RHEL 6 server crashed with error: postgres: page allocation failure. order:0, mode:0x20 (Doc ID 2568647.1)

Last updated on JULY 14, 2020

Applies to:

Symptoms

A physical server running RHEL 6, with kernel 2.6.32-431.29.2.el6.x86_64 crashed unexpectedly, with no hardware errors found.

The vmcore file generated from the crash showed the following back trace:

with call traces showing order zero page allocation failures:
postgres: page allocation failure. order:0, mode:0x20
[] ? ixgbe_poll+0x5fc/0x12c0 [ixgbe]
Pid: 17557, comm: postgres Not tainted 2.6.32-431.29.2.el6.x86_64 #1
[] ? call_softirq+0x1c/0x30
[] ? try_to_free_pages+0x92/0x120
Call Trace:
[] ? net_rx_action+0x103/0x2f0
[] ? do_softirq+0x65/0xa0
[] ? __alloc_pages_nodemask+0x47e/0x8d0
[] ? __do_softirq+0xc1/0x1e0
[] ? __alloc_pages_nodemask+0x74a/0x8d0
[] ? irq_exit+0x85/0x90
[] ? common_interrupt+0xe/0x13
[] ? handle_IRQ_event+0x60/0x170
[] ? __alloc_pages_nodemask+0x113/0x8d0
[] ? do_IRQ+0x75/0xf0
[] ? alloc_pages_vma+0x9a/0x150
[] ? __do_softirq+0x11f/0x1e0
[] ? kmem_getpages+0x62/0x170
[] ? ret_from_intr+0x0/0x11
[] ? handle_pte_fault+0x73d/0xb00
[] ? call_softirq+0x1c/0x30
[] ? fallback_alloc+0x1ba/0x270
[] ? vma_adjust+0x11f/0x5e0
[] ? page_check_address+0xaa/0x1d0
[] ? do_softirq+0x65/0xa0
[] ? cache_grow+0x2cf/0x320
[] ? autoremove_wake_function+0x0/0x40
[] ? page_check_address+0x35/0x1d0
[] ? irq_exit+0x85/0x90
[] ? ____cache_alloc_node+0x99/0x160
[] ? handle_mm_fault+0x22a/0x300
[] ? page_referenced_one+0x53/0x1e0
[] ? do_IRQ+0x75/0xf0
[] ? kmem_cache_alloc_node_trace+0x90/0x200
[] ? __vm_enough_memory+0xf4/0x190
[] ? prio_tree_next+0x216/0x250
[] ? ret_from_intr+0x0/0x11
[] ? __kmalloc_node+0x4d/0x60
[] ? __do_page_fault+0x138/0x480
[] ? _spin_lock+0x21/0x30
[] ? __alloc_skb+0x7a/0x180
[] ? _spin_lock+0x21/0x30
[] ? sys_recvfrom+0x16b/0x180
[] ? page_referenced+0x148/0x360

Cause

My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.