Tuning Component Settings

Helm Charts are a set of Kubernetes manifests that reflect best practices for deploying an application or service on Kubernetes.

After you add the Deis Chart Repository, you can customize the chart using helm inspect values hephy/workflow > values.yaml before using helm install to complete the installation.

There are a few ways to customize the respective component:

  • If the value is exposed in the values.yaml file as derived above, one may modify the section of the component to tune these settings. The modified value(s) will then take effect at chart installation or release upgrade time via either of the two respective commands:

    $ helm install hephy/workflow -n hephy --namespace deis -f values.yaml
    $ helm upgrade hephy -f values.yaml
    
  • If the value hasn't yet been exposed in the values.yaml file, one may edit the component deployment with the tuned setting. Here we edit the deis-controller deployment:

    $ kubectl --namespace deis edit deployment deis-controller
    

    Add/edit the setting via the appropriate environment variable and value under the env section and save. The updated deployment will recreate the component pod with the new/modified setting.

  • Lastly, one may also fetch and edit the chart as served by version control/the chart repository itself:

    $ helm fetch hephy/workflow --untar
    $ $EDITOR workflow/charts/controller/templates/controller-deployment.yaml
    

    Then run helm install ./workflow --namespace deis --name hephy to apply the changes, or helm upgrade hephy ./workflow if the cluster is already running.

Setting Resource limits

You can set resource limits to Workflow components by modifying the values.yaml file fetched earlier. This file has a section for each Workflow component. To set a limit to any Workflow component just add limits_cpu, limits_memory in the section and set them to the appropriate values.

Below is an example of how the builder section of values.yaml might look with CPU and memory limits set:

builder:
  org: "hephyci"
  pullPolicy: "Always"
  dockerTag: "canary"
  limits_cpu: "100m"
  limits_memory: "50Mi"

Customizing the Builder

The following environment variables are tunable for the Builder component:

Setting Description
DEBUG Enable debug log output (default: false)
BUILDER_POD_NODE_SELECTOR A node selector setting for builder job. As it may sometimes consume a lot of node resources, one may want a given builder job to run in a specific node only, so it won't affect critical nodes. for example pool:testing,disk:magnetic

Customizing the Controller

The following environment variables are tunable for the Controller component:

Setting Description
REGISTRATION_MODE set registration to "enabled", "disabled", or "admin_only" (default: "admin_only")
GUNICORN_WORKERS number of gunicorn workers spawned to process requests (default: CPU cores * 4 + 1)
RESERVED_NAMES a comma-separated list of names which applications cannot reserve for routing (default: "deis, deis-builder, deis-workflow-manager")
SLUGRUNNER_IMAGE_NAME the image used to run buildpack application slugs (default: "hephy/slugrunner:v2.6.1")
DEIS_DEPLOY_HOOK_URLS a comma-separated list of URLs to send deploy hooks to.
DEIS_DEPLOY_HOOK_SECRET_KEY a private key used to compute the HMAC signature for deploy hooks.
DEIS_DEPLOY_REJECT_IF_PROCFILE_MISSING rejects a deploy if the previous build had a Procfile but the current deploy is missing it. A 409 is thrown in the API. Prevents accidental process types removal. (default: "false", allowed values: "true", "false")
DEIS_DEPLOY_PROCFILE_MISSING_REMOVE when turned on (default) any missing process type in a Procfile compared to the previous deploy is removed. When set to false will allow an empty Procfile to go through without removing missing process types, note that new images, configs and so on will get updated on all proc types. (default: "true", allowed values: "true", "false")
DEIS_DEFAULT_CONFIG_TAGS set tags for all applications by default, for example: '{"role": "worker"}'. (default: '')
DEIS_IGNORE_SCHEDULING_FAILURE ignore Kubernetes scheduling errors during deployments. With cluster autoscaling, scheduling errors may be temporary and should not fail the deployment (default: "false", allowed values: "true", "false").
KUBERNETES_NAMESPACE_DEFAULT_QUOTA_SPEC set resource quota to application namespace by setting ResourceQuota spec, for example: {"spec":{"hard":{"pods":"10"}}}, restrict app owner to spawn more then 10 pods (default: "", no quota will be applied to namespace)
DEIS_DEBUG set the log level to DEBUG
LOG_VERBOSE activate the verbose logging formatter

LDAP authentication settings

Configuration options for LDAP authentication are detailed here.

The following environment variables are available for enabling LDAP authentication of user accounts in the Controller component:

Setting Description
LDAP_ENDPOINT The URI of the LDAP server. If not specified, LDAP authentication is not enabled (default: "", example: ldap://hostname).
LDAP_BIND_DN The distinguished name to use when binding to the LDAP server (default: "")
LDAP_BIND_PASSWORD The password to use with LDAP_BIND_DN (default: "")
LDAP_USER_BASEDN The distinguished name of the search base for user names (default: "")
LDAP_USER_FILTER The name of the login field in the users search base (default: "username")
LDAP_GROUP_BASEDN The distinguished name of the search base for user's groups names (default: "")
LDAP_GROUP_FILTER The filter for user's groups (default: "", example: objectClass=person)

Global and per application settings

Setting Description
DEIS_DEPLOY_BATCHES the number of pods to bring up and take down sequentially during a scale (default: number of available nodes)
DEIS_DEPLOY_TIMEOUT deploy timeout in seconds per deploy batch (default: 120)
IMAGE_PULL_POLICY the kubernetes image pull policy for application images (default: "IfNotPresent") (allowed values: "Always", "IfNotPresent")
KUBERNETES_DEPLOYMENTS_REVISION_HISTORY_LIMIT how many revisions Kubernetes keeps around of a given Deployment (default: all revisions)
KUBERNETES_POD_TERMINATION_GRACE_PERIOD_SECONDS how many seconds kubernetes waits for a pod to finish work after a SIGTERM before sending SIGKILL (default: 30)

See the Deploying Apps guide for more detailed information on those.

Customizing the Database

The following environment variables are tunable for the Database component:

Setting Description
BACKUP_FREQUENCY how often the database should perform a base backup (default: "12h")
BACKUPS_TO_RETAIN number of base backups the backing store should retain (default: 5)

Customizing Fluentd

The following values can be changed in the values.yaml file or by using the --set flag with the Helm CLI.

Key Default Description
syslog.host "" Host value of a syslog endpoint
syslog.port "" Port value of a syslog endpoint
sources.start_script false Capture kubernetes start script logs
sources.docker false Capture docker daemon logs
sources.etcd false Capture etcd logs
sources.kubelet false Capture kubelet logs
sources.kube_api false Capture Kubernetes API logs
sources.controller false Capture Kubernetes Controller logs
sources.scheduler false Capture Kubernetes Scheduler logs
output.disable_deis false Disable the Deis output plugin
boot.install_build_tools false Install the build tools package. This is useful when using custom plugins
daemon_environment Takes key-value pairs and turns them into environment variables.

For more information about the various environment variables that can be set please see the README

Customizing the Logger

The following environment variables are tunable for the Logger component:

Setting Description
STORAGE_ADAPTER How to store logs that are sent to the logger. Legal values are "file", "memory", and "redis". (default: "redis")
NUMBER_OF_LINES How many lines to store in the ring buffer (default: 1000)

Customizing the Monitor

Grafana

We have exposed some of the more useful configuration values directly in the chart. This allows them to be set using either the values.yaml file or by using the --set flag with the Helm CLI. You can see these options below:

Setting Default Value Description
user "admin" The first user created in the database (this user has admin privileges)
password "admin" Password for the first user.
allow_sign_up "true" Allows users to sign up for an account.

For a list of other options you can set by using environment variables please see the configuration file in Github.

Telegraf

For a list of configuration values that can be set by using environment variables please see the following configuration file.

InfluxDB

You can find a list of values that can be set using environment variables here.

Customizing the Registry

The Registry component can be tuned by following the teamhephy/distribution config doc.

Customizing the Router

The majority of router settings are tunable through annotations, which allows the router to be re-configured with zero downtime post-installation. You can find the list of annotations to tune here.

The following environment variables are tunable for the Router component:

Setting Description
POD_NAMESPACE The pod namespace the router resides in. This is set by the Kubernetes downward API.

Customizing Workflow Manager

The following environment variables are tunable for Workflow Manager:

Setting Description
CHECK_VERSIONS Enables the external version check at https://versions.teamhephy.info/ (default: "true")
POLL_INTERVAL_SEC The interval when Workflow Manager performs a version check, in seconds (default: 43200, or 12 hours)
VERSIONS_API_URL The versions API URL (default: "https://versions-staging.teamhephy.info")
DOCTOR_API_URL The doctor API URL (default: "https://doctor-staging.teamhephy.info")
API_VERSION The version number Workflow Manager sends to the versions API (default: "v2")