bnx2 fault caused network link to go down/up

HP Proliant BL460c G6 .
Ethernet controller: Broadcom Corporation NetXtreme II BCM57711E 10-Gigabit PCIe
Running Red Hat Enterprise Linux Server release 6.3 (Santiago) 
kernel: 2.6.32-279.14.1.el6.x86_64
modinfo bnx2:
filename:       /lib/modules/2.6.32-279.14.1.el6.x86_64/kernel/drivers/net/bnx2.ko
firmware:       bnx2/bnx2-rv2p-09ax-6.0.17.fw
firmware:       bnx2/bnx2-rv2p-09-6.0.17.fw
firmware:       bnx2/bnx2-mips-09-6.2.1b.fw
firmware:       bnx2/bnx2-rv2p-06-6.0.15.fw
firmware:       bnx2/bnx2-mips-06-6.2.3.fw
version:        2.2.1
pcie_aspm=off is currently being passed to the kernel at boot time.

network configuration:Bonding with VLAN Tagging

Inizial crash log
kernel: do_IRQ: 8.228 No irq handler for vector (irq -1)
 Nov 9 17:58:37 kernel: ------------[ cut here ]------------
 Nov 9 17:58:37 kernel: WARNING: at net/sched/sch_generic.c:261 dev_watchdog+0x26d/0x280() (Tainted: G W --------------- )
 Nov 9 17:58:37 kernel: Hardware name: ProLiant BL460c G6

Server is part of a cluster. The NIC went down/up, just long enough for it to think that another server in the cluster was down, so it proceeded to fence it causing an outage.
the problem occurs more frequently!!!
The NIC went down/up, just long enough for it to think that another server in the cluster was down, so it proceeded to fence it causing an outage.

What can I do to fix the problem?