Re: order 1 page allocation failures

2011-09-29 Thread Marcelo Tosatti 
On Tue, Sep 27, 2011 at 04:10:20PM -0600, Thomas Fjellstrom wrote:
 Hi,
 
 I've been having some issues with KVM recently where one or more vms will 
 cause page allocation failure messages, usually with the backtrace including 
 networking functions, example follows:
 
 [362409.429944] kvm: page allocation failure: order:1, mode:0x20
 [362409.429957] Pid: 3453, comm: kvm Not tainted 3.0.0-1-amd64 #1
 [362409.429965] Call Trace:
 [362409.429970]  IRQ  [810b9c90] ? warn_alloc_failed+0x108/0x11b
 [362409.429998]  [810bcd78] ? __alloc_pages_nodemask+0x6e6/0x75c
 [362409.430012]  [810ec0c0] ? kmem_getpages+0x55/0xf0
 [362409.430022]  [810ec87a] ? fallback_alloc+0x129/0x1c1
 [362409.430035]  [8100e28d] ? paravirt_read_tsc+0x5/0x8
 [362409.430045]  [810ed10e] ? kmem_cache_alloc+0x73/0xf0
 [362409.430057]  [812707a2] ? sk_prot_alloc+0x2b/0x128
 [362409.430067]  [81270965] ? sk_clone+0x14/0x2bd
 [362409.430077]  [812ade7d] ? inet_csk_clone+0x10/0x91
 [362409.430088]  [812c1aae] ? tcp_create_openreq_child+0x21/0x41a
 [362409.430099]  [812bf98a] ? tcp_v4_syn_recv_sock+0x33/0x208
 [362409.430110]  [812c2441] ? tcp_check_req+0x1ff/0x2dd
 [362409.430122]  [812adc06] ? inet_csk_search_req+0x35/0xa7
 [362409.430132]  [812bf4f1] ? tcp_v4_do_rcv+0x206/0x32c
 [362409.430144]  [812c15d4] ? tcp_v4_rcv+0x419/0x66c
 [362409.430154]  [8100e74a] ? native_sched_clock+0x28/0x30
 [362409.430173]  [812a5a0c] ? ip_local_deliver_finish+0x14b/0x1bb
 [362409.430186]  [8127cc8f] ? __netif_receive_skb+0x3d7/0x40b
 [362409.430197]  [8127d74b] ? netif_receive_skb+0x52/0x58
 [362409.430220]  [a04b5af6] ? br_nf_pre_routing_finish+0x1d4/0x1e1 
 [bridge]
 [362409.430241]  [a04b5111] ? NF_HOOK_THRESH+0x3b/0x55 [bridge]
 [362409.430260]  [a04b60ed] ? br_nf_pre_routing+0x3be/0x3cb [bridge]
 [362409.430272]  [8129fb69] ? nf_iterate+0x41/0x77
 [362409.430288]  [a04b13a3] ? NF_HOOK.clone.4+0x56/0x56 [bridge]
 [362409.430305]  [a04b13a3] ? NF_HOOK.clone.4+0x56/0x56 [bridge]
 [362409.430315]  [8129fc12] ? nf_hook_slow+0x73/0x111
 [362409.430330]  [a04b13a3] ? NF_HOOK.clone.4+0x56/0x56 [bridge]
 [362409.430342]  [8103f0a4] ? try_to_wake_up+0x199/0x199
 [362409.430358]  [a04b13a3] ? NF_HOOK.clone.4+0x56/0x56 [bridge]
 [362409.430375]  [a04b1389] ? NF_HOOK.clone.4+0x3c/0x56 [bridge]
 [362409.430392]  [a04b1745] ? br_handle_frame+0x1af/0x1c6 [bridge]
 [362409.430408]  [a04b1596] ? br_handle_frame_finish+0x1f3/0x1f3 
 [bridge]
 [362409.430420]  [8127cb7c] ? __netif_receive_skb+0x2c4/0x40b
 [362409.430432]  [8127cd3b] ? process_backlog+0x78/0x157
 [362409.430443]  [8127dd68] ? net_rx_action+0xa4/0x1b2
 [362409.430454]  [81038189] ? test_tsk_need_resched+0xe/0x17
 [362409.430465]  [8104bdd4] ? __do_softirq+0xb9/0x178
 [362409.430476]  [8133cf1c] ? call_softirq+0x1c/0x30
 
 The server has 8G of ram, and usually never uses more than about 4G (sitting 
 at 3.4G right now).

This is a guest problem, please report it to the netfilter/lkml lists
(if its not a known issue with particular kernel version already).

--
To unsubscribe from this list: send the line unsubscribe kvm in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

order 1 page allocation failures

2011-09-27 Thread Thomas Fjellstrom 
Hi,

I've been having some issues with KVM recently where one or more vms will cause 
page allocation failure messages, usually with the backtrace including 
networking functions, example follows:

[362409.429944] kvm: page allocation failure: order:1, mode:0x20
[362409.429957] Pid: 3453, comm: kvm Not tainted 3.0.0-1-amd64 #1
[362409.429965] Call Trace:
[362409.429970]  IRQ  [810b9c90] ? warn_alloc_failed+0x108/0x11b
[362409.429998]  [810bcd78] ? __alloc_pages_nodemask+0x6e6/0x75c
[362409.430012]  [810ec0c0] ? kmem_getpages+0x55/0xf0
[362409.430022]  [810ec87a] ? fallback_alloc+0x129/0x1c1
[362409.430035]  [8100e28d] ? paravirt_read_tsc+0x5/0x8
[362409.430045]  [810ed10e] ? kmem_cache_alloc+0x73/0xf0
[362409.430057]  [812707a2] ? sk_prot_alloc+0x2b/0x128
[362409.430067]  [81270965] ? sk_clone+0x14/0x2bd
[362409.430077]  [812ade7d] ? inet_csk_clone+0x10/0x91
[362409.430088]  [812c1aae] ? tcp_create_openreq_child+0x21/0x41a
[362409.430099]  [812bf98a] ? tcp_v4_syn_recv_sock+0x33/0x208
[362409.430110]  [812c2441] ? tcp_check_req+0x1ff/0x2dd
[362409.430122]  [812adc06] ? inet_csk_search_req+0x35/0xa7
[362409.430132]  [812bf4f1] ? tcp_v4_do_rcv+0x206/0x32c
[362409.430144]  [812c15d4] ? tcp_v4_rcv+0x419/0x66c
[362409.430154]  [8100e74a] ? native_sched_clock+0x28/0x30
[362409.430173]  [812a5a0c] ? ip_local_deliver_finish+0x14b/0x1bb
[362409.430186]  [8127cc8f] ? __netif_receive_skb+0x3d7/0x40b
[362409.430197]  [8127d74b] ? netif_receive_skb+0x52/0x58
[362409.430220]  [a04b5af6] ? br_nf_pre_routing_finish+0x1d4/0x1e1 
[bridge]
[362409.430241]  [a04b5111] ? NF_HOOK_THRESH+0x3b/0x55 [bridge]
[362409.430260]  [a04b60ed] ? br_nf_pre_routing+0x3be/0x3cb [bridge]
[362409.430272]  [8129fb69] ? nf_iterate+0x41/0x77
[362409.430288]  [a04b13a3] ? NF_HOOK.clone.4+0x56/0x56 [bridge]
[362409.430305]  [a04b13a3] ? NF_HOOK.clone.4+0x56/0x56 [bridge]
[362409.430315]  [8129fc12] ? nf_hook_slow+0x73/0x111
[362409.430330]  [a04b13a3] ? NF_HOOK.clone.4+0x56/0x56 [bridge]
[362409.430342]  [8103f0a4] ? try_to_wake_up+0x199/0x199
[362409.430358]  [a04b13a3] ? NF_HOOK.clone.4+0x56/0x56 [bridge]
[362409.430375]  [a04b1389] ? NF_HOOK.clone.4+0x3c/0x56 [bridge]
[362409.430392]  [a04b1745] ? br_handle_frame+0x1af/0x1c6 [bridge]
[362409.430408]  [a04b1596] ? br_handle_frame_finish+0x1f3/0x1f3 
[bridge]
[362409.430420]  [8127cb7c] ? __netif_receive_skb+0x2c4/0x40b
[362409.430432]  [8127cd3b] ? process_backlog+0x78/0x157
[362409.430443]  [8127dd68] ? net_rx_action+0xa4/0x1b2
[362409.430454]  [81038189] ? test_tsk_need_resched+0xe/0x17
[362409.430465]  [8104bdd4] ? __do_softirq+0xb9/0x178
[362409.430476]  [8133cf1c] ? call_softirq+0x1c/0x30
[362409.430481]  EOI  [8100a9ef] ? do_softirq+0x3f/0x84
[362409.430498]  [8127e051] ? netif_rx_ni+0x1e/0x27
[362409.430509]  [a06af7a8] ? tun_get_user+0x390/0x3b8 [tun]
[362409.430520]  [8106] ? bit_waitqueue+0x71/0xa4
[362409.430529]  [81023145] ? _flat_send_IPI_mask+0x6a/0x7c
[362409.430541]  [a06afbad] ? tun_get_socket+0x3b/0x3b [tun]
[362409.430552]  [a06afc0b] ? tun_chr_aio_write+0x5e/0x79 [tun]
[362409.430563]  [810fc193] ? do_sync_readv_writev+0x9a/0xd5
[362409.430574]  [810383fc] ? need_resched+0x1a/0x23
[362409.430585]  [813358d7] ? _cond_resched+0x9/0x20
[362409.430596]  [810fb603] ? copy_from_user+0x18/0x30
[362409.430608]  [811653e9] ? security_file_permission+0x18/0x33
[362409.430618]  [810fc3e1] ? do_readv_writev+0xa4/0x11a
[362409.430627]  [810fcb94] ? fput+0x1a/0x1a2
[362409.430636]  [810fc5c5] ? sys_writev+0x45/0x90
[362409.430647]  [8133bd12] ? system_call_fastpath+0x16/0x1b
[362409.430654] Mem-Info:
[362409.430659] Node 0 DMA per-cpu:
[362409.430667] CPU0: hi:0, btch:   1 usd:   0
[362409.430673] CPU1: hi:0, btch:   1 usd:   0
[362409.430679] CPU2: hi:0, btch:   1 usd:   0
[362409.430686] CPU3: hi:0, btch:   1 usd:   0
[362409.430691] Node 0 DMA32 per-cpu:
[362409.430699] CPU0: hi:  186, btch:  31 usd: 147
[362409.430705] CPU1: hi:  186, btch:  31 usd:  82
[362409.430712] CPU2: hi:  186, btch:  31 usd: 171
[362409.430718] CPU3: hi:  186, btch:  31 usd:  86
[362409.430724] Node 0 Normal per-cpu:
[362409.430730] CPU0: hi:  186, btch:  31 usd: 157
[362409.430737] CPU1: hi:  186, btch:  31 usd:  96
[362409.430743] CPU2: hi:  186, btch:  31 usd: 163
[362409.430749] CPU3: hi:  186, btch:  31 usd: 173
[362409.430764] active_anon:576670 inactive_anon:110462 isolated_anon:0
[362409.430769]  active_file:615766 inactive_file:638287 isolated_file:0
[362409.430774]  unevictable:0 dirty:3203 writeback:0