Date: Tue, 5 Jul 2005 15:42:09 -0700 From: Brooks Davis <brooks@one-eyed-alien.net> To: Sam Leffler <sam@errno.com> Cc: Bakul Shah <bakul@BitBlocks.com>, freebsd-current@freebsd.org Subject: Re: minor WPA problem on a Thinkpad R40 Message-ID: <20050705224209.GC28034@odin.ac.hmc.edu> In-Reply-To: <42CAF849.3020508@errno.com> References: <200507012348.j61NmQEs070640@gate.bitblocks.com> <20050705195945.GA28034@odin.ac.hmc.edu> <42CAF849.3020508@errno.com>
next in thread | previous in thread | raw e-mail | index | archive | help
--bAmEntskrkuBymla Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue, Jul 05, 2005 at 02:14:49PM -0700, Sam Leffler wrote: > Brooks Davis wrote: > >On Fri, Jul 01, 2005 at 04:48:26PM -0700, Bakul Shah wrote: > > > >>After suspend/resume the old wpa_supplicant does not die > >>and ath0 does not come up. Manually running > >> /etc/rc.d/netif stop ath0 > >>or > >> /etc/rc.d/netif start ath0 > >>does the right thing. I am running yesterday's kernel. > >>rc.conf contains > >> > >> ifconfig_ath0=3D"WPA DHCP" > >> > >>Not sure if I am doing something wrong or if this area still > >>needs work. As I understand it, wpa_supplicant should die > >>when an interface goes down just like dhclient should. > > > > > >Hmm, this isn't something I've been able to test. One shot in the dark > >would be adding -w to the wpa_supplicant command line in > >/etc/rc.d/wpa_supplicant. I know wpa_supplicant deals badly with > >yanking and reinserting a nic under normal circumstances so I'm not too > >suprised it's having trouble with suspend/resume. > > > > > >>Another curious behavior is that after a manual > >> ifconfig ath0 down > >>the interface does not stay down. > > > > > >I suspect wpa_supplicant is bringing it back up. I'm not sure what the > >answer is there. >=20 > IMO wpa_supplicant should exit when the interface goes down/away but it= =20 > does not. I sent mail to Jouni about this but he hasn't responded yet.= =20 > I was going to check if there was a way to make it work this way. Not=20 > sure why it works this way except to avoid recalculating various crypto= =20 > state or perhaps to avoid linux hotplug issues. I've we're going to do that, I think we may want either a new target in /etc/rc.d/netif or a new /etc/rc.d/linkstate script. That's probably the right way to go though. -- Brooks --=20 Any statement of the form "X is the one, true Y" is FALSE. PGP fingerprint 655D 519C 26A7 82E7 2529 9BF0 5D8E 8BE9 F238 1AD4 --bAmEntskrkuBymla Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) iD8DBQFCywzAXY6L6fI4GtQRAnLJAJ98+nNQ16gZYEc7ZJ+2FEzvtRnK1QCgrFZY 56VbCCtZ2mjI6Y9x4r5YDsI= =n7Zj -----END PGP SIGNATURE----- --bAmEntskrkuBymla--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050705224209.GC28034>