What do the components in a page allocation failure message mean?
DCA v2
Appliance Version = 3.5.1.0
[root@sdw3 ~]# free -m
total used free shared buffers cached
Mem: 128963 114150 14813 1859 44 89600
-/+ buffers/cache: 24505 104458
Swap: 32765 0 32765
[root@sdw3 ~]# cat /var/log/messages
........
Mar 15 00:15:39 sdw3 kernel: <EOI>
Mar 15 00:15:39 sdw3 kernel: __ratelimit: 185 callbacks suppressed
Mar 15 00:15:39 sdw3 kernel: postgres: page allocation failure. order:2, mode:0x4020
Mar 15 00:15:39 sdw3 kernel: postgres: page allocation failure. order:2, mode:0x4020
Mar 15 00:15:39 sdw3 kernel: Pid: 158882, comm: postgres Not tainted 2.6.32-754.3.5.el6.x86_64 #1
Mar 15 00:15:39 sdw3 kernel: Call Trace:
Mar 15 00:15:39 sdw3 kernel: <IRQ> [<ffffffff81143891>] ? __alloc_pages_nodemask+0x7e1/0x960
Mar 15 00:15:39 sdw3 kernel: [<ffffffff814f0cba>] ? udp_rcv+0x1a/0x20
Mar 15 00:15:39 sdw3 kernel: [<ffffffff814c4000>] ? ip_rcv_finish+0x90/0x460
Mar 15 00:15:39 sdw3 kernel: [<ffffffff814c4648>] ? ip_rcv+0x278/0x360
Mar 15 00:15:39 sdw3 kernel: [<ffffffff8117ec8a>] ? alloc_pages_current+0xaa/0x110
Mar 15 00:15:39 sdw3 kernel: [<ffffffffa0426809>] ? bnad_rxq_refill_page+0x159/0x1d0 [bna]
Mar 15 00:15:39 sdw3 kernel: [<ffffffffa0428a7b>] ? bnad_cq_process+0x48b/0x9c0 [bna]
Mar 15 00:15:39 sdw3 kernel: [<ffffffffa0430a6f>] ? bnad_napi_poll_rx+0x5f/0x160 [bna]
Mar 15 00:15:39 sdw3 kernel: [<ffffffff8148e4d9>] ? net_rx_action+0x109/0x320
Mar 15 00:15:39 sdw3 kernel: [<ffffffff8108973a>] ? __do_softirq+0xea/0x240
Mar 15 00:15:39 sdw3 kernel: [<ffffffff810f8336>] ? handle_IRQ_event+0x66/0x180
Mar 15 00:15:39 sdw3 kernel: [<ffffffff815658cc>] ? call_softirq+0x1c/0x30
Mar 15 00:15:39 sdw3 kernel: [<ffffffff8100e4c5>] ? do_softirq+0x65/0xa0
Mar 15 00:15:39 sdw3 kernel: [<ffffffff810893cd>] ? irq_exit+0x8d/0xa0
Mar 15 00:15:39 sdw3 kernel: [<ffffffff81566725>] ? do_IRQ+0x75/0xf0
Mar 15 00:15:39 sdw3 kernel: [<ffffffff81564e93>] ? ret_from_intr+0x0/0x11
root @ sdw3.gphd.local [/root]# cat /proc/sys/vm/min_free_kbytes
90112
root @ sdw3.gphd.local [/root]# cat /etc/sysctl.conf
# Controls the maximum number of shared memory segments, in pages
kernel.shmall=4000000000
kernel.shmmni=4096
kernel.sem=250 512000 100 2048
kernel.msgmni=2048
kernel.core_pattern=/var/crash/user/core.%e.%p.%t.%s.%u.%g
kernel.pid_max=798720
kernel.threads-max=798720
kernel.unknown_nmi_panic=1
kernel.softlockup_panic=1
kernel.panic_on_unrecovered_nmi=1
fs.nr_open=3000000
net.ipv4.ip_local_port_range=1281 65535
net.ipv4.tcp_tw_recycle=1
net.ipv4.conf.default.arp_filter=1
net.ipv4.conf.all.arp_filter=1
net.ipv4.tcp_max_syn_backlog=200000
net.ipv4.ip_local_reserved_ports=28080
net.core.netdev_max_backlog=200000
net.core.rmem_max=2097152
net.core.wmem_max=2097152
vm.overcommit_memory=2
vm.overcommit_ratio=95
vm.swappiness=2
net.ipv6.conf.all.disable_ipv6=1
net.ipv6.conf.default.disable_ipv6=1
net.ipv6.conf.all.autoconf=0
net.ipv6.conf.default.autoconf=0