RancherOS on AWS host kswapd8 high CPU load

Occasionally my created instances will go haywire looking for swap.

top - 04:05:31 up 17 days,  1:37,  1 user,  load average: 2.01, 2.05, 1.66
Tasks: 134 total,   5 running, 129 sleeping,   0 stopped,   0 zombie
%Cpu(s):  4.7 us, 94.7 sy,  0.0 ni,  0.0 id,  0.0 wa,  0.0 hi,  0.3 si,  0.3 st
KiB Mem:   2047128 total,  1767364 used,   279764 free,   340856 buffers
KiB Swap:        0 total,        0 used,        0 free.   401348 cached Mem

  PID USER      PR  NI    VIRT    RES    SHR S %CPU %MEM     TIME+ COMMAND                                                                                                                                    
   33 root      20   0       0      0      0 R 82.0  0.0   5562:26 kswapd0                                                                                                                                    
 9687 root      20   0  398292  71252  12652 S  5.3  3.5  22:41.47 .cadvisor                          

echo 3 > /proc/sys/vm/drop_caches

seemed to fix it, at least temporarily.