Skip to main content

Run Windows container with Hyper-V isolation mode in Kubernetes

· 3 min read

Windows Container 有两种隔离运行模式 Hyper-V 和 Process, 参见：Isolation Modes

两种模式下的 host 的 OS 版本与 containter 的 OS 版本存在兼容性又不相同，参见：Windows container version compatibility

很明显 Hyper-V 模式的兼容性要比 Process 模式要好，向下兼容，也就是高版本的 host OS 可以运行低版本的 container OS, 反之不行；

而 Process 模式下 Windows Server 中则要求 host OS 与 container OS 的版本完全相同, Windows 10 中则不支持 Process 模式.

某一天，我想在 Kubernetes Windows 节点中以 Hyper-V 模式运行 Container, 于是乎发现1.17 的文档中写道：

Note: In this document, when we talk about Windows containers we mean Windows containers with process isolation. Windows containers with Hyper-V isolation is planned for a future release.

不甘心又 google 了一下，发现：

1. 有人提了 bug, 已经被修复了: https://github.com/kubernetes/kubernetes/issues/58750 2. 代码也 merge 了: https://github.com/kubernetes/kubernetes/pull/58751 3. 有人在测试过程中遇到问题，也解决了: https://github.com/kubernetes/kubernetes/issues/62812

但我测试的过程中却提示：

Error response from daemon: CreateComputeSystem test: The container operating system does not match the host operating system.

我的环境：

Kubernetes Ver: 1.14.8

Kubernetes Node OS Ver: Windows Server Datacenter 10.0.17763.504, 属于 1809 的版本

Container Base Image: windowsservercore-1709

Deployment yaml:

apiVersion: <span style="color: #ff0000;">apps/v1beta2</span>
kind: Deployment
metadata:
  labels:
    app: test
  name: test
  namespace: default
spec:
  replicas: 1
  selector:
    matchLabels:
      app: test
  template:
    metadata:
      annotations:
        <span style="color: #ff0000;">experimental.windows.kubernetes.io/isolation-type: hyperv</span>
      labels:
        app: test
...

然后对比了下 github 别人试成功的 deployment yaml, 发现人家用的是 apps/v1

apiVersion: <span style="color: #ff0000;">apps/v1</span>
kind: Deployment
metadata:
  name: whoami
  labels:
    app: whoami
spec:
  ...

目前在 k8s 中启用 hyperv isolation 的三个条件：

  1. kubelet 启用参数: --feature-gates=HyperVContainer=true
  2. Pod/Deployment apiVersion: apps/v1
  3. spec.template.metadata.annotations[].experimental.windows.kubernetes.io/isolation-type:hyperv

参见: https://kubernetes.io/docs/setup/production-environment/windows/intro-windows-in-kubernetes/#hyper-v-isolation

目前我的云提供商给的 kubernetes 1.14.8 又不支持 apps/v1 ...

于是乎，我要么等提供商升级 kubernetes，要么自己升级 container OS 跟 kubernetes node OS 一样...

Enable LFS (Large File Storage) in GitLab

· 2 min read

Refer to official docs: https://docs.gitlab.com/ce/administration/lfs/manage_large_binaries_with_git_lfs.html

Requirements​

  • Git LFS is supported in GitLab starting with version 8.2
  • Git LFS must be enabled under project settings
  • Git LFS client version 1.0.1 and up

1. Enable LFS on GitLab Server​

a). Configuration for Omnibus installations In​

/etc/gitlab/gitlab.rb:

# Change to true to enable lfs - enabled by default if not defined
gitlab_rails['lfs_enabled]=false

# Optionally, change the storage path location. Defaults to
# `#{gitlab_rails['shared_path']}/lfs-objects`. Which evaluates to
# `/var/opt/gitlab/gitlab-rails/shared/lfs-objects` by default.
gitlab_rails['lfs_storage_path']"/mnt/storage/lfs-objects"

b). Configuration for installations from source In​

config/gitlab.yml:

# Change to true to enable lfs
  lfs:
    enabled: false
    storage_path:/mnt/storage/lfs-objects

2. Submit LFS object to Git project Initialize LFS support in the project, and mark the file extensions for tracking as LFS object:​

$ git clone [email protected]:group/project.git
$ git lfs <span class="nb">install</span> <span class="c"># initialize the Git LFS project
</span>$ git lfs track <span class="s2">"*.out" "*.so" "*.dll" "*.meta" "*.dat" "/bin/**"</span> <span class="c"># select the file extensions that you want to treat as large files
</span>

Details for lfs track partten please refer to : https://git-scm.com/docs/gitignore#_pattern_format

Make sure that .gitattributes is tracked by Git. Otherwise Git LFS will not be working properly for people cloning the project:

$ git add .gitattributes
$ git commit -m "Submit lfs settings"
$ git push -u origin master

An LFS icon is shown on files tracked by Git LFS to denote if a file is stored as a blob or as an LFS pointer. Example:

3. Pull LFS object from Git project Other user need to install git-lfs package and execute git lfs pull after clone the prjoect:​

$ git clone [email protected]:group/project.git
$ git lfs pull

General "git pull" will only download index data of LFS object:

# cat appframed.out
version https://git-lfs.github.com/spec/v1
oid sha256:e8eef08ead6e6f3ad021ba798b35d2da59f804e9dbe8aace7ed4b66d3ede1054
size 6766944

In jenkins, it needs to enable "

Git LFS pull after checkout" addtional behavior for Git repo.

ClustrMaps