Skip to content

G
gitlab-ce
Commit bda435f6 authored by Achilleas Pipinellis's avatar Achilleas Pipinellis
Browse files
Options

Merge branch '37056-improve-kubernetes-installation-docs' into 'master' 

Improve Kubernetes installation docs

Closes #37056

See merge request !13836
parents 52ac5cf2 ae458ad3
Show whitespace changes
Inline Side-by-side
Showing with 79 additions and 65 deletions
doc/install/kubernetes/gitlab_chart.md
View file @ bda435f6
# GitLab Helm Chart
> These Helm charts are in beta. GitLab is working on a [cloud-native](http://docs.gitlab.com/omnibus/package-information/cloud_native.html) set of [Charts](https://gitlab.com/charts/helm.gitlab.io) which will replace these.
> Officially supported cloud providers are Google Container Service and Azure Container Service.
> **Note:**
* GitLab is working on a [cloud native set of Charts](https://gitlab.com/charts/helm.gitlab.io/blob/master/README.md) which will replace these.
* Officially supported cloud providers are Google Container Service and Azure Container Service.
The `gitlab` Helm chart deploys GitLab into your Kubernetes cluster.
......@@ -22,9 +22,7 @@ This chart includes the following:
- [Persistent Volume](https://kubernetes.io/docs/concepts/storage/persistent-volumes/) provisioner support in the underlying infrastructure
- The ability to point a DNS entry or URL at your GitLab install
- The `kubectl` CLI installed locally and authenticated for the cluster
- The Helm Client installed locally
- The Helm Server (Tiller) already installed and running in the cluster, by running `helm init`
- The GitLab Helm Repo [added to your Helm Client](index.md#add-the-gitlab-helm-repository)
- The [Helm client](https://github.com/kubernetes/helm/blob/master/docs/quickstart.md) installed locally on your machine
## Configuring GitLab
......@@ -428,7 +426,7 @@ ingress:
## Installing GitLab using the Helm Chart
> You may see a temporary error message `SchedulerPredicates failed due to PersistentVolumeClaim is not bound` while storage provisions. Once the storage provisions, the pods will automatically restart. This may take a couple minutes depending on your cloud provider. If the error persists, please review the [prerequisites](#prerequisites) to ensure you have enough RAM, CPU, and storage.
Ensure the GitLab repo has been added and re-initialize Helm:
Add the GitLab Helm repository and initialize Helm:
```bash
helm repo add gitlab https://charts.gitlab.io
......
doc/install/kubernetes/gitlab_omnibus.md
View file @ bda435f6
# GitLab-Omnibus Helm Chart
> These Helm charts are in beta. GitLab is working on a [cloud-native](http://docs.gitlab.com/omnibus/package-information/cloud_native.html) set of [Charts](https://gitlab.com/charts/helm.gitlab.io) which will replace these.
> Officially supported cloud providers are Google Container Service and Azure Container Service.
> **Note:**
* This Helm chart is in beta, while [additional features](https://gitlab.com/charts/charts.gitlab.io/issues/68) are being worked on.
* GitLab is working on a [cloud native set of Charts](https://gitlab.com/charts/helm.gitlab.io/blob/master/README.md) which will eventually replace these.
* Officially supported cloud providers are Google Container Service and Azure Container Service.
This work is based partially on: https://github.com/lwolf/kubernetes-gitlab/. GitLab would like to thank Sergey Nuzhdin for his work.
......@@ -29,53 +30,51 @@ Terms:
## Prerequisites
- _At least_ 4 GB of RAM available on your cluster, in chunks of 1 GB. 41GB of storage and 2 CPU are also required.
- _At least_ 4 GB of RAM available on your cluster. 41GB of storage and 2 CPU are also required.
- Kubernetes 1.4+ with Beta APIs enabled
- [Persistent Volume](https://kubernetes.io/docs/concepts/storage/persistent-volumes/) provisioner support in the underlying infrastructure
- An [external IP address](#networking-prerequisites)
- A [wildcard DNS entry](#networking-prerequisites), which resolves to the external IP address
- The `kubectl` CLI installed locally and authenticated for the cluster
- The Helm Client installed locally
- The Helm Server (Tiller) already installed and running in the cluster, by running `helm init`
- The GitLab Helm Repo [added to your Helm Client](index.md#add-the-gitlab-helm-repository)
- The [Helm client](https://github.com/kubernetes/helm/blob/master/docs/quickstart.md) installed locally on your machine
### Networking Prerequisites
This chart configures a GitLab server and Kubernetes cluster which can support dynamic [Review Apps](https://docs.gitlab.com/ee/ci/review_apps/index.html), as well as services like the integrated [Container Registry](https://docs.gitlab.com/ee/user/project/container_registry.html) and [Mattermost](https://docs.gitlab.com/omnibus/gitlab-mattermost/).
To support the GitLab services and dynamic environments, a wildcard DNS entry is required which resolves to the external Load Balancer IP.
To support the GitLab services and dynamic environments, a wildcard DNS entry is required which resolves to the [Load Balancer](#load-balancer-ip) or [External IP](#external-ip). Configuration of the DNS entry will depend upon the DNS service being used.
#### External IP (Recommended)
To provision an external IP on GCP and Azure, simply request a new address from the Networking section. Ensure that the region matches the region your container cluster is created in. Note, it is important that the IP is not assigned at this point in time. It will be automatically assigned once the Helm chart is installed, and assigned to the Load Balancer.
Now that an external IP address has been allocated, ensure that the wildcard DNS entry you would like to use resolves to this IP. Please consult the documentation for your DNS service for more information on creating DNS records.
Finally, set the `baseIP` setting to this IP address when [deploying GitLab](#configuring-and-installing-gitlab).
#### Load Balancer IP
If you do not specify a `baseIP`, an ephemeral IP will be assigned to the Load Balancer or Ingress. You can retrieve this IP by running the following command *after* deploying GitLab:
`kubectl get svc -w --namespace nginx-ingress nginx`
The IP address will be displayed in the `EXTERNAL-IP` field, and should be used to configure the Wildcard DNS entry. For more information on creating a wildcard DNS entry, consult the documentation for the DNS server you are using.
For production deployments of GitLab, we strongly recommend using an [External IP](#external-ip).
## Configuring and Installing GitLab
For most installations, only two parameters are required:
- `baseIP`: the desired [external IP address](#networking-prerequisites)
- `baseDomain`: the [base domain](#networking-prerequisites) with the wildcard host entry resolving to the `baseIP`. For example, `mycompany.io`.
- `legoEmail`: Email address to use when requesting new SSL certificates from Let's Encrypt
Other common configuration options:
- `baseIP`: the desired [external IP address](#networking-prerequisites)
- `gitlab`: Choose the [desired edition](https://about.gitlab.com/products), either `ee` or `ce`. `ce` is the default.
- `gitlabEELicense`: For Enterprise Edition, the [license](https://docs.gitlab.com/ee/user/admin_area/license.html) can be installed directly via the Chart
- `provider`: Optimizes the deployment for a cloud provider. The default is `gke` for GCP, with `acs` also supported for Azure.
- `legoEmail`: Email address to use when requesting new SSL certificates from Let's Encrypt
For additional configuration options, consult the [values.yaml](https://gitlab.com/charts/charts.gitlab.io/blob/master/charts/gitlab-omnibus/values.yaml).
These settings can either be passed directly on the command line:
```bash
helm install --name gitlab --set baseDomain=gitlab.io,baseIP=1.1.1.1,gitlab=ee,gitlabEELicense=$LICENSE,legoEmail=email@gitlab.com gitlab/gitlab-omnibus
```
or within a YAML file:
```bash
helm install --name gitlab -f values.yaml gitlab/gitlab-omnibus
```
> **Note:**
If you are using a machine type with support for less than 4 attached disks, like an Azure trial, you should disable dedicated storage for [Postgres and Redis](#persistent-storage).
### Choosing a different GitLab release version
The version of GitLab installed is based on the `gitlab` setting (see [section](#choosing-gitlab-edition) above), and
......@@ -95,6 +94,8 @@ There is no guarantee that other release versions of GitLab, other than what are
used by default in the chart, will be supported by a chart install.
### Persistent storage
> **Note:**
If you are using a machine type with support for less than 4 attached disks, like an Azure trial, you should disable dedicated storage for [Postgres and Redis](#persistent-storage).
By default, persistent storage is enabled for GitLab and the charts it depends
on (Redis and PostgreSQL).
......@@ -124,9 +125,10 @@ Ingress routing and SSL are automatically configured within this Chart. An NGINX
Let's Encrypt limits a single TLD to five certificate requests within a single week. This means that common DNS wildcard services like [xip.io](http://xip.io) and [nip.io](http://nip.io) are unlikely to work.
## Installing GitLab using the Helm Chart
> You may see a temporary error message `SchedulerPredicates failed due to PersistentVolumeClaim is not bound` while storage provisions. Once the storage provisions, the pods will automatically restart. This may take a couple minutes depending on your cloud provider. If the error persists, please review the [prerequisites](#prerequisites) to ensure you have enough RAM, CPU, and storage.
> **Note:**
You may see a temporary error message `SchedulerPredicates failed due to PersistentVolumeClaim is not bound` while storage provisions. Once the storage provisions, the pods will automatically start. This may take a couple minutes depending on your cloud provider. If the error persists, please review the [prerequisites](#prerequisites) to ensure you have enough RAM, CPU, and storage.
Ensure the GitLab repo has been added and re-initialize Helm:
Add the GitLab Helm repository and initialize Helm:
```bash
helm repo add gitlab https://charts.gitlab.io
......
doc/install/kubernetes/gitlab_runner_chart.md
View file @ bda435f6
# GitLab Runner Helm Chart
> These Helm charts are in beta. GitLab is working on a [cloud-native](http://docs.gitlab.com/omnibus/package-information/cloud_native.html) set of [Charts](https://gitlab.com/charts/helm.gitlab.io) which will replace these.
> Officially supported cloud providers are Google Container Service and Azure Container Service.
> **Note:**
Officially supported cloud providers are Google Container Service and Azure Container Service.
The `gitlab-runner` Helm chart deploys a GitLab Runner instance into your
Kubernetes cluster.
......@@ -17,9 +16,7 @@ This chart configures the Runner to:
- Your GitLab Server's API is reachable from the cluster
- Kubernetes 1.4+ with Beta APIs enabled
- The `kubectl` CLI installed locally and authenticated for the cluster
- The Helm Client installed locally
- The Helm Server (Tiller) already installed and running in the cluster, by running `helm init`
- The GitLab Helm Repo added to your Helm Client. See [Adding GitLab Helm Repo](index.md#add-the-gitlab-helm-repository)
- The [Helm client](https://github.com/kubernetes/helm/blob/master/docs/quickstart.md) installed locally on your machine
## Configuring GitLab Runner using the Helm Chart
......@@ -36,6 +33,8 @@ In order for GitLab Runner to function, your config file **must** specify the fo
- `runnerRegistrationToken` - The Registration Token for adding new Runners to the GitLab Server. This must be
retrieved from your GitLab Instance. See the [GitLab Runner Documentation](../../ci/runners/README.md#creating-and-registering-a-runner) for more information.
Unless you need to specify additional configuration, you are [ready to install](#installing-gitlab-runner-using-the-helm-chart).
### Other configuration
The rest of the configuration is [documented in the `values.yaml`](https://gitlab.com/charts/charts.gitlab.io/blob/master/charts/gitlab-runner/values.yaml) in the chart repository.
......@@ -115,6 +114,17 @@ runners:
```
### Controlling maximum Runner concurrency
A single GitLab Runner deployed on Kubernetes is able to execute multiple jobs in parallel by automatically starting additional Runner pods. The [`concurrent` setting](https://docs.gitlab.com/runner/configuration/advanced-configuration.html#the-global-section) controls the maximum number of pods allowed at a single time, and defaults to `10`.
```yaml
## Configure the maximum number of concurrent jobs
## ref: https://docs.gitlab.com/runner/configuration/advanced-configuration.html#the-global-section
##
concurrent: 10
```
### Running Docker-in-Docker containers with GitLab Runners
See [Running Privileged Containers for the Runners](#running-privileged-containers-for-the-runners) for how to enable it,
......@@ -190,7 +200,7 @@ certsSecretName: <SECRET NAME>
## Installing GitLab Runner using the Helm Chart
Ensure the GitLab repo has been added and re-initialize Helm:
Add the GitLab Helm repository and initialize Helm:
```bash
helm repo add gitlab https://charts.gitlab.io
......
doc/install/kubernetes/index.md
View file @ bda435f6
# Installing GitLab on Kubernetes
> These Helm charts are in beta. GitLab is working on a [cloud-native](http://docs.gitlab.com/omnibus/package-information/cloud_native.html) set of [Charts](https://gitlab.com/charts/helm.gitlab.io) which will replace these.
> Officially supported cloud providers are Google Container Service and Azure Container Service.
The easiest method to deploy GitLab in [Kubernetes](https://kubernetes.io/) is
to take advantage of the official GitLab Helm charts. [Helm] is a package
to take advantage of GitLab's Helm charts. [Helm] is a package
management tool for Kubernetes, allowing apps to be easily managed via their
Charts. A [Chart] is a detailed description of the application including how it
should be deployed, upgraded, and configured.
The GitLab Helm repository is located at https://charts.gitlab.io.
You can report any issues related to GitLab's Helm Charts at
GitLab provides [official Helm Charts](#official-gitlab-helm-charts-recommended) which is the recommended way to run GitLab with Kubernetes.
There are also two other sets of charts:
* Our [upcoming cloud native Charts](#upcoming-cloud-native-helm-charts), which are in development but will eventually replace the current official charts.
* [Community contributed charts](#community-contributed-helm-charts). These charts should be considered deprecated, in favor of the official charts.
## Official GitLab Helm Charts (Recommended)
These charts utilize our [GitLab Omnibus Docker images](https://docs.gitlab.com/omnibus/docker/README.html). You can report any issues and feedback related to these charts at
https://gitlab.com/charts/charts.gitlab.io/issues.
Contributions and improvements are also very welcome.
## Prerequisites
### Deploying GitLab on Kubernetes (Recommended)
> *Note*: This chart will eventually be replaced by the [cloud native charts](#upcoming-cloud-native-helm-charts), which are presently in development.
The best way to deploy GitLab on Kubernetes is to use the [gitlab-omnibus](gitlab_omnibus.md) chart. It includes everything needed to run GitLab, including: a [Runner](https://docs.gitlab.com/runner/), [Container Registry](https://docs.gitlab.com/ee/user/project/container_registry.html#gitlab-container-registry), [automatic SSL](https://github.com/kubernetes/charts/tree/master/stable/kube-lego), and an [Ingress](https://github.com/kubernetes/ingress/tree/master/controllers/nginx). This chart is in beta while [additional features](https://gitlab.com/charts/charts.gitlab.io/issues/68) are being completed.
To use the charts, the Helm tool must be installed and initialized. The best
place to start is by reviewing the [Helm Quick Start Guide][helm-quick].
### Deploying just the GitLab Runner
## Add the GitLab Helm repository
To deploy just the GitLab Runner, utilize the [gitlab-runner](gitlab_runner_chart.md) chart. It offers a quick way to configure and deploy the Runner on Kubernetes, regardless of where your GitLab server may be running.
Once Helm has been installed, the GitLab chart repository must be added:
### Advanced deployment of GitLab (Not recommended)
> *Note*: This chart will eventually be replaced by the [cloud native charts](#upcoming-cloud-native-helm-charts), which are presently in development.
```bash
helm repo add gitlab https://charts.gitlab.io
```
If advanced configuration of GitLab is required, the beta [gitlab](gitlab_chart.md) chart can be used which deploys the GitLab service along with optional Postgres and Redis. It offers extensive configuration, but requires deep knowledge of Kubernetes and Helm to use.
After adding the repository, Helm must be re-initialized:
## Upcoming Cloud Native Helm Charts
```bash
helm init
```
GitLab is working towards a building a [cloud native deployment method](https://gitlab.com/charts/helm.gitlab.io/blob/master/README.md). A key part of this effort is to isolate each service into it's [own Docker container and Helm chart](https://gitlab.com/gitlab-org/omnibus-gitlab/issues/2420), rather than utilizing the all-in-one container image of the [current charts](#official-gitlab-helm-charts-recommended).
## Using the GitLab Helm Charts
By offering individual containers and charts, we will be able to provide a number of benefits:
* Easier horizontal scaling of each service
* Smaller more efficient images
* Potential for rolling updates and canaries within a service
* and plenty more.
GitLab makes available three Helm Charts.
This is a large project and will be worked on over the span of multiple releases. For the most up to date status and release information, please see our [tracking issue](https://gitlab.com/gitlab-org/omnibus-gitlab/issues/2420).
- [gitlab-omnibus](gitlab_omnibus.md): **Recommended** and the easiest way to get started. Includes everything needed to run GitLab, including: a [Runner](https://docs.gitlab.com/runner/), [Container Registry](https://docs.gitlab.com/ee/user/project/container_registry.html#gitlab-container-registry), [automatic SSL](https://github.com/kubernetes/charts/tree/master/stable/kube-lego), and an [Ingress](https://github.com/kubernetes/ingress/tree/master/controllers/nginx).
- [gitlab](gitlab_chart.md): Just the GitLab service, with optional Postgres and Redis.
- [gitlab-runner](gitlab_runner_chart.md): GitLab Runner, to process CI jobs.
## Community Contributed Helm Charts
We are also working on a new set of [cloud native Charts](https://gitlab.com/charts/helm.gitlab.io) which will eventually replace these.
The community has also [contributed GitLab charts](https://github.com/kubernetes/charts/tree/master/stable/gitlab-ce) to the [Helm Stable Repository](https://github.com/kubernetes/charts#repository-structure). These charts should be considered [deprecated](https://github.com/kubernetes/charts/issues/1138) in favor of the [official Charts](#official-gitlab-helm-charts-recommended).
[chart]: https://github.com/kubernetes/charts
[helm-quick]: https://github.com/kubernetes/helm/blob/master/docs/quickstart.md
[helm]: https://github.com/kubernetes/helm/blob/master/README.md
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7