Softlockup and freeze with bonding mode 6 and VLANs
Issue
-
Softlockups and freezes or even crashes occurs when using bonding mode 6 on a system that has VLANs.
BUG: soft lockup - CPU#0 stuck for 10s! [swapper:0] Pid: 0, comm: swapper EIP: 0060:[<c061dc4e>] CPU: 0 EIP is at _spin_lock_bh+0xf/0x18 EFLAGS: 00000286 Not tainted (2.6.18-194.3.1.el5 #1) EAX: c0748000 EBX: f69fe550 ECX: f6f9a52c EDX: f69fe000 ESI: f69fe4fc EDI: f69fe550 EBP: f69418c0 DS: 007b ES: 007b CR0: 8005003b CR2: 0069c270 CR3: 00742000 CR4: 000006d0 [<f8b240ea>] rlb_arp_recv+0x98/0x11d [bonding] [<c05c0aa8>] netif_receive_skb+0x3ac/0x401 [<f8a33bf8>] bnx2_poll_work+0xc3b/0xd45 [bnx2] [<c041000c>] mtrr_bp_init+0x1f7/0x21a [<c041db0c>] kmap_atomic_to_page+0x34/0x54 [<c041f79d>] try_to_wake_up+0x3e8/0x3f2 [<c043887c>] hrtimer_run_queues+0xef/0x176 [<c042d5f5>] lock_timer_base+0x15/0x2f [<f8a3406f>] bnx2_poll+0xbd/0x1ce [bnx2] [<c05c2995>] net_rx_action+0x9c/0x1a7 [<c042a377>] __do_softirq+0x87/0x114 [<c04073cf>] do_softirq+0x52/0x9c [<c044f158>] __do_IRQ+0x0/0xd6 [<c04074ce>] do_IRQ+0xb5/0xc3 [<c0405946>] common_interrupt+0x1a/0x20 [<c0403ce7>] mwait_idle+0x25/0x38 [<c0403ca8>] cpu_idle+0x9f/0xb9 [<c07099fa>] start_kernel+0x37b/0x383 =======================
-
System has eth0 configured as bond0 w/ (3) vlan tagging sub interfaces ==> live traffic and continuous ping tests show recurring packet loss.
Environment
-
Red Hat Enterprise Linux 5.5 or older
-
bonding with mode 6/ALB
- using VLANs
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.