前言：此文档是用来在线下环境harbor利用MinIO做镜像存储的，至于那些说OSS不香吗？或者单机harbor的，不用看了。此文档对你没啥用，如果是采用单机的harbor连接集群MinIO，请看我的另一篇博文。
环境:
应用版本：
helm v3.2.3
kubernetes 1.14.3
nginx-ingress 1.39.1
harbor 2.0
nginx 1.15.3
MinIO RELEASE.2020-05-08T02-40-49Z
### 这里就不讲解kubernetes集群怎么搭建了。我们kubernetes共享存储为了简单，采用的是nfs。我们先讲解一下怎么采用nfs做k8s持久存储。
### 注意执行主机，除了nfs-server是在94那台服务器执行了相关命令，其他的大部分是在master1上面执行
## 一、nfs-client-provisioner
### 1、在nfs-server安装nfs服务
yum -y install nfs-utils rpcbind
mkdir /nfs/data
chmod 777 /nfs/data
echo '/nfs/data *(rw,no_root_squash,sync)' > /etc/exports
exportfs -r
systemctl restart rpcbind && systemctl enable rpcbind
systemctl restart nfs && systemctl enable nfs
rpcinfo -p localhost
showmount -e 10.0.0.94### 2、在其他服务器安装nfs-client
yum install -y nfs-utils
### 3、在k8s-master1上安装nfs-client-provisioner 实现动态持久存储，nfs-client-provisioner 是一个Kubernetes的简易NFS的外部provisioner，本身不提供NFS
cd /usr/local/src && mkdir nfs-client-provisioner && cd nfs-client-provisioner
### 注意deployment.yaml文件中，IP对应的是nfs-server的，PATH路径对应的是nfs-server的/etc/exports的路径
cat > deployment.yaml << EOF
apiVersion: v1
kind: ServiceAccount
metadata:
name: nfs-client-provisioner
---
kind: Deployment
apiVersion: apps/v1
metadata:
name: nfs-client-provisioner
spec:
replicas: 1
selector:
matchLabels:
app: nfs-client-provisioner
strategy:
type: Recreate
template:
metadata:
labels:
app: nfs-client-provisioner
spec:
serviceAccount: nfs-client-provisioner
containers:
- name: nfs-client-provisioner
image: quay-mirror.qiniu.com/external_storage/nfs-client-provisioner:latest
volumeMounts:
- name: nfs-client-root
mountPath: /persistentvolumes
env:
- name: PROVISIONER_NAME
value: fuseim.pri/ifs
- name: NFS_SERVER
value: 10.0.0.94
- name: NFS_PATH
value: /nfs/data
volumes:
- name: nfs-client-root
nfs:
server: 10.0.0.94
path: /nfs/data
EOF
cat > rbac.yaml << EOF
kind: ClusterRole
apiVersion: rbac.authorization.k8s.io/v1
metadata:
name: nfs-client-provisioner-runner
rules:
- apiGroups: [""]
resources: ["persistentvolumes"]
verbs: ["get", "list", "watch", "create", "delete"]
- apiGroups: [""]
resources: ["persistentvolumeclaims"]
verbs: ["get", "list", "watch", "update"]
- apiGroups: [""]
resources: ["endpoints"]
verbs: ["get", "list", "watch", "create", "update", "patch"]
- apiGroups: ["storage.k8s.io"]
resources: ["storageclasses"]
verbs: ["get", "list", "watch"]
- apiGroups: [""]
resources: ["events"]
verbs: ["create", "update", "patch"]
---
kind: ClusterRoleBinding
apiVersion: rbac.authorization.k8s.io/v1
metadata:
name: run-nfs-client-provisioner
subjects:
- kind: ServiceAccount
name: nfs-client-provisioner
namespace: default
- kind: ServiceAccount
name: nfs-client-provisioner
namespace: kube-system
roleRef:
kind: ClusterRole
name: nfs-client-provisioner-runner
apiGroup: rbac.authorization.k8s.io
---
kind: Role
apiVersion: rbac.authorization.k8s.io/v1
metadata:
name: leader-locking-nfs-client-provisioner
rules:
- apiGroups: [""]
resources: ["endpoints"]
verbs: ["get", "list", "watch", "create", "update", "patch"]
---
kind: RoleBinding
apiVersion: rbac.authorization.k8s.io/v1
metadata:
name: leader-locking-nfs-client-provisioner
subjects:
- kind: ServiceAccount
name: nfs-client-provisioner
# replace with namespace where provisioner is deployed
namespace: default
roleRef:
kind: Role
name: leader-locking-nfs-client-provisioner
apiGroup: rbac.authorization.k8s.io
EOFcat > StorageClass.yaml << EOF
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
name: managed-nfs-storage
provisioner: fuseim.pri/ifs
parameters:
archiveOnDelete: "false"
EOF
kubectl apply -f deployment.yaml
kubectl apply -f rbac.yaml
kubectl apply -f StorageClass.yaml
### 稍等片刻，检查nfs-client-provisioner是否正常，出现下面的输出说明正常，如果不正常请检查上面的步骤，是否存在问题
kubectl get pods -n kube-system | grep nfs
nfs-client-provisioner-7778496f89-kthnj 1/1 Running 0 169m
## 二、安装helm3
cd /usr/local/src &&\
wget https://get.helm.sh/helm-v3.2.3-linux-amd64.tar.gz &&\
tar xf helm-v3.2.3-linux-amd64.tar.gz &&\
cp linux-amd64/helm /usr/bin/ &&\
helm version
## 三、安装nginx-controller-manager
helm repo add stable http://mirror.azure.cn/kubernetes/charts
helm pull stable/nginx-ingress &&\
docker pull fungitive/defaultbackend-amd64 &&\
docker tag fungitive/defaultbackend-amd64 k8s.gcr.io/defaultbackend-amd64:1.5 &&\
helm template guoys nginx-ingress-*.tgz | kubectl apply -f -
## 四、安装MinIO
### 1、在准备的4台服务器安装minio Server，官方建议是准备最低4台服务器，并且是单独的磁盘空间存放minio数据
cd /usr/local/src &&\
wget https://dl.min.io/server/minio/release/linux-amd64/minio &&\
chmod +x minio && cp minio /usr/bincat > /etc/systemd/system/minio.service <<EOF
[Unit]
Description=Minio
Documentation=https://docs.minio.io
Wants=network-online.target
After=network-online.target
AssertFileIsExecutable=/usr/bin/minio
[Service]
EnvironmentFile=-/etc/minio/minio.conf
ExecStart=/usr/bin/minio server $ENDPOINTS
# Let systemd restart this service always
Restart=always
# Specifies the maximum file descriptor number that can be opened by this process
LimitNOFILE=65536
# Disable timeout logic and wait until process is stopped
TimeoutStopSec=infinity
SendSIGKILL=no
[Install]
WantedBy=multi-user.target
EOF
### 此处IP地址要与自己的机器地址对应或者采用域名，后缀是minio存储路径
mkdir -p /etc/minio
cat > /etc/minio/minio.conf <<EOF
MINIO_ACCESS_KEY=guoxy
MINIO_SECRET_KEY=guoxy321export
ENDPOINTS="http://10.0.0.91/minio http://10.0.0.92/minio http://10.0.0.93/minio http://10.0.0.94/minio"
EOF
systemctl daemon-reload && systemctl start minio && systemctl enable minio
### 2、在k8s-master1安装mc命令，并创建bucket harbor
cd /usr/local/src && \
wget https://dl.min.io/client/mc/release/linux-amd64/mc && \
chmod +x mc && cp mc /usr/bin/ && \
mc config host add minio "http://10.0.0.91:9000 http://10.0.0.92:9000 http://10.0.0.93:9000/ http://10.0.0.94:9000" guoxy guoxy321export && \
mc mb minio/harbor
## 五、在k8s中安装harbor
### 1、先在k8s中创建harbor要使用的TLS证书的secret，证书如果没有可以let's encrypt申请
kubectl create secret tls guofire.xyz --key privkey.pem --cert fullchain.pem
### 2、克隆harbor-helm
cd /usr/local/src && \
git clone -b 1.4.0 https://github.com/goharbor/harbor-helm
### 3、修改harbor-helm/values.yaml，由于内容太多了，我只把需要修改的内容贴出来
vim harbor-helm/values.yaml
### secretName对应刚刚创建的secret名称，core为harbor访问域名
secretName: "guofire.xyz"
core: harbor.guofire.xyz
notary: notary.guofire.xyz
externalURL: https://harbor.guofire.xyz
### 下面是nfs持久化存储
persistentVolumeClaim:
registry:
storageClass: "managed-nfs-storage"
subPath: "registry"
storageClass: "managed-nfs-storage"
subPath: "chartmuseum"
storageClass: "managed-nfs-storage"
subPath: "jobservice"
storageClass: "managed-nfs-storage"
subPath: "database"
storageClass: "managed-nfs-storage"
subPath: "redis"
storageClass: "managed-nfs-storage"
subPath: "trivy"
### 这往下最重要，regionendpoint地址可以写nginx代理的地址和端口，我这里只写了minio Server其中一台
imageChartStorage:
disableredirect: true
type: s3
filesystem:
rootdirectory: /storage
#maxthreads: 100
s3:
region: us-west-1
bucket: harbor
accesskey: guoys!
secretkey: guoys321export
regionendpoint: http://10.0.0.92:9000
encrypt: false
secure: false
v4auth: true
chunksize: "5242880"
rootdirectory: /
redirect:
disabled: false
maintenance:
uploadpurging:
enabled: false
delete:
enabled: true
### 4、通过helm在k8s中安装harbor
helm install harbor harbor-helm/
### 5、最后稍等3、5分钟，查看harbor应用是否正常
kubectl get pods
### 出现下面类似的输出，基本上说明harbor已经正常启动
NAME READY UP-TO-DATE AVAILABLE AGE
harbor-harbor-chartmuseum 1/1 1 1 13h
harbor-harbor-clair 1/1 1 1 13h
harbor-harbor-core 1/1 1 1 13h
harbor-harbor-jobservice 1/1 1 1 13h
harbor-harbor-notary-server 1/1 1 1 13h
harbor-harbor-notary-signer 1/1 1 1 13h
harbor-harbor-portal 1/1 1 1 13h
harbor-harbor-registry 1/1 1 1 13h
zy-nginx-ingress-controller 1/1 1 1 32h
zy-nginx-ingress-default-backend 1/1 1 1 32h
## 六、安装nginx 4层转发，否则无法通过nginx-ingress访问harbor
### 1、由于nginx-ingress默认是LoadBalancer模式，在线下环境无法正常使用。我们需要改为NodePort
kubectl edit svc guoys-nginx-ingress-controller
### 修改.spec.type的值为NodePort，并保存
### 2、查看nginx-ingress-controller的nodeport端口，记住80和443对应的端口
kubectl get svc | grep 'ingress-controller'
guoys-nginx-ingress-controller NodePort 10.200.248.214 <none> 80:32492/TCP,443:30071/TCP 32h
### 3、安装nginx4层代理
yum install -y gcc make
mkdir /apps
cd /usr/local/src/
wget http://nginx.org/download/nginx-1.15.3.tar.gz
tar xf nginx-1.15.3.tar.gz
cd nginx-1.15.3
./configure --with-stream --without-http --prefix=/apps/nginx --without-http_uwsgi_module --without-http_scgi_module --without-http_fastcgi_module
make && make install
### 下面upstream中的端口一定要跟上面2步骤NodePort的相对应
cat > /apps/nginx/conf/nginx.conf <<EOF
worker_processes 1;
events {
worker_connections 1024;
}
stream {
log_format tcp '$remote_addr [$time_local] '
'$protocol $status $bytes_sent $bytes_received '
'$session_time "$upstream_addr" '
'"$upstream_bytes_sent" "$upstream_bytes_received" "$upstream_connect_time"';
upstream https_default_backend {
server 10.0.0.91:30071;
server 10.0.0.92:30071;
server 10.0.0.93:30071;
}
upstream http_backend {
server 10.0.0.91:32492;
server 10.0.0.92:32492;
server 10.0.0.93:32492;
}
server {
listen 443;
proxy_pass https_default_backend;
access_log logs/access.log tcp;
error_log logs/error.log;
}
server {
listen 80;
proxy_pass http_backend;
}
}
EOF
### 测试并启动nginx
/apps/nginx/sbin/nginx -t
/apps/nginx/sbin/nginx
echo '/apps/nginx/sbin/nginx' >> /etc/rc.local
## 七、最后进行测试，推送镜像到harbor。成功后查看minio的harbor bucket是否存在docker目录。如果存在说明成
本文章为转载内容，我们尊重原作者对文章享有的著作权。如有内容错误或侵权问题，欢迎原作者联系我们进行内容更正或删除文章。
