Advanced configuration

GitLab Runner configuration uses the TOML format.

The file to be edited can be found in:

  1. /etc/gitlab-runner/config.toml on *nix systems when gitlab-runner is executed as root (this is also path for service configuration)
  2. ~/.gitlab-runner/config.toml on *nix systems when gitlab-runner is executed as non-root
  3. ./config.toml on other systems

The global section

This defines global settings of GitLab Runner.

Setting Description
concurrent limits how many jobs globally can be run concurrently. The most upper limit of jobs using all defined runners. 0 does not mean unlimited
log_level Log level (options: debug, info, warn, error, fatal, panic). Note that this setting has lower priority than level set by command line argument --debug, -l or --log-level
log_format Log format (options: runner, text, json). Note that this setting has lower priority than format set by command line argument --log-format
check_interval defines the interval length, in seconds, between new jobs check. The default value is 3; if set to 0 or lower, the default value will be used.
sentry_dsn enable tracking of all system level errors to sentry
listen_address address (<host>:<port>) on which the Prometheus metrics HTTP server should be listening

Example:

concurrent = 4
log_level = "warning"

How check_interval works

If there is more than one [[runners]] section in config.toml (let’s call them workers), the interval between requests to GitLab are more frequent than one could expect. GitLab Runner contains a loop that constantly schedules a request that should be made for a worker against the GitLab instance it’s configured for.

GitLab Runner tries to ensure that subsequent requests for one worker will be done in the specified interval, so the value of check_interval is divided by the number of the [[runners]] sections. The loop will next iterate over all sections, schedule a request for each of them, and will sleep for the calculated amount of time. Things get interesting when the workers are tied to a different GitLab instance. Consider the following example.

If one would set check_interval = 10, and there were 2 workers in total (runner-1 and runner-2), a subsequent request would be made each 10 seconds. The loop would look like:

  1. Get check_interval value (10s)
  2. Get list of workers (runner-1, runner-2)
  3. Calculate the sleep interval (10s / 2 = 5s)
  4. Start an infinite loop:
  5. Request a job for runner-1
  6. Sleep for 5s
  7. Request a job for runner-2
  8. Sleep for 5s
  9. Repeat

So, a request from the Runner’s process is made each 5s. If runner-1 and runner-2 are connected to the same GitLab instance, it means that the request to this GitLab instance will receive a new request from this Runner also each 5s. But as you can see, between the first request for runner-1 and second request for runner-1 there are two sleeps taking 5s, so finally it’s ~10s between subsequent requests for runner-1. The same goes for runner-2. If you define more workers, the sleep interval will be smaller, but a request for a worker will be repeated after all requests for the other workers + their sleeps are called.

The [session_server] section

The section [session_server] is a system runner level configuration, so it should be specified at the root level, not per executor i.e. it should be outside [[runners]] section. The session server allows the user to interact with jobs that the Runner is responsible for. A good example of this is the interactive web terminal.

Both listen_address and advertise_address should be provided in the form of host:port, where host may be an IP address (e.g., 127.0.0.1:8093) or a domain (e.g., my-runner.example.com:8093). The Runner will create a TLS certificate automatically to have a secure connection.

If you want to disable the session server, just delete the [session_server] section and terminal support will be disabled.

Setting Description
listen_address An internal URL to be used for the session server.
advertise_address The URL that the Runner will expose to GitLab to be used to access the session server. Fallbacks to listen_address if not defined.
session_timeout How long in seconds the session can stay active after the job completes (which will block the job from finishing), defaults to 1800 (30 minutes).

Example:

[session_server]
  listen_address = "0.0.0.0:8093"
  advertise_address = "runner-host-name.tld:8093"
  session_timeout = 1800

The [[runners]] section

This defines one runner entry.

Setting Description
name The Runner’s description, just informatory
url GitLab URL
token The Runner’s special token (not to be confused with the registration token)
tls-ca-file File containing the certificates to verify the peer when using HTTPS
tls-cert-file File containing the certificate to authenticate with the peer when using HTTPS
tls-key-file File containing the private key to authenticate with the peer when using HTTPS
limit Limit how many jobs can be handled concurrently by this token. 0 (default) simply means don’t limit
executor Select how a project should be built, see next section
shell The name of shell to generate the script (default value is platform dependent)
builds_dir Directory where builds will be stored in context of selected executor (Locally, Docker, SSH)
cache_dir Directory where build caches will be stored in context of selected executor (locally, Docker, SSH). If the docker executor is used, this directory needs to be included in its volumes parameter.
environment Append or overwrite environment variables
request_concurrency Limit number of concurrent requests for new jobs from GitLab (default 1)
output_limit Set maximum build log size in kilobytes, by default set to 4096 (4MB)
pre_clone_script Commands to be executed on the Runner before cloning the Git repository. this can be used to adjust the Git client configuration first, for example. To insert multiple commands, use a (triple-quoted) multi-line string or “\n” character.
pre_build_script Commands to be executed on the Runner after cloning the Git repository, but before executing the build. To insert multiple commands, use a (triple-quoted) multi-line string or “\n” character.
post_build_script Commands to be executed on the Runner just after executing the build, but before executing after_script. To insert multiple commands, use a (triple-quoted) multi-line string or “\n” character.
clone_url Overwrite the URL for the GitLab instance. Used if the Runner can’t connect to GitLab on the URL GitLab exposes itself.

Example:

[[runners]]
  name = "ruby-2.1-docker"
  url = "https://CI/"
  token = "TOKEN"
  limit = 0
  executor = "docker"
  builds_dir = ""
  shell = ""
  environment = ["ENV=value", "LC_ALL=en_US.UTF-8"]
  clone_url = "http://gitlab.example.local"

How clone_url works

In cases where the GitLab instance is exposed to an URL which can’t be used by the runner, a clone_url can be configured. For example; GitLab is exposed to https://gitlab.example.com, but the runner can’t reach that because of a firewall setup. If the runner can reach the node on 192.168.1.23, the clone_url should be set to "http://192.168.1.23.

Only if the clone_url is set, the runner will construct a clone URL in the form of http://gitlab-ci-token:s3cr3tt0k3n@192.168.1.23/namespace/project.git.

The EXECUTORS

There are a couple of available executors currently.

Executor Description
shell run build locally, default
docker run build using Docker container - this requires the presence of [runners.docker] and Docker Engine installed on the system that the Runner runs
docker-ssh run build using Docker container, but connect to it with SSH - this requires the presence of [runners.docker] , [runners.ssh] and Docker Engine installed on the system that the Runner runs. Note: This will run the docker container on the local machine, it just changes how the commands are run inside that container. If you want to run docker commands on an external machine, then you should change the host parameter in the runners.docker section.
ssh run build remotely with SSH - this requires the presence of [runners.ssh]
parallels run build using Parallels VM, but connect to it with SSH - this requires the presence of [runners.parallels] and [runners.ssh]
virtualbox run build using VirtualBox VM, but connect to it with SSH - this requires the presence of [runners.virtualbox] and [runners.ssh]
docker+machine like docker, but uses auto-scaled Docker machines - this requires the presence of [runners.docker] and [runners.machine]
docker-ssh+machine like docker-ssh, but uses auto-scaled Docker machines - this requires the presence of [runners.docker] and [runners.machine]
kubernetes run build using Kubernetes Pods - this requires the presence of [runners.kubernetes]

The SHELLS

There are a couple of available shells that can be run on different platforms.

Shell Description
bash generate Bash (Bourne-shell) script. All commands executed in Bash context (default for all Unix systems)
sh generate Sh (Bourne-shell) script. All commands executed in Sh context (fallback for bash for all Unix systems)
cmd generate Windows Batch script. All commands are executed in Batch context (default for Windows)
powershell generate Windows PowerShell script. All commands are executed in PowerShell context

The [runners.docker] section

This defines the Docker Container parameters.

Parameter Description
host Specify custom Docker endpoint, by default DOCKER_HOST environment is used or unix:///var/run/docker.sock
hostname Specify custom hostname for Docker container
runtime Specify a runtime for Docker container
tls_cert_path When set it will use ca.pem, cert.pem and key.pem from that folder to make secure TLS connection to Docker (useful in boot2docker)
image Use this image to run builds
memory String value containing the memory limit
memory_swap String value containing the total memory limit
memory_reservation String value containing the memory soft limit
oom_kill_disable Do not kill processes in a container if an out-of-memory (OOM) error occurs
cpuset_cpus String value containing the cgroups CpusetCpus to use
cpus Number of CPUs (available in docker 1.13 or later)
dns A list of DNS servers for the container to use
dns_search A list of DNS search domains
privileged Make container run in Privileged mode (insecure)
disable_entrypoint_overwrite Disable the image entrypoint overwriting
userns_mode Sets the usernamespace mode for the container when usernamespace remapping option is enabled. (available in docker 1.10 or later)
cap_add Add additional Linux capabilities to the container
cap_drop Drop additional Linux capabilities from the container
security_opt Set security options (–security-opt in docker run), takes a list of ‘:’ separated key/values
devices Share additional host devices with the container
cache_dir Specify where Docker caches should be stored (this can be absolute or relative to current working directory). See disable_cache for more information.
disable_cache The Docker executor has 2 levels of caching: a global one (like any other executor) and a local cache based on Docker volumes. This configuration flag acts only on the local one which disables the use of automatically created (not mapped to a host directory) cache volumes. In other words, it only prevents creating a container that holds temporary files of builds, it does not disable the cache if the Runner is configured in distributed cache mode.
network_mode Add container to a custom network
wait_for_services_timeout Specify how long to wait for docker services, set to 0 to disable, default: 30
volumes Specify additional volumes that should be mounted (same syntax as Docker’s -v flag)
extra_hosts Specify hosts that should be defined in container environment
shm_size Specify shared memory size for images (in bytes)
volumes_from Specify a list of volumes to inherit from another container in the form \<container name\>[:\<ro|rw\>]
volume_driver Specify the volume driver to use for the container
links Specify containers which should be linked with building container
services Specify additional services that should be run with build. Please visit Docker Registry for list of available applications. Each service will be run in separate container and linked to the build.
allowed_images Specify wildcard list of images that can be specified in .gitlab-ci.yml. If not present all images are allowed (equivalent to ["*/*:*"])
allowed_services Specify wildcard list of services that can be specified in .gitlab-ci.yml. If not present all images are allowed (equivalent to ["*/*:*"])
pull_policy Specify the image pull policy: never, if-not-present or always (default); read more in the pull policies documentation
sysctls specify the sysctl options
helper_image [ADVANCED] Override the default helper image used to clone repos and upload artifacts. Read the helper image section for more details

Example:

[runners.docker]
  host = ""
  hostname = ""
  tls_cert_path = "/Users/ayufan/.boot2docker/certs"
  image = "ruby:2.1"
  memory = "128m"
  memory_swap = "256m"
  memory_reservation = "64m"
  oom_kill_disable = false
  cpuset_cpus = "0,1"
  dns = ["8.8.8.8"]
  dns_search = [""]
  privileged = false
  userns_mode = "host"
  cap_add = ["NET_ADMIN"]
  cap_drop = ["DAC_OVERRIDE"]
  devices = ["/dev/net/tun"]
  disable_cache = false
  wait_for_services_timeout = 30
  cache_dir = ""
  volumes = ["/data", "/home/project/cache"]
  extra_hosts = ["other-host:127.0.0.1"]
  shm_size = 300000
  volumes_from = ["storage_container:ro"]
  links = ["mysql_container:mysql"]
  services = ["mysql", "redis:2.8", "postgres:9"]
  allowed_images = ["ruby:*", "python:*", "php:*"]
  allowed_services = ["postgres:9.4", "postgres:latest"]
  [runners.docker.sysctls]
    "net.ipv4.ip_forward" = "1"

Volumes in the [runners.docker] section

You can find the complete guide of Docker volume usage here.

Let’s use some examples to explain how it work (assuming you have a working runner).

Example 1: adding a data volume

A data volume is a specially-designated directory within one or more containers that bypasses the Union File System. Data volumes are designed to persist data, independent of the container’s life cycle.

[runners.docker]
  host = ""
  hostname = ""
  tls_cert_path = "/Users/ayufan/.boot2docker/certs"
  image = "ruby:2.1"
  privileged = false
  disable_cache = true
  volumes = ["/path/to/volume/in/container"]

This will create a new volume inside the container at /path/to/volume/in/container.

Example 2: mount a host directory as a data volume

In addition to creating a volume using you can also mount a directory from your Docker daemon’s host into a container. It’s useful when you want to store builds outside the container.

[runners.docker]
  host = ""
  hostname = ""
  tls_cert_path = "/Users/ayufan/.boot2docker/certs"
  image = "ruby:2.1"
  privileged = false
  disable_cache = true
  volumes = ["/path/to/bind/from/host:/path/to/bind/in/container:rw"]

This will use /path/to/bind/from/host of the CI host inside the container at /path/to/bind/in/container.

Using a private container registry

Notes: - This feature requires GitLab Runner 1.8 or higher - For GitLab Runner versions >= 0.6, <1.8 there was a partial support for using private registries, which required manual configuration of credentials on runner’s host. We recommend to upgrade your Runner to at least version 1.8 if you want to use private registries. - Using private registries with the if-not-present pull policy may introduce security implications. To fully understand how pull policies work, read the pull policies documentation.

If you want to use private registries as a source of images for your builds, you can set the authorization configuration in the DOCKER_AUTH_CONFIG variable. It can be set in both GitLab Variables section of a project and in the config.toml file.

For a detailed example, visit the Using Docker images documentation.

The steps performed by the Runner can be summed up to:

  1. The registry name is found from the image name.
  2. If the value is not empty, the executor will search for the authentication configuration for this registry.
  3. Finally, if an authentication corresponding to the specified registry is found, subsequent pulls will make use of it.

Now that the Runner is set up to authenticate against your private registry, learn how to configure .gitlab-ci.yml in order to use that registry.

Support for GitLab integrated registry

Note: To work automatically with private/protected images from GitLab integrated registry it needs at least GitLab CE/EE 8.14 and GitLab Runner 1.8.

Starting with GitLab CE/EE 8.14, GitLab will send credentials for its integrated registry along with the build data. These credentials will be automatically added to registries authorization parameters list.

After this authorization against the registry will be proceed like for configuration added with DOCKER_AUTH_CONFIG variable.

Thanks to this, in your builds you can use any image from you GitLab integrated registry, even if the image is private/protected. To fully understand for which images the builds will have access, read the New CI build permissions model documentation.

Precedence of Docker authorization resolving

As described above, GitLab Runner can authorize Docker against a registry by using credentials sent in different way. To find a proper registry, the following precedence is taken into account:

  1. Credentials configured with DOCKER_AUTH_CONFIG.
  2. Credentials configured locally on Runner’s host with ~/.docker/config.json or ~/.dockercfg files (e.g., by running docker login on the host).
  3. Credentials sent by default with job’s payload (e.g., credentials for integrated registry described above).

The first found credentials for the registry will be used. So for example, if you add some credentials for the integrated registry with the DOCKER_AUTH_CONFIG variable, then the default credentials will be overridden.

Restrict allowed_images to private registry

For certain setups you will restrict access of the build jobs to docker images which comes from your private docker registry. In that case set

[runners.docker]
  ...
  allowed_images = ["my.registry.tld:5000/*:*"]

The [runners.parallels] section

This defines the Parallels parameters.

Parameter Description
base_name name of Parallels VM which will be cloned
template_name custom name of Parallels VM linked template (optional)
disable_snapshots if disabled the VMs will be destroyed after build

Example:

[runners.parallels]
  base_name = "my-parallels-image"
  template_name = ""
  disable_snapshots = false

The [runners.virtualbox] section

This defines the VirtualBox parameters. This executor relies on vboxmanage as executable to control VirtualBox machines so you have to adjust your PATH environment variable on Windows hosts: PATH=%PATH%;C:\Program Files\Oracle\VirtualBox.

Parameter Explanation
base_name name of VirtualBox VM which will be cloned
base_snapshot name or UUID of a specific snapshot of the VM from which to create a linked clone. If this is empty or omitted, the current snapshot will be used. If there is no current snapshot, one will be created unless disable_snapshots is true, in which case a full clone of the base VM will be made.
disable_snapshots if disabled the VMs will be destroyed after build

Example:

[runners.virtualbox]
  base_name = "my-virtualbox-image"
  base_snapshot = "my-image-snapshot"
  disable_snapshots = false

The [runners.ssh] section

This defines the SSH connection parameters.

Parameter Description
host where to connect (overridden when using docker-ssh)
port specify port, default: 22
user specify user
password specify password
identity_file specify file path to SSH private key (id_rsa, id_dsa or id_edcsa). The file needs to be stored unencrypted

Example:

[runners.ssh]
  host = "my-production-server"
  port = "22"
  user = "root"
  password = "production-server-password"
  identity_file = ""

The [runners.machine] section

Note: Added in GitLab Runner v1.1.0.

This defines the Docker Machine based autoscaling feature. More details can be found in the separate runners autoscale documentation.

Parameter Description
IdleCount Number of machines, that need to be created and waiting in Idle state.
IdleTime Time (in seconds) for machine to be in Idle state before it is removed.
OffPeakPeriods Time periods when the scheduler is in the OffPeak mode. An array of cron-style patterns (described below).
OffPeakTimezone Time zone for the times given in OffPeakPeriods. A timezone string like Europe/Berlin (defaults to the locale system setting of the host if omitted or empty).
OffPeakIdleCount Like IdleCount, but for Off Peak time periods.
OffPeakIdleTime Like IdleTime, but for Off Peak time mperiods.
MaxBuilds Builds count after which machine will be removed.
MachineName Name of the machine. It must contain %s, which will be replaced with a unique machine identifier.
MachineDriver Docker Machine driver to use. More details can be found in the Docker Machine configuration section.
MachineOptions Docker Machine options. More details can be found in the Docker Machine configuration section.

Example:

[runners.machine]
  IdleCount = 5
  IdleTime = 600
  OffPeakPeriods = [
    "* * 0-10,18-23 * * mon-fri *",
    "* * * * * sat,sun *"
  ]
  OffPeakTimezone = "Europe/Berlin"
  OffPeakIdleCount = 1
  OffPeakIdleTime = 3600
  MaxBuilds = 100
  MachineName = "auto-scale-%s"
  MachineDriver = "digitalocean"
  MachineOptions = [
      "digitalocean-image=coreos-stable",
      "digitalocean-ssh-user=core",
      "digitalocean-access-token=DO_ACCESS_TOKEN",
      "digitalocean-region=nyc2",
      "digitalocean-size=4gb",
      "digitalocean-private-networking",
      "engine-registry-mirror=http://10.11.12.13:12345"
  ]

OffPeakPeriods syntax

The OffPeakPeriods setting contains an array of string patterns of time periods represented in a cron-style format. The line contains following fields:

[second] [minute] [hour] [day of month] [month] [day of week] [year]

Like in the standard cron configuration file, the fields can contain single values, ranges, lists and asterisks. A detailed description of the syntax can be found here.

The [runners.cache] section

Introduced in GitLab Runner 1.1.0.

This defines the distributed cache feature. More details can be found in the runners autoscale documentation.

Parameter Type Description
Type string One of: s3, gcs.
Path string Name of the path to prepend to the cache URL.
Shared boolean Enables cache sharing between runners, false by default.
Important: With GitLab Runner 11.3.0, the configuration parameters related to S3 were moved to a dedicated [runners.cache.s3] section. The old format of the configuration with S3 configured directly in [runners.cache] section is still supported, but is deprecated with GitLab Runner 11.3.0. This support will be removed in GitLab Runner 12.0.0.

Bellow is a table containing a summary of config.toml, cli options and ENV variables deprecations:

Setting TOML field CLI option for register ENV for register deprecated TOML field deprecated CLI option deprecated ENV
Type [runners.cache] -> Type --cache-type $CACHE_TYPE      
Path [runners.cache] -> Path --cache-path $CACHE_PATH   --cache-s3-cache-path $S3_CACHE_PATH
Shared [runners.cache] -> Shared --cache-shared $CACHE_SHARED   --cache-cache-shared  
S3.ServerAddress [runners.cache.s3] -> ServerAddress --cache-s3-server-address $CACHE_S3_SERVER_ADDRESS [runners.cache] -> ServerAddress   $S3_SERVER_ADDRESS
S3.AccessKey [runners.cache.s3] -> AccessKey --cache-s3-access-key $CACHE_S3_ACCESS_KEY [runners.cache] -> AccessKey   $S3_ACCESS_KEY
S3.SecretKey [runners.cache.s3] -> SecretKey --cache-s3-secret-key $CACHE_S3_SECRET_KEY [runners.cache] -> SecretKey   $S3_SECRET_KEY
S3.BucketName [runners.cache.s3] -> BucketName --cache-s3-bucket-name $CACHE_S3_BUCKET_NAME [runners.cache] -> BucketName   $S3_BUCKET_NAME
S3.BucketLocation [runners.cache.s3] -> BucketLocation --cache-s3-bucket-location $CACHE_S3_BUCKET_LOCATION [runners.cache] -> BucketLocation   $S3_BUCKET_LOCATION
S3.Insecure [runners.cache.s3] -> Insecure --cache-s3-insecure $CACHE_S3_INSECURE [runners.cache] -> Insecure   $S3_INSECURE
GCS.AccessID [runners.cache.gcs] -> AccessID --cache-gcs-access-id $CACHE_GCS_ACCESS_ID      
GCS.PrivateKey [runners.cache.gcs] -> PrivateKey --cache-gcs-private-key $CACHE_GCS_PRIVATE_KEY      
GCS.CredentialsFile [runners.cache.gcs] -> CredentialsFile --cache-gcs-credentials-file $GOOGLE_APPLICATION_CREDENTIALS      
GCS.BucketName [runners.cache.gcs] -> BucketName --cache-gcs-bucket-name $CACHE_GCS_BUCKET_NAME      

The [runners.cache.s3] section

Note: Moved from the [runners.cache] section in GitLab Runner 11.3.0.

Allows to configure S3 storage for cache. This section contains settings related to S3, that previously were present globally in the [runners.cache] section.

Parameter Type Description
ServerAddress string A host:port to the used S3-compatible server.
AccessKey string The access key specified for your S3 instance.
SecretKey string The secret key specified for your S3 instance.
BucketName string Name of the storage bucket where cache will be stored.
BucketLocation string Name of S3 region.
Insecure boolean Set to true if the S3 service is available by HTTP. Set to false by default.

Example:

[runners.cache]
  Type = "s3"
  Path = "path/to/prefix"
  Shared = false
  [runners.cache.s3]
    ServerAddress = "s3.amazonaws.com"
    AccessKey = "AMAZON_S3_ACCESS_KEY"
    SecretKey = "AMAZON_S3_SECRET_KEY"
    BucketName = "runners-cache"
    BucketLocation = "eu-west-1"
    Insecure = false
Note: For Amazon’s S3 service, the ServerAddress should always be s3.amazonaws.com. The Minio S3 client will get bucket metadata and modify the URL to point to the valid region (eg. s3-eu-west-1.amazonaws.com) itself.
Note: If any of ServerAddress, AccessKey or SecretKey aren’t specified, then the S3 client will use the IAM instance profile available to the instance.

The [runners.cache.gcs] section

Introduced in GitLab Runner 11.3.0.

Configure native support for Google Cloud Storage. Read the Google Cloud Storage Authentication documentation to check where these values come from.

Parameter Type Description
CredentialsFile string Path to the Google JSON key file. Currently only the service_account type is supported. If configured, takes precedence over AccessID and PrivateKey configured directly in config.toml.
AccessID string ID of GCP Service Account used to access the storage.
PrivateKey string Private key used to sign GCS requests.
BucketName string Name of the storage bucket where cache will be stored.

Examples:

Credentials configured directly in config.toml file:

[runners.cache]
  Type = "gcs"
  Path = "path/to/prefix"
  Shared = false
  [runners.cache.gcs]
    AccessID = "cache-access-account@test-project-123456.iam.gserviceaccount.com"
    PrivateKey = "-----BEGIN PRIVATE KEY-----\nXXXXXX\n-----END PRIVATE KEY-----\n"
    BucketName = "runners-cache"

Credentials in JSON file downloaded from GCP:

[runners.cache]
  Type = "gcs"
  Path = "path/to/prefix"
  Shared = false
  [runners.cache.gcs]
    CredentialsFile = "/etc/gitlab-runner/service-account.json"
    BucketName = "runners-cache"

The [runners.kubernetes] section

Note: Added in GitLab Runner v1.6.0

This defines the Kubernetes parameters. See Kubernetes executor for additional parameters.

Parameter Type Description
host string Optional Kubernetes master host URL (auto-discovery attempted if not specified)
cert_file string Optional Kubernetes master auth certificate
key_file string Optional Kubernetes master auth private key
ca_file string Optional Kubernetes master auth ca certificate
image string Default docker image to use for builds when none is specified
namespace string Namespace to run Kubernetes jobs in
privileged boolean Run all containers with the privileged flag enabled
node_selector table A table of key=value pairs of string=string. Setting this limits the creation of pods to kubernetes nodes matching all the key=value pairs
image_pull_secrets array A list of secrets that are used to authenticate docker image pulling

Example:

[runners.kubernetes]
	host = "https://45.67.34.123:4892"
	cert_file = "/etc/ssl/kubernetes/api.crt"
	key_file = "/etc/ssl/kubernetes/api.key"
	ca_file = "/etc/ssl/kubernetes/ca.crt"
	image = "golang:1.8"
	privileged = true
	image_pull_secrets = ["docker-registry-credentials"]
	[runners.kubernetes.node_selector]
		gitlab = "true"

Helper image

When one of docker, docker+machine or kubernetes executors is used, GitLab Runner uses a specific container to handle Git, artifacts and cache operations. This container is created from a special image, named helper image.

The helper image is based on Alpine Linux and it’s provided for amd64 and arm architectures. It contains a gitlab-runner-helper binary which is a special compilation of GitLab Runner binary, that contains only a subset of available commands, as well as git, git-lfs, SSL certificates store and basic configuration of Alpine.

When GitLab Runner is installed from the DEB/RPM packages, both images (arm64 and arm based) are installed on the host. When the Runner prepares the environment for the job execution, if the image in specified version (based on Runner’s git revision) is not found on Docker Engine, it is automatically loaded. It works like that for both docker and docker+machine executors.

Things work a little different for the kubernetes executor or when GitLab Runner is installed manually. For manual installations, the gitlab-runner-helper binary is not included and for the kubernetes executor,the API of Kubernetes doesn’t allow to load the gitlab-runner-helper image from a local archive. In both cases, GitLab Runner will download the helper image from Docker Hub, from GitLab’s official repository gitlab/gitlab-runner-helper by using the Runner’s revision and architecture for defining which tag should be downloaded.

Overriding the helper image

In some cases, you may need to override the helper image. There are many reasons for doing this:

  1. To speed up jobs execution: In environments with slower internet connection, downloading over and over again the same image from Docker Hub may generate a significant increase of a job’s timings. Downloading the helper image from a local registry (where the exact copy of gitlab/gitlab-runner-helper:XYZ is stored) may speed things up.

  2. Security concerns: Many people don’t like to download external dependencies that were not checked before. There might be a business rule to use only dependencies that were reviewed and stored in local repositories.

  3. Build environments without internet access: In some cases, jobs are being executed in an environment which has a dedicated, closed network (this doesn’t apply to the kubernetes executor where the image still needs to be downloaded from an external registry that is available at least to the Kubernetes cluster).

  4. Additional software: Some users may want to install some additional software to the helper image, like openssh to support submodules accessible via git+ssh instead of git+http.

In any of the cases described above, it’s possible to configure a custom image using the helper_image configuration field, that is available for the docker, docker+machine and kubernetes executors:

[[runners]]
  (...)
  executor = "docker"
  [runners.docker]
    (...)
    helper_image = "my.registry.local/gitlab/gitlab-runner-helper:tag"

Note that the version of the helper image should be considered as strictly coupled with the version of GitLab Runner. As it was described above, one of the main reasons of providing such images is that Runner is using the gitlab-runner-helper binary, and this binary is compiled from part of GitLab Runner sources which is using an internal API that is expected to be the same in both binaries.

The Runner by default references to a gitlab/gitlab-runner-helper:XYZ image, where XYZ is based on the Runner’s architecture and git revision. Starting with GitLab Runner 11.3 it’s possible to define the version of used image automatically, by using one of the version variables:

[[runners]]
  (...)
  executor = "docker"
  [runners.docker]
    (...)
    helper_image = "my.registry.local/gitlab/gitlab-runner-helper:${CI_RUNNER_REVISION}"

With that configuration, GitLab Runner will instruct the executor to use the image in version ${CI_RUNNER_REVISION}, which is based on its compilation data. After updating the Runner to a new version, this will ensure that the Runner will try to download the proper image. This of course means that the image should be uploaded to the registry before upgrading the Runner, otherwise the jobs will start failing with a “No such image” error.

Note

If you’d like to deploy to multiple servers using GitLab CI, you can create a single script that deploys to multiple servers or you can create many scripts. It depends on what you’d like to do.