Auto commit by PR queue bot

Revert "remove cluster logging e2e test from flaky suite"

skip update when deleting with grace-period=0

Mark L7 tests as Feature:Ingress

Update docs on flaky issues

Make HPA Deployment tests [Feature:Deployment] until GKE has enabled them

Auto commit by PR queue bot

Add timeout, fix potential startup hang

Reference #20015 for [Flaky] NodeOutOfDisk tests

explictly check log tainted string

Auto commit by PR queue bot

Un-disable "Pods should get a host IP" e2e.

Promote DaemonRestart e2es out of Flaky

Adjust the limits in the kubelet resource tracking test

Fix a couple typos

Auto commit by PR queue bot

Auto commit by PR queue bot

Also add a test for the Update() logic.
Reordered tunnels vs. storage initialization (prevent a nil ptr panic)

Don't re-check with gcloud for a firewall's existence to fix a potential flake.

Change docker health checker to using `docker ps`

Auto commit by PR queue bot

`docker ps` is a more reliable health check than docker version from past
experiences.

Add timestamps to the liveness e2e test 

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Knock down the timeouts for gce, gce-slow, gke, and gke-slow

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Add a mutex to guard SetUpAt() and TearDownAt() calls - they should not
run in parallel.  There is a race in these calls when there are two pods
using the same volume, one of them is dying and the other one starting.

TearDownAt() checks that a volume is not needed by any pods and detaches the
volume. It does so by counting how many times is the volume mounted
(GetMountRefs() call below).

When SetUpAt() of the starting pod already attached the volume and did not mount
it yet, TearDownAt() of the dying pod will detach it - GetMountRefs() does not
count with this volume.

These two threads run in parallel:

 dying pod.TearDownAt("myVolume")          starting pod.SetUpAt("myVolume")
   |                                       |
   |                                       AttachDisk("myVolume")
   refs, err := mount.GetMountRefs()       |
   Unmount("myDir")                        |
   if refs == 1 {                          |
   |  |                                    Mount("myVolume", "myDir")
   |  |                                    |
   |  DetachDisk("myVolume")               |
   |                                       start containers - OOPS! The volume is detached!
   |
   finish the pod cleanup


Also, add some logs to cinder plugin for easier debugging in the future, add
a test and update the fake mounter to know about bind mounts.