CFS runqueues are throttled, preventing tasks from being picked up by the scheduler for a long time in Red Hat Enterprise Linux 7

Issue

• System is unresponsive with following call trace:

INFO: task docker-runc-cur:40219 blocked for more than 120 seconds.
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
docker-runc-cur D ffff88edb1718000     0 40219      1 0x00000084
Call Trace:
 [<ffffffff94918f39>] schedule+0x29/0x70
 [<ffffffff9491a825>] rwsem_down_write_failed+0x225/0x3a0
 [<ffffffff944fa284>] ? ima_inode_post_setattr+0xa4/0xd0
 [<ffffffff9431b3b5>] ? allocate_cg_links+0x55/0x80
 [<ffffffff9455f547>] call_rwsem_down_write_failed+0x17/0x30
 [<ffffffff949180ad>] down_write+0x2d/0x3d
 [<ffffffff943edf40>] mpol_rebind_mm+0x20/0x60
 [<ffffffff9432449a>] cpuset_attach+0x1ea/0x3b0
 [<ffffffff9431dc95>] cgroup_attach_task+0x1b5/0x410
 [<ffffffff9431ea50>] attach_task_by_pid+0x160/0x1a0
 [<ffffffff9431eaa6>] cgroup_procs_write+0x16/0x20
 [<ffffffff9431d359>] cgroup_file_write+0x119/0x2d0
 [<ffffffff94422318>] ? __sb_start_write+0x58/0x110
 [<ffffffff944d5577>] ? security_file_permission+0x27/0xa0
 [<ffffffff9441f390>] vfs_write+0xc0/0x1f0
 [<ffffffff944201bf>] SyS_write+0x7f/0xf0
 [<ffffffff9492579b>] system_call_fastpath+0x22/0x27