基于kubernetes1.11安装Harbor私有镜像库(一)

1.简介

本文主要记录基于kubernetes1.11版本安装harbor私有镜像库，及配置nginx, traefik的代理及https相关过程。
此处harbor采用共享存储(GlusterFS)作为镜像的存储系统。
大约的逻辑架构为: user <- - -> nginx(https) <- - -> traefik(https) <- - -> harbor(https) <- - -> GlusterFS(volume);
本节首先说明如何在Centos7上安装规模为三个节点的GlusterFS集群。

2.安装共享存储(GlusterFS)

• 节点准备

• 所有节点配置/etc/hosts

192.168.1.11 gfs-manager
192.168.1.12 gfs-node1
192.168.1.13 gfs-node2

• 所有节点通过yum安装glusterfs

yum install -y centos-release-gluster
yum install -y glusterfs glusterfs-server glusterfs-fuse glusterfs-rdma
# 启动GlusterFS服务
systemctl start glusterd.service
systemctl enable glusterd.service

• 在manager节点上执行，将node节点加入集群

[root@gfs-manager ~]#gluster peer probe gfs-manager
peer probe: success. Probe on localhost not needed
[root@gfs-manager ~]#gluster peer probe gfs-node1
peer probe: success.
[root@gfs-manager ~]#gluster peer probe gfs-node2
peer probe: success.

• 查看集群状态

[root@gfs-manager ~]# gluster peer status
Number of Peers: 2

Hostname: gfs-node1
Uuid: 25f7804c-2b48-4f88-8658-3b9302d06a19
State: Peer in Cluster (Connected)

Hostname: gfs-node2
Uuid: 0c6196d3-318b-46f9-ac40-29a8212d4900
State: Peer in Cluster (Connected)

• 查看volume状态

[root@gfs-manager ~]#gluster volume info
No volumes present

3.创建数据存储目录

假设要创建的目录为/data/gluster/harbordata, 用于挂载为Harbor的存储目录。

• 所有节点执行创建，目录

mkdir -p /data/gluster/harbordata

• manager节点创建GlusterFS磁盘

[root@gfs-manager ~]#gluster volume create harbordata replica 3 gfs-manager:/data/gluster/harbordata gfs-node1:/data/gluster/harbordata gfs-node2:/data/gluster/harbordata force
volume create: harbordata: success: please start the volume to access data

• manager节点启动harbordata

[root@gfs-manager ~]#gluster volume start harbordata
volume start: harbordata: success

• 查看volume状态

[root@gfs-manager ~]# gluster volume info
 
Volume Name: harbordata
Type: Replicate
Volume ID: c4fb0a43-c9e5-4a4e-ba98-cf14a7591ecd
Status: Started
Snapshot Count: 0
Number of Bricks: 1 x 3 = 3
Transport-type: tcp
Bricks:
Brick1: gfs-manager:/data/gluster/harbordata
Brick2: gfs-node1:/data/gluster/harbordata
Brick3: gfs-node2:/data/gluster/harbordata
Options Reconfigured:
performance.write-behind: on

4. GlusterFS参数调优(供参考)

• 配置相关参数：

#开启指定volume的配额： (harbordata 为volume名)
gluster volume quota harbordata enable

#限制harbordata的根目录最大使用 100GB 空间
gluster volume quota harbordata limit-usage / 100GB

#设置cache
gluster volume set harbordata performance.cache-size 2GB

#开启异步操作
gluster volume set harbordata performance.flush-behind on

#设置io线程
gluster volume set harbordata performance.io-thread-count 16

#设置回写 (写数据时间，先写缓存，再写硬盘)
gluster volume set harbordata performance.write-behind on

• 查看当前volume状态

[root@gfs-manager ~]# gluster volume info
 
Volume Name: harbordata
Type: Replicate
Volume ID: c4fb0a43-c9e5-4a4e-ba98-cf14a7591ecd
Status: Started
Snapshot Count: 0
Number of Bricks: 1 x 3 = 3
Transport-type: tcp
Bricks:
Brick1: gfs-manager:/data/gluster/harbordata
Brick2: gfs-node1:/data/gluster/harbordata
Brick3: gfs-node2:/data/gluster/harbordata
Options Reconfigured:
performance.write-behind: on
performance.io-thread-count: 16
performance.flush-behind: on
performance.cache-size: 2GB
features.quota-deem-statfs: on
features.inode-quota: on
features.quota: on
transport.address-family: inet
nfs.disable: on
performance.client-io-threads: off

Habor的安装后续请关注后面的文章。