Date: Wed, 27 Jul 2005 13:01:18 -0700 From: Brooks Davis <brooks@one-eyed-alien.net> To: Jung-uk Kim <jkim@niksun.com> Cc: Wilko Bulte <wb@freebie.xs4all.nl>, freebsd-current@FreeBSD.org, Mike Jakubik <mikej@rogers.com> Subject: Re: dhclient taking all cpu Message-ID: <20050727200118.GE20112@odin.ac.hmc.edu> In-Reply-To: <200507271555.00945.jkim@niksun.com> References: <42E58007.9030202@rogers.com> <20050726233933.GA13679@odin.ac.hmc.edu> <20050727191043.GA17885@odin.ac.hmc.edu> <200507271555.00945.jkim@niksun.com>
next in thread | previous in thread | raw e-mail | index | archive | help
--X3gaHHMYHkYqP6yf Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Jul 27, 2005 at 03:54:58PM -0400, Jung-uk Kim wrote: > On Wednesday 27 July 2005 03:10 pm, Brooks Davis wrote: > > I think I've found it. There was a really odd typo (=3D instead of > > +) in the code that handles undersized captures on the bpf socket.=20 > > Please try the following patch and see if it solves the problem.=20 > > I'm testing here, but I don't have a reliable way to trigger the > > bug. The fix is fairly obvious so I'll commit it to head shortly. >=20 > Good catch! It seems to fix my 'infinite loop' problem. >=20 > One more problem to solve... I have to do the following to make=20 > bge(4) working correctly at boot time: >=20 > ifconfig_bge0=3D"up DHCP" >=20 > Without `up', dhclient fails like this: >=20 > bge0: link state changed to DOWN > bge0: no link .............. giving up > bge0: flags=3D8802<BROADCAST,SIMPLEX,MULTICAST> mtu 1500 > options=3D1a<TXCSUM,VLAN_MTU,VLAN_HWTAGGING> > ether xx:xx:xx:xx:xx:xx > media: Ethernet autoselect (none) > status: no carrier >=20 > because bge(4) does not change link state while the interface is down. = =20 > This problem is more serious with wireless driver because link=20 > doesn't go up until it is associated with some AP. >=20 > I was always wondering which is correct. Do we have to update link=20 > state while interface is down or not? The interface shouldn't really be doing anything if it isn't up. I'm tempted to change the code to always "ifconfig <ifn> up" interfaces configured for DHCP. Longer term (7.x) I may change things to that's all etc/rc.d/netif does for DHCP interfaces. There are a number of issues to work out before that's feasiable though so I'm not doing anything other then pondering at this point. -- 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 --X3gaHHMYHkYqP6yf Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) iD8DBQFC5+gNXY6L6fI4GtQRAr9hAJ4zfmzJQ9wojg0xgBWgrfGbdXcdtwCgjavD /jTXAqlZm6XmbkhQhXSvX5k= =XsCL -----END PGP SIGNATURE----- --X3gaHHMYHkYqP6yf--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050727200118.GE20112>