Operators
Working with Operators in OpenShift Container Platform
Abstract
Chapter 1. Operators overview
Operators are among the most important components of OpenShift Container Platform. Operators are the preferred method of packaging, deploying, and managing services on the control plane. They can also provide advantages to applications that users run.
Operators integrate with Kubernetes APIs and CLI tools such as kubectl
and oc
commands. They provide the means of monitoring applications, performing health checks, managing over-the-air (OTA) updates, and ensuring that applications remain in your specified state.
While both follow similar Operator concepts and goals, Operators in OpenShift Container Platform are managed by two different systems, depending on their purpose:
- Cluster Operators, which are managed by the Cluster Version Operator (CVO), are installed by default to perform cluster functions.
- Optional add-on Operators, which are managed by Operator Lifecycle Manager (OLM), can be made accessible for users to run in their applications.
With Operators, you can create applications to monitor the running services in the cluster. Operators are designed specifically for your applications. Operators implement and automate the common Day 1 operations such as installation and configuration as well as Day 2 operations such as autoscaling up and down and creating backups. All these activities are in a piece of software running inside your cluster.
1.1. For developers
As a developer, you can perform the following Operator tasks:
- Install Operator SDK CLI.
- Create Go-based Operators, Ansible-based Operators, Java-based Operators, and Helm-based Operators.
- Use Operator SDK to build, test, and deploy an Operator.
- Install and subscribe an Operator to your namespace.
- Create an application from an installed Operator through the web console.
Additional resources
1.2. For administrators
As a cluster administrator, you can perform the following Operator tasks:
- Manage custom catalogs.
- Allow non-cluster administrators to install Operators.
- Install an Operator from OperatorHub.
- View Operator status.
- Manage Operator conditions.
- Upgrade installed Operators.
- Delete installed Operators.
- Configure proxy support.
- Use Operator Lifecycle Manager on restricted networks.
To know all about the cluster Operators that Red Hat provides, see Cluster Operators reference.
1.3. Next steps
To understand more about Operators, see What are Operators?
Chapter 2. Understanding Operators
2.1. What are Operators?
Conceptually, Operators take human operational knowledge and encode it into software that is more easily shared with consumers.
Operators are pieces of software that ease the operational complexity of running another piece of software. They act like an extension of the software vendor’s engineering team, monitoring a Kubernetes environment (such as OpenShift Container Platform) and using its current state to make decisions in real time. Advanced Operators are designed to handle upgrades seamlessly, react to failures automatically, and not take shortcuts, like skipping a software backup process to save time.
More technically, Operators are a method of packaging, deploying, and managing a Kubernetes application.
A Kubernetes application is an app that is both deployed on Kubernetes and managed using the Kubernetes APIs and kubectl
or oc
tooling. To be able to make the most of Kubernetes, you require a set of cohesive APIs to extend in order to service and manage your apps that run on Kubernetes. Think of Operators as the runtime that manages this type of app on Kubernetes.
2.1.1. Why use Operators?
Operators provide:
- Repeatability of installation and upgrade.
- Constant health checks of every system component.
- Over-the-air (OTA) updates for OpenShift components and ISV content.
- A place to encapsulate knowledge from field engineers and spread it to all users, not just one or two.
- Why deploy on Kubernetes?
- Kubernetes (and by extension, OpenShift Container Platform) contains all of the primitives needed to build complex distributed systems – secret handling, load balancing, service discovery, autoscaling – that work across on-premises and cloud providers.
- Why manage your app with Kubernetes APIs and
kubectl
tooling? -
These APIs are feature rich, have clients for all platforms and plug into the cluster’s access control/auditing. An Operator uses the Kubernetes extension mechanism, custom resource definitions (CRDs), so your custom object, for example
MongoDB
, looks and acts just like the built-in, native Kubernetes objects. - How do Operators compare with service brokers?
- A service broker is a step towards programmatic discovery and deployment of an app. However, because it is not a long running process, it cannot execute Day 2 operations like upgrade, failover, or scaling. Customizations and parameterization of tunables are provided at install time, versus an Operator that is constantly watching the current state of your cluster. Off-cluster services are a good match for a service broker, although Operators exist for these as well.
2.1.2. Operator Framework
The Operator Framework is a family of tools and capabilities to deliver on the customer experience described above. It is not just about writing code; testing, delivering, and updating Operators is just as important. The Operator Framework components consist of open source tools to tackle these problems:
- Operator SDK
- The Operator SDK assists Operator authors in bootstrapping, building, testing, and packaging their own Operator based on their expertise without requiring knowledge of Kubernetes API complexities.
- Operator Lifecycle Manager
- Operator Lifecycle Manager (OLM) controls the installation, upgrade, and role-based access control (RBAC) of Operators in a cluster. It is deployed by default in OpenShift Container Platform 4.15.
- Operator Registry
- The Operator Registry stores cluster service versions (CSVs) and custom resource definitions (CRDs) for creation in a cluster and stores Operator metadata about packages and channels. It runs in a Kubernetes or OpenShift cluster to provide this Operator catalog data to OLM.
- OperatorHub
- OperatorHub is a web console for cluster administrators to discover and select Operators to install on their cluster. It is deployed by default in OpenShift Container Platform.
These tools are designed to be composable, so you can use any that are useful to you.
2.1.3. Operator maturity model
The level of sophistication of the management logic encapsulated within an Operator can vary. This logic is also in general highly dependent on the type of the service represented by the Operator.
One can however generalize the scale of the maturity of the encapsulated operations of an Operator for certain set of capabilities that most Operators can include. To this end, the following Operator maturity model defines five phases of maturity for generic Day 2 operations of an Operator:
Figure 2.1. Operator maturity model
The above model also shows how these capabilities can best be developed through the Helm, Go, and Ansible capabilities of the Operator SDK.
2.2. Operator Framework packaging format
This guide outlines the packaging format for Operators supported by Operator Lifecycle Manager (OLM) in OpenShift Container Platform.
2.2.1. Bundle format
The bundle format for Operators is a packaging format introduced by the Operator Framework. To improve scalability and to better enable upstream users hosting their own catalogs, the bundle format specification simplifies the distribution of Operator metadata.
An Operator bundle represents a single version of an Operator. On-disk bundle manifests are containerized and shipped as a bundle image, which is a non-runnable container image that stores the Kubernetes manifests and Operator metadata. Storage and distribution of the bundle image is then managed using existing container tools like podman
and docker
and container registries such as Quay.
Operator metadata can include:
- Information that identifies the Operator, for example its name and version.
- Additional information that drives the UI, for example its icon and some example custom resources (CRs).
- Required and provided APIs.
- Related images.
When loading manifests into the Operator Registry database, the following requirements are validated:
- The bundle must have at least one channel defined in the annotations.
- Every bundle has exactly one cluster service version (CSV).
- If a CSV owns a custom resource definition (CRD), that CRD must exist in the bundle.
2.2.1.1. Manifests
Bundle manifests refer to a set of Kubernetes manifests that define the deployment and RBAC model of the Operator.
A bundle includes one CSV per directory and typically the CRDs that define the owned APIs of the CSV in its /manifests
directory.
Example bundle format layout
etcd ├── manifests │ ├── etcdcluster.crd.yaml │ └── etcdoperator.clusterserviceversion.yaml │ └── secret.yaml │ └── configmap.yaml └── metadata └── annotations.yaml └── dependencies.yaml
Additionally supported objects
The following object types can also be optionally included in the /manifests
directory of a bundle:
Supported optional object types
-
ClusterRole
-
ClusterRoleBinding
-
ConfigMap
-
ConsoleCLIDownload
-
ConsoleLink
-
ConsoleQuickStart
-
ConsoleYamlSample
-
PodDisruptionBudget
-
PriorityClass
-
PrometheusRule
-
Role
-
RoleBinding
-
Secret
-
Service
-
ServiceAccount
-
ServiceMonitor
-
VerticalPodAutoscaler
When these optional objects are included in a bundle, Operator Lifecycle Manager (OLM) can create them from the bundle and manage their lifecycle along with the CSV:
Lifecycle for optional objects
- When the CSV is deleted, OLM deletes the optional object.
When the CSV is upgraded:
- If the name of the optional object is the same, OLM updates it in place.
- If the name of the optional object has changed between versions, OLM deletes and recreates it.
2.2.1.2. Annotations
A bundle also includes an annotations.yaml
file in its /metadata
directory. This file defines higher level aggregate data that helps describe the format and package information about how the bundle should be added into an index of bundles:
Example annotations.yaml
annotations: operators.operatorframework.io.bundle.mediatype.v1: "registry+v1" 1 operators.operatorframework.io.bundle.manifests.v1: "manifests/" 2 operators.operatorframework.io.bundle.metadata.v1: "metadata/" 3 operators.operatorframework.io.bundle.package.v1: "test-operator" 4 operators.operatorframework.io.bundle.channels.v1: "beta,stable" 5 operators.operatorframework.io.bundle.channel.default.v1: "stable" 6
- 1
- The media type or format of the Operator bundle. The
registry+v1
format means it contains a CSV and its associated Kubernetes objects. - 2
- The path in the image to the directory that contains the Operator manifests. This label is reserved for future use and currently defaults to
manifests/
. The valuemanifests.v1
implies that the bundle contains Operator manifests. - 3
- The path in the image to the directory that contains metadata files about the bundle. This label is reserved for future use and currently defaults to
metadata/
. The valuemetadata.v1
implies that this bundle has Operator metadata. - 4
- The package name of the bundle.
- 5
- The list of channels the bundle is subscribing to when added into an Operator Registry.
- 6
- The default channel an Operator should be subscribed to when installed from a registry.
In case of a mismatch, the annotations.yaml
file is authoritative because the on-cluster Operator Registry that relies on these annotations only has access to this file.
2.2.1.3. Dependencies
The dependencies of an Operator are listed in a dependencies.yaml
file in the metadata/
folder of a bundle. This file is optional and currently only used to specify explicit Operator-version dependencies.
The dependency list contains a type
field for each item to specify what kind of dependency this is. The following types of Operator dependencies are supported:
olm.package
-
This type indicates a dependency for a specific Operator version. The dependency information must include the package name and the version of the package in semver format. For example, you can specify an exact version such as
0.5.2
or a range of versions such as>0.5.1
. olm.gvk
- With this type, the author can specify a dependency with group/version/kind (GVK) information, similar to existing CRD and API-based usage in a CSV. This is a path to enable Operator authors to consolidate all dependencies, API or explicit versions, to be in the same place.
olm.constraint
- This type declares generic constraints on arbitrary Operator properties.
In the following example, dependencies are specified for a Prometheus Operator and etcd CRDs:
Example dependencies.yaml
file
dependencies: - type: olm.package value: packageName: prometheus version: ">0.27.0" - type: olm.gvk value: group: etcd.database.coreos.com kind: EtcdCluster version: v1beta2
Additional resources
2.2.1.4. About the opm CLI
The opm
CLI tool is provided by the Operator Framework for use with the Operator bundle format. This tool allows you to create and maintain catalogs of Operators from a list of Operator bundles that are similar to software repositories. The result is a container image which can be stored in a container registry and then installed on a cluster.
A catalog contains a database of pointers to Operator manifest content that can be queried through an included API that is served when the container image is run. On OpenShift Container Platform, Operator Lifecycle Manager (OLM) can reference the image in a catalog source, defined by a CatalogSource
object, which polls the image at regular intervals to enable frequent updates to installed Operators on the cluster.
-
See CLI tools for steps on installing the
opm
CLI.
2.2.2. File-based catalogs
File-based catalogs are the latest iteration of the catalog format in Operator Lifecycle Manager (OLM). It is a plain text-based (JSON or YAML) and declarative config evolution of the earlier SQLite database format, and it is fully backwards compatible. The goal of this format is to enable Operator catalog editing, composability, and extensibility.
- Editing
With file-based catalogs, users interacting with the contents of a catalog are able to make direct changes to the format and verify that their changes are valid. Because this format is plain text JSON or YAML, catalog maintainers can easily manipulate catalog metadata by hand or with widely known and supported JSON or YAML tooling, such as the
jq
CLI.This editability enables the following features and user-defined extensions:
- Promoting an existing bundle to a new channel
- Changing the default channel of a package
- Custom algorithms for adding, updating, and removing upgrade edges
- Composability
File-based catalogs are stored in an arbitrary directory hierarchy, which enables catalog composition. For example, consider two separate file-based catalog directories:
catalogA
andcatalogB
. A catalog maintainer can create a new combined catalog by making a new directorycatalogC
and copyingcatalogA
andcatalogB
into it.This composability enables decentralized catalogs. The format permits Operator authors to maintain Operator-specific catalogs, and it permits maintainers to trivially build a catalog composed of individual Operator catalogs. File-based catalogs can be composed by combining multiple other catalogs, by extracting subsets of one catalog, or a combination of both of these.
NoteDuplicate packages and duplicate bundles within a package are not permitted. The
opm validate
command returns an error if any duplicates are found.Because Operator authors are most familiar with their Operator, its dependencies, and its upgrade compatibility, they are able to maintain their own Operator-specific catalog and have direct control over its contents. With file-based catalogs, Operator authors own the task of building and maintaining their packages in a catalog. Composite catalog maintainers, however, only own the task of curating the packages in their catalog and publishing the catalog to users.
- Extensibility
The file-based catalog specification is a low-level representation of a catalog. While it can be maintained directly in its low-level form, catalog maintainers can build interesting extensions on top that can be used by their own custom tooling to make any number of mutations.
For example, a tool could translate a high-level API, such as
(mode=semver)
, down to the low-level, file-based catalog format for upgrade edges. Or a catalog maintainer might need to customize all of the bundle metadata by adding a new property to bundles that meet a certain criteria.While this extensibility allows for additional official tooling to be developed on top of the low-level APIs for future OpenShift Container Platform releases, the major benefit is that catalog maintainers have this capability as well.
As of OpenShift Container Platform 4.11, the default Red Hat-provided Operator catalog releases in the file-based catalog format. The default Red Hat-provided Operator catalogs for OpenShift Container Platform 4.6 through 4.10 released in the deprecated SQLite database format.
The opm
subcommands, flags, and functionality related to the SQLite database format are also deprecated and will be removed in a future release. The features are still supported and must be used for catalogs that use the deprecated SQLite database format.
Many of the opm
subcommands and flags for working with the SQLite database format, such as opm index prune
, do not work with the file-based catalog format. For more information about working with file-based catalogs, see Managing custom catalogs and Mirroring images for a disconnected installation using the oc-mirror plugin.
2.2.2.1. Directory structure
File-based catalogs can be stored and loaded from directory-based file systems. The opm
CLI loads the catalog by walking the root directory and recursing into subdirectories. The CLI attempts to load every file it finds and fails if any errors occur.
Non-catalog files can be ignored using .indexignore
files, which have the same rules for patterns and precedence as .gitignore
files.
Example .indexignore
file
# Ignore everything except non-object .json and .yaml files **/* !*.json !*.yaml **/objects/*.json **/objects/*.yaml
Catalog maintainers have the flexibility to choose their desired layout, but it is recommended to store each package’s file-based catalog blobs in separate subdirectories. Each individual file can be either JSON or YAML; it is not necessary for every file in a catalog to use the same format.
Basic recommended structure
catalog ├── packageA │ └── index.yaml ├── packageB │ ├── .indexignore │ ├── index.yaml │ └── objects │ └── packageB.v0.1.0.clusterserviceversion.yaml └── packageC └── index.json └── deprecations.yaml
This recommended structure has the property that each subdirectory in the directory hierarchy is a self-contained catalog, which makes catalog composition, discovery, and navigation trivial file system operations. The catalog can also be included in a parent catalog by copying it into the parent catalog’s root directory.
2.2.2.2. Schemas
File-based catalogs use a format, based on the CUE language specification, that can be extended with arbitrary schemas. The following _Meta
CUE schema defines the format that all file-based catalog blobs must adhere to:
_Meta
schema
_Meta: { // schema is required and must be a non-empty string schema: string & !="" // package is optional, but if it's defined, it must be a non-empty string package?: string & !="" // properties is optional, but if it's defined, it must be a list of 0 or more properties properties?: [... #Property] } #Property: { // type is required type: string & !="" // value is required, and it must not be null value: !=null }
No CUE schemas listed in this specification should be considered exhaustive. The opm validate
command has additional validations that are difficult or impossible to express concisely in CUE.
An Operator Lifecycle Manager (OLM) catalog currently uses three schemas (olm.package
, olm.channel
, and olm.bundle
), which correspond to OLM’s existing package and bundle concepts.
Each Operator package in a catalog requires exactly one olm.package
blob, at least one olm.channel
blob, and one or more olm.bundle
blobs.
All olm.*
schemas are reserved for OLM-defined schemas. Custom schemas must use a unique prefix, such as a domain that you own.
2.2.2.2.1. olm.package schema
The olm.package
schema defines package-level metadata for an Operator. This includes its name, description, default channel, and icon.
Example 2.1. olm.package
schema
#Package: { schema: "olm.package" // Package name name: string & !="" // A description of the package description?: string // The package's default channel defaultChannel: string & !="" // An optional icon icon?: { base64data: string mediatype: string } }
2.2.2.2.2. olm.channel schema
The olm.channel
schema defines a channel within a package, the bundle entries that are members of the channel, and the upgrade edges for those bundles.
If a bundle entry represents an edge in multiple olm.channel
blobs, it can only appear once per channel.
It is valid for an entry’s replaces
value to reference another bundle name that cannot be found in this catalog or another catalog. However, all other channel invariants must hold true, such as a channel not having multiple heads.
Example 2.2. olm.channel
schema
#Channel: { schema: "olm.channel" package: string & !="" name: string & !="" entries: [...#ChannelEntry] } #ChannelEntry: { // name is required. It is the name of an `olm.bundle` that // is present in the channel. name: string & !="" // replaces is optional. It is the name of bundle that is replaced // by this entry. It does not have to be present in the entry list. replaces?: string & !="" // skips is optional. It is a list of bundle names that are skipped by // this entry. The skipped bundles do not have to be present in the // entry list. skips?: [...string & !=""] // skipRange is optional. It is the semver range of bundle versions // that are skipped by this entry. skipRange?: string & !="" }
When using the skipRange
field, the skipped Operator versions are pruned from the update graph and are longer installable by users with the spec.startingCSV
property of Subscription
objects.
You can update an Operator incrementally while keeping previously installed versions available to users for future installation by using both the skipRange
and replaces
field. Ensure that the replaces
field points to the immediate previous version of the Operator version in question.
2.2.2.2.3. olm.bundle schema
Example 2.3. olm.bundle
schema
#Bundle: { schema: "olm.bundle" package: string & !="" name: string & !="" image: string & !="" properties: [...#Property] relatedImages?: [...#RelatedImage] } #Property: { // type is required type: string & !="" // value is required, and it must not be null value: !=null } #RelatedImage: { // image is the image reference image: string & !="" // name is an optional descriptive name for an image that // helps identify its purpose in the context of the bundle name?: string & !="" }
2.2.2.2.4. olm.deprecations schema
The optional olm.deprecations
schema defines deprecation information for packages, bundles, and channels in a catalog. Operator authors can use this schema to provide relevant messages about their Operators, such as support status and recommended upgrade paths, to users running those Operators from a catalog.
An olm.deprecations
schema entry contains one or more of the following reference
types, which indicates the deprecation scope. After the Operator is installed, any specified messages can be viewed as status conditions on the related Subscription
object.
Type | Scope | Status condition |
---|---|---|
| Represents the entire package |
|
| Represents one channel |
|
| Represents one bundle version |
|
Each reference
type has their own requirements, as detailed in the following example.
Example 2.4. Example olm.deprecations
schema with each reference
type
schema: olm.deprecations package: my-operator 1 entries: - reference: schema: olm.package 2 message: | 3 The 'my-operator' package is end of life. Please use the 'my-operator-new' package for support. - reference: schema: olm.channel name: alpha 4 message: | The 'alpha' channel is no longer supported. Please switch to the 'stable' channel. - reference: schema: olm.bundle name: my-operator.v1.68.0 5 message: | my-operator.v1.68.0 is deprecated. Uninstall my-operator.v1.68.0 and install my-operator.v1.72.0 for support.
- 1
- Each deprecation schema must have a
package
value, and that package reference must be unique across the catalog. There must not be an associatedname
field. - 2
- The
olm.package
schema must not include aname
field, because it is determined by thepackage
field defined earlier in the schema. - 3
- All
message
fields, for anyreference
type, must be a non-zero length and represented as an opaque text blob. - 4
- The
name
field for theolm.channel
schema is required. - 5
- The
name
field for theolm.bundle
schema is required.
The deprecation feature does not consider overlapping deprecation, for example package versus channel versus bundle.
Operator authors can save olm.deprecations
schema entries as a deprecations.yaml
file in the same directory as the package’s index.yaml
file:
Example directory structure for a catalog with deprecations
my-catalog └── my-operator ├── index.yaml └── deprecations.yaml
Additional resources
2.2.2.3. Properties
Properties are arbitrary pieces of metadata that can be attached to file-based catalog schemas. The type
field is a string that effectively specifies the semantic and syntactic meaning of the value
field. The value can be any arbitrary JSON or YAML.
OLM defines a handful of property types, again using the reserved olm.*
prefix.
2.2.2.3.1. olm.package property
The olm.package
property defines the package name and version. This is a required property on bundles, and there must be exactly one of these properties. The packageName
field must match the bundle’s first-class package
field, and the version
field must be a valid semantic version.
Example 2.5. olm.package
property
#PropertyPackage: { type: "olm.package" value: { packageName: string & !="" version: string & !="" } }
2.2.2.3.2. olm.gvk property
The olm.gvk
property defines the group/version/kind (GVK) of a Kubernetes API that is provided by this bundle. This property is used by OLM to resolve a bundle with this property as a dependency for other bundles that list the same GVK as a required API. The GVK must adhere to Kubernetes GVK validations.
Example 2.6. olm.gvk
property
#PropertyGVK: { type: "olm.gvk" value: { group: string & !="" version: string & !="" kind: string & !="" } }
2.2.2.3.3. olm.package.required
The olm.package.required
property defines the package name and version range of another package that this bundle requires. For every required package property a bundle lists, OLM ensures there is an Operator installed on the cluster for the listed package and in the required version range. The versionRange
field must be a valid semantic version (semver) range.
Example 2.7. olm.package.required
property
#PropertyPackageRequired: { type: "olm.package.required" value: { packageName: string & !="" versionRange: string & !="" } }
2.2.2.3.4. olm.gvk.required
The olm.gvk.required
property defines the group/version/kind (GVK) of a Kubernetes API that this bundle requires. For every required GVK property a bundle lists, OLM ensures there is an Operator installed on the cluster that provides it. The GVK must adhere to Kubernetes GVK validations.
Example 2.8. olm.gvk.required
property
#PropertyGVKRequired: { type: "olm.gvk.required" value: { group: string & !="" version: string & !="" kind: string & !="" } }
2.2.2.4. Example catalog
With file-based catalogs, catalog maintainers can focus on Operator curation and compatibility. Because Operator authors have already produced Operator-specific catalogs for their Operators, catalog maintainers can build their catalog by rendering each Operator catalog into a subdirectory of the catalog’s root directory.
There are many possible ways to build a file-based catalog; the following steps outline a simple approach:
Maintain a single configuration file for the catalog, containing image references for each Operator in the catalog:
Example catalog configuration file
name: community-operators repo: quay.io/community-operators/catalog tag: latest references: - name: etcd-operator image: quay.io/etcd-operator/index@sha256:5891b5b522d5df086d0ff0b110fbd9d21bb4fc7163af34d08286a2e846f6be03 - name: prometheus-operator image: quay.io/prometheus-operator/index@sha256:e258d248fda94c63753607f7c4494ee0fcbe92f1a76bfdac795c9d84101eb317
Run a script that parses the configuration file and creates a new catalog from its references:
Example script
name=$(yq eval '.name' catalog.yaml) mkdir "$name" yq eval '.name + "/" + .references[].name' catalog.yaml | xargs mkdir for l in $(yq e '.name as $catalog | .references[] | .image + "|" + $catalog + "/" + .name + "/index.yaml"' catalog.yaml); do image=$(echo $l | cut -d'|' -f1) file=$(echo $l | cut -d'|' -f2) opm render "$image" > "$file" done opm generate dockerfile "$name" indexImage=$(yq eval '.repo + ":" + .tag' catalog.yaml) docker build -t "$indexImage" -f "$name.Dockerfile" . docker push "$indexImage"
2.2.2.5. Guidelines
Consider the following guidelines when maintaining file-based catalogs.
2.2.2.5.1. Immutable bundles
The general advice with Operator Lifecycle Manager (OLM) is that bundle images and their metadata should be treated as immutable.
If a broken bundle has been pushed to a catalog, you must assume that at least one of your users has upgraded to that bundle. Based on that assumption, you must release another bundle with an upgrade edge from the broken bundle to ensure users with the broken bundle installed receive an upgrade. OLM will not reinstall an installed bundle if the contents of that bundle are updated in the catalog.
However, there are some cases where a change in the catalog metadata is preferred:
-
Channel promotion: If you already released a bundle and later decide that you would like to add it to another channel, you can add an entry for your bundle in another
olm.channel
blob. -
New upgrade edges: If you release a new
1.2.z
bundle version, for example1.2.4
, but1.3.0
is already released, you can update the catalog metadata for1.3.0
to skip1.2.4
.
2.2.2.5.2. Source control
Catalog metadata should be stored in source control and treated as the source of truth. Updates to catalog images should include the following steps:
- Update the source-controlled catalog directory with a new commit.
-
Build and push the catalog image. Use a consistent tagging taxonomy, such as
:latest
or:<target_cluster_version>
, so that users can receive updates to a catalog as they become available.
2.2.2.6. CLI usage
For instructions about creating file-based catalogs by using the opm
CLI, see Managing custom catalogs.
For reference documentation about the opm
CLI commands related to managing file-based catalogs, see CLI tools.
2.2.2.7. Automation
Operator authors and catalog maintainers are encouraged to automate their catalog maintenance with CI/CD workflows. Catalog maintainers can further improve on this by building GitOps automation to accomplish the following tasks:
- Check that pull request (PR) authors are permitted to make the requested changes, for example by updating their package’s image reference.
-
Check that the catalog updates pass the
opm validate
command. - Check that the updated bundle or catalog image references exist, the catalog images run successfully in a cluster, and Operators from that package can be successfully installed.
- Automatically merge PRs that pass the previous checks.
- Automatically rebuild and republish the catalog image.
2.2.3. RukPak (Technology Preview)
RukPak is a Technology Preview feature only. Technology Preview features are not supported with Red Hat production service level agreements (SLAs) and might not be functionally complete. Red Hat does not recommend using them in production. These features provide early access to upcoming product features, enabling customers to test functionality and provide feedback during the development process.
For more information about the support scope of Red Hat Technology Preview features, see Technology Preview Features Support Scope.
OpenShift Container Platform 4.12 introduces the platform Operator type as a Technology Preview feature. The platform Operator mechanism relies on the RukPak component, also introduced in OpenShift Container Platform 4.12, and its resources to manage content.
OpenShift Container Platform 4.14 introduces Operator Lifecycle Manager (OLM) 1.0 as a Technology Preview feature, which also relies on the RukPak component.
RukPak is a pluggable solution for packaging and distributing cloud-native content. It supports advanced strategies for installation, updates, and policy.
RukPak provides a content ecosystem for installing a variety of artifacts on a Kubernetes cluster. Artifact examples include Git repositories, Helm charts, and OLM bundles. RukPak can then manage, scale, and upgrade these artifacts in a safe way to enable powerful cluster extensions.
At its core, RukPak is a small set of APIs and controllers. The APIs are packaged as custom resource definitions (CRDs) that express what content to install on a cluster and how to create a running deployment of the content. The controllers watch for the APIs.
Common terminology
- Bundle
- A collection of Kubernetes manifests that define content to be deployed to a cluster
- Bundle image
- A container image that contains a bundle within its filesystem
- Bundle Git repository
- A Git repository that contains a bundle within a directory
- Provisioner
- Controllers that install and manage content on a Kubernetes cluster
- Bundle deployment
- Generates deployed instances of a bundle
Additional resources
2.2.3.1. Bundle
A RukPak Bundle
object represents content to make available to other consumers in the cluster. Much like the contents of a container image must be pulled and unpacked in order for pod to start using them, Bundle
objects are used to reference content that might need to be pulled and unpacked. In this sense, a bundle is a generalization of the image concept and can be used to represent any type of content.
Bundles cannot do anything on their own; they require a provisioner to unpack and make their content available in the cluster. They can be unpacked to any arbitrary storage medium, such as a tar.gz
file in a directory mounted into the provisioner pods. Each Bundle
object has an associated spec.provisionerClassName
field that indicates the Provisioner
object that watches and unpacks that particular bundle type.
Example Bundle
object configured to work with the plain provisioner
apiVersion: core.rukpak.io/v1alpha1 kind: Bundle metadata: name: my-bundle spec: source: type: image image: ref: my-bundle@sha256:xyz123 provisionerClassName: core-rukpak-io-plain
Bundles are considered immutable after they are created.
2.2.3.1.1. Bundle immutability
After a Bundle
object is accepted by the API server, the bundle is considered an immutable artifact by the rest of the RukPak system. This behavior enforces the notion that a bundle represents some unique, static piece of content to source onto the cluster. A user can have confidence that a particular bundle is pointing to a specific set of manifests and cannot be updated without creating a new bundle. This property is true for both standalone bundles and dynamic bundles created by an embedded BundleTemplate
object.
Bundle immutability is enforced by the core RukPak webhook. This webhook watches Bundle
object events and, for any update to a bundle, checks whether the spec
field of the existing bundle is semantically equal to that in the proposed updated bundle. If they are not equal, the update is rejected by the webhook. Other Bundle
object fields, such as metadata
or status
, are updated during the bundle’s lifecycle; it is only the spec
field that is considered immutable.
Applying a Bundle
object and then attempting to update its spec should fail. For example, the following example creates a bundle:
$ oc apply -f -<<EOF apiVersion: core.rukpak.io/v1alpha1 kind: Bundle metadata: name: combo-tag-ref spec: source: type: git git: ref: tag: v0.0.2 repository: https://github.com/operator-framework/combo provisionerClassName: core-rukpak-io-plain EOF
Example output
bundle.core.rukpak.io/combo-tag-ref created
Then, patching the bundle to point to a newer tag returns an error:
$ oc patch bundle combo-tag-ref --type='merge' -p '{"spec":{"source":{"git":{"ref":{"tag":"v0.0.3"}}}}}'
Example output
Error from server (bundle.spec is immutable): admission webhook "vbundles.core.rukpak.io" denied the request: bundle.spec is immutable
The core RukPak admission webhook rejected the patch because the spec of the bundle is immutable. The recommended method to change the content of a bundle is by creating a new Bundle
object instead of updating it in-place.
Further immutability considerations
While the spec
field of the Bundle
object is immutable, it is still possible for a BundleDeployment
object to pivot to a newer version of bundle content without changing the underlying spec
field. This unintentional pivoting could occur in the following scenario:
-
A user sets an image tag, a Git branch, or a Git tag in the
spec.source
field of theBundle
object. - The image tag moves to a new digest, a user pushes changes to a Git branch, or a user deletes and re-pushes a Git tag on a different commit.
- A user does something to cause the bundle unpack pod to be re-created, such as deleting the unpack pod.
If this scenario occurs, the new content from step 2 is unpacked as a result of step 3. The bundle deployment detects the changes and pivots to the newer version of the content.
This is similar to pod behavior, where one of the pod’s container images uses a tag, the tag is moved to a different digest, and then at some point in the future the existing pod is rescheduled on a different node. At that point, the node pulls the new image at the new digest and runs something different without the user explicitly asking for it.
To be confident that the underlying Bundle
spec content does not change, use a digest-based image or a Git commit reference when creating the bundle.
2.2.3.1.2. Plain bundle spec
A plain bundle in RukPak is a collection of static, arbitrary, Kubernetes YAML manifests in a given directory.
The currently implemented plain bundle format is the plain+v0
format. The name of the bundle format, plain+v0
, combines the type of bundle (plain
) with the current schema version (v0
).
The plain+v0
bundle format is at schema version v0
, which means it is an experimental format that is subject to change.
For example, the following shows the file tree in a plain+v0
bundle. It must have a manifests/
directory containing the Kubernetes resources required to deploy an application.
Example plain+v0
bundle file tree
$ tree manifests manifests ├── namespace.yaml ├── service_account.yaml ├── cluster_role.yaml ├── cluster_role_binding.yaml └── deployment.yaml
The static manifests must be located in the manifests/
directory with at least one resource in it for the bundle to be a valid plain+v0
bundle that the provisioner can unpack. The manifests/
directory must also be flat; all manifests must be at the top-level with no subdirectories.
Do not include any content in the manifests/
directory of a plain bundle that are not static manifests. Otherwise, a failure will occur when creating content on-cluster from that bundle. Any file that would not successfully apply with the oc apply
command will result in an error. Multi-object YAML or JSON files are valid, as well.
2.2.3.1.3. Registry bundle spec
A registry bundle, or registry+v1
bundle, contains a set of static Kubernetes YAML manifests organized in the legacy Operator Lifecycle Manager (OLM) bundle format.
Additional resources
2.2.3.2. BundleDeployment
A BundleDeployment
object changes the state of a Kubernetes cluster by installing and removing objects. It is important to verify and trust the content that is being installed and limit access, by using RBAC, to the BundleDeployment
API to only those who require those permissions.
The RukPak BundleDeployment
API points to a Bundle
object and indicates that it should be active. This includes pivoting from older versions of an active bundle. A BundleDeployment
object might also include an embedded spec for a desired bundle.
Much like pods generate instances of container images, a bundle deployment generates a deployed version of a bundle. A bundle deployment can be seen as a generalization of the pod concept.
The specifics of how a bundle deployment makes changes to a cluster based on a referenced bundle is defined by the provisioner that is configured to watch that bundle deployment.
Example BundleDeployment
object configured to work with the plain provisioner
apiVersion: core.rukpak.io/v1alpha1 kind: BundleDeployment metadata: name: my-bundle-deployment spec: provisionerClassName: core-rukpak-io-plain template: metadata: labels: app: my-bundle spec: source: type: image image: ref: my-bundle@sha256:xyz123 provisionerClassName: core-rukpak-io-plain
2.2.3.3. About provisioners
RukPak consists of a series of controllers, known as provisioners, that install and manage content on a Kubernetes cluster. RukPak also provides two primary APIs: Bundle
and BundleDeployment
. These components work together to bring content onto the cluster and install it, generating resources within the cluster.
Two provisioners are currently implemented and bundled with RukPak: the plain provisioner that sources and unpacks plain+v0
bundles, and the registry provisioner that sources and unpacks Operator Lifecycle Manager (OLM) registry+v1
bundles.
Each provisioner is assigned a unique ID and is responsible for reconciling Bundle
and BundleDeployment
objects with a spec.provisionerClassName
field that matches that particular ID. For example, the plain provisioner is able to unpack a given plain+v0
bundle onto a cluster and then instantiate it, making the content of the bundle available in the cluster.
A provisioner places a watch on both Bundle
and BundleDeployment
resources that refer to the provisioner explicitly. For a given bundle, the provisioner unpacks the contents of the Bundle
resource onto the cluster. Then, given a BundleDeployment
resource referring to that bundle, the provisioner installs the bundle contents and is responsible for managing the lifecycle of those resources.
2.3. Operator Framework glossary of common terms
This topic provides a glossary of common terms related to the Operator Framework, including Operator Lifecycle Manager (OLM) and the Operator SDK.
2.3.1. Common Operator Framework terms
2.3.1.1. Bundle
In the bundle format, a bundle is a collection of an Operator CSV, manifests, and metadata. Together, they form a unique version of an Operator that can be installed onto the cluster.
2.3.1.2. Bundle image
In the bundle format, a bundle image is a container image that is built from Operator manifests and that contains one bundle. Bundle images are stored and distributed by Open Container Initiative (OCI) spec container registries, such as Quay.io or DockerHub.
2.3.1.3. Catalog source
A catalog source represents a store of metadata that OLM can query to discover and install Operators and their dependencies.
2.3.1.4. Channel
A channel defines a stream of updates for an Operator and is used to roll out updates for subscribers. The head points to the latest version of that channel. For example, a stable
channel would have all stable versions of an Operator arranged from the earliest to the latest.
An Operator can have several channels, and a subscription binding to a certain channel would only look for updates in that channel.
2.3.1.5. Channel head
A channel head refers to the latest known update in a particular channel.
2.3.1.6. Cluster service version
A cluster service version (CSV) is a YAML manifest created from Operator metadata that assists OLM in running the Operator in a cluster. It is the metadata that accompanies an Operator container image, used to populate user interfaces with information such as its logo, description, and version.
It is also a source of technical information that is required to run the Operator, like the RBAC rules it requires and which custom resources (CRs) it manages or depends on.
2.3.1.7. Dependency
An Operator may have a dependency on another Operator being present in the cluster. For example, the Vault Operator has a dependency on the etcd Operator for its data persistence layer.
OLM resolves dependencies by ensuring that all specified versions of Operators and CRDs are installed on the cluster during the installation phase. This dependency is resolved by finding and installing an Operator in a catalog that satisfies the required CRD API, and is not related to packages or bundles.
2.3.1.8. Index image
In the bundle format, an index image refers to an image of a database (a database snapshot) that contains information about Operator bundles including CSVs and CRDs of all versions. This index can host a history of Operators on a cluster and be maintained by adding or removing Operators using the opm
CLI tool.
2.3.1.9. Install plan
An install plan is a calculated list of resources to be created to automatically install or upgrade a CSV.
2.3.1.10. Multitenancy
A tenant in OpenShift Container Platform is a user or group of users that share common access and privileges for a set of deployed workloads, typically represented by a namespace or project. You can use tenants to provide a level of isolation between different groups or teams.
When a cluster is shared by multiple users or groups, it is considered a multitenant cluster.
2.3.1.11. Operator group
An Operator group configures all Operators deployed in the same namespace as the OperatorGroup
object to watch for their CR in a list of namespaces or cluster-wide.
2.3.1.12. Package
In the bundle format, a package is a directory that encloses all released history of an Operator with each version. A released version of an Operator is described in a CSV manifest alongside the CRDs.
2.3.1.13. Registry
A registry is a database that stores bundle images of Operators, each with all of its latest and historical versions in all channels.
2.3.1.14. Subscription
A subscription keeps CSVs up to date by tracking a channel in a package.
2.3.1.15. Update graph
An update graph links versions of CSVs together, similar to the update graph of any other packaged software. Operators can be installed sequentially, or certain versions can be skipped. The update graph is expected to grow only at the head with newer versions being added.
2.4. Operator Lifecycle Manager (OLM)
2.4.1. Operator Lifecycle Manager concepts and resources
This guide provides an overview of the concepts that drive Operator Lifecycle Manager (OLM) in OpenShift Container Platform.
2.4.1.1. What is Operator Lifecycle Manager?
Operator Lifecycle Manager (OLM) helps users install, update, and manage the lifecycle of Kubernetes native applications (Operators) and their associated services running across their OpenShift Container Platform clusters. It is part of the Operator Framework, an open source toolkit designed to manage Operators in an effective, automated, and scalable way.
Figure 2.2. Operator Lifecycle Manager workflow
OLM runs by default in OpenShift Container Platform 4.15, which aids cluster administrators in installing, upgrading, and granting access to Operators running on their cluster. The OpenShift Container Platform web console provides management screens for cluster administrators to install Operators, as well as grant specific projects access to use the catalog of Operators available on the cluster.
For developers, a self-service experience allows provisioning and configuring instances of databases, monitoring, and big data services without having to be subject matter experts, because the Operator has that knowledge baked into it.
2.4.1.2. OLM resources
The following custom resource definitions (CRDs) are defined and managed by Operator Lifecycle Manager (OLM):
Resource | Short name | Description |
---|---|---|
|
| Application metadata. For example: name, version, icon, required resources. |
|
| A repository of CSVs, CRDs, and packages that define an application. |
|
| Keeps CSVs up to date by tracking a channel in a package. |
|
| Calculated list of resources to be created to automatically install or upgrade a CSV. |
|
|
Configures all Operators deployed in the same namespace as the |
| - |
Creates a communication channel between OLM and an Operator it manages. Operators can write to the |
2.4.1.2.1. Cluster service version
A cluster service version (CSV) represents a specific version of a running Operator on an OpenShift Container Platform cluster. It is a YAML manifest created from Operator metadata that assists Operator Lifecycle Manager (OLM) in running the Operator in the cluster.
OLM requires this metadata about an Operator to ensure that it can be kept running safely on a cluster, and to provide information about how updates should be applied as new versions of the Operator are published. This is similar to packaging software for a traditional operating system; think of the packaging step for OLM as the stage at which you make your rpm
, deb
, or apk
bundle.
A CSV includes the metadata that accompanies an Operator container image, used to populate user interfaces with information such as its name, version, description, labels, repository link, and logo.
A CSV is also a source of technical information required to run the Operator, such as which custom resources (CRs) it manages or depends on, RBAC rules, cluster requirements, and install strategies. This information tells OLM how to create required resources and set up the Operator as a deployment.
2.4.1.2.2. Catalog source
A catalog source represents a store of metadata, typically by referencing an index image stored in a container registry. Operator Lifecycle Manager (OLM) queries catalog sources to discover and install Operators and their dependencies. OperatorHub in the OpenShift Container Platform web console also displays the Operators provided by catalog sources.
Cluster administrators can view the full list of Operators provided by an enabled catalog source on a cluster by using the Administration → Cluster Settings → Configuration → OperatorHub page in the web console.
The spec
of a CatalogSource
object indicates how to construct a pod or how to communicate with a service that serves the Operator Registry gRPC API.
Example 2.9. Example CatalogSource
object
apiVersion: operators.coreos.com/v1alpha1 kind: CatalogSource metadata: generation: 1 name: example-catalog 1 namespace: openshift-marketplace 2 annotations: olm.catalogImageTemplate: 3 "quay.io/example-org/example-catalog:v{kube_major_version}.{kube_minor_version}.{kube_patch_version}" spec: displayName: Example Catalog 4 image: quay.io/example-org/example-catalog:v1 5 priority: -400 6 publisher: Example Org sourceType: grpc 7 grpcPodConfig: securityContextConfig: <security_mode> 8 nodeSelector: 9 custom_label: <label> priorityClassName: system-cluster-critical 10 tolerations: 11 - key: "key1" operator: "Equal" value: "value1" effect: "NoSchedule" updateStrategy: registryPoll: 12 interval: 30m0s status: connectionState: address: example-catalog.openshift-marketplace.svc:50051 lastConnect: 2021-08-26T18:14:31Z lastObservedState: READY 13 latestImageRegistryPoll: 2021-08-26T18:46:25Z 14 registryService: 15 createdAt: 2021-08-26T16:16:37Z port: 50051 protocol: grpc serviceName: example-catalog serviceNamespace: openshift-marketplace
- 1
- Name for the
CatalogSource
object. This value is also used as part of the name for the related pod that is created in the requested namespace. - 2
- Namespace to create the catalog in. To make the catalog available cluster-wide in all namespaces, set this value to
openshift-marketplace
. The default Red Hat-provided catalog sources also use theopenshift-marketplace
namespace. Otherwise, set the value to a specific namespace to make the Operator only available in that namespace. - 3
- Optional: To avoid cluster upgrades potentially leaving Operator installations in an unsupported state or without a continued update path, you can enable automatically changing your Operator catalog’s index image version as part of cluster upgrades.
Set the
olm.catalogImageTemplate
annotation to your index image name and use one or more of the Kubernetes cluster version variables as shown when constructing the template for the image tag. The annotation overwrites thespec.image
field at run time. See the "Image template for custom catalog sources" section for more details. - 4
- Display name for the catalog in the web console and CLI.
- 5
- Index image for the catalog. Optionally, can be omitted when using the
olm.catalogImageTemplate
annotation, which sets the pull spec at run time. - 6
- Weight for the catalog source. OLM uses the weight for prioritization during dependency resolution. A higher weight indicates the catalog is preferred over lower-weighted catalogs.
- 7
- Source types include the following:
-
grpc
with animage
reference: OLM pulls the image and runs the pod, which is expected to serve a compliant API. -
grpc
with anaddress
field: OLM attempts to contact the gRPC API at the given address. This should not be used in most cases. -
configmap
: OLM parses config map data and runs a pod that can serve the gRPC API over it.
-
- 8
- Specify the value of
legacy
orrestricted
. If the field is not set, the default value islegacy
. In a future OpenShift Container Platform release, it is planned that the default value will berestricted
. If your catalog cannot run withrestricted
permissions, it is recommended that you manually set this field tolegacy
. - 9
- Optional: For
grpc
type catalog sources, overrides the default node selector for the pod serving the content inspec.image
, if defined. - 10
- Optional: For
grpc
type catalog sources, overrides the default priority class name for the pod serving the content inspec.image
, if defined. Kubernetes providessystem-cluster-critical
andsystem-node-critical
priority classes by default. Setting the field to empty (""
) assigns the pod the default priority. Other priority classes can be defined manually. - 11
- Optional: For
grpc
type catalog sources, overrides the default tolerations for the pod serving the content inspec.image
, if defined. - 12
- Automatically check for new versions at a given interval to stay up-to-date.
- 13
- Last observed state of the catalog connection. For example:
-
READY
: A connection is successfully established. -
CONNECTING
: A connection is attempting to establish. -
TRANSIENT_FAILURE
: A temporary problem has occurred while attempting to establish a connection, such as a timeout. The state will eventually switch back toCONNECTING
and try again.
See States of Connectivity in the gRPC documentation for more details.
-
- 14
- Latest time the container registry storing the catalog image was polled to ensure the image is up-to-date.
- 15
- Status information for the catalog’s Operator Registry service.
Referencing the name
of a CatalogSource
object in a subscription instructs OLM where to search to find a requested Operator:
Example 2.10. Example Subscription
object referencing a catalog source
apiVersion: operators.coreos.com/v1alpha1 kind: Subscription metadata: name: example-operator namespace: example-namespace spec: channel: stable name: example-operator source: example-catalog sourceNamespace: openshift-marketplace
Additional resources
2.4.1.2.2.1. Image template for custom catalog sources
Operator compatibility with the underlying cluster can be expressed by a catalog source in various ways. One way, which is used for the default Red Hat-provided catalog sources, is to identify image tags for index images that are specifically created for a particular platform release, for example OpenShift Container Platform 4.15.
During a cluster upgrade, the index image tag for the default Red Hat-provided catalog sources are updated automatically by the Cluster Version Operator (CVO) so that Operator Lifecycle Manager (OLM) pulls the updated version of the catalog. For example during an upgrade from OpenShift Container Platform 4.14 to 4.15, the spec.image
field in the CatalogSource
object for the redhat-operators
catalog is updated from:
registry.redhat.io/redhat/redhat-operator-index:v4.14
to:
registry.redhat.io/redhat/redhat-operator-index:v4.15
However, the CVO does not automatically update image tags for custom catalogs. To ensure users are left with a compatible and supported Operator installation after a cluster upgrade, custom catalogs should also be kept updated to reference an updated index image.
Starting in OpenShift Container Platform 4.9, cluster administrators can add the olm.catalogImageTemplate
annotation in the CatalogSource
object for custom catalogs to an image reference that includes a template. The following Kubernetes version variables are supported for use in the template:
-
kube_major_version
-
kube_minor_version
-
kube_patch_version
You must specify the Kubernetes cluster version and not an OpenShift Container Platform cluster version, as the latter is not currently available for templating.
Provided that you have created and pushed an index image with a tag specifying the updated Kubernetes version, setting this annotation enables the index image versions in custom catalogs to be automatically changed after a cluster upgrade. The annotation value is used to set or update the image reference in the spec.image
field of the CatalogSource
object. This helps avoid cluster upgrades leaving Operator installations in unsupported states or without a continued update path.
You must ensure that the index image with the updated tag, in whichever registry it is stored in, is accessible by the cluster at the time of the cluster upgrade.
Example 2.11. Example catalog source with an image template
apiVersion: operators.coreos.com/v1alpha1 kind: CatalogSource metadata: generation: 1 name: example-catalog namespace: openshift-marketplace annotations: olm.catalogImageTemplate: "quay.io/example-org/example-catalog:v{kube_major_version}.{kube_minor_version}" spec: displayName: Example Catalog image: quay.io/example-org/example-catalog:v1.28 priority: -400 publisher: Example Org
If the spec.image
field and the olm.catalogImageTemplate
annotation are both set, the spec.image
field is overwritten by the resolved value from the annotation. If the annotation does not resolve to a usable pull spec, the catalog source falls back to the set spec.image
value.
If the spec.image
field is not set and the annotation does not resolve to a usable pull spec, OLM stops reconciliation of the catalog source and sets it into a human-readable error condition.
For an OpenShift Container Platform 4.15 cluster, which uses Kubernetes 1.28, the olm.catalogImageTemplate
annotation in the preceding example resolves to the following image reference:
quay.io/example-org/example-catalog:v1.28
For future releases of OpenShift Container Platform, you can create updated index images for your custom catalogs that target the later Kubernetes version that is used by the later OpenShift Container Platform version. With the olm.catalogImageTemplate
annotation set before the upgrade, upgrading the cluster to the later OpenShift Container Platform version would then automatically update the catalog’s index image as well.
2.4.1.2.2.2. Catalog health requirements
Operator catalogs on a cluster are interchangeable from the perspective of installation resolution; a Subscription
object might reference a specific catalog, but dependencies are resolved using all catalogs on the cluster.
For example, if Catalog A is unhealthy, a subscription referencing Catalog A could resolve a dependency in Catalog B, which the cluster administrator might not have been expecting, because B normally had a lower catalog priority than A.
As a result, OLM requires that all catalogs with a given global namespace (for example, the default openshift-marketplace
namespace or a custom global namespace) are healthy. When a catalog is unhealthy, all Operator installation or update operations within its shared global namespace will fail with a CatalogSourcesUnhealthy
condition. If these operations were permitted in an unhealthy state, OLM might make resolution and installation decisions that were unexpected to the cluster administrator.
As a cluster administrator, if you observe an unhealthy catalog and want to consider the catalog as invalid and resume Operator installations, see the "Removing custom catalogs" or "Disabling the default OperatorHub catalog sources" sections for information about removing the unhealthy catalog.
Additional resources
2.4.1.2.3. Subscription
A subscription, defined by a Subscription
object, represents an intention to install an Operator. It is the custom resource that relates an Operator to a catalog source.
Subscriptions describe which channel of an Operator package to subscribe to, and whether to perform updates automatically or manually. If set to automatic, the subscription ensures Operator Lifecycle Manager (OLM) manages and upgrades the Operator to ensure that the latest version is always running in the cluster.
Example Subscription
object
apiVersion: operators.coreos.com/v1alpha1 kind: Subscription metadata: name: example-operator namespace: example-namespace spec: channel: stable name: example-operator source: example-catalog sourceNamespace: openshift-marketplace
This Subscription
object defines the name and namespace of the Operator, as well as the catalog from which the Operator data can be found. The channel, such as alpha
, beta
, or stable
, helps determine which Operator stream should be installed from the catalog source.
The names of channels in a subscription can differ between Operators, but the naming scheme should follow a common convention within a given Operator. For example, channel names might follow a minor release update stream for the application provided by the Operator (1.2
, 1.3
) or a release frequency (stable
, fast
).
In addition to being easily visible from the OpenShift Container Platform web console, it is possible to identify when there is a newer version of an Operator available by inspecting the status of the related subscription. The value associated with the currentCSV
field is the newest version that is known to OLM, and installedCSV
is the version that is installed on the cluster.
2.4.1.2.4. Install plan
An install plan, defined by an InstallPlan
object, describes a set of resources that Operator Lifecycle Manager (OLM) creates to install or upgrade to a specific version of an Operator. The version is defined by a cluster service version (CSV).
To install an Operator, a cluster administrator, or a user who has been granted Operator installation permissions, must first create a Subscription
object. A subscription represents the intent to subscribe to a stream of available versions of an Operator from a catalog source. The subscription then creates an InstallPlan
object to facilitate the installation of the resources for the Operator.
The install plan must then be approved according to one of the following approval strategies:
-
If the subscription’s
spec.installPlanApproval
field is set toAutomatic
, the install plan is approved automatically. -
If the subscription’s
spec.installPlanApproval
field is set toManual
, the install plan must be manually approved by a cluster administrator or user with proper permissions.
After the install plan is approved, OLM creates the specified resources and installs the Operator in the namespace that is specified by the subscription.
Example 2.12. Example InstallPlan
object
apiVersion: operators.coreos.com/v1alpha1 kind: InstallPlan metadata: name: install-abcde namespace: operators spec: approval: Automatic approved: true clusterServiceVersionNames: - my-operator.v1.0.1 generation: 1 status: ... catalogSources: [] conditions: - lastTransitionTime: '2021-01-01T20:17:27Z' lastUpdateTime: '2021-01-01T20:17:27Z' status: 'True' type: Installed phase: Complete plan: - resolving: my-operator.v1.0.1 resource: group: operators.coreos.com kind: ClusterServiceVersion manifest: >- ... name: my-operator.v1.0.1 sourceName: redhat-operators sourceNamespace: openshift-marketplace version: v1alpha1 status: Created - resolving: my-operator.v1.0.1 resource: group: apiextensions.k8s.io kind: CustomResourceDefinition manifest: >- ... name: webservers.web.servers.org sourceName: redhat-operators sourceNamespace: openshift-marketplace version: v1beta1 status: Created - resolving: my-operator.v1.0.1 resource: group: '' kind: ServiceAccount manifest: >- ... name: my-operator sourceName: redhat-operators sourceNamespace: openshift-marketplace version: v1 status: Created - resolving: my-operator.v1.0.1 resource: group: rbac.authorization.k8s.io kind: Role manifest: >- ... name: my-operator.v1.0.1-my-operator-6d7cbc6f57 sourceName: redhat-operators sourceNamespace: openshift-marketplace version: v1 status: Created - resolving: my-operator.v1.0.1 resource: group: rbac.authorization.k8s.io kind: RoleBinding manifest: >- ... name: my-operator.v1.0.1-my-operator-6d7cbc6f57 sourceName: redhat-operators sourceNamespace: openshift-marketplace version: v1 status: Created ...
2.4.1.2.5. Operator groups
An Operator group, defined by the OperatorGroup
resource, provides multitenant configuration to OLM-installed Operators. An Operator group selects target namespaces in which to generate required RBAC access for its member Operators.
The set of target namespaces is provided by a comma-delimited string stored in the olm.targetNamespaces
annotation of a cluster service version (CSV). This annotation is applied to the CSV instances of member Operators and is projected into their deployments.
Additional resources
2.4.1.2.6. Operator conditions
As part of its role in managing the lifecycle of an Operator, Operator Lifecycle Manager (OLM) infers the state of an Operator from the state of Kubernetes resources that define the Operator. While this approach provides some level of assurance that an Operator is in a given state, there are many instances where an Operator might need to communicate information to OLM that could not be inferred otherwise. This information can then be used by OLM to better manage the lifecycle of the Operator.
OLM provides a custom resource definition (CRD) called OperatorCondition
that allows Operators to communicate conditions to OLM. There are a set of supported conditions that influence management of the Operator by OLM when present in the Spec.Conditions
array of an OperatorCondition
resource.
By default, the Spec.Conditions
array is not present in an OperatorCondition
object until it is either added by a user or as a result of custom Operator logic.
Additional resources
2.4.2. Operator Lifecycle Manager architecture
This guide outlines the component architecture of Operator Lifecycle Manager (OLM) in OpenShift Container Platform.
2.4.2.1. Component responsibilities
Operator Lifecycle Manager (OLM) is composed of two Operators: the OLM Operator and the Catalog Operator.
Each of these Operators is responsible for managing the custom resource definitions (CRDs) that are the basis for the OLM framework:
Resource | Short name | Owner | Description |
---|---|---|---|
|
| OLM | Application metadata: name, version, icon, required resources, installation, and so on. |
|
| Catalog | Calculated list of resources to be created to automatically install or upgrade a CSV. |
|
| Catalog | A repository of CSVs, CRDs, and packages that define an application. |
|
| Catalog | Used to keep CSVs up to date by tracking a channel in a package. |
|
| OLM |
Configures all Operators deployed in the same namespace as the |
Each of these Operators is also responsible for creating the following resources:
Resource | Owner |
---|---|
| OLM |
| |
| |
| |
| Catalog |
|
2.4.2.2. OLM Operator
The OLM Operator is responsible for deploying applications defined by CSV resources after the required resources specified in the CSV are present in the cluster.
The OLM Operator is not concerned with the creation of the required resources; you can choose to manually create these resources using the CLI or using the Catalog Operator. This separation of concern allows users incremental buy-in in terms of how much of the OLM framework they choose to leverage for their application.
The OLM Operator uses the following workflow:
- Watch for cluster service versions (CSVs) in a namespace and check that requirements are met.
If requirements are met, run the install strategy for the CSV.
NoteA CSV must be an active member of an Operator group for the install strategy to run.
2.4.2.3. Catalog Operator
The Catalog Operator is responsible for resolving and installing cluster service versions (CSVs) and the required resources they specify. It is also responsible for watching catalog sources for updates to packages in channels and upgrading them, automatically if desired, to the latest available versions.
To track a package in a channel, you can create a Subscription
object configuring the desired package, channel, and the CatalogSource
object you want to use for pulling updates. When updates are found, an appropriate InstallPlan
object is written into the namespace on behalf of the user.
The Catalog Operator uses the following workflow:
- Connect to each catalog source in the cluster.
Watch for unresolved install plans created by a user, and if found:
- Find the CSV matching the name requested and add the CSV as a resolved resource.
- For each managed or required CRD, add the CRD as a resolved resource.
- For each required CRD, find the CSV that manages it.
- Watch for resolved install plans and create all of the discovered resources for it, if approved by a user or automatically.
- Watch for catalog sources and subscriptions and create install plans based on them.
2.4.2.4. Catalog Registry
The Catalog Registry stores CSVs and CRDs for creation in a cluster and stores metadata about packages and channels.
A package manifest is an entry in the Catalog Registry that associates a package identity with sets of CSVs. Within a package, channels point to a particular CSV. Because CSVs explicitly reference the CSV that they replace, a package manifest provides the Catalog Operator with all of the information that is required to update a CSV to the latest version in a channel, stepping through each intermediate version.
2.4.3. Operator Lifecycle Manager workflow
This guide outlines the workflow of Operator Lifecycle Manager (OLM) in OpenShift Container Platform.
2.4.3.1. Operator installation and upgrade workflow in OLM
In the Operator Lifecycle Manager (OLM) ecosystem, the following resources are used to resolve Operator installations and upgrades:
-
ClusterServiceVersion
(CSV) -
CatalogSource
-
Subscription
Operator metadata, defined in CSVs, can be stored in a collection called a catalog source. OLM uses catalog sources, which use the Operator Registry API, to query for available Operators as well as upgrades for installed Operators.
Figure 2.3. Catalog source overview
Within a catalog source, Operators are organized into packages and streams of updates called channels, which should be a familiar update pattern from OpenShift Container Platform or other software on a continuous release cycle like web browsers.
Figure 2.4. Packages and channels in a Catalog source
A user indicates a particular package and channel in a particular catalog source in a subscription, for example an etcd
package and its alpha
channel. If a subscription is made to a package that has not yet been installed in the namespace, the latest Operator for that package is installed.
OLM deliberately avoids version comparisons, so the "latest" or "newest" Operator available from a given catalog → channel → package path does not necessarily need to be the highest version number. It should be thought of more as the head reference of a channel, similar to a Git repository.
Each CSV has a replaces
parameter that indicates which Operator it replaces. This builds a graph of CSVs that can be queried by OLM, and updates can be shared between channels. Channels can be thought of as entry points into the graph of updates:
Figure 2.5. OLM graph of available channel updates
Example channels in a package
packageName: example channels: - name: alpha currentCSV: example.v0.1.2 - name: beta currentCSV: example.v0.1.3 defaultChannel: alpha
For OLM to successfully query for updates, given a catalog source, package, channel, and CSV, a catalog must be able to return, unambiguously and deterministically, a single CSV that replaces
the input CSV.
2.4.3.1.1. Example upgrade path
For an example upgrade scenario, consider an installed Operator corresponding to CSV version 0.1.1
. OLM queries the catalog source and detects an upgrade in the subscribed channel with new CSV version 0.1.3
that replaces an older but not-installed CSV version 0.1.2
, which in turn replaces the older and installed CSV version 0.1.1
.
OLM walks back from the channel head to previous versions via the replaces
field specified in the CSVs to determine the upgrade path 0.1.3
→ 0.1.2
→ 0.1.1
; the direction of the arrow indicates that the former replaces the latter. OLM upgrades the Operator one version at the time until it reaches the channel head.
For this given scenario, OLM installs Operator version 0.1.2
to replace the existing Operator version 0.1.1
. Then, it installs Operator version 0.1.3
to replace the previously installed Operator version 0.1.2
. At this point, the installed operator version 0.1.3
matches the channel head and the upgrade is completed.
2.4.3.1.2. Skipping upgrades
The basic path for upgrades in OLM is:
- A catalog source is updated with one or more updates to an Operator.
- OLM traverses every version of the Operator until reaching the latest version the catalog source contains.
However, sometimes this is not a safe operation to perform. There will be cases where a published version of an Operator should never be installed on a cluster if it has not already, for example because a version introduces a serious vulnerability.
In those cases, OLM must consider two cluster states and provide an update graph that supports both:
- The "bad" intermediate Operator has been seen by the cluster and installed.
- The "bad" intermediate Operator has not yet been installed onto the cluster.
By shipping a new catalog and adding a skipped release, OLM is ensured that it can always get a single unique update regardless of the cluster state and whether it has seen the bad update yet.
Example CSV with skipped release
apiVersion: operators.coreos.com/v1alpha1 kind: ClusterServiceVersion metadata: name: etcdoperator.v0.9.2 namespace: placeholder annotations: spec: displayName: etcd description: Etcd Operator replaces: etcdoperator.v0.9.0 skips: - etcdoperator.v0.9.1
Consider the following example of Old CatalogSource and New CatalogSource.
Figure 2.6. Skipping updates
This graph maintains that:
- Any Operator found in Old CatalogSource has a single replacement in New CatalogSource.
- Any Operator found in New CatalogSource has a single replacement in New CatalogSource.
- If the bad update has not yet been installed, it will never be.
2.4.3.1.3. Replacing multiple Operators
Creating New CatalogSource as described requires publishing CSVs that replace
one Operator, but can skip
several. This can be accomplished using the skipRange
annotation:
olm.skipRange: <semver_range>
where <semver_range>
has the version range format supported by the semver library.
When searching catalogs for updates, if the head of a channel has a skipRange
annotation and the currently installed Operator has a version field that falls in the range, OLM updates to the latest entry in the channel.
The order of precedence is:
-
Channel head in the source specified by
sourceName
on the subscription, if the other criteria for skipping are met. -
The next Operator that replaces the current one, in the source specified by
sourceName
. - Channel head in another source that is visible to the subscription, if the other criteria for skipping are met.
- The next Operator that replaces the current one in any source visible to the subscription.
Example CSV with skipRange
apiVersion: operators.coreos.com/v1alpha1 kind: ClusterServiceVersion metadata: name: elasticsearch-operator.v4.1.2 namespace: <namespace> annotations: olm.skipRange: '>=4.1.0 <4.1.2'
2.4.3.1.4. Z-stream support
A z-stream, or patch release, must replace all previous z-stream releases for the same minor version. OLM does not consider major, minor, or patch versions, it just needs to build the correct graph in a catalog.
In other words, OLM must be able to take a graph as in Old CatalogSource and, similar to before, generate a graph as in New CatalogSource:
Figure 2.7. Replacing several Operators
This graph maintains that:
- Any Operator found in Old CatalogSource has a single replacement in New CatalogSource.
- Any Operator found in New CatalogSource has a single replacement in New CatalogSource.
- Any z-stream release in Old CatalogSource will update to the latest z-stream release in New CatalogSource.
- Unavailable releases can be considered "virtual" graph nodes; their content does not need to exist, the registry just needs to respond as if the graph looks like this.
2.4.4. Operator Lifecycle Manager dependency resolution
This guide outlines dependency resolution and custom resource definition (CRD) upgrade lifecycles with Operator Lifecycle Manager (OLM) in OpenShift Container Platform.
2.4.4.1. About dependency resolution
Operator Lifecycle Manager (OLM) manages the dependency resolution and upgrade lifecycle of running Operators. In many ways, the problems OLM faces are similar to other system or language package managers, such as yum
and rpm
.
However, there is one constraint that similar systems do not generally have that OLM does: because Operators are always running, OLM attempts to ensure that you are never left with a set of Operators that do not work with each other.
As a result, OLM must never create the following scenarios:
- Install a set of Operators that require APIs that cannot be provided
- Update an Operator in a way that breaks another that depends upon it
This is made possible with two types of data:
Properties | Typed metadata about the Operator that constitutes the public interface for it in the dependency resolver. Examples include the group/version/kind (GVK) of the APIs provided by the Operator and the semantic version (semver) of the Operator. |
Constraints or dependencies | An Operator’s requirements that should be satisfied by other Operators that might or might not have already been installed on the target cluster. These act as queries or filters over all available Operators and constrain the selection during dependency resolution and installation. Examples include requiring a specific API to be available on the cluster or expecting a particular Operator with a particular version to be installed. |
OLM converts these properties and constraints into a system of Boolean formulas and passes them to a SAT solver, a program that establishes Boolean satisfiability, which does the work of determining what Operators should be installed.
2.4.4.2. Operator properties
All Operators in a catalog have the following properties:
olm.package
- Includes the name of the package and the version of the Operator
olm.gvk
- A single property for each provided API from the cluster service version (CSV)
Additional properties can also be directly declared by an Operator author by including a properties.yaml
file in the metadata/
directory of the Operator bundle.
Example arbitrary property
properties: - type: olm.kubeversion value: version: "1.16.0"
2.4.4.2.1. Arbitrary properties
Operator authors can declare arbitrary properties in a properties.yaml
file in the metadata/
directory of the Operator bundle. These properties are translated into a map data structure that is used as an input to the Operator Lifecycle Manager (OLM) resolver at runtime.
These properties are opaque to the resolver as it does not understand the properties, but it can evaluate the generic constraints against those properties to determine if the constraints can be satisfied given the properties list.
Example arbitrary properties
properties: - property: type: color value: red - property: type: shape value: square - property: type: olm.gvk value: group: olm.coreos.io version: v1alpha1 kind: myresource
This structure can be used to construct a Common Expression Language (CEL) expression for generic constraints.
Additional resources
2.4.4.3. Operator dependencies
The dependencies of an Operator are listed in a dependencies.yaml
file in the metadata/
folder of a bundle. This file is optional and currently only used to specify explicit Operator-version dependencies.
The dependency list contains a type
field for each item to specify what kind of dependency this is. The following types of Operator dependencies are supported:
olm.package
-
This type indicates a dependency for a specific Operator version. The dependency information must include the package name and the version of the package in semver format. For example, you can specify an exact version such as
0.5.2
or a range of versions such as>0.5.1
. olm.gvk
- With this type, the author can specify a dependency with group/version/kind (GVK) information, similar to existing CRD and API-based usage in a CSV. This is a path to enable Operator authors to consolidate all dependencies, API or explicit versions, to be in the same place.
olm.constraint
- This type declares generic constraints on arbitrary Operator properties.
In the following example, dependencies are specified for a Prometheus Operator and etcd CRDs:
Example dependencies.yaml
file
dependencies: - type: olm.package value: packageName: prometheus version: ">0.27.0" - type: olm.gvk value: group: etcd.database.coreos.com kind: EtcdCluster version: v1beta2
2.4.4.4. Generic constraints
An olm.constraint
property declares a dependency constraint of a particular type, differentiating non-constraint and constraint properties. Its value
field is an object containing a failureMessage
field holding a string-representation of the constraint message. This message is surfaced as an informative comment to users if the constraint is not satisfiable at runtime.
The following keys denote the available constraint types:
gvk
-
Type whose value and interpretation is identical to the
olm.gvk
type package
-
Type whose value and interpretation is identical to the
olm.package
type cel
- A Common Expression Language (CEL) expression evaluated at runtime by the Operator Lifecycle Manager (OLM) resolver over arbitrary bundle properties and cluster information
all
,any
,not
-
Conjunction, disjunction, and negation constraints, respectively, containing one or more concrete constraints, such as
gvk
or a nested compound constraint
2.4.4.4.1. Common Expression Language (CEL) constraints
The cel
constraint type supports Common Expression Language (CEL) as the expression language. The cel
struct has a rule
field which contains the CEL expression string that is evaluated against Operator properties at runtime to determine if the Operator satisfies the constraint.
Example cel
constraint
type: olm.constraint value: failureMessage: 'require to have "certified"' cel: rule: 'properties.exists(p, p.type == "certified")'
The CEL syntax supports a wide range of logical operators, such as AND
and OR
. As a result, a single CEL expression can have multiple rules for multiple conditions that are linked together by these logical operators. These rules are evaluated against a dataset of multiple different properties from a bundle or any given source, and the output is solved into a single bundle or Operator that satisfies all of those rules within a single constraint.
Example cel
constraint with multiple rules
type: olm.constraint value: failureMessage: 'require to have "certified" and "stable" properties' cel: rule: 'properties.exists(p, p.type == "certified") && properties.exists(p, p.type == "stable")'
2.4.4.4.2. Compound constraints (all, any, not)
Compound constraint types are evaluated following their logical definitions.
The following is an example of a conjunctive constraint (all
) of two packages and one GVK. That is, they must all be satisfied by installed bundles:
Example all
constraint
schema: olm.bundle name: red.v1.0.0 properties: - type: olm.constraint value: failureMessage: All are required for Red because... all: constraints: - failureMessage: Package blue is needed for... package: name: blue versionRange: '>=1.0.0' - failureMessage: GVK Green/v1 is needed for... gvk: group: greens.example.com version: v1 kind: Green
The following is an example of a disjunctive constraint (any
) of three versions of the same GVK. That is, at least one must be satisfied by installed bundles:
Example any
constraint
schema: olm.bundle name: red.v1.0.0 properties: - type: olm.constraint value: failureMessage: Any are required for Red because... any: constraints: - gvk: group: blues.example.com version: v1beta1 kind: Blue - gvk: group: blues.example.com version: v1beta2 kind: Blue - gvk: group: blues.example.com version: v1 kind: Blue
The following is an example of a negation constraint (not
) of one version of a GVK. That is, this GVK cannot be provided by any bundle in the result set:
Example not
constraint
schema: olm.bundle name: red.v1.0.0 properties: - type: olm.constraint value: all: constraints: - failureMessage: Package blue is needed for... package: name: blue versionRange: '>=1.0.0' - failureMessage: Cannot be required for Red because... not: constraints: - gvk: group: greens.example.com version: v1alpha1 kind: greens
The negation semantics might appear unclear in the not
constraint context. To clarify, the negation is really instructing the resolver to remove any possible solution that includes a particular GVK, package at a version, or satisfies some child compound constraint from the result set.
As a corollary, the not
compound constraint should only be used within all
or any
constraints, because negating without first selecting a possible set of dependencies does not make sense.
2.4.4.4.3. Nested compound constraints
A nested compound constraint, one that contains at least one child compound constraint along with zero or more simple constraints, is evaluated from the bottom up following the procedures for each previously described constraint type.
The following is an example of a disjunction of conjunctions, where one, the other, or both can satisfy the constraint:
Example nested compound constraint
schema: olm.bundle name: red.v1.0.0 properties: - type: olm.constraint value: failureMessage: Required for Red because... any: constraints: - all: constraints: - package: name: blue versionRange: '>=1.0.0' - gvk: group: blues.example.com version: v1 kind: Blue - all: constraints: - package: name: blue versionRange: '<1.0.0' - gvk: group: blues.example.com version: v1beta1 kind: Blue
The maximum raw size of an olm.constraint
type is 64KB to limit resource exhaustion attacks.
2.4.4.5. Dependency preferences
There can be many options that equally satisfy a dependency of an Operator. The dependency resolver in Operator Lifecycle Manager (OLM) determines which option best fits the requirements of the requested Operator. As an Operator author or user, it can be important to understand how these choices are made so that dependency resolution is clear.
2.4.4.5.1. Catalog priority
On OpenShift Container Platform cluster, OLM reads catalog sources to know which Operators are available for installation.
Example CatalogSource
object
apiVersion: "operators.coreos.com/v1alpha1"
kind: "CatalogSource"
metadata:
name: "my-operators"
namespace: "operators"
spec:
sourceType: grpc
grpcPodConfig:
securityContextConfig: <security_mode> 1
image: example.com/my/operator-index:v1
displayName: "My Operators"
priority: 100
- 1
- Specify the value of
legacy
orrestricted
. If the field is not set, the default value islegacy
. In a future OpenShift Container Platform release, it is planned that the default value will berestricted
. If your catalog cannot run withrestricted
permissions, it is recommended that you manually set this field tolegacy
.
A CatalogSource
object has a priority
field, which is used by the resolver to know how to prefer options for a dependency.
There are two rules that govern catalog preference:
- Options in higher-priority catalogs are preferred to options in lower-priority catalogs.
- Options in the same catalog as the dependent are preferred to any other catalogs.
2.4.4.5.2. Channel ordering
An Operator package in a catalog is a collection of update channels that a user can subscribe to in an OpenShift Container Platform cluster. Channels can be used to provide a particular stream of updates for a minor release (1.2
, 1.3
) or a release frequency (stable
, fast
).
It is likely that a dependency might be satisfied by Operators in the same package, but different channels. For example, version 1.2
of an Operator might exist in both the stable
and fast
channels.
Each package has a default channel, which is always preferred to non-default channels. If no option in the default channel can satisfy a dependency, options are considered from the remaining channels in lexicographic order of the channel name.
2.4.4.5.3. Order within a channel
There are almost always multiple options to satisfy a dependency within a single channel. For example, Operators in one package and channel provide the same set of APIs.
When a user creates a subscription, they indicate which channel to receive updates from. This immediately reduces the search to just that one channel. But within the channel, it is likely that many Operators satisfy a dependency.
Within a channel, newer Operators that are higher up in the update graph are preferred. If the head of a channel satisfies a dependency, it will be tried first.
2.4.4.5.4. Other constraints
In addition to the constraints supplied by package dependencies, OLM includes additional constraints to represent the desired user state and enforce resolution invariants.
2.4.4.5.4.1. Subscription constraint
A subscription constraint filters the set of Operators that can satisfy a subscription. Subscriptions are user-supplied constraints for the dependency resolver. They declare the intent to either install a new Operator if it is not already on the cluster, or to keep an existing Operator updated.
2.4.4.5.4.2. Package constraint
Within a namespace, no two Operators may come from the same package.
2.4.4.5.5. Additional resources
2.4.4.6. CRD upgrades
OLM upgrades a custom resource definition (CRD) immediately if it is owned by a singular cluster service version (CSV). If a CRD is owned by multiple CSVs, then the CRD is upgraded when it has satisfied all of the following backward compatible conditions:
- All existing serving versions in the current CRD are present in the new CRD.
- All existing instances, or custom resources, that are associated with the serving versions of the CRD are valid when validated against the validation schema of the new CRD.
Additional resources
2.4.4.7. Dependency best practices
When specifying dependencies, there are best practices you should consider.
- Depend on APIs or a specific version range of Operators
-
Operators can add or remove APIs at any time; always specify an
olm.gvk
dependency on any APIs your Operators requires. The exception to this is if you are specifyingolm.package
constraints instead. - Set a minimum version
The Kubernetes documentation on API changes describes what changes are allowed for Kubernetes-style Operators. These versioning conventions allow an Operator to update an API without bumping the API version, as long as the API is backwards-compatible.
For Operator dependencies, this means that knowing the API version of a dependency might not be enough to ensure the dependent Operator works as intended.
For example:
-
TestOperator v1.0.0 provides v1alpha1 API version of the
MyObject
resource. -
TestOperator v1.0.1 adds a new field
spec.newfield
toMyObject
, but still at v1alpha1.
Your Operator might require the ability to write
spec.newfield
into theMyObject
resource. Anolm.gvk
constraint alone is not enough for OLM to determine that you need TestOperator v1.0.1 and not TestOperator v1.0.0.Whenever possible, if a specific Operator that provides an API is known ahead of time, specify an additional
olm.package
constraint to set a minimum.-
TestOperator v1.0.0 provides v1alpha1 API version of the
- Omit a maximum version or allow a very wide range
Because Operators provide cluster-scoped resources such as API services and CRDs, an Operator that specifies a small window for a dependency might unnecessarily constrain updates for other consumers of that dependency.
Whenever possible, do not set a maximum version. Alternatively, set a very wide semantic range to prevent conflicts with other Operators. For example,
>1.0.0 <2.0.0
.Unlike with conventional package managers, Operator authors explicitly encode that updates are safe through channels in OLM. If an update is available for an existing subscription, it is assumed that the Operator author is indicating that it can update from the previous version. Setting a maximum version for a dependency overrides the update stream of the author by unnecessarily truncating it at a particular upper bound.
NoteCluster administrators cannot override dependencies set by an Operator author.
However, maximum versions can and should be set if there are known incompatibilities that must be avoided. Specific versions can be omitted with the version range syntax, for example
> 1.0.0 !1.2.1
.
Additional resources
- Kubernetes documentation: Changing the API
2.4.4.8. Dependency caveats
When specifying dependencies, there are caveats you should consider.
- No compound constraints (AND)
There is currently no method for specifying an AND relationship between constraints. In other words, there is no way to specify that one Operator depends on another Operator that both provides a given API and has version
>1.1.0
.This means that when specifying a dependency such as:
dependencies: - type: olm.package value: packageName: etcd version: ">3.1.0" - type: olm.gvk value: group: etcd.database.coreos.com kind: EtcdCluster version: v1beta2
It would be possible for OLM to satisfy this with two Operators: one that provides EtcdCluster and one that has version
>3.1.0
. Whether that happens, or whether an Operator is selected that satisfies both constraints, depends on the ordering that potential options are visited. Dependency preferences and ordering options are well-defined and can be reasoned about, but to exercise caution, Operators should stick to one mechanism or the other.- Cross-namespace compatibility
- OLM performs dependency resolution at the namespace scope. It is possible to get into an update deadlock if updating an Operator in one namespace would be an issue for an Operator in another namespace, and vice-versa.
2.4.4.9. Example dependency resolution scenarios
In the following examples, a provider is an Operator which "owns" a CRD or API service.
Example: Deprecating dependent APIs
A and B are APIs (CRDs):
- The provider of A depends on B.
- The provider of B has a subscription.
- The provider of B updates to provide C but deprecates B.
This results in:
- B no longer has a provider.
- A no longer works.
This is a case OLM prevents with its upgrade strategy.
Example: Version deadlock
A and B are APIs:
- The provider of A requires B.
- The provider of B requires A.
- The provider of A updates to (provide A2, require B2) and deprecate A.
- The provider of B updates to (provide B2, require A2) and deprecate B.
If OLM attempts to update A without simultaneously updating B, or vice-versa, it is unable to progress to new versions of the Operators, even though a new compatible set can be found.
This is another case OLM prevents with its upgrade strategy.
2.4.5. Operator groups
This guide outlines the use of Operator groups with Operator Lifecycle Manager (OLM) in OpenShift Container Platform.
2.4.5.1. About Operator groups
An Operator group, defined by the OperatorGroup
resource, provides multitenant configuration to OLM-installed Operators. An Operator group selects target namespaces in which to generate required RBAC access for its member Operators.
The set of target namespaces is provided by a comma-delimited string stored in the olm.targetNamespaces
annotation of a cluster service version (CSV). This annotation is applied to the CSV instances of member Operators and is projected into their deployments.
2.4.5.2. Operator group membership
An Operator is considered a member of an Operator group if the following conditions are true:
- The CSV of the Operator exists in the same namespace as the Operator group.
- The install modes in the CSV of the Operator support the set of namespaces targeted by the Operator group.
An install mode in a CSV consists of an InstallModeType
field and a boolean Supported
field. The spec of a CSV can contain a set of install modes of four distinct InstallModeTypes
:
InstallModeType | Description |
---|---|
| The Operator can be a member of an Operator group that selects its own namespace. |
| The Operator can be a member of an Operator group that selects one namespace. |
| The Operator can be a member of an Operator group that selects more than one namespace. |
|
The Operator can be a member of an Operator group that selects all namespaces (target namespace set is the empty string |
If the spec of a CSV omits an entry of InstallModeType
, then that type is considered unsupported unless support can be inferred by an existing entry that implicitly supports it.
2.4.5.3. Target namespace selection
You can explicitly name the target namespace for an Operator group using the spec.targetNamespaces
parameter:
apiVersion: operators.coreos.com/v1 kind: OperatorGroup metadata: name: my-group namespace: my-namespace spec: targetNamespaces: - my-namespace
You can alternatively specify a namespace using a label selector with the spec.selector
parameter:
apiVersion: operators.coreos.com/v1 kind: OperatorGroup metadata: name: my-group namespace: my-namespace spec: selector: cool.io/prod: "true"
Listing multiple namespaces via spec.targetNamespaces
or use of a label selector via spec.selector
is not recommended, as the support for more than one target namespace in an Operator group will likely be removed in a future release.
If both spec.targetNamespaces
and spec.selector
are defined, spec.selector
is ignored. Alternatively, you can omit both spec.selector
and spec.targetNamespaces
to specify a global Operator group, which selects all namespaces:
apiVersion: operators.coreos.com/v1 kind: OperatorGroup metadata: name: my-group namespace: my-namespace
The resolved set of selected namespaces is shown in the status.namespaces
parameter of an Opeator group. The status.namespace
of a global Operator group contains the empty string (""
), which signals to a consuming Operator that it should watch all namespaces.
2.4.5.4. Operator group CSV annotations
Member CSVs of an Operator group have the following annotations:
Annotation | Description |
---|---|
| Contains the name of the Operator group. |
| Contains the namespace of the Operator group. |
| Contains a comma-delimited string that lists the target namespace selection of the Operator group. |
All annotations except olm.targetNamespaces
are included with copied CSVs. Omitting the olm.targetNamespaces
annotation on copied CSVs prevents the duplication of target namespaces between tenants.
2.4.5.5. Provided APIs annotation
A group/version/kind (GVK) is a unique identifier for a Kubernetes API. Information about what GVKs are provided by an Operator group are shown in an olm.providedAPIs
annotation. The value of the annotation is a string consisting of <kind>.<version>.<group>
delimited with commas. The GVKs of CRDs and API services provided by all active member CSVs of an Operator group are included.
Review the following example of an OperatorGroup
object with a single active member CSV that provides the PackageManifest
resource:
apiVersion: operators.coreos.com/v1 kind: OperatorGroup metadata: annotations: olm.providedAPIs: PackageManifest.v1alpha1.packages.apps.redhat.com name: olm-operators namespace: local ... spec: selector: {} serviceAccount: metadata: creationTimestamp: null targetNamespaces: - local status: lastUpdated: 2019-02-19T16:18:28Z namespaces: - local
2.4.5.6. Role-based access control
When an Operator group is created, three cluster roles are generated. Each contains a single aggregation rule with a cluster role selector set to match a label, as shown below:
Cluster role | Label to match |
---|---|
|
|
|
|
|
|
The following RBAC resources are generated when a CSV becomes an active member of an Operator group, as long as the CSV is watching all namespaces with the AllNamespaces
install mode and is not in a failed state with reason InterOperatorGroupOwnerConflict
:
- Cluster roles for each API resource from a CRD
- Cluster roles for each API resource from an API service
- Additional roles and role bindings
Cluster role | Settings |
---|---|
|
Verbs on
Aggregation labels:
|
|
Verbs on
Aggregation labels:
|
|
Verbs on
Aggregation labels:
|
|
Verbs on
Aggregation labels:
|
Cluster role | Settings |
---|---|
|
Verbs on
Aggregation labels:
|
|
Verbs on
Aggregation labels:
|
|
Verbs on
Aggregation labels:
|
Additional roles and role bindings
-
If the CSV defines exactly one target namespace that contains
*
, then a cluster role and corresponding cluster role binding are generated for each permission defined in thepermissions
field of the CSV. All resources generated are given theolm.owner: <csv_name>
andolm.owner.namespace: <csv_namespace>
labels. -
If the CSV does not define exactly one target namespace that contains
*
, then all roles and role bindings in the Operator namespace with theolm.owner: <csv_name>
andolm.owner.namespace: <csv_namespace>
labels are copied into the target namespace.
2.4.5.7. Copied CSVs
OLM creates copies of all active member CSVs of an Operator group in each of the target namespaces of that Operator group. The purpose of a copied CSV is to tell users of a target namespace that a specific Operator is configured to watch resources created there.
Copied CSVs have a status reason Copied
and are updated to match the status of their source CSV. The olm.targetNamespaces
annotation is stripped from copied CSVs before they are created on the cluster. Omitting the target namespace selection avoids the duplication of target namespaces between tenants.
Copied CSVs are deleted when their source CSV no longer exists or the Operator group that their source CSV belongs to no longer targets the namespace of the copied CSV.
By default, the disableCopiedCSVs
field is disabled. After enabling a disableCopiedCSVs
field, the OLM deletes existing copied CSVs on a cluster. When a disableCopiedCSVs
field is disabled, the OLM adds copied CSVs again.
Disable the
disableCopiedCSVs
field:$ cat << EOF | oc apply -f - apiVersion: operators.coreos.com/v1 kind: OLMConfig metadata: name: cluster spec: features: disableCopiedCSVs: false EOF
Enable the
disableCopiedCSVs
field:$ cat << EOF | oc apply -f - apiVersion: operators.coreos.com/v1 kind: OLMConfig metadata: name: cluster spec: features: disableCopiedCSVs: true EOF
2.4.5.8. Static Operator groups
An Operator group is static if its spec.staticProvidedAPIs
field is set to true
. As a result, OLM does not modify the olm.providedAPIs
annotation of an Operator group, which means that it can be set in advance. This is useful when a user wants to use an Operator group to prevent resource contention in a set of namespaces but does not have active member CSVs that provide the APIs for those resources.
Below is an example of an Operator group that protects Prometheus
resources in all namespaces with the something.cool.io/cluster-monitoring: "true"
annotation:
apiVersion: operators.coreos.com/v1 kind: OperatorGroup metadata: name: cluster-monitoring namespace: cluster-monitoring annotations: olm.providedAPIs: Alertmanager.v1.monitoring.coreos.com,Prometheus.v1.monitoring.coreos.com,PrometheusRule.v1.monitoring.coreos.com,ServiceMonitor.v1.monitoring.coreos.com spec: staticProvidedAPIs: true selector: matchLabels: something.cool.io/cluster-monitoring: "true"
2.4.5.9. Operator group intersection
Two Operator groups are said to have intersecting provided APIs if the intersection of their target namespace sets is not an empty set and the intersection of their provided API sets, defined by olm.providedAPIs
annotations, is not an empty set.
A potential issue is that Operator groups with intersecting provided APIs can compete for the same resources in the set of intersecting namespaces.
When checking intersection rules, an Operator group namespace is always included as part of its selected target namespaces.
Rules for intersection
Each time an active member CSV synchronizes, OLM queries the cluster for the set of intersecting provided APIs between the Operator group of the CSV and all others. OLM then checks if that set is an empty set:
If
true
and the CSV’s provided APIs are a subset of the Operator group’s:- Continue transitioning.
If
true
and the CSV’s provided APIs are not a subset of the Operator group’s:If the Operator group is static:
- Clean up any deployments that belong to the CSV.
-
Transition the CSV to a failed state with status reason
CannotModifyStaticOperatorGroupProvidedAPIs
.
If the Operator group is not static:
-
Replace the Operator group’s
olm.providedAPIs
annotation with the union of itself and the CSV’s provided APIs.
-
Replace the Operator group’s
If
false
and the CSV’s provided APIs are not a subset of the Operator group’s:- Clean up any deployments that belong to the CSV.
-
Transition the CSV to a failed state with status reason
InterOperatorGroupOwnerConflict
.
If
false
and the CSV’s provided APIs are a subset of the Operator group’s:If the Operator group is static:
- Clean up any deployments that belong to the CSV.
-
Transition the CSV to a failed state with status reason
CannotModifyStaticOperatorGroupProvidedAPIs
.
If the Operator group is not static:
-
Replace the Operator group’s
olm.providedAPIs
annotation with the difference between itself and the CSV’s provided APIs.
-
Replace the Operator group’s
Failure states caused by Operator groups are non-terminal.
The following actions are performed each time an Operator group synchronizes:
- The set of provided APIs from active member CSVs is calculated from the cluster. Note that copied CSVs are ignored.
-
The cluster set is compared to
olm.providedAPIs
, and ifolm.providedAPIs
contains any extra APIs, then those APIs are pruned. - All CSVs that provide the same APIs across all namespaces are requeued. This notifies conflicting CSVs in intersecting groups that their conflict has possibly been resolved, either through resizing or through deletion of the conflicting CSV.
2.4.5.10. Limitations for multitenant Operator management
OpenShift Container Platform provides limited support for simultaneously installing different versions of an Operator on the same cluster. Operator Lifecycle Manager (OLM) installs Operators multiple times in different namespaces. One constraint of this is that the Operator’s API versions must be the same.
Operators are control plane extensions due to their usage of CustomResourceDefinition
objects (CRDs), which are global resources in Kubernetes. Different major versions of an Operator often have incompatible CRDs. This makes them incompatible to install simultaneously in different namespaces on a cluster.
All tenants, or namespaces, share the same control plane of a cluster. Therefore, tenants in a multitenant cluster also share global CRDs, which limits the scenarios in which different instances of the same Operator can be used in parallel on the same cluster.
The supported scenarios include the following:
- Operators of different versions that ship the exact same CRD definition (in case of versioned CRDs, the exact same set of versions)
- Operators of different versions that do not ship a CRD, and instead have their CRD available in a separate bundle on the OperatorHub
All other scenarios are not supported, because the integrity of the cluster data cannot be guaranteed if there are multiple competing or overlapping CRDs from different Operator versions to be reconciled on the same cluster.
2.4.5.11. Troubleshooting Operator groups
Membership
An install plan’s namespace must contain only one Operator group. When attempting to generate a cluster service version (CSV) in a namespace, an install plan considers an Operator group invalid in the following scenarios:
- No Operator groups exist in the install plan’s namespace.
- Multiple Operator groups exist in the install plan’s namespace.
- An incorrect or non-existent service account name is specified in the Operator group.
If an install plan encounters an invalid Operator group, the CSV is not generated and the
InstallPlan
resource continues to install with a relevant message. For example, the following message is provided if more than one Operator group exists in the same namespace:attenuated service account query failed - more than one operator group(s) are managing this namespace count=2
where
count=
specifies the number of Operator groups in the namespace.-
If the install modes of a CSV do not support the target namespace selection of the Operator group in its namespace, the CSV transitions to a failure state with the reason
UnsupportedOperatorGroup
. CSVs in a failed state for this reason transition to pending after either the target namespace selection of the Operator group changes to a supported configuration, or the install modes of the CSV are modified to support the target namespace selection.
2.4.6. Multitenancy and Operator colocation
This guide outlines multitenancy and Operator colocation in Operator Lifecycle Manager (OLM).
2.4.6.1. Colocation of Operators in a namespace
Operator Lifecycle Manager (OLM) handles OLM-managed Operators that are installed in the same namespace, meaning their Subscription
resources are colocated in the same namespace, as related Operators. Even if they are not actually related, OLM considers their states, such as their version and update policy, when any one of them is updated.
This default behavior manifests in two ways:
-
InstallPlan
resources of pending updates includeClusterServiceVersion
(CSV) resources of all other Operators that are in the same namespace. - All Operators in the same namespace share the same update policy. For example, if one Operator is set to manual updates, all other Operators' update policies are also set to manual.
These scenarios can lead to the following issues:
- It becomes hard to reason about install plans for Operator updates, because there are many more resources defined in them than just the updated Operator.
- It becomes impossible to have some Operators in a namespace update automatically while other are updated manually, which is a common desire for cluster administrators.
These issues usually surface because, when installing Operators with the OpenShift Container Platform web console, the default behavior installs Operators that support the All namespaces install mode into the default openshift-operators
global namespace.
As a cluster administrator, you can bypass this default behavior manually by using the following workflow:
- Create a namespace for the installation of the Operator.
- Create a custom global Operator group, which is an Operator group that watches all namespaces. By associating this Operator group with the namespace you just created, it makes the installation namespace a global namespace, which makes Operators installed there available in all namespaces.
- Install the desired Operator in the installation namespace.
If the Operator has dependencies, the dependencies are automatically installed in the pre-created namespace. As a result, it is then valid for the dependency Operators to have the same update policy and shared install plans. For a detailed procedure, see "Installing global Operators in custom namespaces".
Additional resources
2.4.7. Operator conditions
This guide outlines how Operator Lifecycle Manager (OLM) uses Operator conditions.
2.4.7.1. About Operator conditions
As part of its role in managing the lifecycle of an Operator, Operator Lifecycle Manager (OLM) infers the state of an Operator from the state of Kubernetes resources that define the Operator. While this approach provides some level of assurance that an Operator is in a given state, there are many instances where an Operator might need to communicate information to OLM that could not be inferred otherwise. This information can then be used by OLM to better manage the lifecycle of the Operator.
OLM provides a custom resource definition (CRD) called OperatorCondition
that allows Operators to communicate conditions to OLM. There are a set of supported conditions that influence management of the Operator by OLM when present in the Spec.Conditions
array of an OperatorCondition
resource.
By default, the Spec.Conditions
array is not present in an OperatorCondition
object until it is either added by a user or as a result of custom Operator logic.
2.4.7.2. Supported conditions
Operator Lifecycle Manager (OLM) supports the following Operator conditions.
2.4.7.2.1. Upgradeable condition
The Upgradeable
Operator condition prevents an existing cluster service version (CSV) from being replaced by a newer version of the CSV. This condition is useful when:
- An Operator is about to start a critical process and should not be upgraded until the process is completed.
- An Operator is performing a migration of custom resources (CRs) that must be completed before the Operator is ready to be upgraded.
Setting the Upgradeable
Operator condition to the False
value does not avoid pod disruption. If you must ensure your pods are not disrupted, see "Using pod disruption budgets to specify the number of pods that must be up" and "Graceful termination" in the "Additional resources" section.
Example Upgradeable
Operator condition
apiVersion: operators.coreos.com/v1 kind: OperatorCondition metadata: name: my-operator namespace: operators spec: conditions: - type: Upgradeable 1 status: "False" 2 reason: "migration" message: "The Operator is performing a migration." lastTransitionTime: "2020-08-24T23:15:55Z"
2.4.7.3. Additional resources
2.4.8. Operator Lifecycle Manager metrics
2.4.8.1. Exposed metrics
Operator Lifecycle Manager (OLM) exposes certain OLM-specific resources for use by the Prometheus-based OpenShift Container Platform cluster monitoring stack.
Name | Description |
---|---|
| Number of catalog sources. |
|
State of a catalog source. The value |
|
When reconciling a cluster service version (CSV), present whenever a CSV version is in any state other than |
| Number of CSVs successfully registered. |
|
When reconciling a CSV, represents whether a CSV version is in a |
| Monotonic count of CSV upgrades. |
| Number of install plans. |
| Monotonic count of warnings generated by resources, such as deprecated resources, included in an install plan. |
| The duration of a dependency resolution attempt. |
| Number of subscriptions. |
|
Monotonic count of subscription syncs. Includes the |
2.4.9. Webhook management in Operator Lifecycle Manager
Webhooks allow Operator authors to intercept, modify, and accept or reject resources before they are saved to the object store and handled by the Operator controller. Operator Lifecycle Manager (OLM) can manage the lifecycle of these webhooks when they are shipped alongside your Operator.
See Defining cluster service versions (CSVs) for details on how an Operator developer can define webhooks for their Operator, as well as considerations when running on OLM.
2.4.9.1. Additional resources
- Types of webhook admission plugins
Kubernetes documentation:
2.5. Understanding OperatorHub
2.5.1. About OperatorHub
OperatorHub is the web console interface in OpenShift Container Platform that cluster administrators use to discover and install Operators. With one click, an Operator can be pulled from its off-cluster source, installed and subscribed on the cluster, and made ready for engineering teams to self-service manage the product across deployment environments using Operator Lifecycle Manager (OLM).
Cluster administrators can choose from catalogs grouped into the following categories:
Category | Description |
---|---|
Red Hat Operators | Red Hat products packaged and shipped by Red Hat. Supported by Red Hat. |
Certified Operators | Products from leading independent software vendors (ISVs). Red Hat partners with ISVs to package and ship. Supported by the ISV. |
Red Hat Marketplace | Certified software that can be purchased from Red Hat Marketplace. |
Community Operators | Optionally-visible software maintained by relevant representatives in the redhat-openshift-ecosystem/community-operators-prod/operators GitHub repository. No official support. |
Custom Operators | Operators you add to the cluster yourself. If you have not added any custom Operators, the Custom category does not appear in the web console on your OperatorHub. |
Operators on OperatorHub are packaged to run on OLM. This includes a YAML file called a cluster service version (CSV) containing all of the CRDs, RBAC rules, deployments, and container images required to install and securely run the Operator. It also contains user-visible information like a description of its features and supported Kubernetes versions.
The Operator SDK can be used to assist developers packaging their Operators for use on OLM and OperatorHub. If you have a commercial application that you want to make accessible to your customers, get it included using the certification workflow provided on the Red Hat Partner Connect portal at connect.redhat.com.
2.5.2. OperatorHub architecture
The OperatorHub UI component is driven by the Marketplace Operator by default on OpenShift Container Platform in the openshift-marketplace
namespace.
2.5.2.1. OperatorHub custom resource
The Marketplace Operator manages an OperatorHub
custom resource (CR) named cluster
that manages the default CatalogSource
objects provided with OperatorHub. You can modify this resource to enable or disable the default catalogs, which is useful when configuring OpenShift Container Platform in restricted network environments.
Example OperatorHub
custom resource
apiVersion: config.openshift.io/v1 kind: OperatorHub metadata: name: cluster spec: disableAllDefaultSources: true 1 sources: [ 2 { name: "community-operators", disabled: false } ]
2.5.3. Additional resources
2.6. Red Hat-provided Operator catalogs
Red Hat provides several Operator catalogs that are included with OpenShift Container Platform by default.
As of OpenShift Container Platform 4.11, the default Red Hat-provided Operator catalog releases in the file-based catalog format. The default Red Hat-provided Operator catalogs for OpenShift Container Platform 4.6 through 4.10 released in the deprecated SQLite database format.
The opm
subcommands, flags, and functionality related to the SQLite database format are also deprecated and will be removed in a future release. The features are still supported and must be used for catalogs that use the deprecated SQLite database format.
Many of the opm
subcommands and flags for working with the SQLite database format, such as opm index prune
, do not work with the file-based catalog format. For more information about working with file-based catalogs, see Managing custom catalogs, Operator Framework packaging format, and Mirroring images for a disconnected installation using the oc-mirror plugin.
2.6.1. About Operator catalogs
An Operator catalog is a repository of metadata that Operator Lifecycle Manager (OLM) can query to discover and install Operators and their dependencies on a cluster. OLM always installs Operators from the latest version of a catalog.
An index image, based on the Operator bundle format, is a containerized snapshot of a catalog. It is an immutable artifact that contains the database of pointers to a set of Operator manifest content. A catalog can reference an index image to source its content for OLM on the cluster.
As catalogs are updated, the latest versions of Operators change, and older versions may be removed or altered. In addition, when OLM runs on an OpenShift Container Platform cluster in a restricted network environment, it is unable to access the catalogs directly from the internet to pull the latest content.
As a cluster administrator, you can create your own custom index image, either based on a Red Hat-provided catalog or from scratch, which can be used to source the catalog content on the cluster. Creating and updating your own index image provides a method for customizing the set of Operators available on the cluster, while also avoiding the aforementioned restricted network environment issues.
Kubernetes periodically deprecates certain APIs that are removed in subsequent releases. As a result, Operators are unable to use removed APIs starting with the version of OpenShift Container Platform that uses the Kubernetes version that removed the API.
If your cluster is using custom catalogs, see Controlling Operator compatibility with OpenShift Container Platform versions for more details about how Operator authors can update their projects to help avoid workload issues and prevent incompatible upgrades.
Support for the legacy package manifest format for Operators, including custom catalogs that were using the legacy format, is removed in OpenShift Container Platform 4.8 and later.
When creating custom catalog images, previous versions of OpenShift Container Platform 4 required using the oc adm catalog build
command, which was deprecated for several releases and is now removed. With the availability of Red Hat-provided index images starting in OpenShift Container Platform 4.6, catalog builders must use the opm index
command to manage index images.
2.6.2. About Red Hat-provided Operator catalogs
The Red Hat-provided catalog sources are installed by default in the openshift-marketplace
namespace, which makes the catalogs available cluster-wide in all namespaces.
The following Operator catalogs are distributed by Red Hat:
Catalog | Index image | Description |
---|---|---|
|
| Red Hat products packaged and shipped by Red Hat. Supported by Red Hat. |
|
| Products from leading independent software vendors (ISVs). Red Hat partners with ISVs to package and ship. Supported by the ISV. |
|
| Certified software that can be purchased from Red Hat Marketplace. |
|
| Software maintained by relevant representatives in the redhat-openshift-ecosystem/community-operators-prod/operators GitHub repository. No official support. |
During a cluster upgrade, the index image tag for the default Red Hat-provided catalog sources are updated automatically by the Cluster Version Operator (CVO) so that Operator Lifecycle Manager (OLM) pulls the updated version of the catalog. For example during an upgrade from OpenShift Container Platform 4.8 to 4.9, the spec.image
field in the CatalogSource
object for the redhat-operators
catalog is updated from:
registry.redhat.io/redhat/redhat-operator-index:v4.8
to:
registry.redhat.io/redhat/redhat-operator-index:v4.9
2.7. Operators in multitenant clusters
The default behavior for Operator Lifecycle Manager (OLM) aims to provide simplicity during Operator installation. However, this behavior can lack flexibility, especially in multitenant clusters. In order for multiple tenants on a OpenShift Container Platform cluster to use an Operator, the default behavior of OLM requires that administrators install the Operator in All namespaces mode, which can be considered to violate the principle of least privilege.
Consider the following scenarios to determine which Operator installation workflow works best for your environment and requirements.
Additional resources
2.7.1. Default Operator install modes and behavior
When installing Operators with the web console as an administrator, you typically have two choices for the install mode, depending on the Operator’s capabilities:
- Single namespace
- Installs the Operator in the chosen single namespace, and makes all permissions that the Operator requests available in that namespace.
- All namespaces
-
Installs the Operator in the default
openshift-operators
namespace to watch and be made available to all namespaces in the cluster. Makes all permissions that the Operator requests available in all namespaces. In some cases, an Operator author can define metadata to give the user a second option for that Operator’s suggested namespace.
This choice also means that users in the affected namespaces get access to the Operators APIs, which can leverage the custom resources (CRs) they own, depending on their role in the namespace:
-
The
namespace-admin
andnamespace-edit
roles can read/write to the Operator APIs, meaning they can use them. -
The
namespace-view
role can read CR objects of that Operator.
For Single namespace mode, because the Operator itself installs in the chosen namespace, its pod and service account are also located there. For All namespaces mode, the Operator’s privileges are all automatically elevated to cluster roles, meaning the Operator has those permissions in all namespaces.
Additional resources
2.7.2. Recommended solution for multitenant clusters
While a Multinamespace install mode does exist, it is supported by very few Operators. As a middle ground solution between the standard All namespaces and Single namespace install modes, you can install multiple instances of the same Operator, one for each tenant, by using the following workflow:
- Create a namespace for the tenant Operator that is separate from the tenant’s namespace.
- Create an Operator group for the tenant Operator scoped only to the tenant’s namespace.
- Install the Operator in the tenant Operator namespace.
As a result, the Operator resides in the tenant Operator namespace and watches the tenant namespace, but neither the Operator’s pod nor its service account are visible or usable by the tenant.
This solution provides better tenant separation, least privilege principle at the cost of resource usage, and additional orchestration to ensure the constraints are met. For a detailed procedure, see "Preparing for multiple instances of an Operator for multitenant clusters".
Limitations and considerations
This solution only works when the following constraints are met:
- All instances of the same Operator must be the same version.
- The Operator cannot have dependencies on other Operators.
- The Operator cannot ship a CRD conversion webhook.
You cannot use different versions of the same Operator on the same cluster. Eventually, the installation of another instance of the Operator would be blocked when it meets the following conditions:
- The instance is not the newest version of the Operator.
- The instance ships an older revision of the CRDs that lack information or versions that newer revisions have that are already in use on the cluster.
As an administrator, use caution when allowing non-cluster administrators to install Operators self-sufficiently, as explained in "Allowing non-cluster administrators to install Operators". These tenants should only have access to a curated catalog of Operators that are known to not have dependencies. These tenants must also be forced to use the same version line of an Operator, to ensure the CRDs do not change. This requires the use of namespace-scoped catalogs and likely disabling the global default catalogs.
2.7.3. Operator colocation and Operator groups
Operator Lifecycle Manager (OLM) handles OLM-managed Operators that are installed in the same namespace, meaning their Subscription
resources are colocated in the same namespace, as related Operators. Even if they are not actually related, OLM considers their states, such as their version and update policy, when any one of them is updated.
For more information on Operator colocation and using Operator groups effectively, see Operator Lifecycle Manager (OLM) → Multitenancy and Operator colocation.
2.8. CRDs
2.8.1. Extending the Kubernetes API with custom resource definitions
Operators use the Kubernetes extension mechanism, custom resource definitions (CRDs), so that custom objects managed by the Operator look and act just like the built-in, native Kubernetes objects. This guide describes how cluster administrators can extend their OpenShift Container Platform cluster by creating and managing CRDs.
2.8.1.1. Custom resource definitions
In the Kubernetes API, a resource is an endpoint that stores a collection of API objects of a certain kind. For example, the built-in Pods
resource contains a collection of Pod
objects.
A custom resource definition (CRD) object defines a new, unique object type, called a kind, in the cluster and lets the Kubernetes API server handle its entire lifecycle.
Custom resource (CR) objects are created from CRDs that have been added to the cluster by a cluster administrator, allowing all cluster users to add the new resource type into projects.
When a cluster administrator adds a new CRD to the cluster, the Kubernetes API server reacts by creating a new RESTful resource path that can be accessed by the entire cluster or a single project (namespace) and begins serving the specified CR.
Cluster administrators that want to grant access to the CRD to other users can use cluster role aggregation to grant access to users with the admin
, edit
, or view
default cluster roles. Cluster role aggregation allows the insertion of custom policy rules into these cluster roles. This behavior integrates the new resource into the RBAC policy of the cluster as if it was a built-in resource.
Operators in particular make use of CRDs by packaging them with any required RBAC policy and other software-specific logic. Cluster administrators can also add CRDs manually to the cluster outside of the lifecycle of an Operator, making them available to all users.
While only cluster administrators can create CRDs, developers can create the CR from an existing CRD if they have read and write permission to it.
2.8.1.2. Creating a custom resource definition
To create custom resource (CR) objects, cluster administrators must first create a custom resource definition (CRD).
Prerequisites
-
Access to an OpenShift Container Platform cluster with
cluster-admin
user privileges.
Procedure
To create a CRD:
Create a YAML file that contains the following field types:
Example YAML file for a CRD
apiVersion: apiextensions.k8s.io/v1 1 kind: CustomResourceDefinition metadata: name: crontabs.stable.example.com 2 spec: group: stable.example.com 3 versions: - name: v1 4 served: true storage: true schema: openAPIV3Schema: type: object properties: spec: type: object properties: cronSpec: type: string image: type: string replicas: type: integer scope: Namespaced 5 names: plural: crontabs 6 singular: crontab 7 kind: CronTab 8 shortNames: - ct 9
- 1
- Use the
apiextensions.k8s.io/v1
API. - 2
- Specify a name for the definition. This must be in the
<plural-name>.<group>
format using the values from thegroup
andplural
fields. - 3
- Specify a group name for the API. An API group is a collection of objects that are logically related. For example, all batch objects like
Job
orScheduledJob
could be in the batch API group (such asbatch.api.example.com
). A good practice is to use a fully-qualified-domain name (FQDN) of your organization. - 4
- Specify a version name to be used in the URL. Each API group can exist in multiple versions, for example
v1alpha
,v1beta
,v1
. - 5
- Specify whether the custom objects are available to a project (
Namespaced
) or all projects in the cluster (Cluster
). - 6
- Specify the plural name to use in the URL. The
plural
field is the same as a resource in an API URL. - 7
- Specify a singular name to use as an alias on the CLI and for display.
- 8
- Specify the kind of objects that can be created. The type can be in CamelCase.
- 9
- Specify a shorter string to match your resource on the CLI.
NoteBy default, a CRD is cluster-scoped and available to all projects.
Create the CRD object:
$ oc create -f <file_name>.yaml
A new RESTful API endpoint is created at:
/apis/<spec:group>/<spec:version>/<scope>/*/<names-plural>/...
For example, using the example file, the following endpoint is created:
/apis/stable.example.com/v1/namespaces/*/crontabs/...
You can now use this endpoint URL to create and manage CRs. The object kind is based on the
spec.kind
field of the CRD object you created.
2.8.1.3. Creating cluster roles for custom resource definitions
Cluster administrators can grant permissions to existing cluster-scoped custom resource definitions (CRDs). If you use the admin
, edit
, and view
default cluster roles, you can take advantage of cluster role aggregation for their rules.
You must explicitly assign permissions to each of these roles. The roles with more permissions do not inherit rules from roles with fewer permissions. If you assign a rule to a role, you must also assign that verb to roles that have more permissions. For example, if you grant the get crontabs
permission to the view role, you must also grant it to the edit
and admin
roles. The admin
or edit
role is usually assigned to the user that created a project through the project template.
Prerequisites
- Create a CRD.
Procedure
Create a cluster role definition file for the CRD. The cluster role definition is a YAML file that contains the rules that apply to each cluster role. An OpenShift Container Platform controller adds the rules that you specify to the default cluster roles.
Example YAML file for a cluster role definition
kind: ClusterRole apiVersion: rbac.authorization.k8s.io/v1 1 metadata: name: aggregate-cron-tabs-admin-edit 2 labels: rbac.authorization.k8s.io/aggregate-to-admin: "true" 3 rbac.authorization.k8s.io/aggregate-to-edit: "true" 4 rules: - apiGroups: ["stable.example.com"] 5 resources: ["crontabs"] 6 verbs: ["get", "list", "watch", "create", "update", "patch", "delete", "deletecollection"] 7 --- kind: ClusterRole apiVersion: rbac.authorization.k8s.io/v1 metadata: name: aggregate-cron-tabs-view 8 labels: # Add these permissions to the "view" default role. rbac.authorization.k8s.io/aggregate-to-view: "true" 9 rbac.authorization.k8s.io/aggregate-to-cluster-reader: "true" 10 rules: - apiGroups: ["stable.example.com"] 11 resources: ["crontabs"] 12 verbs: ["get", "list", "watch"] 13
- 1
- Use the
rbac.authorization.k8s.io/v1
API. - 2 8
- Specify a name for the definition.
- 3
- Specify this label to grant permissions to the admin default role.
- 4
- Specify this label to grant permissions to the edit default role.
- 5 11
- Specify the group name of the CRD.
- 6 12
- Specify the plural name of the CRD that these rules apply to.
- 7 13
- Specify the verbs that represent the permissions that are granted to the role. For example, apply read and write permissions to the
admin
andedit
roles and only read permission to theview
role. - 9
- Specify this label to grant permissions to the
view
default role. - 10
- Specify this label to grant permissions to the
cluster-reader
default role.
Create the cluster role:
$ oc create -f <file_name>.yaml
2.8.1.4. Creating custom resources from a file
After a custom resource definition (CRD) has been added to the cluster, custom resources (CRs) can be created with the CLI from a file using the CR specification.
Prerequisites
- CRD added to the cluster by a cluster administrator.
Procedure
Create a YAML file for the CR. In the following example definition, the
cronSpec
andimage
custom fields are set in a CR ofKind: CronTab
. TheKind
comes from thespec.kind
field of the CRD object:Example YAML file for a CR
apiVersion: "stable.example.com/v1" 1 kind: CronTab 2 metadata: name: my-new-cron-object 3 finalizers: 4 - finalizer.stable.example.com spec: 5 cronSpec: "* * * * /5" image: my-awesome-cron-image
- 1
- Specify the group name and API version (name/version) from the CRD.
- 2
- Specify the type in the CRD.
- 3
- Specify a name for the object.
- 4
- Specify the finalizers for the object, if any. Finalizers allow controllers to implement conditions that must be completed before the object can be deleted.
- 5
- Specify conditions specific to the type of object.
After you create the file, create the object:
$ oc create -f <file_name>.yaml
2.8.1.5. Inspecting custom resources
You can inspect custom resource (CR) objects that exist in your cluster using the CLI.
Prerequisites
- A CR object exists in a namespace to which you have access.
Procedure
To get information on a specific kind of a CR, run:
$ oc get <kind>
For example:
$ oc get crontab
Example output
NAME KIND my-new-cron-object CronTab.v1.stable.example.com
Resource names are not case-sensitive, and you can use either the singular or plural forms defined in the CRD, as well as any short name. For example:
$ oc get crontabs
$ oc get crontab
$ oc get ct
You can also view the raw YAML data for a CR:
$ oc get <kind> -o yaml
For example:
$ oc get ct -o yaml
Example output
apiVersion: v1 items: - apiVersion: stable.example.com/v1 kind: CronTab metadata: clusterName: "" creationTimestamp: 2017-05-31T12:56:35Z deletionGracePeriodSeconds: null deletionTimestamp: null name: my-new-cron-object namespace: default resourceVersion: "285" selfLink: /apis/stable.example.com/v1/namespaces/default/crontabs/my-new-cron-object uid: 9423255b-4600-11e7-af6a-28d2447dc82b spec: cronSpec: '* * * * /5' 1 image: my-awesome-cron-image 2
2.8.2. Managing resources from custom resource definitions
This guide describes how developers can manage custom resources (CRs) that come from custom resource definitions (CRDs).
2.8.2.1. Custom resource definitions
In the Kubernetes API, a resource is an endpoint that stores a collection of API objects of a certain kind. For example, the built-in Pods
resource contains a collection of Pod
objects.
A custom resource definition (CRD) object defines a new, unique object type, called a kind, in the cluster and lets the Kubernetes API server handle its entire lifecycle.
Custom resource (CR) objects are created from CRDs that have been added to the cluster by a cluster administrator, allowing all cluster users to add the new resource type into projects.
Operators in particular make use of CRDs by packaging them with any required RBAC policy and other software-specific logic. Cluster administrators can also add CRDs manually to the cluster outside of the lifecycle of an Operator, making them available to all users.
While only cluster administrators can create CRDs, developers can create the CR from an existing CRD if they have read and write permission to it.
2.8.2.2. Creating custom resources from a file
After a custom resource definition (CRD) has been added to the cluster, custom resources (CRs) can be created with the CLI from a file using the CR specification.
Prerequisites
- CRD added to the cluster by a cluster administrator.
Procedure
Create a YAML file for the CR. In the following example definition, the
cronSpec
andimage
custom fields are set in a CR ofKind: CronTab
. TheKind
comes from thespec.kind
field of the CRD object:Example YAML file for a CR
apiVersion: "stable.example.com/v1" 1 kind: CronTab 2 metadata: name: my-new-cron-object 3 finalizers: 4 - finalizer.stable.example.com spec: 5 cronSpec: "* * * * /5" image: my-awesome-cron-image
- 1
- Specify the group name and API version (name/version) from the CRD.
- 2
- Specify the type in the CRD.
- 3
- Specify a name for the object.
- 4
- Specify the finalizers for the object, if any. Finalizers allow controllers to implement conditions that must be completed before the object can be deleted.
- 5
- Specify conditions specific to the type of object.
After you create the file, create the object:
$ oc create -f <file_name>.yaml
2.8.2.3. Inspecting custom resources
You can inspect custom resource (CR) objects that exist in your cluster using the CLI.
Prerequisites
- A CR object exists in a namespace to which you have access.
Procedure
To get information on a specific kind of a CR, run:
$ oc get <kind>
For example:
$ oc get crontab
Example output
NAME KIND my-new-cron-object CronTab.v1.stable.example.com
Resource names are not case-sensitive, and you can use either the singular or plural forms defined in the CRD, as well as any short name. For example:
$ oc get crontabs
$ oc get crontab
$ oc get ct
You can also view the raw YAML data for a CR:
$ oc get <kind> -o yaml
For example:
$ oc get ct -o yaml
Example output
apiVersion: v1 items: - apiVersion: stable.example.com/v1 kind: CronTab metadata: clusterName: "" creationTimestamp: 2017-05-31T12:56:35Z deletionGracePeriodSeconds: null deletionTimestamp: null name: my-new-cron-object namespace: default resourceVersion: "285" selfLink: /apis/stable.example.com/v1/namespaces/default/crontabs/my-new-cron-object uid: 9423255b-4600-11e7-af6a-28d2447dc82b spec: cronSpec: '* * * * /5' 1 image: my-awesome-cron-image 2
Chapter 3. User tasks
3.1. Creating applications from installed Operators
This guide walks developers through an example of creating applications from an installed Operator using the OpenShift Container Platform web console.
3.1.1. Creating an etcd cluster using an Operator
This procedure walks through creating a new etcd cluster using the etcd Operator, managed by Operator Lifecycle Manager (OLM).
Prerequisites
- Access to an OpenShift Container Platform 4.15 cluster.
- The etcd Operator already installed cluster-wide by an administrator.
Procedure
-
Create a new project in the OpenShift Container Platform web console for this procedure. This example uses a project called
my-etcd
. Navigate to the Operators → Installed Operators page. The Operators that have been installed to the cluster by the cluster administrator and are available for use are shown here as a list of cluster service versions (CSVs). CSVs are used to launch and manage the software provided by the Operator.
TipYou can get this list from the CLI using:
$ oc get csv
On the Installed Operators page, click the etcd Operator to view more details and available actions.
As shown under Provided APIs, this Operator makes available three new resource types, including one for an etcd Cluster (the
EtcdCluster
resource). These objects work similar to the built-in native Kubernetes ones, such asDeployment
orReplicaSet
, but contain logic specific to managing etcd.Create a new etcd cluster:
- In the etcd Cluster API box, click Create instance.
-
The next page allows you to make any modifications to the minimal starting template of an
EtcdCluster
object, such as the size of the cluster. For now, click Create to finalize. This triggers the Operator to start up the pods, services, and other components of the new etcd cluster.
Click the example etcd cluster, then click the Resources tab to see that your project now contains a number of resources created and configured automatically by the Operator.
Verify that a Kubernetes service has been created that allows you to access the database from other pods in your project.
All users with the
edit
role in a given project can create, manage, and delete application instances (an etcd cluster, in this example) managed by Operators that have already been created in the project, in a self-service manner, just like a cloud service. If you want to enable additional users with this ability, project administrators can add the role using the following command:$ oc policy add-role-to-user edit <user> -n <target_project>
You now have an etcd cluster that will react to failures and rebalance data as pods become unhealthy or are migrated between nodes in the cluster. Most importantly, cluster administrators or developers with proper access can now easily use the database with their applications.
3.2. Installing Operators in your namespace
If a cluster administrator has delegated Operator installation permissions to your account, you can install and subscribe an Operator to your namespace in a self-service manner.
3.2.1. Prerequisites
- A cluster administrator must add certain permissions to your OpenShift Container Platform user account to allow self-service Operator installation to a namespace. See Allowing non-cluster administrators to install Operators for details.
3.2.2. About Operator installation with OperatorHub
OperatorHub is a user interface for discovering Operators; it works in conjunction with Operator Lifecycle Manager (OLM), which installs and manages Operators on a cluster.
As a user with the proper permissions, you can install an Operator from OperatorHub by using the OpenShift Container Platform web console or CLI.
During installation, you must determine the following initial settings for the Operator:
- Installation Mode
- Choose a specific namespace in which to install the Operator.
- Update Channel
- If an Operator is available through multiple channels, you can choose which channel you want to subscribe to. For example, to deploy from the stable channel, if available, select it from the list.
- Approval Strategy
You can choose automatic or manual updates.
If you choose automatic updates for an installed Operator, when a new version of that Operator is available in the selected channel, Operator Lifecycle Manager (OLM) automatically upgrades the running instance of your Operator without human intervention.
If you select manual updates, when a newer version of an Operator is available, OLM creates an update request. As a cluster administrator, you must then manually approve that update request to have the Operator updated to the new version.
3.2.3. Installing from OperatorHub by using the web console
You can install and subscribe to an Operator from OperatorHub by using the OpenShift Container Platform web console.
Prerequisites
- Access to an OpenShift Container Platform cluster using an account with Operator installation permissions.
Procedure
- Navigate in the web console to the Operators → OperatorHub page.
Scroll or type a keyword into the Filter by keyword box to find the Operator you want. For example, type
advanced
to find the Advanced Cluster Management for Kubernetes Operator.You can also filter options by Infrastructure Features. For example, select Disconnected if you want to see Operators that work in disconnected environments, also known as restricted network environments.
Select the Operator to display additional information.
NoteChoosing a Community Operator warns that Red Hat does not certify Community Operators; you must acknowledge the warning before continuing.
- Read the information about the Operator and click Install.
On the Install Operator page, configure your Operator installation:
If you want to install a specific version of an Operator, select an Update channel and Version from the lists. You can browse the various versions of an Operator across any channels it might have, view the metadata for that channel and version, and select the exact version you want to install.
NoteThe version selection defaults to the latest version for the channel selected. If the latest version for the channel is selected, the Automatic approval strategy is enabled by default. Otherwise, Manual approval is required when not installing the latest version for the selected channel.
Installing an Operator with Manual approval causes all Operators installed within the namespace to function with the Manual approval strategy and all Operators are updated together. If you want to update Operators independently, install Operators into separate namespaces.
- Choose a specific, single namespace in which to install the Operator. The Operator will only watch and be made available for use in this single namespace.
For clusters on cloud providers with token authentication enabled:
- If the cluster uses AWS STS (STS Mode in the web console), enter the Amazon Resource Name (ARN) of the AWS IAM role of your service account in the role ARN field. To create the role’s ARN, follow the procedure described in Preparing AWS account.
- If the cluster uses Microsoft Entra Workload ID (Workload Identity / Federated Identity Mode in the web console), add the client ID, tenant ID, and subscription ID in the appropriate field.
For Update approval, select either the Automatic or Manual approval strategy.
ImportantIf the web console shows that the cluster uses AWS STS or Microsoft Entra Workload ID, you must set Update approval to Manual.
Subscriptions with automatic update approvals are not recommended because there might be permission changes to make prior to updating. Subscriptions with manual update approvals ensure that administrators have the opportunity to verify the permissions of the later version and take any necessary steps prior to update.
Click Install to make the Operator available to the selected namespaces on this OpenShift Container Platform cluster:
If you selected a Manual approval strategy, the upgrade status of the subscription remains Upgrading until you review and approve the install plan.
After approving on the Install Plan page, the subscription upgrade status moves to Up to date.
- If you selected an Automatic approval strategy, the upgrade status should resolve to Up to date without intervention.
Verification
After the upgrade status of the subscription is Up to date, select Operators → Installed Operators to verify that the cluster service version (CSV) of the installed Operator eventually shows up. The Status should eventually resolve to Succeeded in the relevant namespace.
NoteFor the All namespaces… installation mode, the status resolves to Succeeded in the
openshift-operators
namespace, but the status is Copied if you check in other namespaces.If it does not:
-
Check the logs in any pods in the
openshift-operators
project (or other relevant namespace if A specific namespace… installation mode was selected) on the Workloads → Pods page that are reporting issues to troubleshoot further.
-
Check the logs in any pods in the
When the Operator is installed, the metadata indicates which channel and version are installed.
NoteThe Channel and Version dropdown menus are still available for viewing other version metadata in this catalog context.
3.2.4. Installing from OperatorHub by using the CLI
Instead of using the OpenShift Container Platform web console, you can install an Operator from OperatorHub by using the CLI. Use the oc
command to create or update a Subscription
object.
For SingleNamespace
install mode, you must also ensure an appropriate Operator group exists in the related namespace. An Operator group, defined by an OperatorGroup
object, selects target namespaces in which to generate required RBAC access for all Operators in the same namespace as the Operator group.
In most cases, the web console method of this procedure is preferred because it automates tasks in the background, such as handling the creation of OperatorGroup
and Subscription
objects automatically when choosing SingleNamespace
mode.
Prerequisites
- Access to an OpenShift Container Platform cluster using an account with Operator installation permissions.
-
You have installed the OpenShift CLI (
oc
).
Procedure
View the list of Operators available to the cluster from OperatorHub:
$ oc get packagemanifests -n openshift-marketplace
Example 3.1. Example output
NAME CATALOG AGE 3scale-operator Red Hat Operators 91m advanced-cluster-management Red Hat Operators 91m amq7-cert-manager Red Hat Operators 91m # ... couchbase-enterprise-certified Certified Operators 91m crunchy-postgres-operator Certified Operators 91m mongodb-enterprise Certified Operators 91m # ... etcd Community Operators 91m jaeger Community Operators 91m kubefed Community Operators 91m # ...
Note the catalog for your desired Operator.
Inspect your desired Operator to verify its supported install modes and available channels:
$ oc describe packagemanifests <operator_name> -n openshift-marketplace
Example 3.2. Example output
# ... Kind: PackageManifest # ... Install Modes: 1 Supported: true Type: OwnNamespace Supported: true Type: SingleNamespace Supported: false Type: MultiNamespace Supported: true Type: AllNamespaces # ... Entries: Name: example-operator.v3.7.11 Version: 3.7.11 Name: example-operator.v3.7.10 Version: 3.7.10 Name: stable-3.7 2 # ... Entries: Name: example-operator.v3.8.5 Version: 3.8.5 Name: example-operator.v3.8.4 Version: 3.8.4 Name: stable-3.8 3 Default Channel: stable-3.8 4
TipYou can print an Operator’s version and channel information in YAML format by running the following command:
$ oc get packagemanifests <operator_name> -n <catalog_namespace> -o yaml
If more than one catalog is installed in a namespace, run the following command to look up the available versions and channels of an Operator from a specific catalog:
$ oc get packagemanifest \ --selector=catalog=<catalogsource_name> \ --field-selector metadata.name=<operator_name> \ -n <catalog_namespace> -o yaml
ImportantIf you do not specify the Operator’s catalog, running the
oc get packagemanifest
andoc describe packagemanifest
commands might return a package from an unexpected catalog if the following conditions are met:- Multiple catalogs are installed in the same namespace.
- The catalogs contain the same Operators or Operators with the same name.
If the Operator you intend to install supports the
AllNamespaces
install mode, and you choose to use this mode, skip this step, because theopenshift-operators
namespace already has an appropriate Operator group in place by default, calledglobal-operators
.If the Operator you intend to install supports the
SingleNamespace
install mode, and you choose to use this mode, you must ensure an appropriate Operator group exists in the related namespace. If one does not exist, you can create create one by following these steps:ImportantYou can only have one Operator group per namespace. For more information, see "Operator groups".
Create an
OperatorGroup
object YAML file, for exampleoperatorgroup.yaml
, forSingleNamespace
install mode:Example
OperatorGroup
object forSingleNamespace
install modeapiVersion: operators.coreos.com/v1 kind: OperatorGroup metadata: name: <operatorgroup_name> namespace: <namespace> 1 spec: targetNamespaces: - <namespace> 2
Create the
OperatorGroup
object:$ oc apply -f operatorgroup.yaml
Create a
Subscription
object to subscribe a namespace to an Operator:Create a YAML file for the
Subscription
object, for examplesubscription.yaml
:NoteIf you want to subscribe to a specific version of an Operator, set the
startingCSV
field to the desired version and set theinstallPlanApproval
field toManual
to prevent the Operator from automatically upgrading if a later version exists in the catalog. For details, see the following "ExampleSubscription
object with a specific starting Operator version".Example 3.3. Example
Subscription
objectapiVersion: operators.coreos.com/v1alpha1 kind: Subscription metadata: name: <subscription_name> namespace: <namespace_per_install_mode> 1 spec: channel: <channel_name> 2 name: <operator_name> 3 source: <catalog_name> 4 sourceNamespace: <catalog_source_namespace> 5 config: env: 6 - name: ARGS value: "-v=10" envFrom: 7 - secretRef: name: license-secret volumes: 8 - name: <volume_name> configMap: name: <configmap_name> volumeMounts: 9 - mountPath: <directory_name> name: <volume_name> tolerations: 10 - operator: "Exists" resources: 11 requests: memory: "64Mi" cpu: "250m" limits: memory: "128Mi" cpu: "500m" nodeSelector: 12 foo: bar
- 1
- For default
AllNamespaces
install mode usage, specify theopenshift-operators
namespace. Alternatively, you can specify a custom global namespace, if you have created one. ForSingleNamespace
install mode usage, specify the relevant single namespace. - 2
- Name of the channel to subscribe to.
- 3
- Name of the Operator to subscribe to.
- 4
- Name of the catalog source that provides the Operator.
- 5
- Namespace of the catalog source. Use
openshift-marketplace
for the default OperatorHub catalog sources. - 6
- The
env
parameter defines a list of environment variables that must exist in all containers in the pod created by OLM. - 7
- The
envFrom
parameter defines a list of sources to populate environment variables in the container. - 8
- The
volumes
parameter defines a list of volumes that must exist on the pod created by OLM. - 9
- The
volumeMounts
parameter defines a list of volume mounts that must exist in all containers in the pod created by OLM. If avolumeMount
references avolume
that does not exist, OLM fails to deploy the Operator. - 10
- The
tolerations
parameter defines a list of tolerations for the pod created by OLM. - 11
- The
resources
parameter defines resource constraints for all the containers in the pod created by OLM. - 12
- The
nodeSelector
parameter defines aNodeSelector
for the pod created by OLM.
Example 3.4. Example
Subscription
object with a specific starting Operator versionapiVersion: operators.coreos.com/v1alpha1 kind: Subscription metadata: name: example-operator namespace: example-operator spec: channel: stable-3.7 installPlanApproval: Manual 1 name: example-operator source: custom-operators sourceNamespace: openshift-marketplace startingCSV: example-operator.v3.7.10 2
- 1
- Set the approval strategy to
Manual
in case your specified version is superseded by a later version in the catalog. This plan prevents an automatic upgrade to a later version and requires manual approval before the starting CSV can complete the installation. - 2
- Set a specific version of an Operator CSV.
For clusters on cloud providers with token authentication enabled, configure your
Subscription
object by following these steps:Ensure the
Subscription
object is set to manual update approvals:kind: Subscription # ... spec: installPlanApproval: Manual 1
- 1
- Subscriptions with automatic update approvals are not recommended because there might be permission changes to make prior to updating. Subscriptions with manual update approvals ensure that administrators have the opportunity to verify the permissions of the later version and take any necessary steps prior to update.
Include the relevant cloud provider-specific fields in the
Subscription
object’sconfig
section:If the cluster is in AWS STS mode, include the following fields:
kind: Subscription # ... spec: config: env: - name: ROLEARN value: "<role_arn>" 1
- 1
- Include the role ARN details.
If the cluster is in Microsoft Entra Workload ID mode, include the following fields:
kind: Subscription # ... spec: config: env: - name: CLIENTID value: "<client_id>" 1 - name: TENANTID value: "<tenant_id>" 2 - name: SUBSCRIPTIONID value: "<subscription_id>" 3
Create the
Subscription
object by running the following command:$ oc apply -f subscription.yaml
-
If you set the
installPlanApproval
field toManual
, manually approve the pending install plan to complete the Operator installation. For more information, see "Manually approving a pending Operator update".
At this point, OLM is now aware of the selected Operator. A cluster service version (CSV) for the Operator should appear in the target namespace, and APIs provided by the Operator should be available for creation.
Verification
Check the status of the
Subscription
object for your installed Operator by running the following command:$ oc describe subscription <subscription_name> -n <namespace>
If you created an Operator group for
SingleNamespace
install mode, check the status of theOperatorGroup
object by running the following command:$ oc describe operatorgroup <operatorgroup_name> -n <namespace>
Additional resources
Additional resources
Chapter 4. Administrator tasks
4.1. Adding Operators to a cluster
Using Operator Lifecycle Manager (OLM), cluster administrators can install OLM-based Operators to an OpenShift Container Platform cluster.
For information on how OLM handles updates for installed Operators colocated in the same namespace, as well as an alternative method for installing Operators with custom global Operator groups, see Multitenancy and Operator colocation.
4.1.1. About Operator installation with OperatorHub
OperatorHub is a user interface for discovering Operators; it works in conjunction with Operator Lifecycle Manager (OLM), which installs and manages Operators on a cluster.
As a cluster administrator, you can install an Operator from OperatorHub by using the OpenShift Container Platform web console or CLI. Subscribing an Operator to one or more namespaces makes the Operator available to developers on your cluster.
During installation, you must determine the following initial settings for the Operator:
- Installation Mode
- Choose All namespaces on the cluster (default) to have the Operator installed on all namespaces or choose individual namespaces, if available, to only install the Operator on selected namespaces. This example chooses All namespaces… to make the Operator available to all users and projects.
- Update Channel
- If an Operator is available through multiple channels, you can choose which channel you want to subscribe to. For example, to deploy from the stable channel, if available, select it from the list.
- Approval Strategy
You can choose automatic or manual updates.
If you choose automatic updates for an installed Operator, when a new version of that Operator is available in the selected channel, Operator Lifecycle Manager (OLM) automatically upgrades the running instance of your Operator without human intervention.
If you select manual updates, when a newer version of an Operator is available, OLM creates an update request. As a cluster administrator, you must then manually approve that update request to have the Operator updated to the new version.
Additional resources
4.1.2. Installing from OperatorHub by using the web console
You can install and subscribe to an Operator from OperatorHub by using the OpenShift Container Platform web console.
Prerequisites
-
Access to an OpenShift Container Platform cluster using an account with
cluster-admin
permissions.
Procedure
- Navigate in the web console to the Operators → OperatorHub page.
Scroll or type a keyword into the Filter by keyword box to find the Operator you want. For example, type
advanced
to find the Advanced Cluster Management for Kubernetes Operator.You can also filter options by Infrastructure Features. For example, select Disconnected if you want to see Operators that work in disconnected environments, also known as restricted network environments.
Select the Operator to display additional information.
NoteChoosing a Community Operator warns that Red Hat does not certify Community Operators; you must acknowledge the warning before continuing.
- Read the information about the Operator and click Install.
On the Install Operator page, configure your Operator installation:
If you want to install a specific version of an Operator, select an Update channel and Version from the lists. You can browse the various versions of an Operator across any channels it might have, view the metadata for that channel and version, and select the exact version you want to install.
NoteThe version selection defaults to the latest version for the channel selected. If the latest version for the channel is selected, the Automatic approval strategy is enabled by default. Otherwise, Manual approval is required when not installing the latest version for the selected channel.
Installing an Operator with Manual approval causes all Operators installed within the namespace to function with the Manual approval strategy and all Operators are updated together. If you want to update Operators independently, install Operators into separate namespaces.
Confirm the installation mode for the Operator:
-
All namespaces on the cluster (default) installs the Operator in the default
openshift-operators
namespace to watch and be made available to all namespaces in the cluster. This option is not always available. - A specific namespace on the cluster allows you to choose a specific, single namespace in which to install the Operator. The Operator will only watch and be made available for use in this single namespace.
-
All namespaces on the cluster (default) installs the Operator in the default
For clusters on cloud providers with token authentication enabled:
- If the cluster uses AWS STS (STS Mode in the web console), enter the Amazon Resource Name (ARN) of the AWS IAM role of your service account in the role ARN field. To create the role’s ARN, follow the procedure described in Preparing AWS account.
- If the cluster uses Microsoft Entra Workload ID (Workload Identity / Federated Identity Mode in the web console), add the client ID, tenant ID, and subscription ID in the appropriate field.
For Update approval, select either the Automatic or Manual approval strategy.
ImportantIf the web console shows that the cluster uses AWS STS or Microsoft Entra Workload ID, you must set Update approval to Manual.
Subscriptions with automatic update approvals are not recommended because there might be permission changes to make prior to updating. Subscriptions with manual update approvals ensure that administrators have the opportunity to verify the permissions of the later version and take any necessary steps prior to update.
Click Install to make the Operator available to the selected namespaces on this OpenShift Container Platform cluster:
If you selected a Manual approval strategy, the upgrade status of the subscription remains Upgrading until you review and approve the install plan.
After approving on the Install Plan page, the subscription upgrade status moves to Up to date.
- If you selected an Automatic approval strategy, the upgrade status should resolve to Up to date without intervention.
Verification
After the upgrade status of the subscription is Up to date, select Operators → Installed Operators to verify that the cluster service version (CSV) of the installed Operator eventually shows up. The Status should eventually resolve to Succeeded in the relevant namespace.
NoteFor the All namespaces… installation mode, the status resolves to Succeeded in the
openshift-operators
namespace, but the status is Copied if you check in other namespaces.If it does not:
-
Check the logs in any pods in the
openshift-operators
project (or other relevant namespace if A specific namespace… installation mode was selected) on the Workloads → Pods page that are reporting issues to troubleshoot further.
-
Check the logs in any pods in the
When the Operator is installed, the metadata indicates which channel and version are installed.
NoteThe Channel and Version dropdown menus are still available for viewing other version metadata in this catalog context.
Additional resources
4.1.3. Installing from OperatorHub by using the CLI
Instead of using the OpenShift Container Platform web console, you can install an Operator from OperatorHub by using the CLI. Use the oc
command to create or update a Subscription
object.
For SingleNamespace
install mode, you must also ensure an appropriate Operator group exists in the related namespace. An Operator group, defined by an OperatorGroup
object, selects target namespaces in which to generate required RBAC access for all Operators in the same namespace as the Operator group.
In most cases, the web console method of this procedure is preferred because it automates tasks in the background, such as handling the creation of OperatorGroup
and Subscription
objects automatically when choosing SingleNamespace
mode.
Prerequisites
-
Access to an OpenShift Container Platform cluster using an account with
cluster-admin
permissions. -
You have installed the OpenShift CLI (
oc
).
Procedure
View the list of Operators available to the cluster from OperatorHub:
$ oc get packagemanifests -n openshift-marketplace
Example 4.1. Example output
NAME CATALOG AGE 3scale-operator Red Hat Operators 91m advanced-cluster-management Red Hat Operators 91m amq7-cert-manager Red Hat Operators 91m # ... couchbase-enterprise-certified Certified Operators 91m crunchy-postgres-operator Certified Operators 91m mongodb-enterprise Certified Operators 91m # ... etcd Community Operators 91m jaeger Community Operators 91m kubefed Community Operators 91m # ...
Note the catalog for your desired Operator.
Inspect your desired Operator to verify its supported install modes and available channels:
$ oc describe packagemanifests <operator_name> -n openshift-marketplace
Example 4.2. Example output
# ... Kind: PackageManifest # ... Install Modes: 1 Supported: true Type: OwnNamespace Supported: true Type: SingleNamespace Supported: false Type: MultiNamespace Supported: true Type: AllNamespaces # ... Entries: Name: example-operator.v3.7.11 Version: 3.7.11 Name: example-operator.v3.7.10 Version: 3.7.10 Name: stable-3.7 2 # ... Entries: Name: example-operator.v3.8.5 Version: 3.8.5 Name: example-operator.v3.8.4 Version: 3.8.4 Name: stable-3.8 3 Default Channel: stable-3.8 4
TipYou can print an Operator’s version and channel information in YAML format by running the following command:
$ oc get packagemanifests <operator_name> -n <catalog_namespace> -o yaml
If more than one catalog is installed in a namespace, run the following command to look up the available versions and channels of an Operator from a specific catalog:
$ oc get packagemanifest \ --selector=catalog=<catalogsource_name> \ --field-selector metadata.name=<operator_name> \ -n <catalog_namespace> -o yaml
ImportantIf you do not specify the Operator’s catalog, running the
oc get packagemanifest
andoc describe packagemanifest
commands might return a package from an unexpected catalog if the following conditions are met:- Multiple catalogs are installed in the same namespace.
- The catalogs contain the same Operators or Operators with the same name.
If the Operator you intend to install supports the
AllNamespaces
install mode, and you choose to use this mode, skip this step, because theopenshift-operators
namespace already has an appropriate Operator group in place by default, calledglobal-operators
.If the Operator you intend to install supports the
SingleNamespace
install mode, and you choose to use this mode, you must ensure an appropriate Operator group exists in the related namespace. If one does not exist, you can create create one by following these steps:ImportantYou can only have one Operator group per namespace. For more information, see "Operator groups".
Create an
OperatorGroup
object YAML file, for exampleoperatorgroup.yaml
, forSingleNamespace
install mode:Example
OperatorGroup
object forSingleNamespace
install modeapiVersion: operators.coreos.com/v1 kind: OperatorGroup metadata: name: <operatorgroup_name> namespace: <namespace> 1 spec: targetNamespaces: - <namespace> 2
Create the
OperatorGroup
object:$ oc apply -f operatorgroup.yaml
Create a
Subscription
object to subscribe a namespace to an Operator:Create a YAML file for the
Subscription
object, for examplesubscription.yaml
:NoteIf you want to subscribe to a specific version of an Operator, set the
startingCSV
field to the desired version and set theinstallPlanApproval
field toManual
to prevent the Operator from automatically upgrading if a later version exists in the catalog. For details, see the following "ExampleSubscription
object with a specific starting Operator version".Example 4.3. Example
Subscription
objectapiVersion: operators.coreos.com/v1alpha1 kind: Subscription metadata: name: <subscription_name> namespace: <namespace_per_install_mode> 1 spec: channel: <channel_name> 2 name: <operator_name> 3 source: <catalog_name> 4 sourceNamespace: <catalog_source_namespace> 5 config: env: 6 - name: ARGS value: "-v=10" envFrom: 7 - secretRef: name: license-secret volumes: 8 - name: <volume_name> configMap: name: <configmap_name> volumeMounts: 9 - mountPath: <directory_name> name: <volume_name> tolerations: 10 - operator: "Exists" resources: 11 requests: memory: "64Mi" cpu: "250m" limits: memory: "128Mi" cpu: "500m" nodeSelector: 12 foo: bar
- 1
- For default
AllNamespaces
install mode usage, specify theopenshift-operators
namespace. Alternatively, you can specify a custom global namespace, if you have created one. ForSingleNamespace
install mode usage, specify the relevant single namespace. - 2
- Name of the channel to subscribe to.
- 3
- Name of the Operator to subscribe to.
- 4
- Name of the catalog source that provides the Operator.
- 5
- Namespace of the catalog source. Use
openshift-marketplace
for the default OperatorHub catalog sources. - 6
- The
env
parameter defines a list of environment variables that must exist in all containers in the pod created by OLM. - 7
- The
envFrom
parameter defines a list of sources to populate environment variables in the container. - 8
- The
volumes
parameter defines a list of volumes that must exist on the pod created by OLM. - 9
- The
volumeMounts
parameter defines a list of volume mounts that must exist in all containers in the pod created by OLM. If avolumeMount
references avolume
that does not exist, OLM fails to deploy the Operator. - 10
- The
tolerations
parameter defines a list of tolerations for the pod created by OLM. - 11
- The
resources
parameter defines resource constraints for all the containers in the pod created by OLM. - 12
- The
nodeSelector
parameter defines aNodeSelector
for the pod created by OLM.
Example 4.4. Example
Subscription
object with a specific starting Operator versionapiVersion: operators.coreos.com/v1alpha1 kind: Subscription metadata: name: example-operator namespace: example-operator spec: channel: stable-3.7 installPlanApproval: Manual 1 name: example-operator source: custom-operators sourceNamespace: openshift-marketplace startingCSV: example-operator.v3.7.10 2
- 1
- Set the approval strategy to
Manual
in case your specified version is superseded by a later version in the catalog. This plan prevents an automatic upgrade to a later version and requires manual approval before the starting CSV can complete the installation. - 2
- Set a specific version of an Operator CSV.
For clusters on cloud providers with token authentication enabled, configure your
Subscription
object by following these steps:Ensure the
Subscription
object is set to manual update approvals:kind: Subscription # ... spec: installPlanApproval: Manual 1
- 1
- Subscriptions with automatic update approvals are not recommended because there might be permission changes to make prior to updating. Subscriptions with manual update approvals ensure that administrators have the opportunity to verify the permissions of the later version and take any necessary steps prior to update.
Include the relevant cloud provider-specific fields in the
Subscription
object’sconfig
section:If the cluster is in AWS STS mode, include the following fields:
kind: Subscription # ... spec: config: env: - name: ROLEARN value: "<role_arn>" 1
- 1
- Include the role ARN details.
If the cluster is in Microsoft Entra Workload ID mode, include the following fields:
kind: Subscription # ... spec: config: env: - name: CLIENTID value: "<client_id>" 1 - name: TENANTID value: "<tenant_id>" 2 - name: SUBSCRIPTIONID value: "<subscription_id>" 3
Create the
Subscription
object by running the following command:$ oc apply -f subscription.yaml
-
If you set the
installPlanApproval
field toManual
, manually approve the pending install plan to complete the Operator installation. For more information, see "Manually approving a pending Operator update".
At this point, OLM is now aware of the selected Operator. A cluster service version (CSV) for the Operator should appear in the target namespace, and APIs provided by the Operator should be available for creation.
Verification
Check the status of the
Subscription
object for your installed Operator by running the following command:$ oc describe subscription <subscription_name> -n <namespace>
If you created an Operator group for
SingleNamespace
install mode, check the status of theOperatorGroup
object by running the following command:$ oc describe operatorgroup <operatorgroup_name> -n <namespace>
4.1.4. Preparing for multiple instances of an Operator for multitenant clusters
As a cluster administrator, you can add multiple instances of an Operator for use in multitenant clusters. This is an alternative solution to either using the standard All namespaces install mode, which can be considered to violate the principle of least privilege, or the Multinamespace mode, which is not widely adopted. For more information, see "Operators in multitenant clusters".
In the following procedure, the tenant is a user or group of users that share common access and privileges for a set of deployed workloads. The tenant Operator is the instance of an Operator that is intended for use by only that tenant.
Prerequisites
All instances of the Operator you want to install must be the same version across a given cluster.
ImportantFor more information on this and other limitations, see "Operators in multitenant clusters".
Procedure
Before installing the Operator, create a namespace for the tenant Operator that is separate from the tenant’s namespace. For example, if the tenant’s namespace is
team1
, you might create ateam1-operator
namespace:Define a
Namespace
resource and save the YAML file, for example,team1-operator.yaml
:apiVersion: v1 kind: Namespace metadata: name: team1-operator
Create the namespace by running the following command:
$ oc create -f team1-operator.yaml
Create an Operator group for the tenant Operator scoped to the tenant’s namespace, with only that one namespace entry in the
spec.targetNamespaces
list:Define an
OperatorGroup
resource and save the YAML file, for example,team1-operatorgroup.yaml
:apiVersion: operators.coreos.com/v1 kind: OperatorGroup metadata: name: team1-operatorgroup namespace: team1-operator spec: targetNamespaces: - team1 1
- 1
- Define only the tenant’s namespace in the
spec.targetNamespaces
list.
Create the Operator group by running the following command:
$ oc create -f team1-operatorgroup.yaml
Next steps
Install the Operator in the tenant Operator namespace. This task is more easily performed by using the OperatorHub in the web console instead of the CLI; for a detailed procedure, see Installing from OperatorHub using the web console.
NoteAfter completing the Operator installation, the Operator resides in the tenant Operator namespace and watches the tenant namespace, but neither the Operator’s pod nor its service account are visible or usable by the tenant.
Additional resources
4.1.5. Installing global Operators in custom namespaces
When installing Operators with the OpenShift Container Platform web console, the default behavior installs Operators that support the All namespaces install mode into the default openshift-operators
global namespace. This can cause issues related to shared install plans and update policies between all Operators in the namespace. For more details on these limitations, see "Multitenancy and Operator colocation".
As a cluster administrator, you can bypass this default behavior manually by creating a custom global namespace and using that namespace to install your individual or scoped set of Operators and their dependencies.
Prerequisites
-
You have access to the cluster as a user with the
cluster-admin
role.
Procedure
Before installing the Operator, create a namespace for the installation of your desired Operator. This installation namespace will become the custom global namespace:
Define a
Namespace
resource and save the YAML file, for example,global-operators.yaml
:apiVersion: v1 kind: Namespace metadata: name: global-operators
Create the namespace by running the following command:
$ oc create -f global-operators.yaml
Create a custom global Operator group, which is an Operator group that watches all namespaces:
Define an
OperatorGroup
resource and save the YAML file, for example,global-operatorgroup.yaml
. Omit both thespec.selector
andspec.targetNamespaces
fields to make it a global Operator group, which selects all namespaces:apiVersion: operators.coreos.com/v1 kind: OperatorGroup metadata: name: global-operatorgroup namespace: global-operators
NoteThe
status.namespaces
of a created global Operator group contains the empty string (""
), which signals to a consuming Operator that it should watch all namespaces.Create the Operator group by running the following command:
$ oc create -f global-operatorgroup.yaml
Next steps
Install the desired Operator in your custom global namespace. Because the web console does not populate the Installed Namespace menu during Operator installation with custom global namespaces, this task can only be performed with the OpenShift CLI (
oc
). For a detailed procedure, see Installing from OperatorHub using the CLI.NoteWhen you initiate the Operator installation, if the Operator has dependencies, the dependencies are also automatically installed in the custom global namespace. As a result, it is then valid for the dependency Operators to have the same update policy and shared install plans.
Additional resources
4.1.6. Pod placement of Operator workloads
By default, Operator Lifecycle Manager (OLM) places pods on arbitrary worker nodes when installing an Operator or deploying Operand workloads. As an administrator, you can use projects with a combination of node selectors, taints, and tolerations to control the placement of Operators and Operands to specific nodes.
Controlling pod placement of Operator and Operand workloads has the following prerequisites:
-
Determine a node or set of nodes to target for the pods per your requirements. If available, note an existing label, such as
node-role.kubernetes.io/app
, that identifies the node or nodes. Otherwise, add a label, such asmyoperator
, by using a compute machine set or editing the node directly. You will use this label in a later step as the node selector on your project. -
If you want to ensure that only pods with a certain label are allowed to run on the nodes, while steering unrelated workloads to other nodes, add a taint to the node or nodes by using a compute machine set or editing the node directly. Use an effect that ensures that new pods that do not match the taint cannot be scheduled on the nodes. For example, a
myoperator:NoSchedule
taint ensures that new pods that do not match the taint are not scheduled onto that node, but existing pods on the node are allowed to remain. - Create a project that is configured with a default node selector and, if you added a taint, a matching toleration.
At this point, the project you created can be used to steer pods towards the specified nodes in the following scenarios:
- For Operator pods
-
Administrators can create a
Subscription
object in the project as described in the following section. As a result, the Operator pods are placed on the specified nodes. - For Operand pods
- Using an installed Operator, users can create an application in the project, which places the custom resource (CR) owned by the Operator in the project. As a result, the Operand pods are placed on the specified nodes, unless the Operator is deploying cluster-wide objects or resources in other namespaces, in which case this customized pod placement does not apply.
Additional resources
- Adding taints and tolerations manually to nodes or with compute machine sets
- Creating project-wide node selectors
- Creating a project with a node selector and toleration
4.1.7. Controlling where an Operator is installed
By default, when you install an Operator, OpenShift Container Platform installs the Operator pod to one of your worker nodes randomly. However, there might be situations where you want that pod scheduled on a specific node or set of nodes.
The following examples describe situations where you might want to schedule an Operator pod to a specific node or set of nodes:
-
If an Operator requires a particular platform, such as
amd64
orarm64
- If an Operator requires a particular operating system, such as Linux or Windows
- If you want Operators that work together scheduled on the same host or on hosts located on the same rack
- If you want Operators dispersed throughout the infrastructure to avoid downtime due to network or hardware issues
You can control where an Operator pod is installed by adding node affinity, pod affinity, or pod anti-affinity constraints to the Operator’s Subscription
object. Node affinity is a set of rules used by the scheduler to determine where a pod can be placed. Pod affinity enables you to ensure that related pods are scheduled to the same node. Pod anti-affinity allows you to prevent a pod from being scheduled on a node.
The following examples show how to use node affinity or pod anti-affinity to install an instance of the Custom Metrics Autoscaler Operator to a specific node in the cluster:
Node affinity example that places the Operator pod on a specific node
apiVersion: operators.coreos.com/v1alpha1
kind: Subscription
metadata:
name: openshift-custom-metrics-autoscaler-operator
namespace: openshift-keda
spec:
name: my-package
source: my-operators
sourceNamespace: operator-registries
config:
affinity:
nodeAffinity: 1
requiredDuringSchedulingIgnoredDuringExecution:
nodeSelectorTerms:
- matchExpressions:
- key: kubernetes.io/hostname
operator: In
values:
- ip-10-0-163-94.us-west-2.compute.internal
#...
- 1
- A node affinity that requires the Operator’s pod to be scheduled on a node named
ip-10-0-163-94.us-west-2.compute.internal
.
Node affinity example that places the Operator pod on a node with a specific platform
apiVersion: operators.coreos.com/v1alpha1
kind: Subscription
metadata:
name: openshift-custom-metrics-autoscaler-operator
namespace: openshift-keda
spec:
name: my-package
source: my-operators
sourceNamespace: operator-registries
config:
affinity:
nodeAffinity: 1
requiredDuringSchedulingIgnoredDuringExecution:
nodeSelectorTerms:
- matchExpressions:
- key: kubernetes.io/arch
operator: In
values:
- arm64
- key: kubernetes.io/os
operator: In
values:
- linux
#...
- 1
- A node affinity that requires the Operator’s pod to be scheduled on a node with the
kubernetes.io/arch=arm64
andkubernetes.io/os=linux
labels.
Pod affinity example that places the Operator pod on one or more specific nodes
apiVersion: operators.coreos.com/v1alpha1
kind: Subscription
metadata:
name: openshift-custom-metrics-autoscaler-operator
namespace: openshift-keda
spec:
name: my-package
source: my-operators
sourceNamespace: operator-registries
config:
affinity:
podAffinity: 1
requiredDuringSchedulingIgnoredDuringExecution:
- labelSelector:
matchExpressions:
- key: app
operator: In
values:
- test
topologyKey: kubernetes.io/hostname
#...
- 1
- A pod affinity that places the Operator’s pod on a node that has pods with the
app=test
label.
Pod anti-affinity example that prevents the Operator pod from one or more specific nodes
apiVersion: operators.coreos.com/v1alpha1
kind: Subscription
metadata:
name: openshift-custom-metrics-autoscaler-operator
namespace: openshift-keda
spec:
name: my-package
source: my-operators
sourceNamespace: operator-registries
config:
affinity:
podAntiAffinity: 1
requiredDuringSchedulingIgnoredDuringExecution:
- labelSelector:
matchExpressions:
- key: cpu
operator: In
values:
- high
topologyKey: kubernetes.io/hostname
#...
- 1
- A pod anti-affinity that prevents the Operator’s pod from being scheduled on a node that has pods with the
cpu=high
label.
Procedure
To control the placement of an Operator pod, complete the following steps:
- Install the Operator as usual.
- If needed, ensure that your nodes are labeled to properly respond to the affinity.
Edit the Operator
Subscription
object to add an affinity:apiVersion: operators.coreos.com/v1alpha1 kind: Subscription metadata: name: openshift-custom-metrics-autoscaler-operator namespace: openshift-keda spec: name: my-package source: my-operators sourceNamespace: operator-registries config: affinity: 1 nodeAffinity: requiredDuringSchedulingIgnoredDuringExecution: nodeSelectorTerms: - matchExpressions: - key: kubernetes.io/hostname operator: In values: - ip-10-0-185-229.ec2.internal #...
- 1
- Add a
nodeAffinity
,podAffinity
, orpodAntiAffinity
. See the Additional resources section that follows for information about creating the affinity.
Verification
To ensure that the pod is deployed on the specific node, run the following command:
$ oc get pods -o wide
Example output
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES custom-metrics-autoscaler-operator-5dcc45d656-bhshg 1/1 Running 0 50s 10.131.0.20 ip-10-0-185-229.ec2.internal <none> <none>
4.2. Updating installed Operators
As a cluster administrator, you can update Operators that have been previously installed using Operator Lifecycle Manager (OLM) on your OpenShift Container Platform cluster.
For information on how OLM handles updates for installed Operators colocated in the same namespace, as well as an alternative method for installing Operators with custom global Operator groups, see Multitenancy and Operator colocation.
4.2.1. Preparing for an Operator update
The subscription of an installed Operator specifies an update channel that tracks and receives updates for the Operator. You can change the update channel to start tracking and receiving updates from a newer channel.
The names of update channels in a subscription can differ between Operators, but the naming scheme typically follows a common convention within a given Operator. For example, channel names might follow a minor release update stream for the application provided by the Operator (1.2
, 1.3
) or a release frequency (stable
, fast
).
You cannot change installed Operators to a channel that is older than the current channel.
Red Hat Customer Portal Labs include the following application that helps administrators prepare to update their Operators:
You can use the application to search for Operator Lifecycle Manager-based Operators and verify the available Operator version per update channel across different versions of OpenShift Container Platform. Cluster Version Operator-based Operators are not included.
4.2.2. Changing the update channel for an Operator
You can change the update channel for an Operator by using the OpenShift Container Platform web console.
If the approval strategy in the subscription is set to Automatic, the update process initiates as soon as a new Operator version is available in the selected channel. If the approval strategy is set to Manual, you must manually approve pending updates.
Prerequisites
- An Operator previously installed using Operator Lifecycle Manager (OLM).
Procedure
- In the Administrator perspective of the web console, navigate to Operators → Installed Operators.
- Click the name of the Operator you want to change the update channel for.
- Click the Subscription tab.
- Click the name of the update channel under Update channel.
- Click the newer update channel that you want to change to, then click Save.
For subscriptions with an Automatic approval strategy, the update begins automatically. Navigate back to the Operators → Installed Operators page to monitor the progress of the update. When complete, the status changes to Succeeded and Up to date.
For subscriptions with a Manual approval strategy, you can manually approve the update from the Subscription tab.
4.2.3. Manually approving a pending Operator update
If an installed Operator has the approval strategy in its subscription set to Manual, when new updates are released in its current update channel, the update must be manually approved before installation can begin.
Prerequisites
- An Operator previously installed using Operator Lifecycle Manager (OLM).
Procedure
- In the Administrator perspective of the OpenShift Container Platform web console, navigate to Operators → Installed Operators.
- Operators that have a pending update display a status with Upgrade available. Click the name of the Operator you want to update.
- Click the Subscription tab. Any updates requiring approval are displayed next to Upgrade status. For example, it might display 1 requires approval.
- Click 1 requires approval, then click Preview Install Plan.
- Review the resources that are listed as available for update. When satisfied, click Approve.
- Navigate back to the Operators → Installed Operators page to monitor the progress of the update. When complete, the status changes to Succeeded and Up to date.
4.2.4. Additional resources
4.3. Deleting Operators from a cluster
The following describes how to delete, or uninstall, Operators that were previously installed using Operator Lifecycle Manager (OLM) on your OpenShift Container Platform cluster.
You must successfully and completely uninstall an Operator prior to attempting to reinstall the same Operator. Failure to fully uninstall the Operator properly can leave resources, such as a project or namespace, stuck in a "Terminating" state and cause "error resolving resource" messages to be observed when trying to reinstall the Operator.
For more information, see Reinstalling Operators after failed uninstallation.
4.3.1. Deleting Operators from a cluster using the web console
Cluster administrators can delete installed Operators from a selected namespace by using the web console.
Prerequisites
-
You have access to an OpenShift Container Platform cluster web console using an account with
cluster-admin
permissions.
Procedure
- Navigate to the Operators → Installed Operators page.
- Scroll or enter a keyword into the Filter by name field to find the Operator that you want to remove. Then, click on it.
On the right side of the Operator Details page, select Uninstall Operator from the Actions list.
An Uninstall Operator? dialog box is displayed.
Select Uninstall to remove the Operator, Operator deployments, and pods. Following this action, the Operator stops running and no longer receives updates.
NoteThis action does not remove resources managed by the Operator, including custom resource definitions (CRDs) and custom resources (CRs). Dashboards and navigation items enabled by the web console and off-cluster resources that continue to run might need manual clean up. To remove these after uninstalling the Operator, you might need to manually delete the Operator CRDs.
4.3.2. Deleting Operators from a cluster using the CLI
Cluster administrators can delete installed Operators from a selected namespace by using the CLI.
Prerequisites
-
You have access to an OpenShift Container Platform cluster using an account with
cluster-admin
permissions. -
The OpenShift CLI (
oc
) is installed on your workstation.
Procedure
Ensure the latest version of the subscribed operator (for example,
serverless-operator
) is identified in thecurrentCSV
field.$ oc get subscription.operators.coreos.com serverless-operator -n openshift-serverless -o yaml | grep currentCSV
Example output
currentCSV: serverless-operator.v1.28.0
Delete the subscription (for example,
serverless-operator
):$ oc delete subscription.operators.coreos.com serverless-operator -n openshift-serverless
Example output
subscription.operators.coreos.com "serverless-operator" deleted
Delete the CSV for the Operator in the target namespace using the
currentCSV
value from the previous step:$ oc delete clusterserviceversion serverless-operator.v1.28.0 -n openshift-serverless
Example output
clusterserviceversion.operators.coreos.com "serverless-operator.v1.28.0" deleted
4.3.3. Refreshing failing subscriptions
In Operator Lifecycle Manager (OLM), if you subscribe to an Operator that references images that are not accessible on your network, you can find jobs in the openshift-marketplace
namespace that are failing with the following errors:
Example output
ImagePullBackOff for Back-off pulling image "example.com/openshift4/ose-elasticsearch-operator-bundle@sha256:6d2587129c846ec28d384540322b40b05833e7e00b25cca584e004af9a1d292e"
Example output
rpc error: code = Unknown desc = error pinging docker registry example.com: Get "https://example.com/v2/": dial tcp: lookup example.com on 10.0.0.1:53: no such host
As a result, the subscription is stuck in this failing state and the Operator is unable to install or upgrade.
You can refresh a failing subscription by deleting the subscription, cluster service version (CSV), and other related objects. After recreating the subscription, OLM then reinstalls the correct version of the Operator.
Prerequisites
- You have a failing subscription that is unable to pull an inaccessible bundle image.
- You have confirmed that the correct bundle image is accessible.
Procedure
Get the names of the
Subscription
andClusterServiceVersion
objects from the namespace where the Operator is installed:$ oc get sub,csv -n <namespace>
Example output
NAME PACKAGE SOURCE CHANNEL subscription.operators.coreos.com/elasticsearch-operator elasticsearch-operator redhat-operators 5.0 NAME DISPLAY VERSION REPLACES PHASE clusterserviceversion.operators.coreos.com/elasticsearch-operator.5.0.0-65 OpenShift Elasticsearch Operator 5.0.0-65 Succeeded
Delete the subscription:
$ oc delete subscription <subscription_name> -n <namespace>
Delete the cluster service version:
$ oc delete csv <csv_name> -n <namespace>
Get the names of any failing jobs and related config maps in the
openshift-marketplace
namespace:$ oc get job,configmap -n openshift-marketplace
Example output
NAME COMPLETIONS DURATION AGE job.batch/1de9443b6324e629ddf31fed0a853a121275806170e34c926d69e53a7fcbccb 1/1 26s 9m30s NAME DATA AGE configmap/1de9443b6324e629ddf31fed0a853a121275806170e34c926d69e53a7fcbccb 3 9m30s
Delete the job:
$ oc delete job <job_name> -n openshift-marketplace
This ensures pods that try to pull the inaccessible image are not recreated.
Delete the config map:
$ oc delete configmap <configmap_name> -n openshift-marketplace
- Reinstall the Operator using OperatorHub in the web console.
Verification
Check that the Operator has been reinstalled successfully:
$ oc get sub,csv,installplan -n <namespace>
4.4. Configuring Operator Lifecycle Manager features
The Operator Lifecycle Manager (OLM) controller is configured by an OLMConfig
custom resource (CR) named cluster
. Cluster administrators can modify this resource to enable or disable certain features.
This document outlines the features currently supported by OLM that are configured by the OLMConfig
resource.
4.4.1. Disabling copied CSVs
When an Operator is installed by Operator Lifecycle Manager (OLM), a simplified copy of its cluster service version (CSV) is created by default in every namespace that the Operator is configured to watch. These CSVs are known as copied CSVs and communicate to users which controllers are actively reconciling resource events in a given namespace.
When an Operator is configured to use the AllNamespaces
install mode, versus targeting a single or specified set of namespaces, a copied CSV for the Operator is created in every namespace on the cluster. On especially large clusters, with namespaces and installed Operators potentially in the hundreds or thousands, copied CSVs consume an untenable amount of resources, such as OLM’s memory usage, cluster etcd limits, and networking.
To support these larger clusters, cluster administrators can disable copied CSVs for Operators globally installed with the AllNamespaces
mode.
If you disable copied CSVs, an Operator installed in AllNamespaces
mode has their CSV copied only to the openshift
namespace, instead of every namespace on the cluster. In disabled copied CSVs mode, the behavior differs between the web console and CLI:
-
In the web console, the default behavior is modified to show copied CSVs from the
openshift
namespace in every namespace, even though the CSVs are not actually copied to every namespace. This allows regular users to still be able to view the details of these Operators in their namespaces and create related custom resources (CRs). In the OpenShift CLI (
oc
), regular users can view Operators installed directly in their namespaces by using theoc get csvs
command, but the copied CSVs from theopenshift
namespace are not visible in their namespaces. Operators affected by this limitation are still available and continue to reconcile events in the user’s namespace.To view a full list of installed global Operators, similar to the web console behavior, all authenticated users can run the following command:
$ oc get csvs -n openshift
Procedure
Edit the
OLMConfig
object namedcluster
and set thespec.features.disableCopiedCSVs
field totrue
:$ oc apply -f - <<EOF apiVersion: operators.coreos.com/v1 kind: OLMConfig metadata: name: cluster spec: features: disableCopiedCSVs: true 1 EOF
- 1
- Disabled copied CSVs for
AllNamespaces
install mode Operators
Verification
When copied CSVs are disabled, OLM captures this information in an event in the Operator’s namespace:
$ oc get events
Example output
LAST SEEN TYPE REASON OBJECT MESSAGE 85s Warning DisabledCopiedCSVs clusterserviceversion/my-csv.v1.0.0 CSV copying disabled for operators/my-csv.v1.0.0
When the
spec.features.disableCopiedCSVs
field is missing or set tofalse
, OLM recreates the copied CSVs for all Operators installed with theAllNamespaces
mode and deletes the previously mentioned events.
Additional resources
4.5. Configuring proxy support in Operator Lifecycle Manager
If a global proxy is configured on the OpenShift Container Platform cluster, Operator Lifecycle Manager (OLM) automatically configures Operators that it manages with the cluster-wide proxy. However, you can also configure installed Operators to override the global proxy or inject a custom CA certificate.
Additional resources
- Configuring a custom PKI (custom CA certificate)
- Developing Operators that support proxy settings for Go, Ansible, and Helm
4.5.1. Overriding proxy settings of an Operator
If a cluster-wide egress proxy is configured, Operators running with Operator Lifecycle Manager (OLM) inherit the cluster-wide proxy settings on their deployments. Cluster administrators can also override these proxy settings by configuring the subscription of an Operator.
Operators must handle setting environment variables for proxy settings in the pods for any managed Operands.
Prerequisites
-
Access to an OpenShift Container Platform cluster using an account with
cluster-admin
permissions.
Procedure
- Navigate in the web console to the Operators → OperatorHub page.
- Select the Operator and click Install.
On the Install Operator page, modify the
Subscription
object to include one or more of the following environment variables in thespec
section:-
HTTP_PROXY
-
HTTPS_PROXY
-
NO_PROXY
For example:
Subscription
object with proxy setting overridesapiVersion: operators.coreos.com/v1alpha1 kind: Subscription metadata: name: etcd-config-test namespace: openshift-operators spec: config: env: - name: HTTP_PROXY value: test_http - name: HTTPS_PROXY value: test_https - name: NO_PROXY value: test channel: clusterwide-alpha installPlanApproval: Automatic name: etcd source: community-operators sourceNamespace: openshift-marketplace startingCSV: etcdoperator.v0.9.4-clusterwide
NoteThese environment variables can also be unset using an empty value to remove any previously set cluster-wide or custom proxy settings.
OLM handles these environment variables as a unit; if at least one of them is set, all three are considered overridden and the cluster-wide defaults are not used for the deployments of the subscribed Operator.
-
- Click Install to make the Operator available to the selected namespaces.
After the CSV for the Operator appears in the relevant namespace, you can verify that custom proxy environment variables are set in the deployment. For example, using the CLI:
$ oc get deployment -n openshift-operators \ etcd-operator -o yaml \ | grep -i "PROXY" -A 2
Example output
- name: HTTP_PROXY value: test_http - name: HTTPS_PROXY value: test_https - name: NO_PROXY value: test image: quay.io/coreos/etcd-operator@sha256:66a37fd61a06a43969854ee6d3e21088a98b93838e284a6086b13917f96b0d9c ...
4.5.2. Injecting a custom CA certificate
When a cluster administrator adds a custom CA certificate to a cluster using a config map, the Cluster Network Operator merges the user-provided certificates and system CA certificates into a single bundle. You can inject this merged bundle into your Operator running on Operator Lifecycle Manager (OLM), which is useful if you have a man-in-the-middle HTTPS proxy.
Prerequisites
-
Access to an OpenShift Container Platform cluster using an account with
cluster-admin
permissions. - Custom CA certificate added to the cluster using a config map.
- Desired Operator installed and running on OLM.
Procedure
Create an empty config map in the namespace where the subscription for your Operator exists and include the following label:
apiVersion: v1 kind: ConfigMap metadata: name: trusted-ca 1 labels: config.openshift.io/inject-trusted-cabundle: "true" 2
After creating this config map, it is immediately populated with the certificate contents of the merged bundle.
Update the
Subscription
object to include aspec.config
section that mounts thetrusted-ca
config map as a volume to each container within a pod that requires a custom CA:apiVersion: operators.coreos.com/v1alpha1 kind: Subscription metadata: name: my-operator spec: package: etcd channel: alpha config: 1 selector: matchLabels: <labels_for_pods> 2 volumes: 3 - name: trusted-ca configMap: name: trusted-ca items: - key: ca-bundle.crt 4 path: tls-ca-bundle.pem 5 volumeMounts: 6 - name: trusted-ca mountPath: /etc/pki/ca-trust/extracted/pem readOnly: true
NoteDeployments of an Operator can fail to validate the authority and display a
x509 certificate signed by unknown authority
error. This error can occur even after injecting a custom CA when using the subscription of an Operator. In this case, you can set themountPath
as/etc/ssl/certs
for trusted-ca by using the subscription of an Operator.
4.6. Viewing Operator status
Understanding the state of the system in Operator Lifecycle Manager (OLM) is important for making decisions about and debugging problems with installed Operators. OLM provides insight into subscriptions and related catalog sources regarding their state and actions performed. This helps users better understand the healthiness of their Operators.
4.6.1. Operator subscription condition types
Subscriptions can report the following condition types:
Condition | Description |
---|---|
| Some or all of the catalog sources to be used in resolution are unhealthy. |
| An install plan for a subscription is missing. |
| An install plan for a subscription is pending installation. |
| An install plan for a subscription has failed. |
| The dependency resolution for a subscription has failed. |
Default OpenShift Container Platform cluster Operators are managed by the Cluster Version Operator (CVO) and they do not have a Subscription
object. Application Operators are managed by Operator Lifecycle Manager (OLM) and they have a Subscription
object.
Additional resources
4.6.2. Viewing Operator subscription status by using the CLI
You can view Operator subscription status by using the CLI.
Prerequisites
-
You have access to the cluster as a user with the
cluster-admin
role. -
You have installed the OpenShift CLI (
oc
).
Procedure
List Operator subscriptions:
$ oc get subs -n <operator_namespace>
Use the
oc describe
command to inspect aSubscription
resource:$ oc describe sub <subscription_name> -n <operator_namespace>
In the command output, find the
Conditions
section for the status of Operator subscription condition types. In the following example, theCatalogSourcesUnhealthy
condition type has a status offalse
because all available catalog sources are healthy:Example output
Name: cluster-logging Namespace: openshift-logging Labels: operators.coreos.com/cluster-logging.openshift-logging= Annotations: <none> API Version: operators.coreos.com/v1alpha1 Kind: Subscription # ... Conditions: Last Transition Time: 2019-07-29T13:42:57Z Message: all available catalogsources are healthy Reason: AllCatalogSourcesHealthy Status: False Type: CatalogSourcesUnhealthy # ...
Default OpenShift Container Platform cluster Operators are managed by the Cluster Version Operator (CVO) and they do not have a Subscription
object. Application Operators are managed by Operator Lifecycle Manager (OLM) and they have a Subscription
object.
4.6.3. Viewing Operator catalog source status by using the CLI
You can view the status of an Operator catalog source by using the CLI.
Prerequisites
-
You have access to the cluster as a user with the
cluster-admin
role. -
You have installed the OpenShift CLI (
oc
).
Procedure
List the catalog sources in a namespace. For example, you can check the
openshift-marketplace
namespace, which is used for cluster-wide catalog sources:$ oc get catalogsources -n openshift-marketplace
Example output
NAME DISPLAY TYPE PUBLISHER AGE certified-operators Certified Operators grpc Red Hat 55m community-operators Community Operators grpc Red Hat 55m example-catalog Example Catalog grpc Example Org 2m25s redhat-marketplace Red Hat Marketplace grpc Red Hat 55m redhat-operators Red Hat Operators grpc Red Hat 55m
Use the
oc describe
command to get more details and status about a catalog source:$ oc describe catalogsource example-catalog -n openshift-marketplace
Example output
Name: example-catalog Namespace: openshift-marketplace Labels: <none> Annotations: operatorframework.io/managed-by: marketplace-operator target.workload.openshift.io/management: {"effect": "PreferredDuringScheduling"} API Version: operators.coreos.com/v1alpha1 Kind: CatalogSource # ... Status: Connection State: Address: example-catalog.openshift-marketplace.svc:50051 Last Connect: 2021-09-09T17:07:35Z Last Observed State: TRANSIENT_FAILURE Registry Service: Created At: 2021-09-09T17:05:45Z Port: 50051 Protocol: grpc Service Name: example-catalog Service Namespace: openshift-marketplace # ...
In the preceding example output, the last observed state is
TRANSIENT_FAILURE
. This state indicates that there is a problem establishing a connection for the catalog source.List the pods in the namespace where your catalog source was created:
$ oc get pods -n openshift-marketplace
Example output
NAME READY STATUS RESTARTS AGE certified-operators-cv9nn 1/1 Running 0 36m community-operators-6v8lp 1/1 Running 0 36m marketplace-operator-86bfc75f9b-jkgbc 1/1 Running 0 42m example-catalog-bwt8z 0/1 ImagePullBackOff 0 3m55s redhat-marketplace-57p8c 1/1 Running 0 36m redhat-operators-smxx8 1/1 Running 0 36m
When a catalog source is created in a namespace, a pod for the catalog source is created in that namespace. In the preceding example output, the status for the
example-catalog-bwt8z
pod isImagePullBackOff
. This status indicates that there is an issue pulling the catalog source’s index image.Use the
oc describe
command to inspect a pod for more detailed information:$ oc describe pod example-catalog-bwt8z -n openshift-marketplace
Example output
Name: example-catalog-bwt8z Namespace: openshift-marketplace Priority: 0 Node: ci-ln-jyryyg2-f76d1-ggdbq-worker-b-vsxjd/10.0.128.2 ... Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal Scheduled 48s default-scheduler Successfully assigned openshift-marketplace/example-catalog-bwt8z to ci-ln-jyryyf2-f76d1-fgdbq-worker-b-vsxjd Normal AddedInterface 47s multus Add eth0 [10.131.0.40/23] from openshift-sdn Normal BackOff 20s (x2 over 46s) kubelet Back-off pulling image "quay.io/example-org/example-catalog:v1" Warning Failed 20s (x2 over 46s) kubelet Error: ImagePullBackOff Normal Pulling 8s (x3 over 47s) kubelet Pulling image "quay.io/example-org/example-catalog:v1" Warning Failed 8s (x3 over 47s) kubelet Failed to pull image "quay.io/example-org/example-catalog:v1": rpc error: code = Unknown desc = reading manifest v1 in quay.io/example-org/example-catalog: unauthorized: access to the requested resource is not authorized Warning Failed 8s (x3 over 47s) kubelet Error: ErrImagePull
In the preceding example output, the error messages indicate that the catalog source’s index image is failing to pull successfully because of an authorization issue. For example, the index image might be stored in a registry that requires login credentials.
Additional resources
4.7. Managing Operator conditions
As a cluster administrator, you can manage Operator conditions by using Operator Lifecycle Manager (OLM).
4.7.1. Overriding Operator conditions
As a cluster administrator, you might want to ignore a supported Operator condition reported by an Operator. When present, Operator conditions in the Spec.Overrides
array override the conditions in the Spec.Conditions
array, allowing cluster administrators to deal with situations where an Operator is incorrectly reporting a state to Operator Lifecycle Manager (OLM).
By default, the Spec.Overrides
array is not present in an OperatorCondition
object until it is added by a cluster administrator . The Spec.Conditions
array is also not present until it is either added by a user or as a result of custom Operator logic.
For example, consider a known version of an Operator that always communicates that it is not upgradeable. In this instance, you might want to upgrade the Operator despite the Operator communicating that it is not upgradeable. This could be accomplished by overriding the Operator condition by adding the condition type
and status
to the Spec.Overrides
array in the OperatorCondition
object.
Prerequisites
-
You have access to the cluster as a user with the
cluster-admin
role. -
An Operator with an
OperatorCondition
object, installed using OLM.
Procedure
Edit the
OperatorCondition
object for the Operator:$ oc edit operatorcondition <name>
Add a
Spec.Overrides
array to the object:Example Operator condition override
apiVersion: operators.coreos.com/v2 kind: OperatorCondition metadata: name: my-operator namespace: operators spec: overrides: - type: Upgradeable 1 status: "True" reason: "upgradeIsSafe" message: "This is a known issue with the Operator where it always reports that it cannot be upgraded." conditions: - type: Upgradeable status: "False" reason: "migration" message: "The operator is performing a migration." lastTransitionTime: "2020-08-24T23:15:55Z"
- 1
- Allows the cluster administrator to change the upgrade readiness to
True
.
4.7.2. Updating your Operator to use Operator conditions
Operator Lifecycle Manager (OLM) automatically creates an OperatorCondition
resource for each ClusterServiceVersion
resource that it reconciles. All service accounts in the CSV are granted the RBAC to interact with the OperatorCondition
owned by the Operator.
An Operator author can develop their Operator to use the operator-lib
library such that, after the Operator has been deployed by OLM, it can set its own conditions. For more resources about setting Operator conditions as an Operator author, see the Enabling Operator conditions page.
4.7.2.1. Setting defaults
In an effort to remain backwards compatible, OLM treats the absence of an OperatorCondition
resource as opting out of the condition. Therefore, an Operator that opts in to using Operator conditions should set default conditions before the ready probe for the pod is set to true
. This provides the Operator with a grace period to update the condition to the correct state.
4.7.3. Additional resources
4.8. Allowing non-cluster administrators to install Operators
Cluster administrators can use Operator groups to allow regular users to install Operators.
Additional resources
4.8.1. Understanding Operator installation policy
Operators can require wide privileges to run, and the required privileges can change between versions. Operator Lifecycle Manager (OLM) runs with cluster-admin
privileges. By default, Operator authors can specify any set of permissions in the cluster service version (CSV), and OLM consequently grants it to the Operator.
To ensure that an Operator cannot achieve cluster-scoped privileges and that users cannot escalate privileges using OLM, Cluster administrators can manually audit Operators before they are added to the cluster. Cluster administrators are also provided tools for determining and constraining which actions are allowed during an Operator installation or upgrade using service accounts.
Cluster administrators can associate an Operator group with a service account that has a set of privileges granted to it. The service account sets policy on Operators to ensure they only run within predetermined boundaries by using role-based access control (RBAC) rules. As a result, the Operator is unable to do anything that is not explicitly permitted by those rules.
By employing Operator groups, users with enough privileges can install Operators with a limited scope. As a result, more of the Operator Framework tools can safely be made available to more users, providing a richer experience for building applications with Operators.
Role-based access control (RBAC) for Subscription
objects is automatically granted to every user with the edit
or admin
role in a namespace. However, RBAC does not exist on OperatorGroup
objects; this absence is what prevents regular users from installing Operators. Preinstalling Operator groups is effectively what gives installation privileges.
Keep the following points in mind when associating an Operator group with a service account:
-
The
APIService
andCustomResourceDefinition
resources are always created by OLM using thecluster-admin
role. A service account associated with an Operator group should never be granted privileges to write these resources. - Any Operator tied to this Operator group is now confined to the permissions granted to the specified service account. If the Operator asks for permissions that are outside the scope of the service account, the install fails with appropriate errors so the cluster administrator can troubleshoot and resolve the issue.
4.8.1.1. Installation scenarios
When determining whether an Operator can be installed or upgraded on a cluster, Operator Lifecycle Manager (OLM) considers the following scenarios:
- A cluster administrator creates a new Operator group and specifies a service account. All Operator(s) associated with this Operator group are installed and run against the privileges granted to the service account.
- A cluster administrator creates a new Operator group and does not specify any service account. OpenShift Container Platform maintains backward compatibility, so the default behavior remains and Operator installs and upgrades are permitted.
- For existing Operator groups that do not specify a service account, the default behavior remains and Operator installs and upgrades are permitted.
- A cluster administrator updates an existing Operator group and specifies a service account. OLM allows the existing Operator to continue to run with their current privileges. When such an existing Operator is going through an upgrade, it is reinstalled and run against the privileges granted to the service account like any new Operator.
- A service account specified by an Operator group changes by adding or removing permissions, or the existing service account is swapped with a new one. When existing Operators go through an upgrade, it is reinstalled and run against the privileges granted to the updated service account like any new Operator.
- A cluster administrator removes the service account from an Operator group. The default behavior remains and Operator installs and upgrades are permitted.
4.8.1.2. Installation workflow
When an Operator group is tied to a service account and an Operator is installed or upgraded, Operator Lifecycle Manager (OLM) uses the following workflow:
-
The given
Subscription
object is picked up by OLM. - OLM fetches the Operator group tied to this subscription.
- OLM determines that the Operator group has a service account specified.
- OLM creates a client scoped to the service account and uses the scoped client to install the Operator. This ensures that any permission requested by the Operator is always confined to that of the service account in the Operator group.
- OLM creates a new service account with the set of permissions specified in the CSV and assigns it to the Operator. The Operator runs as the assigned service account.
4.8.2. Scoping Operator installations
To provide scoping rules to Operator installations and upgrades on Operator Lifecycle Manager (OLM), associate a service account with an Operator group.
Using this example, a cluster administrator can confine a set of Operators to a designated namespace.
Prerequisites
-
You have access to the cluster as a user with the
cluster-admin
role. -
You have installed the OpenShift CLI (
oc
).
Procedure
Create a new namespace:
$ cat <<EOF | oc create -f - apiVersion: v1 kind: Namespace metadata: name: scoped EOF
Allocate permissions that you want the Operator(s) to be confined to. This involves creating a new service account, relevant role(s), and role binding(s).
$ cat <<EOF | oc create -f - apiVersion: v1 kind: ServiceAccount metadata: name: scoped namespace: scoped EOF
The following example grants the service account permissions to do anything in the designated namespace for simplicity. In a production environment, you should create a more fine-grained set of permissions:
$ cat <<EOF | oc create -f - apiVersion: rbac.authorization.k8s.io/v1 kind: Role metadata: name: scoped namespace: scoped rules: - apiGroups: ["*"] resources: ["*"] verbs: ["*"] --- apiVersion: rbac.authorization.k8s.io/v1 kind: RoleBinding metadata: name: scoped-bindings namespace: scoped roleRef: apiGroup: rbac.authorization.k8s.io kind: Role name: scoped subjects: - kind: ServiceAccount name: scoped namespace: scoped EOF
Create an
OperatorGroup
object in the designated namespace. This Operator group targets the designated namespace to ensure that its tenancy is confined to it.In addition, Operator groups allow a user to specify a service account. Specify the service account created in the previous step:
$ cat <<EOF | oc create -f - apiVersion: operators.coreos.com/v1 kind: OperatorGroup metadata: name: scoped namespace: scoped spec: serviceAccountName: scoped targetNamespaces: - scoped EOF
Any Operator installed in the designated namespace is tied to this Operator group and therefore to the service account specified.
Create a
Subscription
object in the designated namespace to install an Operator:$ cat <<EOF | oc create -f - apiVersion: operators.coreos.com/v1alpha1 kind: Subscription metadata: name: etcd namespace: scoped spec: channel: singlenamespace-alpha name: etcd source: <catalog_source_name> 1 sourceNamespace: <catalog_source_namespace> 2 EOF
Any Operator tied to this Operator group is confined to the permissions granted to the specified service account. If the Operator requests permissions that are outside the scope of the service account, the installation fails with relevant errors.
4.8.2.1. Fine-grained permissions
Operator Lifecycle Manager (OLM) uses the service account specified in an Operator group to create or update the following resources related to the Operator being installed:
-
ClusterServiceVersion
-
Subscription
-
Secret
-
ServiceAccount
-
Service
-
ClusterRole
andClusterRoleBinding
-
Role
andRoleBinding
To confine Operators to a designated namespace, cluster administrators can start by granting the following permissions to the service account:
The following role is a generic example and additional rules might be required based on the specific Operator.
kind: Role rules: - apiGroups: ["operators.coreos.com"] resources: ["subscriptions", "clusterserviceversions"] verbs: ["get", "create", "update", "patch"] - apiGroups: [""] resources: ["services", "serviceaccounts"] verbs: ["get", "create", "update", "patch"] - apiGroups: ["rbac.authorization.k8s.io"] resources: ["roles", "rolebindings"] verbs: ["get", "create", "update", "patch"] - apiGroups: ["apps"] 1 resources: ["deployments"] verbs: ["list", "watch", "get", "create", "update", "patch", "delete"] - apiGroups: [""] 2 resources: ["pods"] verbs: ["list", "watch", "get", "create", "update", "patch", "delete"]
In addition, if any Operator specifies a pull secret, the following permissions must also be added:
kind: ClusterRole 1
rules:
- apiGroups: [""]
resources: ["secrets"]
verbs: ["get"]
---
kind: Role
rules:
- apiGroups: [""]
resources: ["secrets"]
verbs: ["create", "update", "patch"]
- 1
- Required to get the secret from the OLM namespace.
4.8.3. Operator catalog access control
When an Operator catalog is created in the global catalog namespace openshift-marketplace
, the catalog’s Operators are made available cluster-wide to all namespaces. A catalog created in other namespaces only makes its Operators available in that same namespace of the catalog.
On clusters where non-cluster administrator users have been delegated Operator installation privileges, cluster administrators might want to further control or restrict the set of Operators those users are allowed to install. This can be achieved with the following actions:
- Disable all of the default global catalogs.
- Enable custom, curated catalogs in the same namespace where the relevant Operator groups have been preinstalled.
Additional resources
4.8.4. Troubleshooting permission failures
If an Operator installation fails due to lack of permissions, identify the errors using the following procedure.
Procedure
Review the
Subscription
object. Its status has an object referenceinstallPlanRef
that points to theInstallPlan
object that attempted to create the necessary[Cluster]Role[Binding]
object(s) for the Operator:apiVersion: operators.coreos.com/v1 kind: Subscription metadata: name: etcd namespace: scoped status: installPlanRef: apiVersion: operators.coreos.com/v1 kind: InstallPlan name: install-4plp8 namespace: scoped resourceVersion: "117359" uid: 2c1df80e-afea-11e9-bce3-5254009c9c23
Check the status of the
InstallPlan
object for any errors:apiVersion: operators.coreos.com/v1 kind: InstallPlan status: conditions: - lastTransitionTime: "2019-07-26T21:13:10Z" lastUpdateTime: "2019-07-26T21:13:10Z" message: 'error creating clusterrole etcdoperator.v0.9.4-clusterwide-dsfx4: clusterroles.rbac.authorization.k8s.io is forbidden: User "system:serviceaccount:scoped:scoped" cannot create resource "clusterroles" in API group "rbac.authorization.k8s.io" at the cluster scope' reason: InstallComponentFailed status: "False" type: Installed phase: Failed
The error message tells you:
-
The type of resource it failed to create, including the API group of the resource. In this case, it was
clusterroles
in therbac.authorization.k8s.io
group. - The name of the resource.
-
The type of error:
is forbidden
tells you that the user does not have enough permission to do the operation. - The name of the user who attempted to create or update the resource. In this case, it refers to the service account specified in the Operator group.
The scope of the operation:
cluster scope
or not.The user can add the missing permission to the service account and then iterate.
NoteOperator Lifecycle Manager (OLM) does not currently provide the complete list of errors on the first try.
-
The type of resource it failed to create, including the API group of the resource. In this case, it was
4.9. Managing custom catalogs
Cluster administrators and Operator catalog maintainers can create and manage custom catalogs packaged using the bundle format on Operator Lifecycle Manager (OLM) in OpenShift Container Platform.
Kubernetes periodically deprecates certain APIs that are removed in subsequent releases. As a result, Operators are unable to use removed APIs starting with the version of OpenShift Container Platform that uses the Kubernetes version that removed the API.
If your cluster is using custom catalogs, see Controlling Operator compatibility with OpenShift Container Platform versions for more details about how Operator authors can update their projects to help avoid workload issues and prevent incompatible upgrades.
Additional resources
4.9.1. Prerequisites
-
You have installed the
opm
CLI.
4.9.2. File-based catalogs
File-based catalogs are the latest iteration of the catalog format in Operator Lifecycle Manager (OLM). It is a plain text-based (JSON or YAML) and declarative config evolution of the earlier SQLite database format, and it is fully backwards compatible.
As of OpenShift Container Platform 4.11, the default Red Hat-provided Operator catalog releases in the file-based catalog format. The default Red Hat-provided Operator catalogs for OpenShift Container Platform 4.6 through 4.10 released in the deprecated SQLite database format.
The opm
subcommands, flags, and functionality related to the SQLite database format are also deprecated and will be removed in a future release. The features are still supported and must be used for catalogs that use the deprecated SQLite database format.
Many of the opm