="YES"
wlan_tkip_load="YES"

and the following in /etc/rc.conf:

ifconfig_em0="up"
ifconfig_iwn0="ether 00:1f:16:0c:46:82"
wlans_iwn0="wlan0"
ifconfig_wlan0="WPA"
cloned_interfaces="lagg0"
ifconfig_lagg0="laggproto failover laggport em0 laggport wlan0 DHCP"

On boot, lagg0 gets created and correctly includes em0 and wlan0 as
laggports. Dhcp starts on lagg0, but if I ping a hostname, I get a timeout,
and if I ping an ip address, I get send failed, something like no room in
buffer (not on that comp now).

If I do a netif restart, it brings back eth0 and lagg0, but lagg0 doesn't
include wlan0, which seems to get started later (and not by netif? The
output says it stops lo, em, iwn, lagg, tun, ipfw, and wlan, but only starts
lo, em, iwn, and lagg, and directly after, ifconfig shows everything up,
just lagg0 not using wlan0). If I do a dhclient wlan0 (assuming
wpa_supplicant finds something it likes), I get a working network
connection, but not if I manually patch it through lagg0 through lagg0.

Is there a way to ensure that wlan gets up before lagg? Why would I get that
buffer error through lagg?

Also, if I have to switch wireless routers and use wpa_cli, can I get
dhclient  to notice and automatically request a new ip?

-- 
Cheers,
Leif


More information about the freebsd-questions mailing list