Ceph - OSD nodes have 'page allocation failure' messages in system logs and causing slow requests because of heartbeat failures.

Solution Verified - Updated -

Issue

  • Ceph - OSD nodes have page allocation failure messages in system logs and causing slow requests because of heartbeat failures.
- In dmesg:

[2487558.078129] swapper/2: page allocation failure: order:2, mode:0x104020
[2487558.078471] python: page allocation failure: order:2, mode:0x104020
- In /var/log/messages:

May 11 12:42:55 node01 kernel: kswapd0: page allocation failure: order:2, mode:0x104020
May 11 12:42:55 node01 kernel: CPU: 36 PID: 225 Comm: kswapd0 Tainted: G        W      ------------   3.10.0-514.10.2.el7.x86_64 #1

Environment

  • Red Hat Enterprise Linux 7.3 - kernel-3.10.0-514.10.2.el7
  • Red Hat Ceph Storage 2.2
  • Virtual memory kernel tunable - vm.min_free_kbytes

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase of over 48,000 articles and solutions.

Current Customers and Partners

Log in for full access

Log In