Failed to shutdown container with hcsshim error "WaitTimeout: hcsshim: timeout waiting for notification"

Hi, I am running Kubernete on Windows, and runtime is docker. When I delete Windows Pod, following errors are printed

time="2021-06-29T18:50:48.483759300-07:00" level=error msg="hcsshim::ComputeSystem::WaitTimeout - End Operation - Error" cid=d775ab1799c663245df
b553d6667bd27f7bb3c39ff7786cc55f21dc91555cadb error="WaitTimeout d775ab1799c663245dfb553d6667bd27f7bb3c39ff7786cc55f21dc91555cadb: hcsshim: time
out waiting for notification"                                                                                                                   
time="2021-06-29T18:50:48.483759300-07:00" level=error msg="failed to shutdown container, and subsequent terminate also failed" container=d775ab
1799c663245dfb553d6667bd27f7bb3c39ff7786cc55f21dc91555cadb error="container d775ab1799c663245dfb553d6667bd27f7bb3c39ff7786cc55f21dc91555cadb enc
ountered an error during WaitTimeout: hcsshim: timeout waiting for notification" module=libcontainerd namespace=moby                            
time="2021-06-29T18:50:48.483759300-07:00" level=warning msg="failed to shutdown container" container=d775ab1799c663245dfb553d6667bd27f7bb3c39ff
7786cc55f21dc91555cadb error="container d775ab1799c663245dfb553d6667bd27f7bb3c39ff7786cc55f21dc91555cadb encountered an error during WaitTimeout
: hcsshim: timeout waiting for notification: subsequent terminate failed container d775ab1799c663245dfb553d6667bd27f7bb3c39ff7786cc55f21dc91555c
adb encountered an error during WaitTimeout: hcsshim: timeout waiting for notification" module=libcontainerd namespace=moby process=init        
time="2021-06-29T18:50:48.483759300-07:00" level=info msg="sending event" container=d775ab1799c663245dfb553d6667bd27f7bb3c39ff7786cc55f21dc91555
cadb event=exit event-info="{d775ab1799c663245dfb553d6667bd27f7bb3c39ff7786cc55f21dc91555cadb init 816 4294967295 2021-06-29 18:40:48.4397434 -0
700 PDT m=+1797.570635201 false failed to shutdown container: container d775ab1799c663245dfb553d6667bd27f7bb3c39ff7786cc55f21dc91555cadb encount
ered an error during WaitTimeout: hcsshim: timeout waiting for notification: subsequent terminate failed container d775ab1799c663245dfb553d6667b
d27f7bb3c39ff7786cc55f21dc91555cadb encountered an error during WaitTimeout: hcsshim: timeout waiting for notification}" module=libcontainerd na
mespace=moby                                                                                                                                    
time="2021-06-29T18:50:48.483759300-07:00" level=error msg="failed to terminate hccshim container" container=d775ab1799c663245dfb553d6667bd27f7b
b3c39ff7786cc55f21dc91555cadb error="container d775ab1799c663245dfb553d6667bd27f7bb3c39ff7786cc55f21dc91555cadb encountered an error during Term
inate: hcsshim: the handle has already been closed" module=libcontainerd namespace=moby pid=816 process=init signal=9

The error might cause kubelet fail to sync message to kube-apiserver, and cause the Windows Node status to be not ready.

My Windows version is Windows Server 2019 (1809), and docker version is 19.03.5

Start with the reported hcsshim WaitTimeout and subsequent handle-closed errors while deleting a Windows pod, using the stated Windows Server 2019 (1809) and Docker 19.03.5 versions. Reproduce the container shutdown failure and determine whether it explains the kubelet synchronization and Windows Node readiness symptoms; done requires a confirmed cause and a validated fix or workaround.