task oracle_110944_k:110944 blocked for more than 120 seconds

Jul 14 09:21:38 db systemd-logind: Failed to abandon session scope: Connection timed out
Jul 14 09:26:38 db systemd-logind: Failed to start session scope session-2283.scope: Connection timed out
Jul 14 09:28:45 db systemd: Failed to start Session 2283 of user root.
Jul 14 09:29:52 db systemd-logind: Failed to start session scope session-2285.scope: Connection timed out
Jul 14 09:31:08 db systemd: Failed to start Session 2285 of user root.
Jul 14 09:32:04 db systemd-logind: Failed to start session scope session-2286.scope: Connection timed out
Jul 14 09:32:46 db kernel: INFO: task oracle_110944_k:110944 blocked for more than 120 seconds.
Jul 14 09:33:38 db kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Jul 14 09:33:38 db kernel: oracle_110944_k D ffff94e117266eb0     0 110944      1 0x00000080
Jul 14 09:33:58 db kernel: Call Trace:
Jul 14 09:34:29 db kernel: [] schedule_preempt_disabled+0x29/0x70
Jul 14 09:35:44 db kernel: [] __mutex_lock_slowpath+0xc7/0x1d0
Jul 14 09:35:44 db kernel: [] mutex_lock+0x1f/0x2f
Jul 14 09:35:44 db kernel: [] do_last+0x28f/0x12c0
Jul 14 09:35:44 db kernel: [] path_openat+0xd7/0x640
Jul 14 09:35:44 db kernel: [] ? __do_fault.isra.58+0x8a/0x100
Jul 14 09:35:44 db kernel: [] do_filp_open+0x4d/0xb0

Jul 14 09:48:14 db systemd: Started Session 2299 of user root.
Jul 14 09:50:32 db systemd-logind: Failed to start session scope session-2300.scope: Connection timed out
Jul 14 09:52:34 db systemd-logind: Failed to start session scope session-2301.scope: Connection timed out
Jul 14 09:52:48 db su: (to oracle) root on tty1
Jul 14 09:53:12 db systemd-logind: Failed to start session scope session-2302.scope: Connection timed out
Jul 14 09:53:36 db kernel: [drm:vmw_kms_update_layout_ioctl [vmwgfx]] *ERROR* Topology is beyond supported limits.
Jul 14 17:54:46 db kernel: Initializing cgroup subsys cpuset
Jul 14 17:54:46 db kernel: Initializing cgroup subsys cpu
Jul 14 17:54:46 db kernel: Initializing cgroup subsys cpuacct
Jul 14 17:54:46 db kernel: Linux version 3.10.0-862.el7.x86_64 ([email protected]) (gcc version 4.8.5 20150623 (Red Hat 4.8.5-28) (GCC) ) #1 SMP Fri Apr 20 16:44:24 UTC 2018
Jul 14 17:54:46 db kernel: Command line: BOOT_IMAGE=/vmlinuz-3.10.0-862.el7.x86_64 root=/dev/mapper/centos-root ro crashkernel=auto rd.lvm.lv=centos/root rd.lvm.lv=centos/swap rhgb quiet LANG=en_US.UTF-8
 
 
 
Jul 14 09:20:07 db systemd: Started Session 2280 of user root.
Jul 14 09:20:07 db systemd: Started Session 2281 of user root.
Jul 14 09:21:38 db systemd-logind: Failed to abandon session scope: Connection timed out
Jul 14 09:26:38 db systemd-logind: Failed to start session scope session-2283.scope: Connection timed out
Jul 14 09:28:45 db systemd: Failed to start Session 2283 of user root.
Jul 14 09:29:52 db systemd-logind: Failed to start session scope session-2285.scope: Connection timed out
Jul 14 09:31:08 db systemd: Failed to start Session 2285 of user root.
Jul 14 09:32:04 db systemd-logind: Failed to start session scope session-2286.scope: Connection timed out
Jul 14 09:32:46 db kernel: INFO: task oracle_110944_k:110944 blocked for more than 120 seconds.
Jul 14 09:33:38 db kernel: "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Jul 14 09:33:38 db kernel: oracle_110944_k D ffff94e117266eb0     0 110944      1 0x00000080
Jul 14 09:33:58 db kernel: Call Trace:
Jul 14 09:34:29 db kernel: [] schedule_preempt_disabled+0x29/0x70
Jul 14 09:35:44 db kernel: [] __mutex_lock_slowpath+0xc7/0x1d0
Jul 14 09:35:44 db kernel: [] mutex_lock+0x1f/0x2f
Jul 14 09:35:44 db kernel: [] do_last+0x28f/0x12c0
Jul 14 09:35:44 db kernel: [] path_openat+0xd7/0x640
Jul 14 09:35:44 db kernel: [] ? __do_fault.isra.58+0x8a/0x100

处理办法

1、首先操作 # echo 0 > /proc/sys/kernel/hung_task_timeout_secs

2、根据mos:Linux: Node Reboots After "kernel: INFO: task ocssd.bin: blocked for more than 120 seconds." in /var/log/messages (Doc ID 1423693.1)

做如下操作：

echo 'vm.min_free_kbytes = 262144' >> /etc/sysctl.conf
echo 'vm.swappiness=100' >> /etc/sysctl.conf
sysctl -p

3、配置hugepage并禁用Auto Memory Management (AMM)