Jenkins angent on k8s is slow when checkout codes - kubernetes

Question

My jenkins agent is deploy on the k8s, here is the agent yaml:
---
apiVersion: v1
kind: Pod
metadata:
labels:
jenkins: slave
cluster: dev-monitor-platform
spec:
affinity:
nodeAffinity:
requiredDuringSchedulingIgnoredDuringExecution:
nodeSelectorTerms:
- matchExpressions:
- key: jenkins
operator: In
values:
- ci
securityContext:
runAsUser: 0
runAsGroup: 0
fsGroup: 0
containers:
- name: slave-docker
image: harbor.mycompany.net/jenkins/docker:19.03-git
imagePullPolicy: IfNotPresent
resources:
requests:
cpu: 50m
memory: 256Mi
limits:
cpu: 100m
memory: 512Mi
securityContext:
privileged: true
command:
- cat
tty: true
volumeMounts:
- mountPath: /var/run/docker.sock
name: docker-sock
- mountPath: /root/.m2
name: jenkins-maven-m2
- mountPath: /home/jenkins/
name: workspace-volume
readOnly: false
- name: jnlp
image: harbor.mycompany.net/jenkins/inbound-agent:alpine-jdk11
imagePullPolicy: IfNotPresent
resources:
requests:
cpu: 50m
memory: 256Mi
limits:
cpu: 100m
memory: 512Mi
volumeMounts:
- mountPath: /home/jenkins/
name: workspace-volume
readOnly: false
volumes:
- name: docker-sock
hostPath:
path: /var/run/docker.sock
- name: workspace-volume
emptyDir: {}
- name: jenkins-maven-m2
nfs:
path: /export/mid-devops/jenkins/m2
server: xxx.xxx.xxx.xxx
The master itself pull code is fast:
However when the agent try to pull pipeline codes, it always gets stuck for about One-and-a-half minutes:
and when agent pull application codes, it is fast.
This problem happens every time, I have no idea.
I expect the agent don't get stuck when checkout codes