bond - 2008-01-17

Has anyone encountered such a problem:

-----begin--------
Jan 17 08:08:38 localhost kernel: sky2 eth1: Link is down.
Jan 17 08:08:38 localhost kernel: bonding: bond0: link status definitely down for interface eth1, disabling it
Jan 17 08:08:38 localhost kernel: bonding: bond0: making interface eth2 the new active one.
Jan 17 08:08:38 localhost kernel: device eth2 entered promiscuous mode
Jan 17 08:08:38 localhost kernel: audit(1200553718.413:2): dev=eth2 prom=256 old_prom=0 auid=4294967295
Jan 17 08:08:38 localhost kernel: BUG: sleeping function called from invalid context at kernel/mutex.c:207
Jan 17 08:08:38 localhost kernel: in_atomic():1, irqs_disabled():0
Jan 17 08:08:38 localhost kernel: INFO: lockdep is turned off.
Jan 17 08:08:38 localhost kernel: Pid: 3349, comm: bond0 Not tainted 2.6.24-rc8 #1
Jan 17 08:08:38 localhost kernel:  [<c0444d26>] debug_show_held_locks+0x16/0x1e
Jan 17 08:08:38 localhost kernel:  [<c0422e52>] __might_sleep+0xc7/0xcc
Jan 17 08:08:38 localhost kernel:  [<c0733116>] mutex_lock_nested+0x1c/0x2b9
Jan 17 08:08:38 localhost kernel:  [<c0443fd0>] __lock_release+0x1a/0x4e
Jan 17 08:08:38 localhost kernel:  [<c06c95d0>] rtnl_unlock+0xa/0xf
Jan 17 08:08:38 localhost kernel:  [<c0443fd0>] __lock_release+0x1a/0x4e
Jan 17 08:08:38 localhost kernel:  [<c06c95d0>] rtnl_unlock+0xa/0xf
Jan 17 08:08:38 localhost kernel:  [<c06c3bd0>] netdev_run_todo+0x11/0x174
Jan 17 08:08:38 localhost kernel:  [<fca1745a>] bond_mii_monitor+0x5b/0x85 [bonding]
Jan 17 08:08:38 localhost kernel:  [<fca173ff>] bond_mii_monitor+0x0/0x85 [bonding]
Jan 17 08:08:38 localhost kernel:  [<c04343a4>] run_workqueue+0xe3/0x1b4
Jan 17 08:08:38 localhost kernel:  [<c043434f>] run_workqueue+0x8e/0x1b4
Jan 17 08:08:38 localhost kernel:  [<c0732a84>] __sched_text_start+0x41c/0x436
Jan 17 08:08:38 localhost kernel:  [<c0434475>] worker_thread+0x0/0x9b
Jan 17 08:08:38 localhost kernel:  [<c0434506>] worker_thread+0x91/0x9b
Jan 17 08:08:38 localhost kernel:  [<c0437ad0>] autoremove_wake_function+0x0/0x33
Jan 17 08:08:38 localhost kernel:  [<c0732a84>] __sched_text_start+0x41c/0x436
Jan 17 08:08:38 localhost kernel:  [<c0437ad0>] autoremove_wake_function+0x0/0x33
Jan 17 08:08:38 localhost kernel:  [<c0434475>] worker_thread+0x0/0x9b
Jan 17 08:08:38 localhost kernel:  [<c04376c7>] kthread+0x34/0x55
Jan 17 08:08:38 localhost kernel:  [<c0437693>] kthread+0x0/0x55
Jan 17 08:08:38 localhost kernel:  [<c0405c63>] kernel_thread_helper+0x7/0x10
Jan 17 08:08:38 localhost kernel:  =======================
-------end------

I configured the bonding driver on the 2.6.24-rc8 kernel with 2 network interfaces. After configuring the IP layer of the bond0 interface I cannot reach the other side of the bond,  i.e. I cannot reach it with any kind of protocol - even ping doesn't work.
After I pull the RJ45 plug out of one of the network cards - the kernel dumps the message above and then the transmission over the IP protocol is available.
Is there any problem with the locking subsystem in bonding or maybe I am missing something?