After running on kernel 4.9 for quite some time, we suddenly experienced hangs. Tried compiling 4.11.0, but the problem remains. The only think that changed was a couple more rules in nft, traffic is the same. 4.8.6 kernel works fine (but has other problems). Attached is the debugging output, hope it can help you track down the issue: May 3 12:46:59 fw01 kernel: [ 271.875363] ------------[ cut here ]------------ May 3 12:46:59 fw01 kernel: [ 271.875369] WARNING: CPU: 0 PID: 3 at net/sched/sch_generic.c:316 dev_watchdog+0x212/0x220 May 3 12:46:59 fw01 kernel: [ 271.875370] NETDEV WATCHDOG: eth1 (igb): transmit queue 1 timed out May 3 12:46:59 fw01 kernel: [ 271.875370] Modules linked in: May 3 12:46:59 fw01 kernel: [ 271.875373] CPU: 0 PID: 3 Comm: kworker/0:0 Not tainted 4.11.0 #1 May 3 12:46:59 fw01 kernel: [ 271.875374] Hardware name: Dell Inc. PowerEdge R210/05KX61, BIOS 1.3.4 05/24/2010 May 3 12:46:59 fw01 kernel: [ 271.875377] Workqueue: events_long gc_worker May 3 12:46:59 fw01 kernel: [ 271.875378] Call Trace: May 3 12:46:59 fw01 kernel: [ 271.875380] <IRQ> May 3 12:46:59 fw01 kernel: [ 271.875383] dump_stack+0x4d/0x6e May 3 12:46:59 fw01 kernel: [ 271.875386] __warn+0xcc/0xf0 May 3 12:46:59 fw01 kernel: [ 271.875387] warn_slowpath_fmt+0x4a/0x50 May 3 12:46:59 fw01 kernel: [ 271.875390] ? tick_program_event+0x3f/0x70 May 3 12:46:59 fw01 kernel: [ 271.875391] dev_watchdog+0x212/0x220 May 3 12:46:59 fw01 kernel: [ 271.875392] ? dev_deactivate_queue.constprop.31+0x60/0x60 May 3 12:46:59 fw01 kernel: [ 271.875395] call_timer_fn+0x30/0x140 May 3 12:46:59 fw01 kernel: [ 271.875397] run_timer_softirq+0x1be/0x3f0 May 3 12:46:59 fw01 kernel: [ 271.875399] ? handle_irq_event_percpu+0x40/0x50 May 3 12:46:59 fw01 kernel: [ 271.875401] __do_softirq+0xbe/0x280 May 3 12:46:59 fw01 kernel: [ 271.875404] do_softirq_own_stack+0x1c/0x30 May 3 12:46:59 fw01 kernel: [ 271.875404] </IRQ> May 3 12:46:59 fw01 kernel: [ 271.875406] do_softirq+0x42/0x50 May 3 12:46:59 fw01 kernel: [ 271.875407] __local_bh_enable_ip+0x75/0x80 May 3 12:46:59 fw01 kernel: [ 271.875410] _raw_spin_unlock_bh+0x15/0x20 May 3 12:46:59 fw01 kernel: [ 271.875412] nf_nat_cleanup_conntrack+0x1bd/0x230 May 3 12:46:59 fw01 kernel: [ 271.875413] ? nf_nat_l3proto_register+0x70/0x70 May 3 12:46:59 fw01 kernel: [ 271.875415] __nf_ct_ext_destroy+0x38/0x50 May 3 12:46:59 fw01 kernel: [ 271.875417] nf_conntrack_free+0x17/0x50 May 3 12:46:59 fw01 kernel: [ 271.875418] destroy_conntrack+0x74/0x90 May 3 12:46:59 fw01 kernel: [ 271.875421] nf_conntrack_destroy+0x12/0x20 May 3 12:46:59 fw01 kernel: [ 271.875422] nf_ct_gc_expired+0x45/0x90 May 3 12:46:59 fw01 kernel: [ 271.875423] gc_worker+0xb3/0x180 May 3 12:46:59 fw01 kernel: [ 271.875425] process_one_work+0x143/0x3e0 May 3 12:46:59 fw01 kernel: [ 271.875426] worker_thread+0x126/0x480 May 3 12:46:59 fw01 kernel: [ 271.875429] kthread+0x104/0x140 May 3 12:46:59 fw01 kernel: [ 271.875430] ? process_one_work+0x3e0/0x3e0 May 3 12:46:59 fw01 kernel: [ 271.875431] ? kthread_park+0x90/0x90 May 3 12:46:59 fw01 kernel: [ 271.875432] ret_from_fork+0x29/0x40 May 3 12:46:59 fw01 kernel: [ 271.875434] ---[ end trace 3da51f3ef83370a7 ]--- May 3 12:47:03 fw01 kernel: [ 275.629072] igb 0000:01:00.0 eth0: igb: eth0 NIC Link is Down May 3 12:47:03 fw01 kernel: [ 275.629135] igb 0000:01:00.1 eth1: Reset adapter May 3 12:47:03 fw01 kernel: [ 275.629316] igb 0000:01:00.0 eth0: Reset adapter May 3 12:47:12 fw01 kernel: [ 285.159798] igb 0000:01:00.1 eth1: igb: eth1 NIC Link is Up 1000 Mbps Half Duplex, Flow Control: RX/TX May 3 12:47:12 fw01 kernel: [ 285.159801] igb 0000:01:00.1: EEE Disabled: unsupported at half duplex. Re-enable using ethtool when at full duplex. May 3 12:47:15 fw01 kernel: [ 287.842412] igb 0000:01:00.1: exceed max 2 second May 3 12:48:21 fw01 kernel: [ 353.875946] igb 0000:01:00.0 eth0: Reset adapter -- Bj(/)rnar -- To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
