ocs_ci.ocs.resources package

Submodules

ocs_ci.ocs.resources.backingstore module

class ocs_ci.ocs.resources.backingstore.BackingStore(name, method, type, uls_name=None, secret_name=None, mcg_obj=None, vol_num=None, vol_size=None)

Bases: object

A class that represents BackingStore objects

delete(retry=True, timeout=120)

Deletes the current backingstore by using OC/CLI commands

Parameters:

• retry (bool) – Whether to retry the deletion if it fails

• timeout (int) – Timeout to wait if retry is true

ocs_ci.ocs.resources.backingstore.backingstore_factory(request, cld_mgr, mcg_obj, cloud_uls_factory)

Create a Backing Store factory. Calling this fixture creates a new Backing Store(s).

Parameters:

• request (object) – Pytest built-in fixture

• cld_mgr (CloudManager) – Cloud Manager object containing all connections to clouds

• mcg_obj (MCG) – MCG object containing data and utils related to MCG

• cloud_uls_factory – Factory for underlying storage creation

Returns:

Factory method - each call to this function creates

a backingstore

Return type:

func

ocs_ci.ocs.resources.bucket_policy module

class ocs_ci.ocs.resources.bucket_policy.HttpResponseParser(http_response)

Bases: object

A simple class for parsing HTTP responses

class ocs_ci.ocs.resources.bucket_policy.NoobaaAccount(mcg, name, email, buckets=None, admin_access=False, s3_access=True, full_bucket_access=True, backingstore_name='noobaa-default-backing-store')

Bases: object

Class for Noobaa account

access_key = None

access_key_id = None

account_name = None

email_id = None

s3_endpoint = None

s3_resource = None

token = None

ocs_ci.ocs.resources.bucket_policy.gen_bucket_policy(user_list, actions_list, resources_list, effect='Allow', sid='statement')

Function prepares bucket policy parameters in syntax and format provided by AWS bucket policy

Parameters:

• user_list (list) – List of user accounts to access bucket policy

• actions_list (list) – List of actions in bucket policy eg: Get, Put objects etc

• resources_list (list) – List of resources. Eg: Bucket name, specific object in a bucket etc

• effect (str) – Permission given to the bucket policy ie: Allow(default) or Deny

• sid (str) – Statement name. Can be any string. Default: “Statement”

Returns:

Bucket policy in json format

Return type:

dict

ocs_ci.ocs.resources.bucketclass module

class ocs_ci.ocs.resources.bucketclass.BucketClass(name, backingstores, namespacestores, placement_policy, replication_policy, namespace_policy)

Bases: object

A class that represents BucketClass objects

delete()

ocs_ci.ocs.resources.bucketclass.bucket_class_factory(request, mcg_obj, backingstore_factory, namespace_store_factory)

Create a bucket class factory. Calling this fixture creates a new custom bucket class. For a custom backingstore(s), provide the ‘backingstore_dict’ parameter.

Parameters:

• request (object) – Pytest built-in fixture

• mcg_obj (MCG) – An MCG object containing the MCG S3 connection credentials

• backingstore_factory – Factory for backing store creation

ocs_ci.ocs.resources.cache_drop module

class ocs_ci.ocs.resources.cache_drop.OSDCashDrop

Bases: OCP

This module is for deploying pod which enable to drop the OSD’s cache.

Usage:

import OSDCashDrop

cd = OSDCashDrop() # create new cache_drop object cd.deploy() # deploy the cache_drop pod …. # run test cd.drop_cache() # drop the OSD’s cache …. # run the test again cd.cleanup() # delete the cache_drop pod

cleanup()

Delete the pod from the cluster

deploy()

Deploy the cache drop pod and wait until it is up

drop_cache()

Drop the OSD’s cache by sending http request to the pod

Raises:

exception – if the request to drop the cache failed

property ip

return the cache drop IP

ocs_ci.ocs.resources.catalog_source module

CatalogSource related functionalities

class ocs_ci.ocs.resources.catalog_source.CatalogSource(resource_name='', namespace=None, *args, **kwargs)

Bases: OCP

This class represent CatalogSource and contains all related methods we need to do with it.

check_state(state)

Check state of catalog source

Parameters:

state (str) – State of CatalogSource object

Returns:

True if state of object is the same as desired one, False

otherwise.

Return type:

bool

get_image_name()

Fetch image name from catalog source resource

Returns:

especially version info extracted from image

name

Return type:

image info (str)

get_image_url()

Fetch image url from catalog source resource

Returns:

URL of image

Return type:

image url (str)

wait_for_state(state, timeout=480, sleep=5)

Wait till state of catalog source resource is the same as required one passed in the state parameter.

Parameters:

• state (str) – Desired state of catalog source object

• timeout (int) – Timeout in seconds to wait for desired state

• sleep (int) – Time in seconds to sleep between attempts

Raises:

ResourceWrongStatusException – In case the catalog source is not in expected state.

ocs_ci.ocs.resources.catalog_source.disable_default_sources()

Disable default sources

ocs_ci.ocs.resources.catalog_source.disable_specific_source(source_name)

Disable specific default source

Parameters:

source_name (str) – Source name (e.g. redhat-operators)

ocs_ci.ocs.resources.catalog_source.enable_default_sources()

Enable default sources

ocs_ci.ocs.resources.catalog_source.enable_specific_source(source_name)

Enable specific default source

Parameters:

source_name (str) – Source name (e.g. redhat-operators)

ocs_ci.ocs.resources.cloud_manager module

class ocs_ci.ocs.resources.cloud_manager.AwsSTSClient(auth_dict, verify=True, endpoint='https://s3.amazonaws.com', *args, **kwargs)

Bases: S3Client

class ocs_ci.ocs.resources.cloud_manager.AzureClient(account_name=None, credential=None, auth_dict=None, *args, **kwargs)

Bases: CloudClient

Implementation of a Azure Client using the Azure API

create_azure_secret()

Create a Kubernetes secret to allow NooBaa to create Azure-based backingstores

get_all_uls_names()

Returns a set containing all the container names that the client has access to

internal_create_uls(name, region)

Creates the Underlying Storage using the Azure API

Parameters:

name (str) – The Underlying Storage name to be created

internal_delete_uls(name)

Deletes the Underlying Storage using the Azure API

Parameters:

name (str) – The Underlying Storage name to be deleted

Returns:

True if deleted successfully

Return type:

bool

verify_uls_exists(uls_name)

Verifies whether a Underlying Storage with the given uls_name exists

Parameters:

uls_name (str) – The Underlying Storage name to be verified

Returns:

True if Underlying Storage exists, False otherwise

Return type:

bool

class ocs_ci.ocs.resources.cloud_manager.AzureWithLogsClient(account_name=None, credential=None, auth_dict=None, *args, **kwargs)

Bases: AzureClient

Implementation of an Azure Client using the Azure API to an existing storage account with bucket logs enabled

create_azure_secret()

Create a Kubernetes secret to allow NooBaa to create Azure-based backingstores

Note that this method overides the parent method to include the additional fields that are needed for the bucket logs feature

class ocs_ci.ocs.resources.cloud_manager.CloudClient(*args, **kwargs)

Bases: ABC

Base abstract class for Cloud based API calls

client = None

create_uls(name, region)

Super method that first logs the Underlying Storage creation and then calls the appropriate implementation

delete_uls(name)

Super method that first logs the Underlying Storage deletion and then calls the appropriate implementation

get_all_uls_names()

abstract internal_create_uls(name, region)

abstract internal_delete_uls(name)

verify_uls_exists(uls_name)

verify_uls_state(uls_name, is_available)

class ocs_ci.ocs.resources.cloud_manager.CloudManager

Bases: ABC

Class containing all client types

class ocs_ci.ocs.resources.cloud_manager.GoogleClient(auth_dict, *args, **kwargs)

Bases: CloudClient

Implementation of a Google Client using the Google API

create_gcp_secret()

Create a Kubernetes secret to allow NooBaa to create Google-based backingstores

get_all_uls_names()

Returns a set containing all the bucket names that the client has access to

internal_create_uls(name, region=None)

Creates the Underlying Storage using the Google API

Parameters:

• name (str) – The Underlying Storage name to be created

• region (str) – The region to create the Underlying Storage

internal_delete_uls(name)

Deletes the Underlying Storage using the Google API

Parameters:

name (str) – The Underlying Storage name to be deleted

Returns:

True if deleted successfully

Return type:

bool

verify_uls_exists(uls_name)

Verifies whether a Underlying Storage with the given uls_name exists

Parameters:

uls_name (str) – The Underlying Storage name to be verified

Returns:

True if Underlying Storage exists, False otherwise

Return type:

bool

class ocs_ci.ocs.resources.cloud_manager.S3Client(auth_dict, verify=True, endpoint='https://s3.amazonaws.com', *args, **kwargs)

Bases: CloudClient

Implementation of a S3 Client using the S3 API

create_s3_secret(secret_prefix, data_prefix)

Create a Kubernetes secret to allow NooBaa to create AWS-based backingstores

get_all_uls_names()

Returns a set containing all the bucket names that the client has access to

internal_create_uls(name, region=None)

Creates the Underlying Storage using the S3 API

Parameters:

• name (str) – The Underlying Storage name to be created

• region (str) – The region to create the Underlying Storage,

• us-east-1 (if none will be created on) –

• **IMPORTANT** –

• is (Passing us-east-1 as the region will cause an error if used since it) –

• AWS (the default region for) –

internal_delete_uls(name)

Deletes the Underlying Storage using the S3 API

Parameters:

name (str) – The Underlying Storage name to be deleted

Returns:

True if deleted successfully

Return type:

bool

toggle_aws_bucket_readwrite(aws_bucket_name, block=True)

Toggles a bucket’s IO using a bucket policy

Parameters:

• aws_bucket_name – The name of the bucket that should be manipulated

• block – Whether to block RW or un-block. True | False

verify_uls_exists(uls_name)

Verifies whether a Underlying Storage with the given uls_name exists

Parameters:

uls_name (str) – The Underlying Storage name to be verified

Returns:

True if Underlying Storage exists, False otherwise

Return type:

bool

ocs_ci.ocs.resources.cloud_uls module

ocs_ci.ocs.resources.cloud_uls.cloud_uls_factory(request, cld_mgr)

Create an Underlying Storage factory. Calling this fixture creates a new underlying storage(s).

Parameters:

• request (object) – Pytest built-in fixture

• cld_mgr (CloudManager) – Cloud Manager object containing all connections to clouds

Returns:

Factory method - each call to this function creates

an Underlying Storage factory

Return type:

func

ocs_ci.ocs.resources.csv module

CSV related functionalities

class ocs_ci.ocs.resources.csv.CSV(resource_name='', *args, **kwargs)

Bases: OCP

This class represent ClusterServiceVersion (CSV) and contains all related methods we need to do with CSV.

ocs_ci.ocs.resources.csv.check_all_csvs_are_succeeded(namespace, timeout=600)

Check if all CSVs in namespace are in succeeded phase

Parameters:

namespace (str) – namespace of CSV

Returns:

True if all CSVs are in succeeded phase

Return type:

bool

ocs_ci.ocs.resources.csv.get_csv_name_start_with_prefix(csv_prefix, namespace)

Get CSV name start with prefix

Parameters:

• csv_prefix (str) – prefix from name

• namespace (str) – namespace of CSV

Returns:

CSV name

Return type:

str

ocs_ci.ocs.resources.csv.get_csvs_start_with_prefix(csv_prefix, namespace)

Get CSVs start with prefix

Parameters:

• csv_prefix (str) – prefix from name

• namespace (str) – namespace of CSV

Returns:

found CSVs

Return type:

list

ocs_ci.ocs.resources.deployment module

General Deployment object

class ocs_ci.ocs.resources.deployment.Deployment(**kwargs)

Bases: OCS

A basic Deployment kind resource

property available_replicas

Returns number of available replicas for the deployment

Returns:

Number of replicas

Return type:

int

property pods

Returns list of pods of the Deployment resource

Returns:

Deployment’s pods

Return type:

list

property replicas

Returns number of replicas for the deployment as defined in its spec

Returns:

Number of replicas

Return type:

int

property revision

Returns revision of a Deployment resource

Returns:

revision number

Return type:

str

scale(replicas, resource_name=None)

Scale deployment to required number of replicas

Parameters:

• replicas (int) – number of required replicas

• resource_name (str) – name of resouce to querry the revision

set_revision(revision, resource_name=None)

Set revision to a Deployment or a similar resource that supports kubernetes revisions.

Parameters:

• revision (int) – revision number of the resource

• resource_name (str) – name of resouce to querry the revision

wait_for_available_replicas(timeout=15, sleep=3)

Wait for number of available replicas reach number of desired replicas.

Parameters:

• timeout (int) – Timeout in seconds

• sleep (int) – Sleep interval in seconds

ocs_ci.ocs.resources.deployment.get_deployments_having_label(label, namespace)

Fetches deployment resources with given label in given namespace

Parameters:

• label (str) – label which deployments might have

• namespace (str) – Namespace in which to be looked up

Returns:

deployment OCP instances

Return type:

list

ocs_ci.ocs.resources.deployment.get_mon_deployments(mon_label='app=rook-ceph-mon', namespace=None)

Fetches info about mon deployments in the cluster

Parameters:

• mon_label (str) – label associated with mon deployments (default: defaults.MON_APP_LABEL)

• namespace (str) – Namespace in which ceph cluster lives (default: config.ENV_DATA[“cluster_namespace”])

Returns:

Mon deployment OCS instances

Return type:

list

ocs_ci.ocs.resources.deployment.get_osd_deployments(osd_label='app=rook-ceph-osd', namespace=None)

Fetches info about osd deployments in the cluster

Parameters:

• osd_label (str) – label associated with osd deployments (default: defaults.OSD_APP_LABEL)

• namespace (str) – Namespace in which ceph cluster lives (default: config.ENV_DATA[“cluster_namespace”])

Returns:

OSD deployment OCS instances

Return type:

list

ocs_ci.ocs.resources.drpc module

DRPlacementControl related functionalities

class ocs_ci.ocs.resources.drpc.DRPC(namespace, resource_name='', *args, **kwargs)

Bases: OCP

This class represent DRPlacementControl (DRPC) and contains all related methods we need to do with DRPC.

property drpolicy

property drpolicy_obj

get_peer_ready_status()

wait_for_peer_ready_status()

ocs_ci.ocs.resources.drpc.get_drpc_name(namespace)

Get the DRPC resource name in the given namespace

Parameters:

namespace (str) – Name of the namespace

Returns:

DRPC resource name

Return type:

str

ocs_ci.ocs.resources.fips module

ocs_ci.ocs.resources.fips.check_fips_enabled(fips_location='/proc/sys/crypto/fips_enabled')

Checks if FIPS is activated on all pods

Args:

fips_location: File that refers to fips, written 1 if enabled,

0 otherwise

Raises:

FipsNotInstalledException:

If the value of fips location file does not include 1

in all pods within the given namespace.

ocs_ci.ocs.resources.install_plan module

Install plan related functionalities

class ocs_ci.ocs.resources.install_plan.InstallPlan(resource_name='', namespace=None, *args, **kwargs)

Bases: OCP

This class represent InstallPlan and contains all the related functionality.

approve()

Approve install plan.

ocs_ci.ocs.resources.install_plan.get_install_plans_for_approve(namespace, raise_exception=False)

Get all install plans for approve

Parameters:

• namespace (str) – namespace of CSV

• raise_exception (bool) – True if the function should raise the exception when no install plan found for approve.

Returns:

found install plans for approve

Return type:

list

Raises:

NoInstallPlanForApproveFoundException – in case raise_exception is True and no install plan for approve found.

ocs_ci.ocs.resources.install_plan.wait_for_install_plan_and_approve(namespace, timeout=960)

Wait for install plans ready for approve and approve them.

Parameters:

• namespace (str) – namespace of install plan.

• timeout (int) – timeout in seconds.

Raises:

TimeoutExpiredError – in case no install plan found in specified timeout.

ocs_ci.ocs.resources.job module

ocs_ci.ocs.resources.job.get_all_jobs(namespace=None)

Get all the jobs in a specific namespace

Parameters:

namespace (str) – Name of cluster namespace(default: config.ENV_DATA[“cluster_namespace”])

Returns:

list of dictionaries of the job OCS instances.

Return type:

list

ocs_ci.ocs.resources.job.get_job_obj(name, namespace=None)

Get OCS instance for job of given job name.

Parameters:

• name (str) – The name of the job

• namespace (str) – The namespace to look in

Returns:

A job OCS instance

Return type:

OCS

ocs_ci.ocs.resources.job.get_job_pods(job_name, namespace, names_only=False)

Get list of pods of given job (via job-name pod selector).

Parameters:

• job_name (str) – name of the job to wait for

• namespace (str) – name of the namespace where the job is running

Returns:

list of pod names (if names_only is True) or full item dicts

Return type:

list

ocs_ci.ocs.resources.job.get_jobs_with_prefix(prefix, namespace=None)

Get all the jobs that start with a specific prefix

Parameters:

• prefix (str) – The prefix to search in the job names

• namespace (str) – Name of cluster namespace (default: config.ENV_DATA[“cluster_namespace”] if None provided)

Returns:

list of dictionaries of the job OCS instances that start with the prefix

Return type:

list

ocs_ci.ocs.resources.job.log_output_of_job_pods(job_name, namespace)

Log (via standard logger) output of all pods of given job. Expected to be used in case of error, when evidence needs to be captured in logs.

Parameters:

• job_name (str) – name of the job to wait for

• namespace (str) – name of the namespace where the job is running

ocs_ci.ocs.resources.job.wait_for_job_completion(job_name, namespace, timeout=600, sleep_time=30)

Wait for given k8s Job to complete.

Parameters:

• job_name (str) – name of the job to wait for

• namespace (str) – name of the namespace where the job is running

• timeout (int) – timeout in seconds

• sleep_time (int) – sleep time between consequent job status checks in seconds

Raises:

TimeoutExpiredError – When job fails to complete in given time

ocs_ci.ocs.resources.machineconfig module

ocs_ci.ocs.resources.machineconfig.deploy_machineconfig(tmp_path, mc_name, mc_dict, mcp_num=2)

Deploy given MachineConfig dict and wait for the configuration to be deployed on all MachineConfigPools. By default we assume there are just two pools.

Parameters:

• tmp_path (pathlib.Path) – Directory where a temporary yaml file will be created. In test context, use pytest fixture tmp_path.

• mc_name (str) – name prefix for object config yaml file which will be created for the machineconfig before it’s deployment

• mc_dict (list) – list of dictionaries with MachineConfig resource(s) to deploy

• mcp_num (int) – number of MachineConfigPool resources in the cluster

ocs_ci.ocs.resources.mcg module

class ocs_ci.ocs.resources.mcg.MCG(*args, **kwargs)

Bases: object

Wrapper class for the Multi Cloud Gateway’s S3 service

access_key = None

access_key_id = None

check_backingstore_state(backingstore_name, desired_state, timeout=600)

Checks whether the backing store reached a specific state :param backingstore_name: Name of the backing store to be checked :type backingstore_name: str :param desired_state: The desired state of the backing store :type desired_state: str :param timeout: Number of seconds for timeout which will be used :type timeout: int :param in the checks used in this function.:

Returns:

Whether the backing store has reached the desired state

Return type:

bool

check_data_reduction(bucketname, expected_reduction_in_bytes)

Checks whether the data reduction on the MCG server works properly :param bucketname: An example bucket name that contains compressed/deduped data :param expected_reduction_in_bytes: amount of data that is supposed to be reduced after data :param compression and deduplication.:

Returns:

True if the data reduction mechanics work, False otherwise

Return type:

bool

check_if_mirroring_is_done(bucket_name, timeout=300)

Check whether all object chunks in a bucket are mirrored across all backing stores.

Parameters:

• bucket_name – The name of the bucket that should be checked

• timeout – timeout in seconds to check if mirroring

Raises:

AssertionError – In case mirroring is not done in defined time.

check_ns_resource_validity(ns_resource_name, target_bucket_name, endpoint)

Check namespace resource validity

Parameters:

• ns_resource_name (str) – The name of the to be verified namespace resource

• target_bucket_name (str) – The name of the expected target bucket (uls)

• endpoint – The expected endpoint path

cli_create_bucketclass(name, backingstores, placement_policy, namespace_policy=None, replication_policy=None)

Creates a new NooBaa bucket class using the noobaa cli :param name: The name to be given to the bucket class :type name: str :param backingstores: The backing stores to use as part of the policy :type backingstores: list :param placement_policy: The placement policy to be used - Mirror | Spread :type placement_policy: str :param namespace_policy: The namespace policy to be used :type namespace_policy: dict :param replication_policy: The replication policy dictionary :type replication_policy: dict

Returns:

The bucket class resource

Return type:

OCS

cli_get_all_bucket_names()

Returns:

A set of all bucket names

Return type:

set

cli_verify_bucket_exists(bucketname)

Verifies whether a bucket with the given bucketname exists :param bucketname: The bucket name to be verified

Returns:

True if bucket exists, False otherwise

Return type:

bool

create_connection(cld_mgr, platform, conn_name=None)

Creates a new NooBaa connection to an AWS backend

Parameters:

• cld_mgr (obj) – A cloud manager instance

• platform (str) – Platform to use for new connection

• conn_name (str) – The connection name to be used If None provided then the name will be generated

Returns:

False if the connection creation failed

Return type:

bool

create_namespace_resource(ns_resource_name, conn_name, region, cld_mgr, cloud_uls_factory, platform)

Creates a new namespace resource

Parameters:

• ns_resource_name (str) – The name to be given to the new namespace resource

• conn_name (str) – The external connection name to be used

• region (str) – The region name to be used

• cld_mgr – A cloud manager instance

• cloud_uls_factory – The cloud uls factory

• platform (str) – The platform resource name

Returns:

The name of the created target_bucket_name (cloud uls)

Return type:

str

data_to_mask = None

delete_ns_connection(ns_connection_name)

Delete external connection

Parameters:

ns_connection_name (str) – The name of the to be deleted external connection

delete_ns_resource(ns_resource_name)

Delete namespace resource

Parameters:

ns_resource_name (str) – The name of the to be deleted namespace resource

exec_mcg_cmd(cmd, namespace=None, use_yes=False, **kwargs)

Executes an MCG CLI command through the noobaa-operator pod’s CLI binary

Parameters:

• cmd (str) – The command to run

• namespace (str) – The namespace to run the command in

Returns:

stdout of the command

Return type:

str

get_admin_default_resource_name()

Get the default resource name of the admin account

Returns:

The default resource name

Return type:

str

get_bucket_info(bucket_name)

Parameters:

bucket_name (str) – Name of searched bucket

Returns:

Information about the bucket

Return type:

dict

get_default_bc_backingstore_name()

Get the default backingstore name of the default bucketclass

Returns:

The default backingstore name

Return type:

str

get_mcg_cli_version()

Get the MCG CLI version by parsing the output of the mcg-cli version command.

Example output of the mcg-cli version command:

INFO[0000] CLI version: 5.12.0 INFO[0000] noobaa-image: noobaa/noobaa-core:master-20220913 INFO[0000] operator-image: noobaa/noobaa-operator:5.12.0

Returns:

Object of semantic version.

Return type:

semantic_version.base.Version

get_noobaa_admin_credentials_from_secret()

Get the NooBaa admin credentials from the OCP secret

Returns:

Dictionary containing the following keys:

AWS_ACCESS_KEY_ID (str): NooBaa admin S3 access key ID AWS_SECRET_ACCESS_KEY (str): NooBaa admin S3 secret access key email (str): NooBaa admin user email password (str): NooBaa admin user password

Return type:

credentials_dict (dict)

mgmt_endpoint = None

namespace = None

noobaa_password = None

noobaa_token = None

noobaa_user = None

oc_create_bucketclass(name, backingstores, placement_policy, namespace_policy, replication_policy)

Creates a new NooBaa bucket class using a template YAML :param name: The name to be given to the bucket class :type name: str :param backingstores: The backing stores to use as part of the policy :type backingstores: list :param placement_policy: The placement policy to be used - Mirror | Spread :type placement_policy: str :param namespace_policy: The namespace policy to be used :type namespace_policy: dict :param replication_policy: The replication policy dictionary :type replication_policy: dict

Returns:

The bucket class resource

Return type:

OCS

oc_verify_bucket_exists(bucketname)

Verifies whether a bucket with the given bucketname exists :param bucketname: The bucket name to be verified

Returns:

True if bucket exists, False otherwise

Return type:

bool

ocp_resource = None

read_system()

Returns:

A dictionary with information about MCG resources

Return type:

dict

region = None

request_aws_credentials()

Uses a CredentialsRequest CR to create an AWS IAM that allows the program to interact with S3

Returns:

The CredentialsRequest resource

Return type:

OCS

reset_admin_pw(new_password)

Reset the NooBaa admin password

Parameters:

new_password (str) – New password to set for the NooBaa admin user

reset_core_pod()

Delete the noobaa-core pod and wait for it to come up again

retrieve_nb_token(timeout=300, sleep=30)

Try to retrieve a NB RPC token and decode its JSON

s3_endpoint = None

s3_get_all_bucket_names()

Returns:

A set of all bucket names

Return type:

set

s3_get_all_buckets()

Returns:

A list of all s3.Bucket objects

Return type:

list

s3_internal_endpoint = None

s3_list_all_objects_in_bucket(bucketname)

Returns:

A list of all bucket objects

Return type:

list

s3_resource = None

s3_verify_bucket_exists(bucketname)

Verifies whether a bucket with the given bucketname exists :param bucketname: The bucket name to be verified

Returns:

True if bucket exists, False otherwise

Return type:

bool

send_rpc_query(api, method, params=None)

Templates and sends an RPC query to the MCG mgmt endpoint

Parameters:

• api – The name of the API to use

• method – The method to use inside the API

• params – A dictionary containing the command payload

Returns:

The server’s response

status

Verify the status of NooBaa, and its default backing store and bucket class

Returns:

return False if any of the above components of noobaa is not in READY state

Return type:

bool

update_s3_creds()

Set the S3 credentials of the NooBaa admin user from the noobaa-admin secret, and update the S3 resource and client

ocs_ci.ocs.resources.mcg_lifecycle_policies module

class ocs_ci.ocs.resources.mcg_lifecycle_policies.ExpirationRule(days, filter={}, use_date=False, is_enabled=True, expire_solo_delete_markers=False)

Bases: LifecycleRule

A class for handling the parsing of an MCG object expiration rule

as_dict()

class ocs_ci.ocs.resources.mcg_lifecycle_policies.LifecycleFilter(prefix=None, tags=None, minBytes=None, maxBytes=None)

Bases: object

A class for handling S3 lifecycle filter parsing

as_dict()

class ocs_ci.ocs.resources.mcg_lifecycle_policies.LifecyclePolicy(*args)

Bases: object

A class for handling MCG lifecycle policy config parsing

as_dict()

class ocs_ci.ocs.resources.mcg_lifecycle_policies.LifecycleRule(filter={}, is_enabled=True)

Bases: ABC

An abstract class for handling MCG lifecycle rule parsing

as_dict()

property id

ocs_ci.ocs.resources.mcg_params module

class ocs_ci.ocs.resources.mcg_params.NSFS(method: str = 'CLI', pvc_name: str | None = None, pvc_size: int = 20, sub_path: str | None = None, fs_backend: str | None = None, mount_existing_dir: bool = False, existing_dir_mode: int = 777, mount_path: str = '/nsfs', uid: int = 5678, gid: int = 1234, interface_pod: Pod | None = None, bucket_obj: ObjectBucket | None = None, bucket_name: str | None = None, mounted_bucket_path: str | None = None, s3_creds: dict | None = None, nss: NamespaceStore | None = None)

Bases: object

An NSFS dataclass to represent NSFS test parametrization and provide a central state store

Parametrization parameters:

method (str): The method to use for namespacestore creation. OC | CLI pvc_name (str): Name of the PVC that will host the namespace filesystem pvc_size (int): Size of the PVC in Gi sub_path (str): The path to a sub directory inside the PVC FS which the NSS will use as its root directory mount_existing_dir (bool): Whether to mount an existing directory or create a new one existing_dir_mode (int): The mode of the existing directory fs_backend (str): The file system backend type - CEPH_FS | GPFS | NFSv4. Defaults to None. mount_path (str): The path to the mount point of the NSFS uid (int): The UID of the user that will be used to create the NSFS gid (int): The GID of the user that will be used to create the NSFS

bucket_name: str = None

bucket_obj: ObjectBucket = None

existing_dir_mode: int = 777

fs_backend: str = None

gid: int = 1234

State parameters; These should not be modified unless needed, and will be (over/)written after the NSFS object will be passed to the bucket factory.

interface_pod (Pod): The pod that will be used to interact with the NSFS bucket_name (str): The name of the NSFS bucket mounted_bucket_path (str): The path to where the bucket is “mounted” in the FS s3_creds (str): The NSFS S3 credentials nss (NamespaceStore): The namespacestore that the NSFS uses

interface_pod: Pod = None

method: str = 'CLI'

mount_existing_dir: bool = False

mount_path: str = '/nsfs'

mounted_bucket_path: str = None

nss: NamespaceStore = None

pvc_name: str = None

pvc_size: int = 20

s3_creds: dict = None

sub_path: str = None

uid: int = 5678

ocs_ci.ocs.resources.mcg_replication_policy module

class ocs_ci.ocs.resources.mcg_replication_policy.AwsLogBasedReplicationPolicy(destination_bucket, sync_deletions=False, logs_bucket='', prefix='', logs_location_prefix='')

Bases: LogBasedReplicationPolicy

A class to handle the AWS log-based bucket replication policy JSON structure.

to_dict()

class ocs_ci.ocs.resources.mcg_replication_policy.AzureLogBasedReplicationPolicy(destination_bucket, sync_deletions=False, prefix='')

Bases: LogBasedReplicationPolicy

A class to handle the Azure log-based bucket replication policy JSON structure.

to_dict()

class ocs_ci.ocs.resources.mcg_replication_policy.LogBasedReplicationPolicy(destination_bucket, sync_deletions=False, prefix='')

Bases: McgReplicationPolicy, ABC

An abstract subclass of ReplicationPolicy that includes log-based replication information.

abstract to_dict()

class ocs_ci.ocs.resources.mcg_replication_policy.McgReplicationPolicy(destination_bucket, prefix='')

Bases: object

A class to handle the MCG bucket replication policy JSON structure.

to_dict()

ocs_ci.ocs.resources.mockup_bucket_logger module

class ocs_ci.ocs.resources.mockup_bucket_logger.MockupBucketLogger(awscli_pod, mcg_obj, bucket_factory, platform, region)

Bases: object

This class facilitates S3 operations on an MCG bucket while writing corresponding mockup logs directly to a ULS bucket via a Namespacestore MCG bucket.

The use of a Namespacestore MCG bucket for the logs enables storage provider agnosticity.

delete_all_objects_and_log(bucket_name)

Deletes all objects from the MCG bucket and write matching mockup logs

delete_objs_and_log(bucket_name, objs)

Delete list of objects from the MCG bucket and write matching mockup logs

Parameters:

• bucket_name (str) – Name of the MCG bucket

• objs (list) – List of the objects to delete

property standard_test_obj_list

upload_arbitrary_object_and_log(bucket_name)

Uploads an arbitrary object to the MCG bucket and upload a matching mockup log

upload_test_objs_and_log(bucket_name)

Uploads files from files_dir to the MCG bucket and write matching mockup logs to the logs bucket

Parameters:

files_dir (str) – Full path to a directory on awscli_pod

class ocs_ci.ocs.resources.mockup_bucket_logger.S3MockupLog(bucket_name, object_key, operation)

Bases: object

This class represents a mockup log file in the AWS log file format

OP_CODES = {'DELETE': 204, 'GET': 206, 'PUT': 200}

property bucket_name

property file_name

property object_key

property operation

property time

Get the time in the AWS logs format. i.e “[06/Feb/2019:00:00:38 +0000]”

Returns:

The current time in the AWS bucket log format

Return type:

str

ocs_ci.ocs.resources.namespacestore module

class ocs_ci.ocs.resources.namespacestore.NamespaceStore(name, method, uls_name=None, secret_name=None, mcg_obj=None)

Bases: object

A class that represents NamespaceStore objects

cli_verify_health()

Verify the health of the namespace store by running the noobaa namespacestore status command

Returns:

Based on whether the namespace store is healthy or not

Return type:

bool

delete(retry=True)

Deletes the current namespacestore by using OC/CLI commands

Parameters:

retry (bool) – Whether to retry the deletion if it fails

oc_verify_health()

Verify the health of the namespace store by checking the status of the CR

Returns:

Based on whether the namespace store is healthy or not

Return type:

bool

verify_health(timeout=180, interval=5)

Health verification function that tries to verify a namespacestores’s health until a given time limit is reached

Parameters:

• timeout (int) – Timeout for the check, in seconds

• interval (int) – Interval to wait between checks, in seconds

Returns:

True if the bucket is healthy, False otherwise

Return type:

(bool)

ocs_ci.ocs.resources.namespacestore.cli_create_namespacestore(nss_name, platform, mcg_obj, uls_name=None, cld_mgr=None, nss_tup=None)

Create a namespace filesystem namespacestore using YAMLs

Parameters:

• nss_name (str) – Name of the namespacestore

• platform (str) – Platform to create the namespacestore on

• mcg_obj (MCG) – An MCG object used for executing the MCG CLI commands

• uls_name (str) – Name of the ULS bucket / PVC to use for the namespacestore

• cld_mgr (CloudManager) – CloudManager object used for supplying the needed connection credentials

• nss_tup (tuple) – A tuple containing the NSFS namespacestore details, in this order: pvc_name (str): Name of the PVC that will host the namespace filesystem pvc_size (int): Size in Gi of the PVC that will host the namespace filesystem sub_path (str): The path to a sub directory inside the PVC FS which the NSS will use as the root directory fs_backend (str): The file system backend type - CEPH_FS | GPFS | NFSv4. Defaults to None.

ocs_ci.ocs.resources.namespacestore.namespace_store_factory(request, cld_mgr, mcg_obj_session, cloud_uls_factory_session, pvc_factory_session)

Create a NamespaceStore factory. Calling this fixture lets the user create namespace stores.

Parameters:

• request (object) – Pytest built-in fixture

• cld_mgr (CloudManager) – Cloud Manager object containing all connections to clouds

• mcg_obj (MCG) – MCG object containing data and utils related to MCG

• cloud_uls_factory – Factory for creation of underlying storage

Returns:

Factory method - allows the user to create namespace stores

Return type:

func

ocs_ci.ocs.resources.namespacestore.oc_create_namespacestore(nss_name, platform, mcg_obj, uls_name=None, cld_mgr=None, nss_tup=None, nsfs_pvc_name=None)

Create a namespacestore using the MCG CLI

Parameters:

• nss_name (str) – Name of the namespacestore

• platform (str) – Platform to create the namespacestore on

• mcg_obj (MCG) – A redundant MCG object, used for uniformity between OC and CLI calls

• uls_name (str) – Name of the ULS bucket to use for the namespacestore

• cld_mgr (CloudManager) – CloudManager object used for supplying the needed connection credentials

• nss_tup (tuple) – A tuple containing the NSFS namespacestore details, in this order: pvc_name (str): Name of the PVC that will host the namespace filesystem pvc_size (int): Size in Gi of the PVC that will host the namespace filesystem sub_path (str): The path to a sub directory inside the PVC FS which the NSS will use as the root directory fs_backend (str): The file system backend type - CEPH_FS | GPFS | NFSv4. Defaults to None.

ocs_ci.ocs.resources.namespacestore.template_pvc(name, namespace='openshift-storage', storageclass='ocs-storagecluster-cephfs', access_mode='ReadWriteMany', size=20)

Create a PVC using the MCG CLI

Parameters:

• name (str) – Name of the PVC

• namespace (str) – Namespace to create the PVC in

• access_mode (str) – Access mode for the PVC

• size (str) – Size of the PVC in GiB

ocs_ci.ocs.resources.objectbucket module

class ocs_ci.ocs.resources.objectbucket.MCGCLIBucket(*args, **kwargs)

Bases: ObjectBucket

Implementation of an MCG bucket using the NooBaa CLI

internal_delete()

Deletes the bucket using the NooBaa CLI

Raises:

NotFoundError – In case the OBC was not found

property internal_status

Returns the OBC status as printed by the NB CLI

Returns:

OBC status

Return type:

str

internal_verify_deletion()

Abstract deletion verification method

internal_verify_health()

Verifies that the bucket is healthy using the CLI

Returns:

True if the bucket is healthy, False otherwise

Return type:

bool

class ocs_ci.ocs.resources.objectbucket.MCGNamespaceBucket(*args, **kwargs)

Bases: ObjectBucket

Implementation of an MCG bucket using the S3 API

internal_delete()

Deletes the bucket using the S3 API

property internal_status

Returns the OBC mode as shown in the NB UI and retrieved via RPC

Returns:

The bucket’s mode

Return type:

str

internal_verify_deletion()

Abstract deletion verification method

internal_verify_health()

Verifies that the bucket is healthy by checking its mode

Returns:

True if the bucket is healthy, False otherwise

Return type:

bool

class ocs_ci.ocs.resources.objectbucket.MCGOCBucket(*args, **kwargs)

Bases: OCBucket

Implementation of an MCG bucket using the OC CLI

class ocs_ci.ocs.resources.objectbucket.MCGS3Bucket(*args, **kwargs)

Bases: ObjectBucket

Implementation of an MCG bucket using the S3 API

internal_delete()

Deletes the bucket using the S3 API

Raises:

NotFoundError – In case the bucket was not found

property internal_status

Returns the OBC mode as shown in the NB UI and retrieved via RPC

Returns:

The bucket’s mode

Return type:

str

internal_verify_deletion()

Abstract deletion verification method

internal_verify_health()

Verifies that the bucket is healthy by checking its mode

Returns:

True if the bucket is healthy, False otherwise

Return type:

bool

class ocs_ci.ocs.resources.objectbucket.OBC(obc_name)

Bases: object

Wrapper class for Object Bucket Claim credentials

access_key = None

access_key_id = None

bucket_name = None

namespace = None

ob_name = None

obc_account = None

obc_name = None

s3_endpoint = None

s3_resource = None

class ocs_ci.ocs.resources.objectbucket.OCBucket(*args, **kwargs)

Bases: ObjectBucket

internal_delete(verify=True)

Deletes the bucket using the OC CLI

Raises:

NotFoundError – In case the OBC was not found

property internal_status

Returns the OBC’s phase

Returns:

OBC phase

Return type:

str

internal_verify_deletion()

Abstract deletion verification method

internal_verify_health()

Verifies that the bucket is healthy by checking its phase

Returns:

True if the bucket is healthy, False otherwise

Return type:

bool

class ocs_ci.ocs.resources.objectbucket.ObjectBucket(name, mcg=None, rgw=None, bucketclass=None, replication_policy=None, quota=None, *args, **kwargs)

Bases: ABC

Base abstract class for MCG buckets

delete(verify=True)

Super method that first logs the bucket deletion and then calls the appropriate implementation

abstract internal_delete()

Abstract internal deletion method

abstract internal_status()

Abstract status method

abstract internal_verify_deletion()

Abstract deletion verification method

abstract internal_verify_health()

Abstract health verification method

mcg = None

name = None

property status

A method that first logs the bucket’s status and then calls the appropriate implementation

verify_deletion(timeout=60, interval=5)

Super method used for logging the deletion verification process and then calls the appropriate implementatation

verify_health(timeout=60, interval=5, **kwargs)

Health verification function that tries to verify the a bucket’s health by using its appropriate internal_verify_health function until a given time limit is reached

Parameters:

• timeout (int) – Timeout for the check, in seconds

• interval (int) – Interval to wait between checks, in seconds

class ocs_ci.ocs.resources.objectbucket.RGWOCBucket(*args, **kwargs)

Bases: OCBucket

Implementation of an RGW bucket using the S3 API

ocs_ci.ocs.resources.objectconfigfile module

Representation of general Kubernetes/OpenShift objects config file.

This allows one to work with multiple objects of different kind at once, as explained in Imperative Management of Kubernetes Objects Using Configuration Files.

Usage:

First you prepare list of dictionaries of k8s objects such as Deployment or PVC which describes your workload/project to be deployed in OCP. Then create instance of ObjectConfFile class with the list to be able to create the resource in the cluster (to run it), or delete it later when it’s no longer needed.

class ocs_ci.ocs.resources.objectconfigfile.ObjectConfFile(name, obj_dict_list, project, tmp_path)

Bases: object

This class represents particular k8s object config file which describes multiple k8s resources.

Methods of this class implements Imperative Management of Kubernetes Objects Using Configuration Files.

apply(namespace=None)

Run oc apply on in this object file.

Parameters:

• namespace (str) – Name of the namespace where to deploy, overriding

• specify (self.project.namespace value (in a similar way how you can) –

• apply. (any value to -n option of oc) –

create(namespace=None)

Run oc create on in this object file.

Parameters:

• namespace (str) – Name of the namespace where to deploy, overriding

• specify (self.project.namespace value (in a similar way how you can) –

• create. (any value to -n option of oc) –

delete(namespace=None)

Run oc delete on in this object file.

Parameters:

• namespace (str) – Name of the namespace where to deploy, overriding

• specify (self.project.namespace value (in a similar way how you can) –

• delete. (any value to -n option of oc) –

describe(namespace=None)

Run oc describe on in this object file.

Parameters:

• namespace (str) – Name of the namespace where to deploy, overriding

• specify (self.project.namespace value (in a similar way how you can) –

• describe. (any value to -n option of oc) –

get(namespace=None)

Run oc get on in this object file.

Parameters:

• namespace (str) – Name of the namespace where to deploy, overriding

• specify (self.project.namespace value (in a similar way how you can) –

• get. (any value to -n option of oc) –

wait_for_delete(resource_name='', timeout=60, sleep=3, namespace=None)

Wait for a resource to be deleted

Parameters:

• resource_name (str) – The name of the resource to wait for (e.g.kube_obj_name)

• timeout (int) – Time in seconds to wait

• sleep (int) – Sampling time in seconds

• namespace (str) – Name of the namespace where to deploy, overriding self.project.namespace value (in a similar way how you can specify any value to -n option of oc get.

Raises:

• CommandFailed – If failed to verify the resource deletion

• TimeoutError – If resource is not deleted within specified timeout

Returns:

True in case resource deletion is successful

Return type:

bool

ocs_ci.ocs.resources.objectconfigfile.link_spec_volume(spec_dict, volume_name, pvc_name)

Find volume of given name in given spec dict, and set given pvc name as a pvc for the volume in the spec.

Parameters:

• spec_dict (dict) – dictionary with a container/template spec

• volume_name (str) – name of the volume in the spec dict to link

• pvc_name (str) – name of the target pvc (for the given volume)

Raises:

NotFoundError when given volume is not found in given spec –

ocs_ci.ocs.resources.ocs module

General OCS object

class ocs_ci.ocs.resources.ocs.OCS(**kwargs)

Bases: object

Base OCSClass

add_label(label)

Addss a new label

Parameters:

label (str) – New label to be assigned for this pod E.g: “label=app=’rook-ceph-mds’”

property api_version

apply(**data)

create(do_reload=True)

delete(wait=True, force=False)

Delete the OCS object if its not already deleted (using the internal is_deleted flag)

Parameters:

• wait (bool) – Wait for object to be deleted

• force (bool) – Force delete object

Returns:

True if deleted, False otherwise

Return type:

bool

delete_temp_yaml_file()

describe()

get(out_yaml_format=True)

property is_deleted

property kind

property name

property namespace

reload()

Reloading the OCS instance with the new information from its actual data. After creating a resource from a yaml file, the actual yaml file is being changed and more information about the resource is added.

set_deleted()

status()

ocs_ci.ocs.resources.ocs.check_if_cluster_was_upgraded()

Check whether the OCS cluster went through upgrade

Returns:

True if the OCS cluster went through upgrade, False otherwise

Return type:

bool

ocs_ci.ocs.resources.ocs.get_ocs_csv()

Get the OCS CSV object

Returns:

OCS CSV object

Return type:

CSV

Raises:

CSVNotFound – In case no CSV found.

ocs_ci.ocs.resources.ocs.get_version_info(namespace=None)

ocs_ci.ocs.resources.osd_resize module

ocs_ci.ocs.resources.osd_resize.base_ceph_verification_steps_post_resize_osd(old_osd_pods, old_osd_pvcs, old_osd_pvs, expected_storage_size)

Check the Ceph verification steps post resize OSD. It will perform the following steps: 1. Check the resources state post resize OSD 2. Check the resources size post resize OSD 3. Check the Ceph state post resize OSD

Parameters:

• old_osd_pods (list) – The old osd pod objects before resizing the osd

• old_osd_pvcs (list) – The old osd PVC objects before resizing the osd

• old_osd_pvs (list) – The old osd PV objects before resizing the osd

• expected_storage_size (str) – The expected storage size after resizing the osd

Raises:

StorageSizeNotReflectedException – If the current storage size, PVCs, PVs, and ceph capacity are not in the expected size

ocs_ci.ocs.resources.osd_resize.ceph_verification_steps_post_resize_osd(old_osd_pods, old_osd_pvcs, old_osd_pvs, expected_storage_size, num_of_tries=6)

Try to execute the function ‘base_ceph_verification_steps_post_resize_osd’ a number of tries until success, ignoring the exception ‘StorageSizeNotReflectedException’. In every iteration, if we get the exception ‘StorageSizeNotReflectedException’, it will restart the osd pods and try again until it reaches the maximum tries.

Parameters:

• old_osd_pods (list) – The old osd pod objects before resizing the osd

• old_osd_pvcs (list) – The old osd PVC objects before resizing the osd

• old_osd_pvs (list) – The old osd PV objects before resizing the osd

• expected_storage_size (str) – The expected storage size after resizing the osd

• num_of_tries (int) – The number of tries to try executing the function ‘base_ceph_verification_steps_post_resize_osd’.

Raises:

StorageSizeNotReflectedException – If the current storage size, PVCs, PVs, and ceph capacity are not in the expected size

ocs_ci.ocs.resources.osd_resize.check_ceph_health_after_resize_osd(ceph_health_tries=40, ceph_rebalance_timeout=900)

Check Ceph health after resize osd

Parameters:

• ceph_health_tries (int) – The number of tries to wait for the Ceph health to be OK.

• ceph_rebalance_timeout (int) – The time to wait for the Ceph cluster rebalanced.

ocs_ci.ocs.resources.osd_resize.check_ceph_state_post_resize_osd()

Check the Ceph state post resize osd. The function checks the Ceph device classes and osd tree.

Raises:

CephHealthException – In case the Ceph device classes and osd tree checks didn’t finish successfully

ocs_ci.ocs.resources.osd_resize.check_resources_state_post_resize_osd(old_osd_pods, old_osd_pvcs, old_osd_pvs)

Check that the pods, PVCs, and PVs are in the expected state post resizing the osd. It will perform the following steps: 1. Check that the old osd pods are in a terminating state or deleted 2. Check that the new osd pods running, and we have exactly the same number of osd pods as the old ones. 3. Check that the PVCs are in a Bound state 4. Check that the old PVC and PV names are equal to the current PVC and PV names

Parameters:

• old_osd_pods (list) – The old osd pod objects before resizing the osd

• old_osd_pvcs (list) – The old osd PVC objects before resizing the osd

• old_osd_pvs (list) – The old osd PV objects before resizing the osd

Raises:

ResourceWrongStatusException – If the following occurs: 1. The OSD pods failed to reach the status Terminated or to be deleted 2. The old PVC and PV names are not equal to the current PVC and PV names

ocs_ci.ocs.resources.osd_resize.check_storage_size_is_reflected(expected_storage_size)

Check that the expected storage size is reflected in the current storage size, PVCs, PVs, and ceph capacity.

Parameters:

expected_storage_size (str) – The expected storage size

Raises:

StorageSizeNotReflectedException – If the current storage size, PVCs, PVs, and ceph capacity are not in the expected size

ocs_ci.ocs.resources.packagemanifest module

Package manifest related functionalities

class ocs_ci.ocs.resources.packagemanifest.PackageManifest(resource_name='', namespace='openshift-marketplace', install_plan_namespace=None, subscription_plan_approval='Automatic', **kwargs)

Bases: OCP

This class represent PackageManifest and contains all related methods.

get(**kwargs)

Overloaded get method from OCP class.

Raises:

ResourceNotFoundError – In case the selector and resource_name specified and no such resource found.

get_channels()

Returns channels for package manifest

Returns:

available channels for package manifest

Return type:

list

Raises:

ResourceNameNotSpecifiedException – in case the name is not specified.

get_current_csv(channel=None, csv_pattern='ocs-operator')

Returns current csv for default or specified channel

Parameters:

• channel (str) – Channel of the CSV

• csv_pattern (str) – CSV name pattern - needed for manual subscription plan

Returns:

Current CSV name

Return type:

str

Raises:

• ResourceNameNotSpecifiedException – in case the name is not specified.

• ChannelNotFound – in case the required channel doesn’t exist.

get_default_channel()

Returns default channel for package manifest

Returns:

default channel name

Return type:

str

Raises:

ResourceNameNotSpecifiedException – in case the name is not specified.

get_installed_csv_from_install_plans(pattern)

Get currently installed CSV out latest approved install plans.

Parameters:

patter (str) – pattern of CSV name to look for.

Raises:

• CSVNotFound – In case no CSV found from approved install plans.

• NoInstallPlanForApproveFoundException – In case no install plan for approve found.

wait_for_resource(resource_name='', timeout=60, sleep=3, label=None)

Wait for a packagemanifest exists.

Parameters:

• resource_name (str) – The name of the resource to wait for. If not specified the self.resource_name will be used. At least on of those has to be set!

• timeout (int) – Time in seconds to wait

• sleep (int) – Sampling time in seconds

Raises:

• ResourceNameNotSpecifiedException – in case the name is not specified.

• TimeoutExpiredError – in case the resource not found in timeout

ocs_ci.ocs.resources.packagemanifest.get_selector_for_ocs_operator()

This is the helper function which returns selector for package manifest. It’s needed because of conflict with live content and multiple package manifests with the ocs-operator name. In case we are using internal builds we label catalog source or operator source and using the same selector for package manifest.

Returns:

Selector for package manifest if we are on internal

builds, otherwise it returns None

Return type:

str

ocs_ci.ocs.resources.pod module

Pod related functionalities and context info

Each pod in the openshift cluster will have a corresponding pod object

class ocs_ci.ocs.resources.pod.Pod(**kwargs)

Bases: OCS

Handles per pod related context

add_role(role)

Adds a new role for this pod

Parameters:

role (str) – New role to be assigned for this pod

copy_from_pod_oc_exec(target_path, src_path, timeout=600, chunk_size=2000)

Copies to local path file from the pod using standard output stream via ‘oc exec’. Good for log/json/yaml/text files, not good for large files/binaries with one-line plain string * Hand data from the pod over oc exec cat, other standard output ways will fail for the files > 1 Mb

Parameters:

• target_path (str) – local path

• src_path (str) – path within pod what you want to copy

• timeout (int) – timeout in seconds until size of the local file will not reach the initial file size. 2000 lines is a maximum chunk size tested successfully

• chunk_size (int) – file will be copied by chunks, by number of lines

copy_to_pod_cat(src_path, target_path, timeout=60)

Copies to pod path from the local path file using standard output stream

Parameters:

• src_path (str) – local path

• target_path (str) – path within pod where you want to copy

• timeout (int) – timeout in seconds

Returns:

stdout of the command

Return type:

str

copy_to_pod_rsync(src_path, target_path, container=None)

Copies to pod path from the local path

Parameters:

• src_path (str) – local path

• target_path (str) – path within pod where you want to copy

• container (str) – if multi-container pod you can specify the container name, by default its None

Returns:

stdout of the command

Return type:

str

copy_to_server(server, authkey, localpath, remotepath, user=None)

Upload a file from pod to server

Parameters:

• server (str) – Name of the server to upload

• authkey (str) – Authentication file (.pem file)

• localpath (str) – Local file/dir in pod to upload

• remotepath (str) – Target path on the remote server

• user (str) – User name to connect to server

exec_ceph_cmd(ceph_cmd, format='json-pretty', out_yaml_format=True, timeout=600)

Execute a Ceph command on the Ceph tools pod

Parameters:

• ceph_cmd (str) – The Ceph command to execute on the Ceph tools pod

• format (str) – The returning output format of the Ceph command

• out_yaml_format (bool) – whether to return yaml loaded python object OR to return raw output

• timeout (int) – timeout for the exec_cmd_on_pod, defaults to 600 seconds

Returns:

Ceph command output

Return type:

dict

Raises:

CommandFailed – In case the pod is not a toolbox pod

exec_cmd_on_node(server, authkey, cmd, user=None)

Run command on a remote server from pod

Parameters:

• server (str) – Name of the server to run the command

• authkey (str) – Authentication file (.pem file)

• cmd (str) – command to run on server from pod

• user (str) – User name to connect to server

exec_cmd_on_pod(command, out_yaml_format=True, secrets=None, timeout=600, container_name=None, cluster_config=None, **kwargs)

Execute a command on a pod (e.g. oc rsh)

Parameters:

• command (str) – The command to execute on the given pod

• out_yaml_format (bool) – whether to return yaml loaded python object OR to return raw output

• secrets (list) – A list of secrets to be masked with asterisks This kwarg is popped in order to not interfere with subprocess.run(**kwargs)

• timeout (int) – timeout for the exec_oc_cmd, defaults to 600 seconds

• container_name (str) – The container name

• cluster_config (MultiClusterConfig) – In case of multicluser scenario, this object will hold specific cluster’s Config

Returns:

This object represents a returned yaml file

Return type:

Munch Obj

exec_s3_cmd_on_pod(command, mcg_obj=None)

Execute an S3 command on a pod

Parameters:

• mcg_obj (MCG) – An MCG object containing the MCG S3 connection credentials

• command (str) – The command to execute on the given pod

Returns:

This object represents a returned yaml file

Return type:

Munch Obj

exec_sh_cmd_on_pod(command, sh='bash', timeout=600, **kwargs)

Execute a pure bash command on a pod via oc exec where you can use bash syntaxt like &&, ||, ;, for loop and so on.

Parameters:

• command (str) – The command to execute on the given pod

• timeout (int) – timeout for the exec_oc_cmd, defaults to 600 seconds

Returns:

stdout of the command

Return type:

str

fillup_fs(size, fio_filename=None, performance_pod=False)

Execute FIO on a pod to fillup a file This will run sequential IO of 1MB block size to fill up the fill with data This operation will run in background and will store the results in ‘self.thread.result()’. In order to wait for the output and not continue with the test until FIO is done, call self.thread.result() right after calling run_io. See tests/manage/test_pvc_deletion_during_io.py::test_run_io for usage of FIO

Parameters:

• size (str) – Size in MB, e.g. ‘200M’

• fio_filename (str) – Name of fio file created on app pod’s mount point

• performance_pod (bool) – True if this pod is the performance pod created with PERF_POD_YAML

get_fio_results(timeout=600)

Get FIO execution results

Returns:

Dictionary represents the FIO execution results

Return type:

dict

Raises:

Exception – In case of exception from FIO

get_labels()

Get labels from pod

Raises:

NotFoundError – If resource not found

Returns:

All the openshift labels on a given pod

Return type:

dict

get_memory(container_name)

Get the pod memory size

Parameters:

container_name (str) – The name of the container to look for

Returns:

The container memory size (e.g. ‘5Gi’)

Return type:

str

get_node()

Gets the node name

Returns:

Node name

Return type:

str

get_storage_path(storage_type='fs')

Get the pod volume mount path or device path

Returns:

The mount path of the volume on the pod (e.g. /var/lib/www/html/) if storage_type is fs

else device path of raw block pv

Return type:

str

install_packages(packages)

Install packages in a Pod

Parameters:

packages (list) – List of packages to install

property labels

property name

property namespace

property restart_count

property roles

run_git_clone(skip_install=True)

Execute git clone on a pod to simulate a Jenkins user

Parameters:

skip_install (bool) – By default True, skips git package installation in pod

run_io(storage_type, size, io_direction='rw', rw_ratio=75, jobs=1, runtime=60, depth=4, rate='1m', rate_process='poisson', fio_filename=None, bs='4K', end_fsync=0, invalidate=None, buffer_compress_percentage=None, buffer_pattern=None, readwrite=None, direct=0, verify=False, fio_installed=False, timeout=0)

Execute FIO on a pod This operation will run in background and will store the results in ‘self.thread.result()’. In order to wait for the output and not continue with the test until FIO is done, call self.thread.result() right after calling run_io. See tests/manage/test_pvc_deletion_during_io.py::test_run_io for usage of FIO

Parameters:

• storage_type (str) – ‘fs’ or ‘block’

• size (str) – Size in MB, e.g. ‘200M’

• io_direction (str) – Determines the operation: ‘ro’, ‘wo’, ‘rw’ (default: ‘rw’)

• rw_ratio (int) – Determines the reads and writes using a <rw_ratio>%/100-<rw_ratio>% (e.g. the default is 75 which means it is 75%/25% which equivalent to 3 reads are performed for every 1 write)

• jobs (int) – Number of jobs to execute FIO

• runtime (int) – Number of seconds IO should run for

• depth (int) – IO depth

• rate (str) – rate of IO default 1m, e.g. 16k

• rate_process (str) – kind of rate process default poisson, e.g. poisson

• fio_filename (str) – Name of fio file created on app pod’s mount point

• bs (str) – Block size, e.g. 4K

• end_fsync (int) – If 1, fio will sync file contents when a write stage has completed. Fio default is 0

• invalidate (bool) – Invalidate the buffer/page cache parts of the files to be used prior to starting I/O

• buffer_compress_percentage (int) – If this is set, then fio will attempt to provide I/O buffer content (on WRITEs) that compresses to the specified level

• buffer_pattern (str) – fio will fill the I/O buffers with this pattern

• readwrite (str) – Type of I/O pattern default is randrw from yaml

• direct (int) – If value is 1, use non-buffered I/O. This is usually O_DIRECT. Fio default is 0.

• verify (bool) – This method verifies file contents after each iteration of the job. e.g. crc32c, md5

• fio_installed (bool) – True if fio is already installed on the pod

• timeout (int) – The timeout in seconds to wait for fio to be completed

wait_for_pod_delete(resource_name=None, timeout=60)

Waits for the pod delete

Parameters:

• resource_name (str) – name of the pod

• timeout (int) – time to wait in seconds

Returns:

True

Return type:

bool

workload_setup(storage_type, jobs=1, fio_installed=False)

Do setup on pod for running FIO

Parameters:

• storage_type (str) – ‘fs’ or ‘block’

• jobs (int) – Number of jobs to execute FIO

• fio_installed (bool) – True if fio is already installed on the pod

ocs_ci.ocs.resources.pod.cal_md5sum(pod_obj, file_name, block=False, raw_path=False)

Calculates the md5sum of the file

Parameters:

• pod_obj (Pod) – The object of the pod

• file_name (str) – The name of the file for which md5sum to be calculated

• block (bool) – True if the volume mode of PVC used on pod is ‘Block’. file_name will be the devicePath in this case.

• raw_path (bool) – True if file_name includes the filepath and should be used as-is i.e. - a pod without a PVC attached to it

Returns:

The md5sum of the file

Return type:

str

ocs_ci.ocs.resources.pod.calculate_md5sum_of_pod_files(pods_for_integrity_check, pod_file_name)

Calculate the md5sum of the pod files, and save it in the pod objects

Parameters:

• pods_for_integrity_check (list) – The list of the pod objects to calculate the md5sum

• pod_file_name (str) – The pod file name to save the md5sum

ocs_ci.ocs.resources.pod.check_ceph_cmd_execute_successfully()

Check that a Ceph command executes successfully

Returns:

True, if the Ceph command executed successfully. False, otherwise

Return type:

bool

ocs_ci.ocs.resources.pod.check_file_existence(pod_obj, file_path)

Check if file exists inside the pod

Parameters:

• pod_obj (Pod) – The object of the pod

• file_path (str) – The full path of the file to look for inside the pod

Returns:

True if the file exist, False otherwise

Return type:

bool

ocs_ci.ocs.resources.pod.check_pods_after_node_replacement()

Check the pods status after the node replacement process.

Returns:

True if all the pods are running after a specific time. False otherwise.

Return type:

bool

ocs_ci.ocs.resources.pod.check_pods_in_running_state(namespace='openshift-storage', pod_names=None, raise_pod_not_found_error=False, skip_for_status=None)

Checks whether the pods in a given namespace are in Running state or not. The pods which are in ‘Completed’ state will be skipped when checking for all pods in the namespace openshift-storage. ‘Completed’ will be the expected state of such pods.

Parameters:

• namespace (str) – Name of cluster namespace(default: config.ENV_DATA[“cluster_namespace”])

• pod_names (list) – List of the pod names to check. If not provided, it will check all the pods in the given namespace

• raise_pod_not_found_error (bool) – If True, it raises an exception, if one of the pods in the pod names are not found. If False, it ignores the case of pod not found and returns the pod objects of the rest of the pod names. The default value is False

• skip_for_status (list) – List of pod status that should be skipped. If the status of a pod is in the given list, the check for ‘Running’ status of that particular pod will be skipped. eg: [“Pending”, “Completed”]

Returns:

True, if all pods in Running state. False, otherwise

Return type:

Boolean

ocs_ci.ocs.resources.pod.check_pods_in_statuses(expected_statuses, pod_names=None, namespace='openshift-storage', raise_pod_not_found_error=False, exclude_pod_name_prefixes=None)

checks whether the pods in a given namespace are in the expected statuses or not

Parameters:

• expected_statuses (list) – The expected statuses of the pods

• pod_names (list) – List of the pod names to check. If not provided, it will check all the pods in the given namespace

• namespace (str) – Name of cluster namespace(default: config.ENV_DATA[“cluster_namespace”])

• raise_pod_not_found_error (bool) – If True, it raises an exception, if one of the pods in the pod names are not found. If False, it ignores the case of pod not found and check the pod objects of the rest of the pod names. The default value is False

• exclude_pod_name_prefixes (list) – The list of the pod name prefixes to exclude from the pods to check

Returns:

True, if the pods are in the expected statuses. False, otherwise

Return type:

Boolean

ocs_ci.ocs.resources.pod.check_safe_to_destroy_status(osd_id)

check if it is safe to destroy the osd

Parameters:

osd_id (str) – osd id

Returns:

True, if it is safe to destroy the osd. False, otherwise

Return type:

bool

ocs_ci.ocs.resources.pod.check_toleration_on_pods(toleration_key='node.ocs.openshift.io/storage')

Function to check toleration on pods

Parameters:

toleration_key (str) – The toleration key to check

ocs_ci.ocs.resources.pod.delete_all_osd_removal_jobs(namespace='openshift-storage')

Delete all the osd removal jobs in a specific namespace

Parameters:

namespace (str) – Name of cluster namespace(default: config.ENV_DATA[“cluster_namespace”])

Returns:

True, if all the jobs deleted successfully. False, otherwise

Return type:

bool

ocs_ci.ocs.resources.pod.delete_deploymentconfig_pods(pod_obj)

Delete a DeploymentConfig pod and all the pods that are controlled by it

Parameters:

pod_obj (Pod) – Pod object

ocs_ci.ocs.resources.pod.delete_osd_removal_job(osd_id)

Delete the ocs-osd-removal job.

Parameters:

osd_id (str) – The osd id

Returns:

True, if the ocs-osd-removal job deleted successfully. False, otherwise

Return type:

bool

ocs_ci.ocs.resources.pod.delete_pods(pod_objs, wait=True)

Deletes list of the pod objects

Parameters:

• pod_objs (list) – List of the pod objects to be deleted

• wait (bool) – Determines if the delete command should wait for completion

ocs_ci.ocs.resources.pod.download_file_from_pod(pod_name, remotepath, localpath, namespace=None, cluster_config=None)

Download a file from a pod

Parameters:

• pod_name (str) – Name of the pod

• remotepath (str) – Target path on the pod

• localpath (str) – Local file to upload

• namespace (str) – The namespace of the pod

ocs_ci.ocs.resources.pod.exit_osd_maintenance_mode(osd_deployment)

Exit from osd maintenance mode

Parameters:

osd_deployment (OCS) – List of OSD deployment OCS instances

ocs_ci.ocs.resources.pod.get_admin_key_from_ceph_tools()

Fetches admin key secret from ceph :returns: admin keyring encoded with base64 as a string

ocs_ci.ocs.resources.pod.get_alertmanager_managed_ocs_alertmanager_pods(label='alertmanager=managed-ocs-alertmanager', namespace=None)

Get alertmanager-managed-ocs-alertmanager pods in the cluster

Parameters:

• label (str) – Label associated with alertmanager-managed-ocs-alertmanager pods

• namespace (str) – Namespace in which alertmanager-managed-ocs-alertmanager pods are residing

Returns:

Pod objects of alertmanager-managed-ocs-alertmanager pods

Return type:

list

ocs_ci.ocs.resources.pod.get_all_pods(namespace=None, selector=None, selector_label='app', exclude_selector=False, wait=False, field_selector=None)

Get all pods in a namespace.

Parameters:

• namespace (str) – Name of the namespace If namespace is None - get all pods

• selector (list) – List of the resource selector to search with. Example: [‘alertmanager’,’prometheus’]

• selector_label (str) – Label of selector (default: app).

• exclude_selector (bool) – If list of the resource selector not to search with

• field_selector (str) – Selector (field query) to filter on, supports ‘=’, ‘==’, and ‘!=’. (e.g. status.phase=Running)

Returns:

List of Pod objects

Return type:

list

ocs_ci.ocs.resources.pod.get_ceph_daemon_id(pod_obj)

Get Ceph Daemon ID of osd, mds, mon, rgw, mgr

Parameters:

pod_obj (POD Obj) – pod object

Returns:

ceph_daemon_id

Return type:

str

ocs_ci.ocs.resources.pod.get_ceph_tools_pod(skip_creating_pod=False, wait=False, namespace=None)

Get the Ceph tools pod

Parameters:

• skip_creating_pod (bool) – True if user doesn’t want to create new tool box if it doesn’t exist

• wait (bool) – True if you want to wait for the tool pods to be Running

• namespace – Namespace of OCS

Returns:

The Ceph tools pod object

Return type:

Pod object

Raises:

ToolBoxNotFoundException – In case of tool box not found

ocs_ci.ocs.resources.pod.get_cephfsplugin_provisioner_pods(cephfsplugin_provisioner_label='app=csi-cephfsplugin-provisioner', namespace=None)

Fetches info about CSI Cephfs plugin provisioner pods in the cluster

Parameters:

• cephfsplugin_provisioner_label (str) – label associated with cephfs provisioner pods (default: defaults.CSI_CEPHFSPLUGIN_PROVISIONER_LABEL)

• namespace (str) – Namespace in which ceph cluster lives (default: config.ENV_DATA[“cluster_namespace”])

Returns:

csi-cephfsplugin-provisioner Pod objects

Return type:

list

ocs_ci.ocs.resources.pod.get_containers_names_by_pod(pod: OCP) → set

Gets the names of all containers in given pod or pods

Parameters:

pod (ocp.OCP) – instance of OCP object that represents a pod (kind=POD)

Returns:

hash set of names of all containers in given pod or pods

Return type:

set

ocs_ci.ocs.resources.pod.get_crashcollector_pods(crashcollector_label='app=rook-ceph-crashcollector', namespace=None)

Fetches info about crashcollector pods in the cluster

Parameters:

• crashcollector_label (str) – label associated with mon pods (default: defaults.CRASHCOLLECTOR_APP_LABEL)

• namespace (str) – Namespace in which ceph cluster lives (default: config.ENV_DATA[“cluster_namespace”])

Returns:

of crashcollector pod objects

Return type:

list

ocs_ci.ocs.resources.pod.get_csi_provisioner_pod(interface)

Get the provisioner pod based on interface :returns: The provisioner pod object based on iterface :rtype: Pod object

ocs_ci.ocs.resources.pod.get_csi_snapshoter_pod()

Get the csi snapshot controller pod

Returns:

csi snapshot controller pod

Return type:

Pod object

ocs_ci.ocs.resources.pod.get_debug_pods(debug_nodes, namespace='openshift-storage')

Get debug pods created for the nodes in debug

Parameters:

• debug_nodes (list) – List of nodes in debug mode

• namespace (str) – By default ‘openshift-storage’ namespace

Returns:

of Pod objects

Return type:

List

ocs_ci.ocs.resources.pod.get_deployment_name(pod_name)

Get the deployment of the pod.

Parameters:

pod_name (str) – The pod’s name.

Returns:

The deployment of the specific pod name

ocs_ci.ocs.resources.pod.get_deployments_having_label(label, namespace)

Fetches deployment resources with given label in given namespace

Parameters:

• label (str) – label which deployments might have

• namespace (str) – Namespace in which to be looked up

Returns:

deployment OCP instances

Return type:

list

ocs_ci.ocs.resources.pod.get_device_path(pod_obj)

get device path from pod in block mode :param pod_obj: The object of the pod :type pod_obj: Pod

Returns:

device path

Return type:

str

ocs_ci.ocs.resources.pod.get_file_path(pod_obj, file_name)

Get the full path of the file

Parameters:

• pod_obj (Pod) – The object of the pod

• file_name (str) – The name of the file for which path to get

Returns:

The full path of the file

Return type:

str

ocs_ci.ocs.resources.pod.get_fio_rw_iops(pod_obj)

Execute FIO on a pod

Parameters:

pod_obj (Pod) – The object of the pod

ocs_ci.ocs.resources.pod.get_lvm_operator_pod(label='app.kubernetes.io/name=lvms-operator', namespace='openshift-storage')

Get lvm operator controller manager pod in the lvm cluster

Parameters:

• label (str) – Label associated with lvm operator controller manager pod

• namespace (str) – Namespace in which lvm operator controler manager pod is residing

Returns:

Pod object of lvm operator controller manager pod

Return type:

Pod

ocs_ci.ocs.resources.pod.get_lvm_vg_manager_pod(label='app.kubernetes.io/name=vg-manager', namespace='openshift-storage')

Get vg manager pod in the lvm cluster

Parameters:

• label (str) – Label associated with vg manager pod

• namespace (str) – Namespace in which vg manager pod is residing

Returns:

Pod object of vg manager pod

Return type:

Pod

ocs_ci.ocs.resources.pod.get_mds_pods(mds_label='app=rook-ceph-mds', namespace=None)

Fetches info about mds pods in the cluster

Parameters:

• mds_label (str) – label associated with mds pods (default: defaults.MDS_APP_LABEL)

• namespace (str) – Namespace in which ceph cluster lives (default: config.ENV_DATA[“cluster_namespace”])

Returns:

of mds pod objects

Return type:

list

ocs_ci.ocs.resources.pod.get_mgr_pods(mgr_label='app=rook-ceph-mgr', namespace=None)

Fetches info about mgr pods in the cluster

Parameters:

• mgr_label (str) – label associated with mgr pods (default: defaults.MGR_APP_LABEL)

• namespace (str) – Namespace in which ceph cluster lives (default: config.ENV_DATA[“cluster_namespace”])

Returns:

of mgr pod objects

Return type:

list

ocs_ci.ocs.resources.pod.get_mon_label(mon_pod_obj)

Gets the mon pod label

Parameters:

mon_pod_obj (Pod) – The pod object

Returns:

The mon pod label (eg: a)

Return type:

str

ocs_ci.ocs.resources.pod.get_mon_pod_by_pvc_name(pvc_name: str)

Function to get monitor pod by pvc_name label

Parameters:

pvc_name (str) – name of the pvc the monitor pod is related to

ocs_ci.ocs.resources.pod.get_mon_pod_id(mon_pod)

Get the mon pod id

Parameters:

mon_pod (ocs_ci.ocs.resources.pod.Pod) – The mon pod object

Returns:

The mon pod id

Return type:

str

ocs_ci.ocs.resources.pod.get_mon_pods(mon_label='app=rook-ceph-mon', namespace=None)

Fetches info about mon pods in the cluster

Parameters:

• mon_label (str) – label associated with mon pods (default: defaults.MON_APP_LABEL)

• namespace (str) – Namespace in which ceph cluster lives (default: config.ENV_DATA[“cluster_namespace”])

Returns:

of mon pod objects

Return type:

list

ocs_ci.ocs.resources.pod.get_noobaa_core_pod()

Fetches Noobaa core pod details

Returns:

Noobaa core pod object

Return type:

Pod object

ocs_ci.ocs.resources.pod.get_noobaa_db_pod()

Get noobaa db pod obj

Returns:

Noobaa db pod object

Return type:

Pod object

ocs_ci.ocs.resources.pod.get_noobaa_endpoint_pods()

Fetches noobaa endpoint pod details

Returns:

List containing noobaa endpoint pod objects

Return type:

List

ocs_ci.ocs.resources.pod.get_noobaa_operator_pod(ocs_label='noobaa-operator=deployment', namespace=None)

Fetches info about noobaa operator pod in the cluster

Parameters:

• ocs_label (str) – label associated with noobaa_operator pod (default: defaults.NOOBAA_OPERATOR_POD_LABEL)

• namespace (str) – Namespace in which ceph cluster lives (default: none)

Returns:

noobaa_operator pod object

Return type:

Pod object

ocs_ci.ocs.resources.pod.get_noobaa_pods(noobaa_label='app=noobaa', namespace=None)

Fetches info about noobaa pods in the cluster

Parameters:

• noobaa_label (str) – label associated with osd pods (default: defaults.NOOBAA_APP_LABEL)

• namespace (str) – Namespace in which ceph cluster lives (default: config.ENV_DATA[“cluster_namespace”])

Returns:

of noobaa pod objects

Return type:

list

ocs_ci.ocs.resources.pod.get_not_running_pods(selector=None, namespace='openshift-storage')

Get all the non-running pods in a given namespace and give selector

Returns:

all the pods that are not Running

Return type:

List

ocs_ci.ocs.resources.pod.get_ocs_operator_pod(ocs_label='name=ocs-operator', namespace=None)

Fetches info about rgw pods in the cluster

Parameters:

• ocs_label (str) – label associated with ocs_operator pod (default: defaults.OCS_OPERATOR_LABEL)

• namespace (str) – Namespace in which ceph cluster lives (default: none)

Returns:

ocs_operator pod object

Return type:

Pod object

ocs_ci.ocs.resources.pod.get_ocs_osd_controller_manager_pod(label='control-plane=controller-manager', namespace=None)

Get ocs-osd-controller-manager pod in the cluster

Parameters:

• label (str) – Label associated with ocs-osd-controller-manager pod

• namespace (str) – Namespace in which ocs-osd-controller-manager pod is residing

Returns:

Pod object of ocs-osd-controller-manager pod

Return type:

Pod

ocs_ci.ocs.resources.pod.get_ocs_provider_server_pod(label='app=ocsProviderApiServer', namespace=None)

Get ocs-provider-server pod in the cluster

Parameters:

• label (str) – Label associated with ocs-provider-server pod

• namespace (str) – Namespace in which ocs-provider-server pod is residing

Returns:

Pod object of ocs-provider-server pod

Return type:

Pod

ocs_ci.ocs.resources.pod.get_odf_operator_controller_manager(ocs_label='control-plane=controller-manager', namespace=None)

Fetches info about odf operator control manager pod in the cluster

Parameters:

• ocs_label (str) – label associated with ocs_operator pod (default: defaults.ODF_OPERATOR_CONTROL_MANAGER_LABEL)

• namespace (str) – Namespace in which ceph cluster lives (default: none)

Returns:

odf_operator_controller_manager pod object

Return type:

Pod object

ocs_ci.ocs.resources.pod.get_operator_pods(operator_label='app=rook-ceph-operator', namespace=None)

Fetches info about rook-ceph-operator pods in the cluster

Parameters:

• operator_label (str) – Label associated with rook-ceph-operator pod

• namespace (str) – Namespace in which ceph cluster lives

Returns:

of rook-ceph-operator pod objects

Return type:

list

ocs_ci.ocs.resources.pod.get_osd_deployments(osd_label='app=rook-ceph-osd', namespace=None)

Fetches info about osd deployments in the cluster

Parameters:

• osd_label (str) – label associated with osd deployments (default: defaults.OSD_APP_LABEL)

• namespace (str) – Namespace in which ceph cluster lives (default: config.ENV_DATA[“cluster_namespace”])

Returns:

OSD deployment OCS instances

Return type:

list

ocs_ci.ocs.resources.pod.get_osd_pod_id(osd_pod)

Get the osd pod id

Parameters:

osd_pod (ocs_ci.ocs.resources.pod.Pod) – The osd pod object

Returns:

The osd pod id

Return type:

str

ocs_ci.ocs.resources.pod.get_osd_pods(osd_label='app=rook-ceph-osd', namespace=None)

Fetches info about osd pods in the cluster

Parameters:

• osd_label (str) – label associated with osd pods (default: defaults.OSD_APP_LABEL)

• namespace (str) – Namespace in which ceph cluster lives (default: config.ENV_DATA[“cluster_namespace”])

Returns:

of osd pod objects

Return type:

list

ocs_ci.ocs.resources.pod.get_osd_pods_having_ids(osd_ids)

Get the osd pods having specific ids

Parameters:

osd_ids (list) – The list of the osd ids

Returns:

The osd pods having the osd ids

Return type:

list

ocs_ci.ocs.resources.pod.get_osd_prepare_pods(osd_prepare_label='app=rook-ceph-osd-prepare', namespace='openshift-storage')

Fetches info about osd prepare pods in the cluster

Parameters:

• osd_prepare_label (str) – label associated with osd prepare pods (default: constants.OSD_PREPARE_APP_LABEL)

• namespace (str) – Namespace in which ceph cluster lives (default: config.ENV_DATA[“cluster_namespace”])

Returns:

OSD prepare pod objects

Return type:

list

ocs_ci.ocs.resources.pod.get_osd_removal_pod_name(timeout=60)

Get the osd removal pod name

Parameters:

timeout (int) – The time to wait for getting the osd removal pod name

Returns:

The osd removal pod name

Return type:

str

ocs_ci.ocs.resources.pod.get_plugin_pods(interface, namespace=None)

Fetches info of csi-cephfsplugin pods or csi-rbdplugin pods

Parameters:

• interface (str) – Interface type. eg: CephBlockPool, CephFileSystem

• namespace (str) – Name of cluster namespace

Returns:

csi-cephfsplugin pod objects or csi-rbdplugin pod objects

Return type:

list

ocs_ci.ocs.resources.pod.get_plugin_provisioner_leader(interface, namespace=None, leader_type='provisioner')

Get csi-cephfsplugin-provisioner or csi-rbdplugin-provisioner leader pod

Parameters:

• interface (str) – Interface type. eg: CephBlockPool, CephFileSystem

• namespace (str) – Name of cluster namespace

• leader_type (str) – Parameter to check the lease. eg: ‘snapshotter’ to select external-snapshotter leader holder

Returns:

csi-cephfsplugin-provisioner or csi-rbdplugin-provisioner leader

pod

Return type:

Pod

ocs_ci.ocs.resources.pod.get_pod_ceph_daemon_type(pod_obj)

Get the ceph daemon type of the pod object

Parameters:

pod_obj (Pod) – the pod object

Returns:

The pod’s ceph daemon type

Return type:

str

ocs_ci.ocs.resources.pod.get_pod_count(label, namespace=None)

ocs_ci.ocs.resources.pod.get_pod_ip(pod_obj)

Get the pod ip

Parameters:

pod_obj (Pod) – The pod object

Returns:

The pod ip

Return type:

str

ocs_ci.ocs.resources.pod.get_pod_logs(pod_name, container=None, namespace='openshift-storage', previous=False, all_containers=False, since=None)

Get logs from a given pod

pod_name (str): Name of the pod container (str): Name of the container namespace (str): Namespace of the pod previous (bool): True, if pod previous log required. False otherwise. all_containers (bool): fetch logs from all containers of the resource since (str): only return logs newer than a relative duration like 5s, 2m, or 3h.

Returns:

Output from ‘oc get logs <pod_name> command

Return type:

str

ocs_ci.ocs.resources.pod.get_pod_node(pod_obj)

Get the node that the pod is running on

Parameters:

pod_obj (OCS) – The pod object

Returns:

The node object

Return type:

ocs_ci.ocs.ocp.OCP

Raises:

NotFoundError when the node name is not found –

ocs_ci.ocs.resources.pod.get_pod_obj(name, namespace=None)

Returns the pod obj for the given pod

Parameters:

name (str) – Name of the resources

Returns:

A pod object

Return type:

obj

ocs_ci.ocs.resources.pod.get_pod_objs(pod_names, raise_pod_not_found_error=False, namespace='openshift-storage')

Get the pod objects of the specified pod names

Parameters:

• pod_names (list) – The list of the pod names to get their pod objects

• namespace (str) – Name of cluster namespace(default: config.ENV_DATA[“cluster_namespace”])

• raise_pod_not_found_error (bool) – If True, it raises an exception, if one of the pods in the pod names are not found. If False, it ignores the case of pod not found and returns the pod objects of the rest of the pod names. The default value is False

Returns:

The pod objects of the specified pod names

Return type:

list

Raises:

ResourceNotFoundError – If ‘raise_pod_not_found_error’ is True, and not all the pod names were found

ocs_ci.ocs.resources.pod.get_pod_restarts_count(namespace='openshift-storage', label=None)

Gets the dictionary of pod and its restart count for all the pods in a given namespace

Returns:

dictionary of pod name and its corresponding restart count

Return type:

dict

ocs_ci.ocs.resources.pod.get_pods_having_label(label, namespace='openshift-storage', cluster_config=None, statuses=None)

Fetches pod resources with given label in given namespace

Parameters:

• label (str) – label which pods might have

• namespace (str) – Namespace in which to be looked up

• cluster_config (MultiClusterConfig) – In case of multicluster, this object will hold specif cluster config

• statuses (list) – List of pod statuses. Fetch only pods in any of the status mentioned in the statuses list

Returns:

of pods info

Return type:

list

ocs_ci.ocs.resources.pod.get_pods_in_statuses(status_options, namespace='openshift-storage', exclude_pod_name_prefixes=None)

Get all the pods in specific statuses

Parameters:

• status_options (list) – The list of the status options.

• namespace (str) – Name of cluster namespace(default: config.ENV_DATA[“cluster_namespace”])

• exclude_pod_name_prefixes (list) – The list of the pod name prefixes to exclude from the pods to get

Returns:

All the pods that their status in the ‘status_options’ list.

Return type:

list

ocs_ci.ocs.resources.pod.get_prometheus_managed_ocs_prometheus_pod(label='prometheus=managed-ocs-prometheus', namespace=None)

Get prometheus-managed-ocs-prometheus pod in the cluster

Parameters:

• label (str) – Label associated with prometheus-managed-ocs-prometheus pod

• namespace (str) – Namespace in which prometheus-managed-ocs-prometheus pod is residing

Returns:

Pod object of prometheus-managed-ocs-prometheus pod

Return type:

Pod

ocs_ci.ocs.resources.pod.get_prometheus_operator_pod(label='app.kubernetes.io/name=prometheus-operator', namespace=None)

Get prometheus-operator pod in the cluster

Parameters:

• label (str) – Label associated with prometheus-operator pod

• namespace (str) – Namespace in which prometheus-operator pod is residing

Returns:

Pod object of prometheus-operator pod

Return type:

Pod

ocs_ci.ocs.resources.pod.get_pvc_name(pod_obj)

Function to get pvc_name from pod_obj

Parameters:

pod_obj (str) – The pod object

Returns:

The pvc name of a given pod_obj,

Return type:

str

Raises:

UnavailableResourceException – If no pvc attached

ocs_ci.ocs.resources.pod.get_rbdfsplugin_provisioner_pods(rbdplugin_provisioner_label='app=csi-rbdplugin-provisioner', namespace=None)

Fetches info about CSI Cephfs plugin provisioner pods in the cluster

Parameters:

• rbdplugin_provisioner_label (str) – label associated with RBD provisioner pods (default: defaults.CSI_RBDPLUGIN_PROVISIONER_LABEL)

• namespace (str) – Namespace in which ceph cluster lives (default: config.ENV_DATA[“cluster_namespace”])

Returns:

csi-rbdplugin-provisioner Pod objects

Return type:

list

ocs_ci.ocs.resources.pod.get_rgw_pods(rgw_label='app=rook-ceph-rgw', namespace=None)

Fetches info about rgw pods in the cluster

Parameters:

• rgw_label (str) – label associated with rgw pods (default: defaults.RGW_APP_LABEL)

• namespace (str) – Namespace in which ceph cluster lives (default: none)

Returns:

Pod objects of rgw pods

Return type:

list

ocs_ci.ocs.resources.pod.get_rook_ceph_pod_names()

Get all the rook ceph pod names

Returns:

List of the rook ceph pod names

Return type:

list

ocs_ci.ocs.resources.pod.get_running_state_pods(namespace='openshift-storage')

Checks the running state pods in a given namespace.

Returns:

List: all the pod objects that are in running state only

ocs_ci.ocs.resources.pod.get_topolvm_controller_pod(label='app.kubernetes.io/component=topolvm-controller', namespace='openshift-storage')

Get topolvm controller pod in the lvm cluster

Parameters:

• label (str) – Label associated with topolvm controller pod

• namespace (str) – Namespace in which topolvm controler pod is residing

Returns:

Pod object of topolvm controller pod

Return type:

Pod

ocs_ci.ocs.resources.pod.get_topolvm_node_pod(label='app.kubernetes.io/component=topolvm-node', namespace='openshift-storage')

Get topolvm node pod in the lvm cluster

Parameters:

• label (str) – Label associated with topolvm node pod

• namespace (str) – Namespace in which topolvm node pod is residing

Returns:

Pod object of topolvm node pod

Return type:

Pod

ocs_ci.ocs.resources.pod.get_used_space_on_mount_point(pod_obj)

Get the used space on a mount point

Parameters:

pod_obj (POD) – The pod object

Returns:

Percentage represent the used space on the mount point

Return type:

int

ocs_ci.ocs.resources.pod.list_ceph_images(pool_name='rbd')

Parameters:

pool_name (str) – Name of the pool to get the ceph images

Returns (List): List of RBD images in the pool

ocs_ci.ocs.resources.pod.list_of_nodes_running_pods(selector, namespace='openshift-storage')

The function returns the list of nodes for the given selector

Parameters:

selector (str) – The resource selector to search with

Returns:

a list of nodes that runs the given selector pods

Return type:

list

ocs_ci.ocs.resources.pod.pod_resource_utilization_raw_output_from_adm_top(namespace='openshift-storage')

Gets the pod’s memory utilization using adm top command.

Parameters:

namespace (str) – The pod’s namespace where the adm top command has to be run

Returns:

Raw output of adm top pods command

Return type:

str

ocs_ci.ocs.resources.pod.restart_pods_having_label(label, namespace='openshift-storage')

Restart the pods having particular label

Parameters:

• label (str) – Label of the pod

• namespace (str) – namespace where the pods are running

ocs_ci.ocs.resources.pod.restart_pods_in_statuses(status_options, namespace='openshift-storage', wait=True)

Restart all the pods in specific statuses

Parameters:

• status_options (list) – The list of the status options.

• namespace (str) – Name of cluster namespace(default: config.ENV_DATA[“cluster_namespace”])

• wait (bool) – Determines if the delete command should wait for completion

Returns:

Restart all the pods that their status in the ‘status_options’ list.

Return type:

list

ocs_ci.ocs.resources.pod.run_io_and_verify_mount_point(pod_obj, bs='10M', count='950')

Run I/O on mount point

Parameters:

• pod_obj (Pod) – The object of the pod

• bs (str) – Read and write up to bytes at a time

• count (str) – Copy only N input blocks

Returns:

Used percentage on mount point

Return type:

used_percentage (str)

ocs_ci.ocs.resources.pod.run_io_in_bg(pod_obj, expect_to_fail=False, fedora_dc=False)

Run I/O in the background

Parameters:

• pod_obj (Pod) – The object of the pod

• expect_to_fail (bool) – True for the command to be expected to fail (disruptive operations), False otherwise

• fedora_dc (bool) – set to False by default. If set to True, it runs IO in background on a fedora dc pod.

Returns:

A thread of the I/O execution

Return type:

Thread

ocs_ci.ocs.resources.pod.run_osd_removal_job(osd_ids=None)

Run the ocs-osd-removal job

Parameters:

osd_ids (list) – The osd IDs.

Returns:

The ocs-osd-removal job object

Return type:

ocs_ci.ocs.resources.ocs.OCS

ocs_ci.ocs.resources.pod.search_pattern_in_pod_logs(pod_name, pattern, namespace='openshift-storage', container=None, all_containers=False)

Searches for the given regular expression pattern in the logs of a pod and returns all matching lines.

Parameters:

• pod_name (str) – The name of the pod.

• pattern (str) – The regular expression pattern to search for.

• namespace (str, optional) – The namespace of the pod. Defaults to None.

• container (str, optional) – The name of the container to search logs for. Defaults to None.

• all_containers (bool, optional) – Whether to search logs for all containers in the pod. Defaults to False.

Returns:

A list of matched lines with the pattern.

ocs_ci.ocs.resources.pod.set_osd_maintenance_mode(osd_deployment)

Set osd in maintenance mode for running ceph-objectstore commands

Parameters:

osd_deployment (OCS) – List of OSD deployment OCS instances

ocs_ci.ocs.resources.pod.upload(pod_name, localpath, remotepath, namespace=None)

Upload a file to pod

Parameters:

• pod_name (str) – Name of the pod

• localpath (str) – Local file to upload

• remotepath (str) – Target path on the pod

ocs_ci.ocs.resources.pod.validate_pods_are_respinned_and_running_state(pod_objs_list)

Verifies the list of the pods are respinned and in running state

Parameters:

pod_objs_list (list) – List of the pods obj

Returns:

True if the pods are respinned and running, False otherwise

Return type:

bool

Raises:

ResourceWrongStatusException – In case the resources hasn’t reached the Running state

ocs_ci.ocs.resources.pod.verify_data_integrity(pod_obj, file_name, original_md5sum, block=False)

Verifies existence and md5sum of file created from first pod

Parameters:

• pod_obj (Pod) – The object of the pod

• file_name (str) – The name of the file for which md5sum to be calculated

• original_md5sum (str) – The original md5sum of the file

• block (bool) – True if the volume mode of PVC used on pod is ‘Block’. file_name will be the devicePath in this case.

Returns:

True if the file exists and md5sum matches

Return type:

bool

Raises:

AssertionError – If file doesn’t exist or md5sum mismatch

ocs_ci.ocs.resources.pod.verify_data_integrity_after_expansion_for_block_pvc(pod_obj, pvc_obj, fio_size)

Verifies data integrity the block PVC obj, by comparing the md5sum of data written using FIO before expansion and after expansion.

Parameters:

• pod_obj (Pod) – POD object for which md5sum of data written during FIO needs to be verified.

• pvc_obj (PVC) – Original PVC object before expansion.

• fio_size (int) – Size in MB of FIO.

Raises:

AssertionError – Raises an exception if current md5sum does not match the original md5sum.

ocs_ci.ocs.resources.pod.verify_data_integrity_for_multi_pvc_objs(pod_objs, pvc_objs, file_name)

Verifies existence and md5sum of file created during IO, for all the pods.

Parameters:

• pod_objs (list) – List of POD objects for which existence and md5sum of file created during IO needs to be verified.

• pvc_objs (list) – List of original PVC objects.

• file_name (str) – The name of the file for which md5sum is to be calculated.

Raises:

AssertionError – Raises an exception if current md5sum does not match the original md5sum.

ocs_ci.ocs.resources.pod.verify_md5sum_on_pod_files(pods_for_integrity_check, pod_file_name)

Verify the md5sum of the pod files

Parameters:

• pods_for_integrity_check (list) – The list of the pod objects to verify the md5sum

• pod_file_name (str) – The pod file name to verify its md5sum

Raises:

AssertionError – If file doesn’t exist or md5sum mismatch

ocs_ci.ocs.resources.pod.verify_node_name(pod_obj, node_name)

Verifies that the pod is running on a particular node

Parameters:

• pod_obj (Pod) – The pod object

• node_name (str) – The name of node to check

Returns:

True if the pod is running on a particular node, False otherwise

Return type:

bool

ocs_ci.ocs.resources.pod.verify_osd_removal_job_completed_successfully(osd_id)

Verify that the ocs-osd-removal job completed successfully

Parameters:

osd_id (str) – The osd id

Returns:

True, if the ocs-osd-removal job completed successfully. False, otherwise

Return type:

bool

ocs_ci.ocs.resources.pod.verify_pods_upgraded(old_images, selector, count=1, timeout=720)

Verify that all pods do not have old image.

Parameters:

• old_images (set) – Set with old images.

• selector (str) – Selector (e.g. app=ocs-osd)

• count (int) – Number of resources for selector.

• timeout (int) – Timeout in seconds to wait for pods to be upgraded.

Raises:

TimeoutException – If the pods didn’t get upgraded till the timeout.

ocs_ci.ocs.resources.pod.wait_for_ceph_cmd_execute_successfully(timeout=300)

Wait for a Ceph command to execute successfully

Parameters:

timeout (int) – The time to wait for a Ceph command to execute successfully

Returns:

True, if the Ceph command executed successfully. False, otherwise

Return type:

bool

ocs_ci.ocs.resources.pod.wait_for_change_in_pods_statuses(pod_names, current_statuses=None, namespace='openshift-storage', timeout=300, sleep=20)

Wait for the pod statuses in a specific namespace to change.

Parameters:

• pod_names (list) – List of the pod names to check if their status changed.

• namespace (str) – the namespace ot the pods

• current_statuses (list) – The current pod statuses. These are the pod statuses to check if they changed during each iteration.

• timeout (int) – time to wait for pod statuses to change

• sleep (int) – Time in seconds to sleep between attempts

Returns:

True, if the pod statuses have changed. False, otherwise

Return type:

bool

ocs_ci.ocs.resources.pod.wait_for_dc_app_pods_to_reach_running_state(dc_pod_obj, timeout=120, exclude_state=None)

Wait for DC app pods to reach running state

Parameters:

• dc_pod_obj (list) – list of dc app pod objects

• timeout (int) – Timeout in seconds to wait for pods to be in Running state.

• exclude_state (str) – A resource state to ignore

ocs_ci.ocs.resources.pod.wait_for_new_osd_pods_to_come_up(number_of_osd_pods_before)

ocs_ci.ocs.resources.pod.wait_for_noobaa_pods_running(timeout=300, sleep=10)

Wait until all the noobaa pods have reached status RUNNING

Parameters:

timeout (int) – Timeout in seconds

ocs_ci.ocs.resources.pod.wait_for_osd_pods_having_ids(osd_ids, timeout=180, sleep=10)

Wait for the osd pods having specific ids

Parameters:

• osd_ids (list) – The list of the osd ids

• timeout (int) – Time to wait for the osd pods having the specified ids

• sleep (int) – Time in seconds to sleep between attempts

Returns:

The osd pods having the specified ids

Return type:

list

Raises:

TimeoutExpiredError – In case it didn’t find all the osd pods with the specified ids

ocs_ci.ocs.resources.pod.wait_for_pods_by_label_count(label, exptected_count, namespace='openshift-storage', timeout=200, sleep=10)

Wait for the expected number of pods with the given selector.

Parameters:

• selector (str) – The resource selector to search with

• exptected_count (int) – The expected number of pods with the given selector

• namespace (str) – the namespace ot the pods

• timeout (int) – time to wait for pods to be running

• sleep (int) – Time in seconds to sleep between attempts

ocs_ci.ocs.resources.pod.wait_for_pods_deletion(label, timeout=120, sleep=5, namespace='openshift-storage')

Wait for the pods with particular label to be deleted until the given timeout

Parameters:

• label (str) – Pod label

• timeout (int) – Timeout

• namespace (str) – Namespace in which pods are running

Raises:

TimeoutExpiredError –

ocs_ci.ocs.resources.pod.wait_for_pods_to_be_in_statuses(expected_statuses, pod_names=None, namespace='openshift-storage', raise_pod_not_found_error=False, exclude_pod_name_prefixes=None, timeout=180, sleep=10)

Wait for the pods in a given namespace to be in the expected statuses

Parameters:

• expected_statuses (list) – The expected statuses of the pods

• pod_names (list) – List of the pod names to check. If not provided, it will check all the pods in the given namespace

• namespace (str) – Name of cluster namespace(default: config.ENV_DATA[“cluster_namespace”])

• raise_pod_not_found_error (bool) – If True, it raises an exception, if one of the pods in the pod names are not found. If False, it ignores the case of pod not found and check the pod objects of the rest of the pod names. The default value is False

• exclude_pod_name_prefixes (list) – The list of the pod name prefixes to exclude from the pods to check

• timeout (int) – time to wait for the pods to be in the expected statuses

• sleep (int) – Time in seconds to sleep between attempts

Returns:

True, if all pods are in the expected statuses. False, otherwise

Return type:

Boolean

ocs_ci.ocs.resources.pod.wait_for_pods_to_be_running(namespace='openshift-storage', pod_names=None, raise_pod_not_found_error=False, timeout=200, sleep=10)

Wait for all the pods in a specific namespace to be running.

Parameters:

• namespace (str) – the namespace ot the pods

• pod_names (list) – List of the pod names to check. If not provided, it will check all the pods in the given namespace

• raise_pod_not_found_error (bool) – If True, it raises an exception(in the function ‘check_pods_in_running_state’), if one of the pods in the pod names are not found. If False, it ignores the case of pod not found and returns the pod objects of the rest of the pod names. The default value is False

• timeout (int) – time to wait for pods to be running

• sleep (int) – Time in seconds to sleep between attempts

Returns:

True, if all pods in Running state. False, otherwise

Return type:

bool

ocs_ci.ocs.resources.pod.wait_for_storage_pods(timeout=200)

Check all OCS pods status, they should be in Running or Completed state

Parameters:

timeout (int) – Number of seconds to wait for pods to get into correct state

ocs_ci.ocs.resources.pv module

ocs_ci.ocs.resources.pv.check_pvs_present_for_ocs_expansion(sc='localblock')

Check for pvs present for OCS cluster expansion

Parameters:

sc (str) – Name of SC

Returns:

True if pv present false if not

Return type:

bool

ocs_ci.ocs.resources.pv.delete_released_pvs_in_sc(sc_name)

Delete the released PVs in a specific Storage class

Parameters:

sc_name (str) – The storage class name

Returns:

The number of PVs that have been deleted successfully.

Return type:

int

ocs_ci.ocs.resources.pv.get_all_pvs(selector=None)

Gets all pv in openshift-storage namespace

Parameters:

selector (str) – The Selector name

Returns:

Dict of all pv in openshift-storage namespace

Return type:

dict

ocs_ci.ocs.resources.pv.get_node_pv_objs(sc_name, node_name)

Get the pv objects that associated to a node in a specific Storage class

Parameters:

• sc_name (str) – The storage class name

• node_name (str) – The node name

Returns:

list of dictionaries of the pv objects that associated to the node name

Return type:

list

ocs_ci.ocs.resources.pv.get_pv_name(pv_obj)

Get the name of the pv object

Parameters:

pv_obj (dict) – A dictionary that represent the pv object

Returns:

The status of the pv object

Return type:

str

ocs_ci.ocs.resources.pv.get_pv_objs_in_sc(sc_name)

Get the pv objects in a specific Storage class

Parameters:

sc_name (str) – The storage class name

Returns:

list of dictionaries of the pv objects

Return type:

list

ocs_ci.ocs.resources.pv.get_pv_size(pv_obj, convert_size=1024)

Get the size of a pv object in GB

Parameters:

• pv_obj (dict) – A dictionary that represent the pv object

• convert_size (int) – set convert by 1024 or 1000

Returns:

The size of the pv object

Return type:

int

ocs_ci.ocs.resources.pv.get_pv_status(pv_obj)

Get the status of the pv object

Parameters:

pv_obj (dict) – A dictionary that represent the pv object

Returns:

The status of the pv object

Return type:

str

ocs_ci.ocs.resources.pv.verify_new_pvs_available_in_sc(old_pv_objs, sc_name, num_of_new_pvs=1, timeout=120)

Verify that the new pv, that has been created in a specific storage class, is available.

Parameters:

• old_pv_objs (list) – List of dictionaries of the pv objects

• sc_name (str) – The name of the storage class

• num_of_new_pvs (int) – Number of the new pvs that should be available in the storage class

• timeout (int) – time to wait for the new pv to come up

Returns:

True if the new pv is available. False, otherwise.

Return type:

bool

ocs_ci.ocs.resources.pvc module

General PVC object

class ocs_ci.ocs.resources.pvc.PVC(**kwargs)

Bases: OCS

A basic PersistentVolumeClaim kind resource

property backed_pv

Returns the backed PV name of pvc_name in namespace

Returns:

PV name

Return type:

str

property backed_pv_obj

Returns the backed PV object of pvc_name in namespace

Returns:

An OCS instance for PV

Return type:

OCS

property backed_sc

Returns the storage class of pvc object in namespace

Returns:

Storage class name

Return type:

str

create_reclaim_space_cronjob(schedule=None)

Create ReclaimSpaceCronJob to invoke reclaim space operation on RBD volume

Returns:

An OCS object representing ReclaimSpaceCronJob

Return type:

ocs_ci.ocs.resources.ocs.OCS

create_reclaim_space_job()

Create ReclaimSpaceJob to invoke reclaim space operation on RBD volume

Returns:

An OCS object representing ReclaimSpaceJob

Return type:

ocs_ci.ocs.resources.ocs.OCS

create_snapshot(snapshot_name=None, wait=False)

Take snapshot of the PVC

Parameters:

• snapshot_name (str) – Name to be provided for snapshot

• wait (bool) – True to wait for snapshot to be ready, False otherwise

Returns:

Kind Snapshot

Return type:

OCS

get_attached_pods()

Get the pods attached to the PVC represented by this object instance

Returns:

A list of pod objects attached to the PVC

Return type:

list

property get_pv_volume_handle_name

Fetch volume handle name from PV

Returns:

volume handle name from pv

Return type:

str

property get_pvc_access_mode

Function to get pvc access_mode

Returns:

The accessModes Value of pvc_obj

Return type:

(str)

property get_pvc_vol_mode

Function to get pvc volume_mode

Returns:

The volumeMode Value of pvc_obj

Return type:

str

property get_rbd_image_name

Fetch image name associated with the RBD PVC

Returns:

Image name associated with the RBD PVC

Return type:

str

property image_uuid

Fetch image uuid associated with PVC

Returns:

Image uuid associated with PVC

Return type:

str

property provisioner

property reclaim_policy

Get the reclaim policy of PV associated with the PVC

Returns:

Reclaim policy. eg: Reclaim, Delete

Return type:

str

resize_pvc(new_size, verify=False)

Modify the capacity of PVC

Parameters:

• new_size (int) – New size of PVC in Gi

• verify (bool) – True to verify the change is reflected on PVC, False otherwise

Returns:

True if operation succeeded, False otherwise

Return type:

bool

property size

Returns the PVC size pvc_name in namespace

Returns:

PVC size

Return type:

int

property status

Returns the PVC status

Returns:

PVC status

Return type:

str

ocs_ci.ocs.resources.pvc.create_pvc_clone(sc_name, parent_pvc, clone_yaml, namespace, pvc_name=None, do_reload=True, storage_size=None, access_mode=None, volume_mode=None)

Create a cloned pvc from existing pvc

Parameters:

• sc_name (str) – The name of storage class.

• parent_pvc (str) – Name of the parent pvc, whose clone is to be created.

• namespace (str) – Namespace of the parent pvc

• pvc_name (str) – The name of the PVC being created

• do_reload (bool) – True for wait for reloading PVC after its creation, False otherwise

• storage_size (str) – Size of the clone, if not passed will use the default “storage” value from pvc-clone.yaml. eg: ‘5Gi’

• access_mode (str) – This decides the access mode to be used for the cloned PVC. eg: ReadWriteOnce, ReadOnlyMany, ReadWriteMany

• volume_mode (str) – Volume mode for PVC. This should match the volume mode of parent PVC

Returns:

PVC instance

Return type:

PVC

ocs_ci.ocs.resources.pvc.create_pvc_snapshot(pvc_name, snap_yaml, snap_name, namespace, sc_name=None, wait=False)

Create snapshot of a PVC

Parameters:

• pvc_name (str) – Name of the PVC

• snap_yaml (str) – The path of snapshot yaml

• snap_name (str) – The name of the snapshot to be created

• namespace (str) – The namespace for the snapshot creation

• sc_name (str) – The name of the snapshot class

• wait (bool) – True to wait for snapshot to be ready, False otherwise

Returns:

OCS object

ocs_ci.ocs.resources.pvc.create_restore_pvc(sc_name, snap_name, namespace, size, pvc_name, volume_mode=None, restore_pvc_yaml='/home/docs/checkouts/readthedocs.org/user_builds/ocs-ci/checkouts/latest/ocs_ci/templates/CSI/rbd/pvc-restore.yaml', access_mode='ReadWriteOnce')

Create PVC from snapshot

Parameters:

• sc_name (str) – The name of the storageclass

• snap_name (str) – The name of the snapshot from which pvc would

• created (be) –

• namespace (str) – The namespace for the PVC creation

• size (str) – Size of pvc being created

• pvc_name (str) – The name of the PVC being created

• volume_mode (str) – Volume mode for rbd RWX pvc i.e. ‘Block’

• restore_pvc_yaml (str) – The location of pvc-restore.yaml

• access_mode (str) – The access mode to be used for the PVC

Returns:

PVC instance

Return type:

PVC

ocs_ci.ocs.resources.pvc.delete_pvcs(pvc_objs, concurrent=False)

Deletes list of the pvc objects

Parameters:

• pvc_objs (list) – List of the pvc objects to be deleted

• concurrent (bool) – Determines if the delete operation should be executed with multiple thread in parallel

Returns:

True if deletion is successful

Return type:

bool

ocs_ci.ocs.resources.pvc.get_all_pvc_objs(namespace=None, selector=None)

Gets all PVCs objects in given namespace

Parameters:

• namespace (str) – Name of namespace

• selector (str) – The label selector to look for

Returns:

Instances of PVC

Return type:

list

ocs_ci.ocs.resources.pvc.get_all_pvcs(namespace=None, selector=None)

Gets all pvc in given namespace

Parameters:

• namespace (str) – Name of namespace (‘all-namespaces’ to get all namespaces)

• selector (str) – The label selector to look for

Returns:

Dict of all pvc in namespaces

Return type:

dict

ocs_ci.ocs.resources.pvc.get_all_pvcs_in_storageclass(storage_class)

This function returen all the PVCs in a given storage class

Parameters:

storage_class (str) – name of the storage class

Returns:

list of PVC objects

Return type:

out

ocs_ci.ocs.resources.pvc.get_deviceset_pvcs()

Get the deviceset PVCs

Returns:

The deviceset PVCs OCS objects

Return type:

list

Raises:

AssertionError – In case the deviceset PVCs are not found

ocs_ci.ocs.resources.pvc.get_deviceset_pvs()

Get the deviceset PVs

Returns:

the deviceset PVs OCS objects

Return type:

list

Raises:

AssertionError – In case the deviceset PVCs are not found

ocs_ci.ocs.resources.pvc.get_pvc_objs(pvc_names, namespace='openshift-storage')

Get the PVC objects of the specified names

Parameters:

• pvc_names (list) – The list of the pvc names to get their objects

• namespace (str) – Name of cluster namespace(default: config.ENV_DATA[“cluster_namespace”])

Returns:

The PVC objects of the specified names

Return type:

list

ocs_ci.ocs.resources.pvc.get_pvc_size(pvc_obj, convert_size=1024)

Returns the PVC size in GB

Parameters:

• pvc_obj (ocs_ci.ocs.resources.ocs.OCS) – The pvc object

• convert_size (int) – set convert by 1024 or 1000

Returns:

PVC size

Return type:

int

ocs_ci.ocs.resources.pvc.scale_down_pods_and_remove_pvcs(sc_name)

Removes the mon and osd pods and also removes PVC’s

Parameters:

sc_name (str) – Name of storage class backing PVC

ocs_ci.ocs.resources.rgw module

class ocs_ci.ocs.resources.rgw.RGW(namespace=None)

Bases: object

Wrapper class for interaction with a cluster’s RGW service

get_credentials(secret_name='rook-ceph-object-user-ocs-storagecluster-cephobjectstore-noobaa-ceph-objectstore-user')

Get Endpoint, Access key and Secret key from OCS secret. Endpoint is taken from rgw exposed service. Use rgw_endpoint fixture in test to get it exposed.

Parameters:

secret_name (str) – Name of secret to be used for getting RGW credentials

Returns:

Endpoint, Access key, Secret key

Return type:

tuple

ocs_ci.ocs.resources.storage_cluster module

StorageCluster related functionalities

class ocs_ci.ocs.resources.storage_cluster.StorageCluster(resource_name='', *args, **kwargs)

Bases: OCP

This class represent StorageCluster and contains all related methods we need to do with StorageCluster.

ocs_ci.ocs.resources.storage_cluster.add_capacity(osd_size_capacity_requested, add_extra_disk_to_existing_worker=True)

Add storage capacity to the cluster

Parameters:

• osd_size_capacity_requested (int) – Requested osd size capacity

• add_extra_disk_to_existing_worker (bool) – Add Disk if True

Returns:

Returns True if all OSDs are in Running state

Return type:

new storage device set count (int)

Note: “StoragedeviceSets->count” represents the set of 3 OSDs. That is, if there are 3 OSDs in the system then count will be 1. If there are 6 OSDs then count is 2 and so on. By changing this value,we can add extra devices to the cluster. For example, if we want to expand the cluster by 3 more osds in a cluster that already has 3 osds, we can set count as 2. So, with each increase of count by 1, we get 3 OSDs extra added to the cluster. This is how we are going to ‘add capacity’ via automation. As we know that OCS has 3 way replica. That is, same data is placed in 3 OSDs. Because of this, the total usable capacity for apps from 3 OSDs will be the size of one OSD (all osds are of same size). If we want to add more capacity to the cluster then we need to add 3 OSDs of same size as that of the original OSD. add_capacity needs to accept the ‘capacity_to_add’ as an argument. From this we need to arrive at storagedeviceSets -> count and then “Patch” this count to get the required capacity to add. To do so, we use following formula: storageDeviceSets->count = (capacity reqested / osd capacity ) + existing count storageDeviceSets

ocs_ci.ocs.resources.storage_cluster.add_capacity_lso(ui_flag=False)

Add capacity on LSO cluster.

In this procedure we need to add the disk before add capacity via UI. Because the UI backend check the pv and available state and base on it change the count param on StorageCluster.

Parameters:

ui_flag (bool) – add capacity via ui [true] or via cli [false]

ocs_ci.ocs.resources.storage_cluster.basic_verification(ocs_registry_image=None)

Basic verification which is needed for Full deployment and MCG only deployment

Parameters:

ocs_registry_image (str) – Specific image to check if it was installed properly.

ocs_ci.ocs.resources.storage_cluster.ceph_config_dump()

Get the ‘ceph config dump’ output.

Returns:

The output of the ‘ceph config dump’ command as a dict.

Return type:

dict

ocs_ci.ocs.resources.storage_cluster.ceph_mon_dump()

Get the ‘ceph mon dump’ output.

Returns:

The output of the ‘ceph mon dump’ command as a dictionary.

Return type:

dict

ocs_ci.ocs.resources.storage_cluster.check_custom_storageclass_presence(interface=None)

Verify if the custom-defined storage class names are present in the oc get sc output.

Returns:

Returns True if all custom-defined storage class names are present in the oc get sc output , otherwise False.

Return type:

bool

ocs_ci.ocs.resources.storage_cluster.get_all_storageclass()

Function for getting all storageclass excluding ‘gp2’ and ‘flex’

Returns:

list of storageclass

Return type:

list

ocs_ci.ocs.resources.storage_cluster.get_ceph_clients()

Get the yamls of all ceph clients. Runs on provider cluster

Returns:

yamls of all ceph clients

Return type:

list

ocs_ci.ocs.resources.storage_cluster.get_consumer_storage_provider_endpoint()

Get the consumer “storageProviderEndpoint” from the ocs storage cluster

Returns:

The consumer “storageProviderEndpoint”

Return type:

str

ocs_ci.ocs.resources.storage_cluster.get_device_class()

Fetches the device class from storage cluster

Returns:

Device class name

Return type:

str

ocs_ci.ocs.resources.storage_cluster.get_deviceset_count()

Get storageDeviceSets count from storagecluster

Returns:

storageDeviceSets count

Return type:

int

ocs_ci.ocs.resources.storage_cluster.get_in_transit_encryption_config_state()

Returns the state of in-transit encryption for the OCS cluster.

Returns:

True if in-transit encryption is enabled, False if it is disabled, or None if an error occurred.

Return type:

bool

ocs_ci.ocs.resources.storage_cluster.get_osd_count()

Get osd count from Storage cluster.

Returns:

osd count (In the case of external mode it returns 0)

Return type:

int

ocs_ci.ocs.resources.storage_cluster.get_osd_replica_count()

Get OSD replication count from storagecluster cr

Returns:

Returns OSD replication count

Return type:

replica_count (int)

ocs_ci.ocs.resources.storage_cluster.get_osd_size()

Get osd size from Storage cluster

Returns:

osd size

Return type:

int

ocs_ci.ocs.resources.storage_cluster.get_rook_ceph_mon_per_endpoint_ip()

Get a dictionary of the rook ceph mon per endpoint ip

Returns:

A dictionary of the rook ceph mon per endpoint ip

Return type:

dict

ocs_ci.ocs.resources.storage_cluster.get_storage_cluster(namespace=None)

Get storage cluster name

Parameters:

namespace (str) – Namespace of the resource

Returns:

Storage cluster object handler

Return type:

storage cluster (obj)

ocs_ci.ocs.resources.storage_cluster.get_storage_cluster_state(sc_name, namespace=None)

Get the storage cluster state

Parameters:

• sc_name (str) – The storage cluster name

• namespace (str) – Namespace of the resource. The default value is: ‘config.ENV_DATA[“cluster_namespace”]’ if None provided

Returns:

The storage cluster state

Return type:

str

ocs_ci.ocs.resources.storage_cluster.get_storage_size()

Get the storagecluster storage size

Returns:

The storagecluster storage size

Return type:

str

ocs_ci.ocs.resources.storage_cluster.get_storageclass_names_from_storagecluster_spec()

Retrieve storage class names from the storage cluster’s spec.

This function queries the storage cluster’s specification and returns a dictionary containing the storage class names for various resources, such as cephFilesystems, cephObjectStores, cephBlockPools, cephNonResilientPools, nfs, and encryption.

Returns:

A dictionary containing the storage class names for various resources.

The keys are the names of the resources, and the values are the respective storage class names. If a resource does not have a storage class name, it will be set to None.

Return type:

dict

ocs_ci.ocs.resources.storage_cluster.in_transit_encryption_verification()

Verifies in-transit encryption is enabled and ceph mons are configured with ‘v2’ protocol version.

Raises:

ValueError – if in-transit encryption is not configured or ceph mon protocol is not configured with ‘v2’ version.

ocs_ci.ocs.resources.storage_cluster.mcg_only_install_verification(ocs_registry_image=None)

Verification for successful MCG only deployment

Parameters:

ocs_registry_image (str) – Specific image to check if it was installed properly.

ocs_ci.ocs.resources.storage_cluster.ocs_install_verification(timeout=600, skip_osd_distribution_check=False, ocs_registry_image=None, post_upgrade_verification=False, version_before_upgrade=None)

Perform steps necessary to verify a successful OCS installation

Parameters:

• timeout (int) – Number of seconds for timeout which will be used in the checks used in this function.

• skip_osd_distribution_check (bool) – If true skip the check for osd distribution.

• ocs_registry_image (str) – Specific image to check if it was installed properly.

• post_upgrade_verification (bool) – Set to True if this function is called after upgrade.

• version_before_upgrade (float) – Set to OCS version before upgrade

ocs_ci.ocs.resources.storage_cluster.osd_encryption_verification()

Verify if OSD encryption at rest if successfully deployed on OCS

Raises:

• UnsupportedFeatureError – OCS version is smaller than 4.6

• EnvironmentError – The OSD is not encrypted

ocs_ci.ocs.resources.storage_cluster.patch_storage_cluster_for_custom_storage_class(storage_class_type, storage_class_name=None, action='add')

Patch the storage cluster for a custom storage class.

This function updates the storage cluster’s storage class settings based on the provided storage class type.

Parameters:

• storage_class_type (str) – The type of storage class (“nfs”, “encryption”, etc.).

• storage_class_name (str, optional) – The name of the custom storage class to be set. If None, a default name will be generated.

• action (str, optional) – The action to perform (“add” or “remove”).

Returns:

Result of the patch operation.

Return type:

bool

ocs_ci.ocs.resources.storage_cluster.resize_osd(new_osd_size, check_size=True)

Resize the OSD(e.g., from 512 to 1024, 1024 to 2048, etc.)

Parameters:

• new_osd_size (str) – The new osd size(e.g, 512Gi, 1024Gi, 1Ti, 2Ti, etc.)

• check_size (bool) – Check that the given osd size is valid

Returns:

True in case if changes are applied. False otherwise

Return type:

bool

Raises:

ValueError – In case the osd size is not valid(start with digits and follow by string) or the new osd size is less than the current osd size

ocs_ci.ocs.resources.storage_cluster.set_deviceset_count(count)

Set osd count for Storage cluster.

Parameters:

count (int) – the count param is storagecluster

ocs_ci.ocs.resources.storage_cluster.set_in_transit_encryption(enabled=True)

Enable or disable in-transit encryption for the default storage cluster.

Parameters:

enabled (bool, optional) – A boolean indicating whether to enable or disable in-transit encryption. Defaults to True, i.e., enabling in-transit encryption.

Returns:

True if in-transit encryption was successfully enabled or disabled, False otherwise.

Return type:

bool

ocs_ci.ocs.resources.storage_cluster.setup_ceph_debug()

Set Ceph to run in debug log level using a ConfigMap. This functionality is available starting OCS 4.7.

ocs_ci.ocs.resources.storage_cluster.validate_serviceexport()

validate the serviceexport resource Number of osds and mons should match

ocs_ci.ocs.resources.storage_cluster.verify_backing_store()

Verify backingstore

ocs_ci.ocs.resources.storage_cluster.verify_consumer_resources()

Verify resources specific to managed OCS consumer: 1. MGR endpoint 2. monitoring endpoint in cephcluster yaml 3. Verify the default Storageclassclaims

ocs_ci.ocs.resources.storage_cluster.verify_consumer_storagecluster(sc_data)

Verify that Storagecluster is has: 1. externalStorage: enable: true 2. storageProviderEndpoint: IP:31659 3. onboardingTicket is present 4. catsrc existence 5. requested capacity matches granted capacity 6. requested and granted capacity fields have a valid value

Args: sc_data (dict): storagecluster data dictionary

ocs_ci.ocs.resources.storage_cluster.verify_device_class_in_osd_tree(ct_pod, device_class)

Verifies device class in ceph osd tree output

Parameters:

• ct_pod (OCP) – Object of the Ceph tools pod

• device_class (str) – Name of the device class

ocs_ci.ocs.resources.storage_cluster.verify_kms_ca_only()

Verify KMS deployment with only CA Certificate without Client Certificate and without Client Private Key

ocs_ci.ocs.resources.storage_cluster.verify_managed_secrets()

Verify that ocs-converged-pagerduty, ocs-converged-smtp, ocs-converged-deadmanssnitch, addon-ocs-provider-parameters, alertmanager-managed-ocs-alertmanager-generated, rook-ceph-mon secrets exist in openshift-storage namespace. For a provider cluster verify existence of onboarding-ticket-key and ocs-provider-server secrets. For a consumer cluster verify existence of 5 rook-ceph-client secrets

ocs_ci.ocs.resources.storage_cluster.verify_managed_service_networkpolicy()

Verify Networkpolicy and EgressNetworkpolicy creation Temporarily left out for V2 offering

ocs_ci.ocs.resources.storage_cluster.verify_managed_service_resources()

Verify creation and status of resources specific to OSD and ROSA deployments: 1. ocs-operator, ose-prometheus-operator csvs are Succeeded 2. 1 prometheus and 1 alertmanager pods are in Running state 3. Managedocs components alertmanager, prometheus, storageCluster are in Ready state 4. Verify that noobaa-operator replicas is set to 0 5. Verify managed ocs secrets 6. If cluster is Provider, verify resources specific to provider clusters 7. Verify that version of Prometheus is 4.10 8. Verify security restrictions are in place 9. [temporarily left out] Verify Networkpolicy and EgressNetworkpolicy creation

ocs_ci.ocs.resources.storage_cluster.verify_managedocs_security()

Check ocs-osd-deployer-operator permissions: 1. Verify runAsUser is not 0 2. Verify SecurityContext.allowPrivilegeEscalation is set to false 3. Verify SecurityContext.capabilities.drop contains ALL

ocs_ci.ocs.resources.storage_cluster.verify_max_openshift_version()

Verify the maximum OpenShift version supported for ODF

ocs_ci.ocs.resources.storage_cluster.verify_mcg_only_pods()

Verify pods in MCG Only deployment

ocs_ci.ocs.resources.storage_cluster.verify_multus_network()

Verify Multus network(s) created successfully and are present on relevant pods.

ocs_ci.ocs.resources.storage_cluster.verify_networks_in_ceph_pod(pod_networks, net_name, net_namespace)

Verify network configuration on ceph pod

Parameters:

• pod_networks (str) – the value of k8s.v1.cni.cncf.io/networks param

• net_name (str) – the network-attachment-definitions name

• net_namespace (str) – the network-attachment-definitions namespace

Returns:

return True if net_name and net_namespce exist in pod_networks otherwise False

Return type:

bool

ocs_ci.ocs.resources.storage_cluster.verify_noobaa_endpoint_count()

Verify noobaa endpoints

ocs_ci.ocs.resources.storage_cluster.verify_ocs_csv(ocs_registry_image=None)

OCS CSV verification ( succeeded state )

Parameters:

ocs_registry_image (str) – Specific image to check if it was installed properly.

ocs_ci.ocs.resources.storage_cluster.verify_osd_tree_schema(ct_pod, deviceset_pvcs)

Verify Ceph OSD tree schema

Parameters:

• ct_pod (OCP) – Object of the Ceph tools pod

• deviceset_pvcs (list) – List of strings of deviceset PVC names

ocs_ci.ocs.resources.storage_cluster.verify_provider_resources()

Verify resources specific to managed OCS provider: 1. Ocs-provider-server pod is Running 2. cephcluster is Ready and its hostNetworking is set to True 3. Security groups are set up correctly

ocs_ci.ocs.resources.storage_cluster.verify_provider_storagecluster(sc_data)

Verify that storagecluster of the provider passes the following checks: 1. allowRemoteStorageConsumers: true 2. hostNetwork: true 3. matchExpressions: key: node-role.kubernetes.io/worker operator: Exists key: node-role.kubernetes.io/infra operator: DoesNotExist 4. storageProviderEndpoint: IP:31659 5. annotations: uninstall.ocs.openshift.io/cleanup-policy: delete uninstall.ocs.openshift.io/mode: graceful

Parameters:

sc_data (dict) – storagecluster data dictionary

ocs_ci.ocs.resources.storage_cluster.verify_sc_images(storage_cluster)

Verifying images in storage cluster such as ceph, noobaaDB and noobaaCore

Parameters:

storage_cluster (obj) – storage_cluster ocp object

ocs_ci.ocs.resources.storage_cluster.verify_storage_cluster()

Verify storage cluster status

ocs_ci.ocs.resources.storage_cluster.verify_storage_cluster_images()

Verify images in storage cluster

ocs_ci.ocs.resources.storage_cluster.verify_storage_cluster_version(storage_cluster)

Verifies the storage cluster version

Parameters:

storage_cluster (obj) – storage cluster object

ocs_ci.ocs.resources.storage_cluster.verify_storage_device_class(device_class)

Verifies the parameters of storageClassDeviceSets in CephCluster.

For internal deployments, if user is not specified any DeviceClass in the StorageDeviceSet, then tunefastDeviceClass will be true and crushDeviceClass will set to “ssd”

Parameters:

device_class (str) – Name of the device class

ocs_ci.ocs.resources.storage_cluster.verify_storage_system()

Verify storage system status

ocs_ci.ocs.resources.storage_cluster.wait_for_consumer_rook_ceph_mon_endpoints_in_provider_wnodes(timeout=180, sleep=10)

Wait for the rook ceph mon endpoint ips to be found in the provider worker node ips

Parameters:

• timeout (int) – The time to wait for the rook ceph mon endpoint ips to be found in the provider worker node ips

• sleep (int) – Time in seconds to sleep between attempts

Returns:

True, If all the rook ceph mon endpoint ips are found in the

provider worker nodes. False, otherwise.

Return type:

bool

ocs_ci.ocs.resources.storage_cluster.wait_for_consumer_storage_provider_endpoint_in_provider_wnodes(timeout=180, sleep=10)

Wait for the consumer “storageProviderEndpoint” ip to be found in the provider worker node ips

Parameters:

• timeout (int) – timeout in seconds to wait for the consumer “storageProviderEndpoint” ip to be found in the provider worker node ips

• sleep (int) – Time in seconds to sleep between attempts

Returns:

True, if the consumer “storageProviderEndpoint” ip is found in the

provider worker node ips. False, otherwise.

ocs_ci.ocs.resources.storageclassclaim module

StorageClassClaim related functionalities

class ocs_ci.ocs.resources.storageclassclaim.StorageClassClaim(**kwargs)

Bases: OCS

StorageClassClaim kind resource

property status

Returns the storageclassclaim status

Returns:

Storageclassclaim status

Return type:

str

property storageclassclaim_type

Returns the type of the storageclassclaim

Returns:

Storageclassclaim type

Return type:

str

ocs_ci.ocs.resources.storageclassclaim.create_storageclassclaim(interface_type, storage_class_claim_name=None, namespace=None, storageclient_name=None, storageclient_namespace=None)

Create a storageclassclaim

Parameters:

• interface_type (str) – The type of the interface (e.g. CephBlockPool, CephFileSystem)

• storage_class_claim_name (str) – The name of storageclassclaim to create

• namespace (str) – The namespace in which the storageclassclaim should be created

• storageclient_name (str) – Name of the storageclient

• storageclient_namespace (str) – Namespace of storageclient

Returns:

An OCS instance for the storageclassclaim

Return type:

OCS

ocs_ci.ocs.resources.storageconsumer module

A module for all StorageConsumer functionalities and abstractions.

class ocs_ci.ocs.resources.storageconsumer.StorageConsumer(consumer_name, consumer_context=None)

Bases: object

Base StorageConsumer class

get_heartbeat_cronjob()

Returns:

status reporter cronjob OCS object

Return type:

object

get_ocs_version()

Get ocs version from storageconsumer resource.

Returns:

consumer ocs version

Return type:

string

resume_heartbeat()

Resume status reporter cron job.

set_ocs_version(version)

Update ocs consumer version in storageconsumer resource. This change assumes that the hearthbeat is stopped so that the version is not overwritten by it.

Parameters:

version (str) – OCS version to be set

stop_heartbeat()

Suspend status reporter cron job.

ocs_ci.ocs.resources.stretchcluster module

class ocs_ci.ocs.resources.stretchcluster.StretchCluster(**kwargs)

Bases: OCS

A basic StrethCluster class to objectify stretch cluster related operations, methods and properties

cephfs_failure_checks(start_time, end_time, wait_for_read_completion=True)

Checks cephFs workloads for write or read pause between start_time and end_time

Parameters:

• start_time (datetime) – Start time of the failure

• end_time (datetime) – End time of the failure

• wait_for_read_completion (bool) – True if needs to be waited for the read operation to complete else False

property cephfs_log_file_map

property cephfs_logreader_pods

property cephfs_logwriter_pods

property cephfs_old_log

check_ceph_accessibility(timeout, delay=5, grace=15)

Check for ceph access for the ‘timeout’ seconds

Parameters:

• timeout (int) – timeout in seconds

• delay (int) – how often ceph access should be checked in seconds

• grace (int) – grace time to wait for the ceph to respond in seconds

Returns:

True of no ceph accessibility issues else False

Return type:

Bool

check_for_data_corruption(label, namespace='sc-project')

Check for data corruption

Parameters:

• label (str) – Label for workload type

• namespace (str) – namespace

Returns:

True if no data corruption else False

Return type:

Bool

check_for_data_loss(label)

Check for data loss

Parameters:

label (str) – Label for workload type

Returns:

True if no data loss else False

Return type:

Bool

check_for_read_pause(label, start_time, end_time)

This checks for any read pause has occurred during the given window of start_time and end_time

Parameters:

• label (str) – label for the workload (RBD or CephFS)

• start_time (datetime) – datetime object representing the start time

• end_time (datetime) – datetime object representing the end time

Returns:

number of logreader instances has seen read pause

Return type:

Int

check_for_write_pause(label, start_time, end_time)

Checks for write pause between start time and end time

Parameters:

• label (str) – Label for the workload

• start_time (datetime) – datetime object representing the start time

• end_time (datetime) – datetime object representing the end time

Returns:

number of instances has seen write pause

Return type:

Int

get_logfile_map(label)

Update map of logfiles created by each workload types

Parameters:

label (str) – Label for the workload

get_logwriter_reader_pods(label, exp_num_replicas=None, statuses=None, namespace='sc-project')

Update logwriter and reader pods for the mentioned workload type

Parameters:

• label (str) – Label for the workload type

• exp_num_replicas (int) – Expected number of replicas

• statuses (List) – List of statuses that is expected

• namespace (str) – namespace

get_nodes_in_zone(zone)

This will return the list containing OCS objects represeting the nodes having mentioned label

Parameters:

• zone (str) – string represeting zone that node

• to (belongs) –

Returns:

Node (OCS) objects

Return type:

List

get_ocs_nodes_in_zone(zone)

Get the OCS nodes in a particular zone

Parameters:

zone (str) – Zone that node belongs to

Returns:

Node(OCS) objects

Return type:

List

get_out_of_quorum_nodes()

Get the zone nodes where the mon’s are not in quorum

Returns:

List of non-quorum node names

get_workload_pvc_obj(workload_label)

Gets the PVC object for the volume attached to the workload type mentioned by label

Parameters:

workload_label (str) – Label for the workload

Returns:

PVC object

post_failure_checks(start_time, end_time, types=['rbd', 'cephfs'], wait_for_read_completion=True)

Post failure checks that will check for any failure during start_time and end_time

Parameters:

• start_time (datetime) – Start time of the failure

• end_time (datetime) – End time of the failure

• types (list) – List containing workload types, default., [“rbd”, “cephfs”]

• wait_for_read_completion (bool) – True if needs to be waited for the read operation to complete else False

rbd_failure_checks(start_time, end_time, **kwargs)

Checks RBD workloads for write or read pause between start_time and end_time

Parameters:

• start_time (datetime) – Start time of the failure

• end_time (datetime) – End time of the failure

property rbd_log_file_map

property rbd_logwriter_pods

property rbd_old_log

reset_conn_score()

Reset connection scores for all the mon’s

validate_conn_score(conn_score_map, quorum_ranks)

Validate connection score of each mons from the connection score map

Parameters:

• conn_score_map (dict) – Dict map representing connection score for each mons

• quorum_ranks (list) – Expected mon quorum ranks at the moment

ocs_ci.ocs.resources.test_packagemanifest module

ocs_ci.ocs.resources.test_packagemanifest.test_no_resource_found_for_packagemanifest()

Test that when we run into issue #1338, when no PackageManifest object found.

This unit test serves two purposes: - to show what exactly happens to PackageManifest during issue #1338 - demonstrate that PackageManifest API remains unchanged

ocs_ci.ocs.resources.test_packagemanifest.test_pm_null()

Test that creation of PackageManifest object without any constructor agruments works (object is created, no exceptions are raised).

ocs_ci.ocs.resources.test_packagemanifest.test_pm_null_get_default_channel()

ocs_ci.ocs.resources.topology module

ocs_ci.ocs.resources.topology.drop_topology_constraint(spec_dict, topology_key)

Removes constraints for given topology key from given dict spec with topologySpreadConstraints. If the constraint is not present, the spec dict is not changed.

Parameters:

• tsc_dict (dict) – topologySpreadConstraints spec

• topology_key (string) – name of the topology key

Module contents