Colt45
Tux's lil' helper
Joined: 05 Sep 2007
Posts: 122
Location: Central Washington
|
 Posted: Tue Jun 05, 2018 9:32 am Post subject: Ethernet hw csum failure |
 |
|
So I havent been able to figure this one out. I cant determine if its some hardware fault or some wierd driver issue. I never noticed this before, but right now the logs are being flooded with this, and its causing enormous jitter on the ping.
| Code: | [ 336.941376] wan0: hw csum failure
[ 336.941380] CPU: 1 PID: 0 Comm: swapper/1 Tainted: G W 4.13.15 #2
[ 336.941382] Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./To be filled by O.E.M., BIOS 080014 12/07/2007
[ 336.941384] Call Trace:
[ 336.941386] <IRQ>
[ 336.941391] dump_stack+0x4d/0x6c
[ 336.941396] netdev_rx_csum_fault+0x33/0x40
[ 336.941401] __skb_checksum_complete+0xb1/0xc0
[ 336.941406] nf_ip_checksum+0xbb/0xe0
[ 336.941410] tcp_error+0xee/0x230
[ 336.941415] ? refcount_dec_and_test+0x11/0x20
[ 336.941419] ? consume_skb+0x37/0x40
[ 336.941424] nf_conntrack_in+0xce/0x370
[ 336.941429] ? consume_skb+0x27/0x40
[ 336.941433] ? __netif_receive_skb_core+0x8d6/0xaa0
[ 336.941438] ipv4_conntrack_in+0x17/0x20
[ 336.941442] nf_hook_slow+0x25/0xa0
[ 336.941445] ip_rcv+0x31e/0x3d0
[ 336.941449] ? inet_del_offload+0x50/0x50
[ 336.941454] __netif_receive_skb_core+0x4f1/0xaa0
[ 336.941460] ? napi_gro_receive+0x78/0xa0
[ 336.941466] ? sky2_poll+0xb43/0xda0 [sky2]
[ 336.941470] __netif_receive_skb+0x1c/0x70
[ 336.941475] netif_receive_skb_internal+0x45/0x340
[ 336.941480] netif_receive_skb+0x9/0x10
[ 336.941484] ifb_ri_tasklet+0x17e/0x240
[ 336.941490] tasklet_action+0x91/0xb0
[ 336.941494] __do_softirq+0xd8/0x1e0
[ 336.941498] irq_exit+0x76/0x80
[ 336.941503] do_IRQ+0x59/0xe0
[ 336.941507] common_interrupt+0x86/0x86
[ 336.941511] RIP: 0010:amd_e400_idle+0xd/0x30
[ 336.941514] RSP: 0018:ffffc9000006bea8 EFLAGS: 00000246 ORIG_RAX: ffffffffffffff4e
[ 336.941517] RAX: 0000004000002d40 RBX: ffff88007b0db840 RCX: 0000000000000000
[ 336.941519] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
[ 336.941521] RBP: ffffc9000006beb0 R08: 0000000000000046 R09: 0000000000000000
[ 336.941523] R10: 0000000100009027 R11: 0000000000000000 R12: ffff88007b0db840
[ 336.941525] R13: ffff88007b0db840 R14: 0000000000000000 R15: 0000000000000000
[ 336.941527] </IRQ>
[ 336.941533] ? arch_cpu_idle+0xa/0x10
[ 336.941538] default_idle_call+0x1e/0x30
[ 336.941541] do_idle+0x15a/0x180
[ 336.941545] cpu_startup_entry+0x6e/0x80
[ 336.941548] start_secondary+0x13e/0x160
[ 336.941552] secondary_startup_64+0x9f/0x9f
|
Ignore the outdated kernel; it still does it on a gentoo sources from a few days ago. 4.9. I just forgot to tell grub to boot the new one and I CBF to reboot again for the 20th time.
This is a VLAN port, and its by far the busiest on that physical port, eth0. So as part of diag'ing this I moved it over to eth1 and reconfuggered the switch since I use vlan for everything. Still does it, for the same Vlan, WAN0. I changed cables, changed ports on the switch, Ive even pulled RAM. I honestly have no clue, but Ive been up too late to bother with it anymore, so Ill try again tomorrow. I think Ill stick a card in the PCI slot and try that. Not looking forward to reconfiguring everything for that. |
|
Kernel & Hardware
