Skip to main content

Installing GitHub Enterprise Server on AWS

To install GitHub Enterprise Server on Amazon Web Services (AWS), you must launch an Amazon Elastic Compute Cloud (EC2) instance and create and attach a separate Amazon Elastic Block Store (EBS) data volume.

Prerequisites

This guide assumes you are familiar with the following AWS concepts:

For an architectural overview, see the "AWS Architecture Diagram for Deploying GitHub Enterprise Server".

This guide recommends the principle of least privilege when setting up 您的 GitHub Enterprise Server 实例 on AWS. For more information, refer to the AWS Identity and Access Management (IAM) documentation.

Hardware considerations

Minimum requirements

We recommend different hardware configurations depending on the number of user licenses for 您的 GitHub Enterprise Server 实例. If you provision more resources than the minimum requirements, your instance will perform and scale better.

User licensesvCPUsMemoryAttached storageRoot storage
Trial, demo, or 10 light users432 GB150 GB200 GB
10 to 3,000848 GB300 GB200 GB
3,000 to 50001264 GB500 GB200 GB
5,000 to 80001696 GB750 GB200 GB
8,000 to 10,000+20160 GB1000 GB200 GB

If you plan to enable GitHub Actions for the users of your instance, more resources are required.

vCPU内存最大并行数
32128 GB1000 个作业
64256 GB1300 个作业
96384 GB2200 个作业

For more information about these requirements, see "Getting started with GitHub Actions for GitHub Enterprise Server."

有关为现有实例调整资源的更多信息,请参阅“增加存储容量”和“增加 CPU 或内存资源”。

Storage

We recommend a high-performance SSD with high input/output operations per second (IOPS) and low latency for GitHub Enterprise Server. Workloads are I/O intensive. If you use a bare metal hypervisor, we recommend directly attaching the disk or using a disk from a storage area network (SAN).

Your instance requires a persistent data disk separate from the root disk. For more information, see "System overview."

To configure GitHub Actions, you must provide external blob storage. For more information, see "Getting started with GitHub Actions for GitHub Enterprise Server."

The available space on the root filesystem will be 50% of the total disk size. You can resize your instance's root disk by building a new instance or using an existing instance. For more information, see "System overview" and "Increasing storage capacity."

CPU and memory

The CPU and memory resources that GitHub Enterprise Server requires depend on the levels of activity for users, automations, and integrations.

If you plan to enable GitHub Actions for the users of your GitHub Enterprise Server instance, you may need to provision additional CPU and memory resources for your instance. For more information, see "Getting started with GitHub Actions for GitHub Enterprise Server."

增加 CPU 资源时,我们建议为实例预配的每个 vCPU(最多 16 个 vCPU)增加至少 6.5 GB 的内存。 如果您使用的 vCPU 超过 16 个,则无需为每个 vCPU 添加 6.5 GB 内存,但应监控您的实例以确保其有足够的内存。

Warning: We recommend that users configure webhook events to notify external systems of activity on GitHub Enterprise Server. Automated checks for changes, or polling, will negatively impact the performance and scalability of your instance. For more information, see "About webhooks."

For more information about monitoring the capacity and performance of GitHub Enterprise Server, see "Monitoring your appliance."

You can increase your instance's CPU or memory resources. For more information, see "Increasing CPU or memory resources."

Determining the instance type

Before launching 您的 GitHub Enterprise Server 实例 on AWS, you'll need to determine the machine type that best fits the needs of your organization. To review the minimum requirements for GitHub Enterprise Server, see "Minimum requirements."

注意:您可以随时通过调整实例大小来扩展 CPU 或内存。 但由于调整 CPU 或内存的大小需要对用户停机,因此我们建议超配资源来应对扩展。

GitHub 建议对 GitHub Enterprise Server 使用内存优化的实例。 更多信息请参阅 Amazon EC2 网站上的 Amazon EC2 实例类型

Selecting the GitHub Enterprise Server AMI

You can select an Amazon Machine Image (AMI) for GitHub Enterprise Server using the GitHub Enterprise Server portal or the AWS CLI.

AMIs for GitHub Enterprise Server are available in the AWS GovCloud (US-East and US-West) region. This allows US customers with specific regulatory requirements to run GitHub Enterprise Server in a federally compliant cloud environment. For more information on AWS's compliance with federal and other standards, see AWS's GovCloud (US) page and AWS's compliance page.

Using the GitHub Enterprise Server portal to select an AMI

  1. 导航到要用于新实例的映像。

  2. Under "GitHub in the Cloud", select the "Select your platform" dropdown menu, and click Amazon Web Services.

  3. Select the "Select your AWS region" drop-down menu, and click your desired region.

  4. Take note of the AMI ID that is displayed.

Using the AWS CLI to select an AMI

  1. Using the AWS CLI, get a list of GitHub Enterprise Server images published by GitHub's AWS owner IDs (025577942450 for GovCloud, and 895557238572 for other regions). For more information, see "describe-images" in the AWS documentation.
    aws ec2 describe-images \
    --owners OWNER ID \
    --query 'sort_by(Images,&Name)[*].{Name:Name,ImageID:ImageId}' \
    --output=text
  2. Take note of the AMI ID for the latest GitHub Enterprise Server image.

Creating a security group

If you're setting up your AMI for the first time, you will need to create a security group and add a new security group rule for each port in the table below. For more information, see the AWS guide "Using Security Groups."

  1. Using the AWS CLI, create a new security group. For more information, see "create-security-group" in the AWS documentation.

    $ aws ec2 create-security-group --group-name SECURITY_GROUP_NAME --description "SECURITY GROUP DESCRIPTION"
  2. Take note of the security group ID (sg-xxxxxxxx) of your newly created security group.

  3. Create a security group rule for each of the ports in the table below. For more information, see "authorize-security-group-ingress" in the AWS documentation.

    $ aws ec2 authorize-security-group-ingress --group-id SECURITY_GROUP_ID --protocol PROTOCOL --port PORT_NUMBER --cidr SOURCE IP RANGE

    This table identifies what each port is used for.

    端口服务描述
    22SSHGit over SSH 访问。 支持克隆、获取和推送操作到公共/私有仓库。
    25SMTP支持加密 (STARTTLS) 的 SMTP。
    80HTTPWeb 应用程序访问。 当 SSL 启用时,所有请求都会重定向到 HTTPS 端口。
    122SSH实例 shell 访问。 默认 SSH 端口 (22) 专用于应用程序 git+ssh 网络流量。
    161/UDPSNMP为网络监视协议操作所需。
    443HTTPSWeb 应用程序和 Git over HTTPS 访问。
    1194/UDPVPN采用高可用性配置的安全复制网络隧道。
    8080HTTP基于纯文本 Web 的 管理控制台。 除非手动禁用 SSL,否则不需要。
    8443HTTPS基于安全 Web 的 管理控制台。 进行基本安装和配置时需要。
    9418Git简单的 Git 协议端口。 仅克隆和获取操作到公共仓库。 未加密的网络通信。 如果在实例上启用了私有模式,则仅当您也启用了匿名 Git 读取访问时才需要打开此端口。 更多信息请参阅“在企业中实施仓库管理策略”。

Creating the GitHub Enterprise Server instance

To create the instance, you'll need to launch an EC2 instance with your GitHub Enterprise Server AMI and attach an additional storage volume for your instance data. For more information, see "Hardware considerations."

Note: You can encrypt the data disk to gain an extra level of security and ensure that any data you write to your instance is protected. There is a slight performance impact when using encrypted disks. If you decide to encrypt your volume, we strongly recommend doing so before starting your instance for the first time. For more information, see the Amazon guide on EBS encryption.

Warning: If you decide to enable encryption after you've configured your instance, you will need to migrate your data to the encrypted volume, which will incur some downtime for your users.

Launching an EC2 instance

In the AWS CLI, launch an EC2 instance using your AMI and the security group you created. Attach a new block device to use as a storage volume for your instance data, and configure the size based on your user license count. For more information, see "run-instances" in the AWS documentation.

aws ec2 run-instances \
  --security-group-ids SECURITY_GROUP_ID \
  --instance-type INSTANCE_TYPE \
  --image-id AMI_ID \
  --block-device-mappings '[{"DeviceName":"/dev/xvdf","Ebs":{"VolumeSize":SIZE,"VolumeType":"TYPE"}}]' \
  --region REGION \
  --ebs-optimized

Allocating an Elastic IP and associating it with the instance

If this is a production instance, we strongly recommend allocating an Elastic IP (EIP) and associating it with the instance before proceeding to GitHub Enterprise Server configuration. Otherwise, the public IP address of the instance will not be retained after instance restarts. For more information, see "Allocating an Elastic IP Address" and "Associating an Elastic IP Address with a Running Instance" in the Amazon documentation.

Both primary and replica instances should be assigned separate EIPs in production High Availability configurations. For more information, see "Configuring GitHub Enterprise Server for High Availability."

Configuring the GitHub Enterprise Server instance

  1. 复制虚拟机的公共 DNS 名称,然后将其粘贴到 web 浏览器中。
  2. 在提示时上传许可文件并设置管理控制台密码。 更多信息请参阅“管理 GitHub Enterprise 的许可”。
  3. 管理控制台 中,配置并保存您所需的设置。 For more information, see "Configuring the GitHub Enterprise Server appliance."
  4. 实例将自动重启。
  5. 单击 Visit your instance(访问您的实例)

Further reading