How to add kubernetes liveness probe - centos

Question

I am writing a simple YAML file to apply liveness probe using a TCP port on Centos.6
I pulled a centos:6 image from public repository
started a container using the image.
installed mysql, and started it to verify a opened port (3306)
committed to local repository as "mysql-test:v0.1"
and apply a pod as below
apiVersion: v1
kind: Pod
metadata:
labels:
test: mysql-test
name: mysql-test-exec
spec:
containers:
- name: mysql-test
args:
- /sbin/service
- mysqld
- start
image: mysql-test:v0.1
livenessProbe:
tcpSocket:
port: 3306
initialDelaySeconds: 15
periodSeconds: 20
But, the status of the pod is CrashLoopBackOff, and the status of the container on work02 is Exited.
1) master node
root#kubeadm-master01:~# kubectl get pods
NAME READY STATUS RESTARTS AGE
mysql-test-exec 0/1 CrashLoopBackOff 6 8m
root#kubeadm-master01:~# kubectl describe pod mysql-test-exec
.
.
.
Events:
FirstSeen LastSeen Count From SubObjectPath Type Reason Message
--------- -------- ----- ---- ------------- -------- ------ -------
1m 1m 1 default-scheduler Normal Scheduled Successfully assigned mysql-test-exec to kubeadm-work02
1m 1m 1 kubelet, kubeadm-work02 spec.containers{mysql-test} Normal Created Created container with id abbad6585021151cd86fdfb3a9f733245f603686c90f533f23
44397c97c36918
1m 1m 1 kubelet, kubeadm-work02 spec.containers{mysql-test} Normal Started Started container with id abbad6585021151cd86fdfb3a9f733245f603686c90f533f23
44397c97c36918
1m 1m 1 kubelet, kubeadm-work02 spec.containers{mysql-test} Normal Started Started container with id a1062083089eed109fe8f41344136631bb9d4c08a2c6454dc7
7f677f01a48666
1m 1m 1 kubelet, kubeadm-work02 spec.containers{mysql-test} Normal Created Created container with id a1062083089eed109fe8f41344136631bb9d4c08a2c6454dc7
7f677f01a48666
1m 1m 3 kubelet, kubeadm-work02 Warning FailedSync Error syncing pod, skipping: failed to "StartContainer" for "mysql-test" wit
h CrashLoopBackOff: "Back-off 10s restarting failed container=mysql-test pod=mysql-test-exec_default(810c37bd-7a8c-11e7-9224-525400603584)"
1m 1m 1 kubelet, kubeadm-work02 spec.containers{mysql-test} Normal Created Created container with id 79512aeaf8a6b4692e11b344adb24763343bb2a06c9003222097962822d42202
1m 1m 1 kubelet, kubeadm-work02 spec.containers{mysql-test} Normal Started Started container with id 79512aeaf8a6b4692e11b344adb24763343bb2a06c9003222097962822d42202
1m 43s 3 kubelet, kubeadm-work02 Warning FailedSync Error syncing pod, skipping: failed to "StartContainer" for "mysql-test" with CrashLoopBackOff: "Bac
k-off 20s restarting failed container=mysql-test pod=mysql-test-exec_default(810c37bd-7a8c-11e7-9224-525400603584)"
29s 29s 1 kubelet, kubeadm-work02 spec.containers{mysql-test} Normal Started Started container with id 4427a3b8e5320b284ac764c1152def4ba749e4f656b3c464a472514bccf2e30e
1m 29s 4 kubelet, kubeadm-work02 spec.containers{mysql-test} Normal Pulled Container image "centos-mysql:v0.1" already present on machine
29s 29s 1 kubelet, kubeadm-work02 spec.containers{mysql-test} Normal Created Created container with id 4427a3b8e5320b284ac764c1152def4ba749e4f656b3c464a472514bccf2e30e
1m 10s 9 kubelet, kubeadm-work02 spec.containers{mysql-test} Warning BackOff Back-off restarting failed container
27s 10s 3 kubelet, kubeadm-work02 Warning FailedSync Error syncing pod, skipping: failed to "StartContainer" for "mysql-test" with CrashLoopBackOff: "Bac
k-off 40s restarting failed container=mysql-test pod=mysql-test-exec_default(810c37bd-7a8c-11e7-9224-525400603584)"
2) work node
root#kubeadm-work02:~# docker logs f64e20bf33a8
Starting mysqld: [ OK ]

Answer 1

I have to remove the args to work with docker image. below deployment works for me.
apiVersion: v1
kind: Pod
metadata:
labels:
test: mysql-test
name: mysql-test-exec
spec:
containers:
- name: mysql-test
image: mysql:5.6
env:
- name: MYSQL_ROOT_PASSWORD
value: mysql456
livenessProbe:
tcpSocket:
port: 3306
initialDelaySeconds: 15
periodSeconds: 20