Metering
Configuring and using Metering in OpenShift Container Platform
Abstract
Chapter 1. About Metering
1.1. Metering overview
Metering is a general purpose data analysis tool that enables you to write reports to process data from different data sources. As a cluster administrator, you can use metering to analyze what is happening in your cluster. You can either write your own, or use predefined SQL queries to define how you want to process data from the different data sources you have available.
Metering focuses primarily on in-cluster metric data using Prometheus as a default data source, enabling users of metering to do reporting on pods, namespaces, and most other Kubernetes resources.
You can install metering on OpenShift Container Platform 4.x clusters and above.
1.1.1. Metering resources
Metering has many resources which can be used to manage the deployment and installation of metering, as well as the reporting functionality metering provides.
Metering is managed using the following CustomResourceDefinitions (CRDs):
MeteringConfig | Configures the metering stack for deployment. Contains customizations and configuration options to control each component that makes up the metering stack. |
Reports | Controls what query to use, when, and how often the query should be run, and where to store the results. |
ReportQueries | Contains the SQL queries used to perform analysis on the data contained within ReportDataSources. |
ReportDataSources | Controls the data available to ReportQueries and Reports. Allows configuring access to different databases for use within metering. |
Chapter 2. Installing metering
Review the following sections before installing metering into your cluster.
To get started installing metering, first install the Metering Operator from OperatorHub. Next, configure your instance of metering by creating a CustomResource
, referred to here as your MeteringConfig. Installing the Metering Operator creates a default MeteringConfig that you can modify using the examples in the documentation. After creating your MeteringConfig, install the metering stack. Last, verify your installation.
2.1. Prerequisites
Metering requires the following components:
- A StorageClass for dynamic volume provisioning. Metering supports a number of different storage solutions.
- 4GB memory and 4 CPU cores available cluster capacity and at least one node with 2 CPU cores and 2GB memory capacity available.
The minimum resources needed for the largest single Pod installed by metering are 2GB of memory and 2 CPU cores.
- Memory and CPU consumption may often be lower, but will spike when running reports, or collecting data for larger clusters.
2.2. Installing the Metering Operator
You can install metering by deploying the Metering Operator. The Metering Operator creates and manages the components of the metering stack.
You cannot create a Project starting with openshift-
using the web console or by using the oc new-project
command in the CLI.
2.2.1. Installing metering using the web console
You can use the OpenShift Container Platform web console to install the Metering Operator.
Procedure
Create a namespace object YAML file for the Metering Operator with the
oc create -f <file-name>.yaml
command. You must use the CLI to create the namespace. For example,metering-namespace.yaml
:apiVersion: v1 kind: Namespace metadata: name: openshift-metering 1 annotations: openshift.io/node-selector: "" 2 labels: openshift.io/cluster-monitoring: "true"
-
In the OpenShift Container Platform web console, click Operators → OperatorHub. Filter for
metering
to find the Metering Operator. - Click the Metering card, review the package description, and then click Install.
- Select an Update Channel, Installation Mode, and Approval Strategy.
- Click Subscribe.
Verify that the Metering Operator is installed by switching to the Operators → Installed Operators page. The Metering Operator has a Status of Succeeded when the installation is complete.
NoteIt might take several minutes for the Metering Operator to appear.
- Click Metering on the Installed Operators page for Operator Details. From the Details page you can create different resources related to metering.
To complete the metering installation, create a MeteringConfig resource to configure metering and install the components of the metering stack.
2.2.2. Installing metering using the CLI
You can use the OpenShift Container Platform CLI to install the Metering Operator.
Procedure
Create a namespace object YAML file for the Metering Operator. You must use the CLI to create the namespace. For example,
metering-namespace.yaml
:apiVersion: v1 kind: Namespace metadata: name: openshift-metering 1 annotations: openshift.io/node-selector: "" 2 labels: openshift.io/cluster-monitoring: "true"
Create the namespace object:
$ oc create -f <file-name>.yaml
For example:
$ oc create -f openshift-metering.yaml
Create the OperatorGroup object YAML file. For example,
metering-og
:apiVersion: operators.coreos.com/v1 kind: OperatorGroup metadata: name: openshift-metering 1 namespace: openshift-metering 2 spec: targetNamespaces: - openshift-metering
Create a Subscription object YAML file to subscribe a namespace to the Metering Operator. This object targets the most recently released version in the
redhat-operators
CatalogSource. For example,metering-sub.yaml
:apiVersion: operators.coreos.com/v1alpha1 kind: Subscription metadata: name: metering-ocp 1 namespace: openshift-metering 2 spec: channel: "4.3" 3 source: "redhat-operators" 4 sourceNamespace: "openshift-marketplace" name: "metering-ocp" installPlanApproval: "Automatic" 5
- 1
- The name is arbitrary.
- 2
- You must specify the
openshift-metering
namespace. - 3
- Specify 4.3 as the channel.
- 4
- Specify the
redhat-operators
CatalogSource, which contains themetering-ocp
package manifests. If your OpenShift Container Platform is installed on a restricted network, also known as a disconnected cluster, specify the name of the CatalogSource object you created when you configured the Operator LifeCycle Manager (OLM). - 5
- Specify "Automatic" install plan approval.
2.3. Installing the metering stack
After adding the Metering Operator to your cluster you can install the components of metering by installing the metering stack.
2.4. Prerequisites
- Review the configuration options
Create a MeteringConfig resource. You can begin the following process to generate a default MeteringConfig, then use the examples in the documentation to modify this default file for your specific installation. Review the following topics to create your MeteringConfig resource:
- For configuration options, review About configuring metering.
- At a minimum, you need to configure persistent storage and configure the Hive metastore.
There can only be one MeteringConfig resource in the openshift-metering
namespace. Any other configuration is not supported.
Procedure
-
From the web console, ensure you are on the Operator Details page for the Metering Operator in the
openshift-metering
project. You can navigate to this page by clicking Operators → Installed Operators, then selecting the Metering Operator. Under Provided APIs, click Create Instance on the Metering Configuration card. This opens a YAML editor with the default MeteringConfig file where you can define your configuration.
NoteFor example configuration files and all supported configuration options, review the configuring metering documentation.
- Enter your MeteringConfig into the YAML editor and click Create.
The MeteringConfig resource begins to create the necessary resources for your metering stack. You can now move on to verifying your installation.
2.5. Verifying the metering installation
You can verify the metering installation by performing any of the following checks:
Check the Metering Operator ClusterServiceVersion (CSV) for the metering version. This can be done through either the web console or CLI.
Procedure (UI)
-
Navigate to Operators → Installed Operators in the
openshift-metering
namespace. - Click Metering Operator.
- Click Subscription for Subscription Details.
- Check the Installed Version.
Procedure (CLI)
Check the Metering Operator CSV in the
openshift-metering
namespace:$ oc --namespace openshift-metering get csv
In the following example, the 4.3 Metering Operator installation is successful:
NAME DISPLAY VERSION REPLACES PHASE elasticsearch-operator.4.3.0-202006231303.p0 Elasticsearch Operator 4.3.0-202006231303.p0 Succeeded metering-operator.v4.3.0 Metering 4.3.0 Succeeded
-
Navigate to Operators → Installed Operators in the
Check that all required Pods in the
openshift-metering
namespace are created. This can be done through either the web console or CLI.NoteMany Pods rely on other components to function before they themselves can be considered ready. Some Pods may restart if other Pods take too long to start. This is to be expected during the Metering Operator installation.
Procedure (UI)
- Navigate to Workloads → Pods in the metering namespace and verify that Pods are being created. This can take several minutes after installing the metering stack.
Procedure (CLI)
Check that all required Pods in the
openshift-metering
namespace are created:$ oc -n openshift-metering get pods
The output shows that all Pods are created in the
Ready
column:NAME READY STATUS RESTARTS AGE hive-metastore-0 2/2 Running 0 3m28s hive-server-0 3/3 Running 0 3m28s metering-operator-68dd64cfb6-2k7d9 2/2 Running 0 5m17s presto-coordinator-0 2/2 Running 0 3m9s reporting-operator-5588964bf8-x2tkn 2/2 Running 0 2m40s
Verify that the
ReportDataSources
are beginning to import data, indicated by a valid timestamp in theEARLIEST METRIC
column. This might take several minutes. Filter out the "-raw"ReportDataSources
, which do not import data:$ oc get reportdatasources -n openshift-metering | grep -v raw
$ oc get reportdatasources -n openshift-metering | grep -v raw NAME EARLIEST METRIC NEWEST METRIC IMPORT START IMPORT END LAST IMPORT TIME AGE node-allocatable-cpu-cores 2019-08-05T16:52:00Z 2019-08-05T18:52:00Z 2019-08-05T16:52:00Z 2019-08-05T18:52:00Z 2019-08-05T18:54:45Z 9m50s node-allocatable-memory-bytes 2019-08-05T16:51:00Z 2019-08-05T18:51:00Z 2019-08-05T16:51:00Z 2019-08-05T18:51:00Z 2019-08-05T18:54:45Z 9m50s node-capacity-cpu-cores 2019-08-05T16:51:00Z 2019-08-05T18:29:00Z 2019-08-05T16:51:00Z 2019-08-05T18:29:00Z 2019-08-05T18:54:39Z 9m50s node-capacity-memory-bytes 2019-08-05T16:52:00Z 2019-08-05T18:41:00Z 2019-08-05T16:52:00Z 2019-08-05T18:41:00Z 2019-08-05T18:54:44Z 9m50s persistentvolumeclaim-capacity-bytes 2019-08-05T16:51:00Z 2019-08-05T18:29:00Z 2019-08-05T16:51:00Z 2019-08-05T18:29:00Z 2019-08-05T18:54:43Z 9m50s persistentvolumeclaim-phase 2019-08-05T16:51:00Z 2019-08-05T18:29:00Z 2019-08-05T16:51:00Z 2019-08-05T18:29:00Z 2019-08-05T18:54:28Z 9m50s persistentvolumeclaim-request-bytes 2019-08-05T16:52:00Z 2019-08-05T18:30:00Z 2019-08-05T16:52:00Z 2019-08-05T18:30:00Z 2019-08-05T18:54:34Z 9m50s persistentvolumeclaim-usage-bytes 2019-08-05T16:52:00Z 2019-08-05T18:30:00Z 2019-08-05T16:52:00Z 2019-08-05T18:30:00Z 2019-08-05T18:54:36Z 9m49s pod-limit-cpu-cores 2019-08-05T16:52:00Z 2019-08-05T18:30:00Z 2019-08-05T16:52:00Z 2019-08-05T18:30:00Z 2019-08-05T18:54:26Z 9m49s pod-limit-memory-bytes 2019-08-05T16:51:00Z 2019-08-05T18:40:00Z 2019-08-05T16:51:00Z 2019-08-05T18:40:00Z 2019-08-05T18:54:30Z 9m49s pod-persistentvolumeclaim-request-info 2019-08-05T16:51:00Z 2019-08-05T18:40:00Z 2019-08-05T16:51:00Z 2019-08-05T18:40:00Z 2019-08-05T18:54:37Z 9m49s pod-request-cpu-cores 2019-08-05T16:51:00Z 2019-08-05T18:18:00Z 2019-08-05T16:51:00Z 2019-08-05T18:18:00Z 2019-08-05T18:54:24Z 9m49s pod-request-memory-bytes 2019-08-05T16:52:00Z 2019-08-05T18:08:00Z 2019-08-05T16:52:00Z 2019-08-05T18:08:00Z 2019-08-05T18:54:32Z 9m49s pod-usage-cpu-cores 2019-08-05T16:52:00Z 2019-08-05T17:57:00Z 2019-08-05T16:52:00Z 2019-08-05T17:57:00Z 2019-08-05T18:54:10Z 9m49s pod-usage-memory-bytes 2019-08-05T16:52:00Z 2019-08-05T18:08:00Z 2019-08-05T16:52:00Z 2019-08-05T18:08:00Z 2019-08-05T18:54:20Z 9m49s
After all Pods are ready and you have verified that data is being imported, you can begin using metering to collect data and report on your cluster.
2.6. Additional resources
- For more information on configuration steps and available storage platforms, see Configuring persistent storage.
- For the steps to configure Hive, see Configuring the Hive metastore.
Chapter 3. Configuring metering
3.1. About configuring metering
A CustomResource
called your MeteringConfig
specifies all the configuration details for your metering installation. When you first install the metering stack, a default MeteringConfig
is generated. Use the examples in the documentation to modify this default file. Keep in mind the following key points:
- At a minimum, you need to configure persistent storage and configure the Hive metastore.
- Most default configuration settings work, but larger deployments or highly customized deployments should review all configuration options carefully.
- Some configuration options can not be modified after installation.
For configuration options that can be modified after installation, make the changes in your MeteringConfig
and reapply the file.
3.2. Common configuration options
3.2.1. Resource requests and limits
You can adjust the CPU, memory, or storage resource requests and/or limits for pods and volumes. The default-resource-limits.yaml
below provides an example of setting resource request and limits for each component.
apiVersion: metering.openshift.io/v1 kind: MeteringConfig metadata: name: "operator-metering" spec: reporting-operator: spec: resources: limits: cpu: 1 memory: 500Mi requests: cpu: 500m memory: 100Mi presto: spec: coordinator: resources: limits: cpu: 4 memory: 4Gi requests: cpu: 2 memory: 2Gi worker: replicas: 0 resources: limits: cpu: 8 memory: 8Gi requests: cpu: 4 memory: 2Gi hive: spec: metastore: resources: limits: cpu: 4 memory: 2Gi requests: cpu: 500m memory: 650Mi storage: class: null create: true size: 5Gi server: resources: limits: cpu: 1 memory: 1Gi requests: cpu: 500m memory: 500Mi
3.2.2. Node selectors
You can run the metering components on specific sets of nodes. Set the nodeSelector
on a metering component to control where the component is scheduled. The node-selectors.yaml
file below provides an example of setting node selectors for each component.
Add the openshift.io/node-selector: ""
namespace annotation to the metering namespace YAML file before configuring specific node selectors for the operand Pods. Specify ""
as the annotation value.
apiVersion: metering.openshift.io/v1 kind: MeteringConfig metadata: name: "operator-metering" spec: reporting-operator: spec: nodeSelector: "node-role.kubernetes.io/infra": "" 1 presto: spec: coordinator: nodeSelector: "node-role.kubernetes.io/infra": "" 2 worker: nodeSelector: "node-role.kubernetes.io/infra": "" 3 hive: spec: metastore: nodeSelector: "node-role.kubernetes.io/infra": "" 4 server: nodeSelector: "node-role.kubernetes.io/infra": "" 5
Add the openshift.io/node-selector: ""
namespace annotation to the metering namespace YAML file before configuring specific node selectors for the operand Pods. When the openshift.io/node-selector
annotation is set on the project, the value is used in preference to the value of the spec.defaultNodeSelector
field in the cluster-wide Scheduler object.
Verification
You can verify the metering node selectors by performing any of the following checks:
Verify that all Pods for metering are correctly scheduled on the IP of the node that is configured in the MeteringConfig custom resource:
Procedure
Check all pods in the
openshift-metering
namespace:$ oc --namespace openshift-metering get pods -o wide
The output shows the
NODE
and correspondingIP
for each Pod running in theopenshift-metering
namespace:NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES hive-metastore-0 1/2 Running 0 4m33s 10.129.2.26 ip-10-0-210-167.us-east-2.compute.internal <none> <none> hive-server-0 2/3 Running 0 4m21s 10.128.2.26 ip-10-0-150-175.us-east-2.compute.internal <none> <none> metering-operator-964b4fb55-4p699 2/2 Running 0 7h30m 10.131.0.33 ip-10-0-189-6.us-east-2.compute.internal <none> <none> nfs-server 1/1 Running 0 7h30m 10.129.2.24 ip-10-0-210-167.us-east-2.compute.internal <none> <none> presto-coordinator-0 2/2 Running 0 4m8s 10.131.0.35 ip-10-0-189-6.us-east-2.compute.internal <none> <none> reporting-operator-869b854c78-8g2x5 1/2 Running 0 7h27m 10.128.2.25 ip-10-0-150-175.us-east-2.compute.internal <none> <none>
Compare the nodes in the
openshift-metering
namespace to each nodeNAME
in your cluster:$ oc get nodes
NAME STATUS ROLES AGE VERSION ip-10-0-147-106.us-east-2.compute.internal Ready master 14h v1.18.3+6025c28 ip-10-0-150-175.us-east-2.compute.internal Ready worker 14h v1.18.3+6025c28 ip-10-0-175-23.us-east-2.compute.internal Ready master 14h v1.18.3+6025c28 ip-10-0-189-6.us-east-2.compute.internal Ready worker 14h v1.18.3+6025c28 ip-10-0-205-158.us-east-2.compute.internal Ready master 14h v1.18.3+6025c28 ip-10-0-210-167.us-east-2.compute.internal Ready worker 14h v1.18.3+6025c28
Verify that the node selector configuration in the MeteringConfig custom resource does not interfere with the cluster-wide node selector configuration such that no metering operand Pods are scheduled.
Procedure
Check the cluster-wide Scheduler object for the
spec.defaultNodeSelector
field, which shows where Pods are scheduled by default:$ oc get schedulers.config.openshift.io cluster -o yaml
3.3. Configuring persistent storage
Metering requires persistent storage to persist data collected by the metering-operator
and to store the results of reports. A number of different storage providers and storage formats are supported. Select your storage provider and modify the example configuration files to configure persistent storage for your metering installation.
3.3.1. Storing data in Amazon S3
Metering can use an existing Amazon S3 bucket or create a bucket for storage.
Metering does not manage or delete any S3 bucket data. When uninstalling metering, any S3 buckets used to store metering data must be manually cleaned up.
To use Amazon S3 for storage, edit the spec.storage
section in the example s3-storage.yaml
file below.
apiVersion: metering.openshift.io/v1 kind: MeteringConfig metadata: name: "operator-metering" spec: storage: type: "hive" hive: type: "s3" s3: bucket: "bucketname/path/" 1 region: "us-west-1" 2 secretName: "my-aws-secret" 3 # Set to false if you want to provide an existing bucket, instead of # having metering create the bucket on your behalf. createBucket: true 4
- 1
- Specify the name of the bucket where you would like to store your data. You may optionally specify the path within the bucket.
- 2
- Specify the region of your bucket.
- 3
- The name of a secret in the metering namespace containing the AWS credentials in the
data.aws-access-key-id
anddata.aws-secret-access-key
fields. See the examples that follow for more details. - 4
- Set this field to
false
if you want to provide an existing S3 bucket, or if you do not want to provide IAM credentials that haveCreateBucket
permissions.
Use the example secret below as a template.
The values of the aws-access-key-id
and aws-secret-access-key
must be base64 encoded.
apiVersion: v1 kind: Secret metadata: name: your-aws-secret data: aws-access-key-id: "dGVzdAo=" aws-secret-access-key: "c2VjcmV0Cg=="
You can use the following command to create the secret.
This command automatically base64 encodes your aws-access-key-id
and aws-secret-access-key
values.
oc create secret -n openshift-metering generic your-aws-secret --from-literal=aws-access-key-id=your-access-key --from-literal=aws-secret-access-key=your-secret-key
The aws-access-key-id
and aws-secret-access-key
credentials must have read and write access to the bucket. For an example of an IAM policy granting the required permissions, see the aws/read-write.json
file below.
{ "Version": "2012-10-17", "Statement": [ { "Sid": "1", "Effect": "Allow", "Action": [ "s3:AbortMultipartUpload", "s3:DeleteObject", "s3:GetObject", "s3:HeadBucket", "s3:ListBucket", "s3:ListMultipartUploadParts", "s3:PutObject" ], "Resource": [ "arn:aws:s3:::operator-metering-data/*", "arn:aws:s3:::operator-metering-data" ] } ] }
If you left spec.storage.hive.s3.createBucket
set to true
, or unset, then you should use the aws/read-write-create.json
file below, which contains permissions for creating and deleting buckets.
{ "Version": "2012-10-17", "Statement": [ { "Sid": "1", "Effect": "Allow", "Action": [ "s3:AbortMultipartUpload", "s3:DeleteObject", "s3:GetObject", "s3:HeadBucket", "s3:ListBucket", "s3:CreateBucket", "s3:DeleteBucket", "s3:ListMultipartUploadParts", "s3:PutObject" ], "Resource": [ "arn:aws:s3:::operator-metering-data/*", "arn:aws:s3:::operator-metering-data" ] } ] }
3.3.2. Storing data in S3-compatible storage
To use S3-compatible storage such as Noobaa, edit the spec.storage
section in the example s3-compatible-storage.yaml
file below.
apiVersion: metering.openshift.io/v1 kind: MeteringConfig metadata: name: "operator-metering" spec: storage: type: "hive" hive: type: "s3Compatible" s3Compatible: bucket: "bucketname" 1 endpoint: "http://example:port-number" 2 secretName: "my-aws-secret" 3
Use the example secret below as a template.
apiVersion: v1 kind: Secret metadata: name: your-aws-secret data: aws-access-key-id: "dGVzdAo=" aws-secret-access-key: "c2VjcmV0Cg=="
3.3.3. Storing data in Microsoft Azure
To store data in Azure blob storage you must use an existing container. Edit the spec.storage
section in the example azure-blob-storage.yaml
file below.
apiVersion: metering.openshift.io/v1 kind: MeteringConfig metadata: name: "operator-metering" spec: storage: type: "hive" hive: type: "azure" azure: container: "bucket1" 1 secretName: "my-azure-secret" 2 rootDirectory: "/testDir" 3
Use the example secret below as a template.
apiVersion: v1 kind: Secret metadata: name: your-azure-secret data: azure-storage-account-name: "dGVzdAo=" azure-secret-access-key: "c2VjcmV0Cg=="
You can use the following command to create the secret.
oc create secret -n openshift-metering generic your-azure-secret --from-literal=azure-storage-account-name=your-storage-account-name --from-literal=azure-secret-access-key=your-secret-key
3.3.4. Storing data in Google Cloud Storage
To store your data in Google Cloud Storage you must use an existing bucket. Edit the spec.storage
section in the example gcs-storage.yaml
file below.
apiVersion: metering.openshift.io/v1 kind: MeteringConfig metadata: name: "operator-metering" spec: storage: type: "hive" hive: type: "gcs" gcs: bucket: "metering-gcs/test1" 1 secretName: "my-gcs-secret" 2
Use the example secret below as a template:
apiVersion: v1 kind: Secret metadata: name: your-gcs-secret data: gcs-service-account.json: "c2VjcmV0Cg=="
You can use the following command to create the secret.
oc create secret -n openshift-metering generic your-gcs-secret --from-file gcs-service-account.json=/path/to/your/service-account-key.json
3.4. Configuring the Hive metastore
Hive metastore is responsible for storing all the metadata about the database tables created in Presto and Hive. By default, the metastore stores this information in a local embedded Derby database in a PersistentVolume
attached to the pod.
Generally, the default configuration of the Hive metastore works for small clusters, but users may wish to improve performance or move storage requirements out of cluster by using a dedicated SQL database for storing the Hive metastore data.
3.4.1. Configuring PersistentVolumes
By default, Hive requires one PersistentVolume to operate.
hive-metastore-db-data
is the main PersistentVolumeClaim (PVC) required by default. This PVC is used by the Hive metastore to store metadata about tables, such as table name, columns, and location. Hive metastore is used by Presto and the Hive server to look up table metadata when processing queries. You remove this requirement by using MySQL or PostgreSQL for the Hive metastore database.
To install, Hive metastore requires that dynamic volume provisioning be enabled via a StorageClass, a persistent volume of the correct size must be manually pre-created, or that you use a pre-existing MySQL or PostgreSQL database.
3.4.1.1. Configuring the storage class for Hive metastore
To configure and specify a StorageClass for the hive-metastore-db-data
PVC, specify the StorageClass in your MeteringConfig. An example StorageClass section is included in metastore-storage.yaml
file below.
apiVersion: metering.openshift.io/v1
kind: MeteringConfig
metadata:
name: "operator-metering"
spec:
hive:
spec:
metastore:
storage:
# Default is null, which means using the default storage class if it exists.
# If you wish to use a different storage class, specify it here
# class: "null" 1
size: "5Gi"
- 1
- Uncomment this line and replace
null
with the name of the StorageClass to use. Leaving the valuenull
will cause metering to use the default StorageClass for the cluster.
3.4.1.2. Configuring the volume sizes for the Hive Metastore
Use the metastore-storage.yaml
file below as a template.
apiVersion: metering.openshift.io/v1
kind: MeteringConfig
metadata:
name: "operator-metering"
spec:
hive:
spec:
metastore:
storage:
# Default is null, which means using the default storage class if it exists.
# If you wish to use a different storage class, specify it here
# class: "null"
size: "5Gi" 1
- 1
- Replace the value for
size
with your desired capacity. The example file shows "5Gi".
3.4.2. Use MySQL or PostgreSQL for the Hive metastore
The default installation of metering configures Hive to use an embedded Java database called Derby. This is unsuited for larger environments and can be replaced with either a MySQL or PostgreSQL database. Use the following example configuration files if your deployment requires a MySQL or PostgreSQL database for Hive.
There are 4 configuration options you can use to control the database used by Hive metastore: url, driver, username, and password.
Use the example configuration file below to use a MySQL database for Hive:
spec: hive: spec: metastore: storage: create: false config: db: url: "jdbc:mysql://mysql.example.com:3306/hive_metastore" driver: "com.mysql.jdbc.Driver" username: "REPLACEME" password: "REPLACEME"
You can pass additional JDBC parameters using the spec.hive.config.url
. For more details see the MySQL Connector/J documentation.
Use the example configuration file below to use a PostgreSQL database for Hive:
spec: hive: spec: metastore: storage: create: false config: db: url: "jdbc:postgresql://postgresql.example.com:5432/hive_metastore" driver: "org.postgresql.Driver" username: "REPLACEME" password: "REPLACEME"
You can pass additional JDBC parameters using the URL. For more details see the PostgreSQL JDBC driver documentation.
3.5. Configuring the reporting-operator
The reporting-operator
is responsible for collecting data from Prometheus, storing the metrics in Presto, running report queries against Presto, and exposing their results via an HTTP API. Configuring the Operator is primarily done through your MeteringConfig
file.
3.5.1. Prometheus connection
When you install metering on OpenShift Container Platform, Prometheus is available at https://prometheus-k8s.openshift-monitoring.svc:9091/.
To secure the connection to Prometheus, the default metering installation uses the OpenShift Container Platform certificate authority. If your Prometheus instance uses a different CA, the CA can be injected through a ConfigMap. See the following example.
spec: reporting-operator: spec: config: prometheus: certificateAuthority: useServiceAccountCA: false configMap: enabled: true create: true name: reporting-operator-certificate-authority-config filename: "internal-ca.crt" value: | -----BEGIN CERTIFICATE----- (snip) -----END CERTIFICATE-----
Alternatively, to use the system certificate authorities for publicly valid certificates, set both useServiceAccountCA
and configMap.enabled
to false
.
The reporting-operator
can also be configured to use a specified bearer token to auth with Prometheus. See the following example.
spec: reporting-operator: spec: config: prometheus: metricsImporter: auth: useServiceAccountToken: false tokenSecret: enabled: true create: true value: "abc-123"
3.5.2. Exposing the reporting API
On OpenShift Container Platform the default metering installation automatically exposes a Route, making the reporting API available. This provides the following features:
- Automatic DNS
- Automatic TLS based on the cluster CA
Also, the default installation makes it possible to use the OpenShift service for serving certificates to protect the reporting API with TLS. The OpenShift OAuth proxy is deployed as a side-car container for reporting-operator
, which protects the reporting API with authentication.
3.5.2.1. Using OpenShift Authentication
By default, the reporting API is secured with TLS and authentication. This is done by configuring the reporting-operator
to deploy a pod containing both the reporting-operator’s
container, and a sidecar container running OpenShift auth-proxy.
In order to access the reporting API, the metering operator exposes a route. Once that route has been installed, you can run the following command to get the route’s hostname.
METERING_ROUTE_HOSTNAME=$(oc -n openshift-metering get routes metering -o json | jq -r '.status.ingress[].host')
Next, set up authentication using either a service account token or basic authentication with a username/password.
3.5.2.1.1. Authenticate using a service account token
With this method, you use the token in the reporting Operator’s service account, and pass that bearer token to the Authorization header in the following command:
TOKEN=$(oc -n openshift-metering serviceaccounts get-token reporting-operator) curl -H "Authorization: Bearer $TOKEN" -k "https://$METERING_ROUTE_HOSTNAME/api/v1/reports/get?name=[Report Name]&namespace=openshift-metering&format=[Format]"
Be sure to replace the name=[Report Name]
and format=[Format]
parameters in the URL above. The format
parameter can be json, csv, or tabular.
3.5.2.1.2. Authenticate using a username and password
We are able to do basic authentication using a username and password combination, which is specified in the contents of a htpasswd file. By default, we create a secret containing an empty htpasswd data. You can, however, configure the reporting-operator.spec.authProxy.htpasswd.data
and reporting-operator.spec.authProxy.htpasswd.createSecret
keys to use this method.
Once you have specified the above in your MeteringConfig, you can run the following command:
curl -u testuser:password123 -k "https://$METERING_ROUTE_HOSTNAME/api/v1/reports/get?name=[Report Name]&namespace=openshift-metering&format=[Format]"
Be sure to replace testuser:password123
with a valid username and password combination.
3.5.2.2. Manually Configuring Authentication
In order to manually configure, or disable OAuth in the reporting-operator
, you must set spec.tls.enabled: false
in your MeteringConfig.
This also disables all TLS/authentication between the reporting-operator
, presto, and hive. You would need to manually configure these resources yourself.
Authentication can be enabled by configuring the following options. Enabling authentication configures the reporting-operator
pod to run the OpenShift auth-proxy as a sidecar container in the pod. This adjusts the ports so that the reporting-operator
API isn’t exposed directly, but instead is proxied to via the auth-proxy sidecar container.
- reporting-operator.spec.authProxy.enabled
- reporting-operator.spec.authProxy.cookie.createSecret
- reporting-operator.spec.authProxy.cookie.seed
You need to set reporting-operator.spec.authProxy.enabled
and reporting-operator.spec.authProxy.cookie.createSecret
to true
and reporting-operator.spec.authProxy.cookie.seed
to a 32-character random string.
You can generate a 32-character random string using the following command.
$ openssl rand -base64 32 | head -c32; echo.
3.5.2.2.1. Token authentication
When the following options are set to true
, authentication using a bearer token is enabled for the reporting REST API. Bearer tokens can come from serviceAccounts or users.
- reporting-operator.spec.authProxy.subjectAccessReview.enabled
- reporting-operator.spec.authProxy.delegateURLs.enabled
When authentication is enabled, the Bearer token used to query the reporting API of the user or serviceAccount
must be granted access using one of the following roles:
- report-exporter
- reporting-admin
- reporting-viewer
- metering-admin
- metering-viewer
The metering-operator
is capable of creating RoleBindings for you, granting these permissions by specifying a list of subjects in the spec.permissions
section. For an example, see the following advanced-auth.yaml
example configuration.
apiVersion: metering.openshift.io/v1 kind: MeteringConfig metadata: name: "operator-metering" spec: permissions: # anyone in the "metering-admins" group can create, update, delete, etc any # metering.openshift.io resources in the namespace. # This also grants permissions to get query report results from the reporting REST API. meteringAdmins: - kind: Group name: metering-admins # Same as above except read only access and for the metering-viewers group. meteringViewers: - kind: Group name: metering-viewers # the default serviceaccount in the namespace "my-custom-ns" can: # create, update, delete, etc reports. # This also gives permissions query the results from the reporting REST API. reportingAdmins: - kind: ServiceAccount name: default namespace: my-custom-ns # anyone in the group reporting-readers can get, list, watch reports, and # query report results from the reporting REST API. reportingViewers: - kind: Group name: reporting-readers # anyone in the group cluster-admins can query report results # from the reporting REST API. So can the user bob-from-accounting. reportExporters: - kind: Group name: cluster-admins - kind: User name: bob-from-accounting reporting-operator: spec: authProxy: # htpasswd.data can contain htpasswd file contents for allowing auth # using a static list of usernames and their password hashes. # # username is 'testuser' password is 'password123' # generated htpasswdData using: `htpasswd -nb -s testuser password123` # htpasswd: # data: | # testuser:{SHA}y/2sYAj5yrQIN4TL0YdPdmGNKpc= # # change REPLACEME to the output of your htpasswd command htpasswd: data: | REPLACEME
Alternatively, you can use any role which has rules granting get
permissions to reports/export
. This means get
access to the export
sub-resource of the Report resources in the namespace of the reporting-operator
. For example: admin
and cluster-admin
.
By default, the reporting-operator
and metering-operator
serviceAccounts
both have these permissions, and their tokens can be used for authentication.
3.5.2.2.2. Basic authentication (username/password)
For basic authentication you can supply a username and password in reporting-operator.spec.authProxy.htpasswd.data
. The username and password must be the same format as those found in an htpasswd file. When set, you can use HTTP basic authentication to provide your username and password that has a corresponding entry in the htpasswdData
contents.
3.6. Configure AWS billing correlation
Metering can correlate cluster usage information with AWS detailed billing information, attaching a dollar amount to resource usage. For clusters running in EC2, you can enable this by modifying the example aws-billing.yaml
file below.
apiVersion: metering.openshift.io/v1 kind: MeteringConfig metadata: name: "operator-metering" spec: openshift-reporting: spec: awsBillingReportDataSource: enabled: true # Replace these with where your AWS billing reports are # stored in S3. bucket: "<your-aws-cost-report-bucket>" 1 prefix: "<path/to/report>" region: "<your-buckets-region>" reporting-operator: spec: config: aws: secretName: "<your-aws-secret>" 2 presto: spec: config: aws: secretName: "<your-aws-secret>" 3 hive: spec: config: aws: secretName: "<your-aws-secret>" 4
To enable AWS billing correlation, first ensure the AWS Cost and Usage Reports are enabled. For more information, see Turning on the AWS Cost and Usage Report in the AWS documentation.
- 1
- Update the bucket, prefix, and region to the location of your AWS Detailed billing report.
- 2 3 4
- All
secretName
fields should be set to the name of a secret in the metering namespace containing AWS credentials in thedata.aws-access-key-id
anddata.aws-secret-access-key
fields. See the example secret file below for more details.
apiVersion: v1 kind: Secret metadata: name: <your-aws-secret> data: aws-access-key-id: "dGVzdAo=" aws-secret-access-key: "c2VjcmV0Cg=="
To store data in S3, the aws-access-key-id
and aws-secret-access-key
credentials must have read and write access to the bucket. For an example of an IAM policy granting the required permissions, see the aws/read-write.json
file below.
{ "Version": "2012-10-17", "Statement": [ { "Sid": "1", "Effect": "Allow", "Action": [ "s3:AbortMultipartUpload", "s3:DeleteObject", "s3:GetObject", "s3:HeadBucket", "s3:ListBucket", "s3:ListMultipartUploadParts", "s3:PutObject" ], "Resource": [ "arn:aws:s3:::operator-metering-data/*", 1 "arn:aws:s3:::operator-metering-data" 2 ] } ] } { "Version": "2012-10-17", "Statement": [ { "Sid": "1", "Effect": "Allow", "Action": [ "s3:AbortMultipartUpload", "s3:DeleteObject", "s3:GetObject", "s3:HeadBucket", "s3:ListBucket", "s3:ListMultipartUploadParts", "s3:PutObject" ], "Resource": [ "arn:aws:s3:::operator-metering-data/*", 3 "arn:aws:s3:::operator-metering-data" 4 ] } ] }
This can be done either pre-installation or post-installation. Disabling it post-installation can cause errors in the reporting-operator
.
Chapter 4. Reports
4.1. About Reports
A Report is an API object that provides a method to manage periodic ETL (Extract Transform and Load) jobs using SQL queries. They are composed using other Metering resources such as ReportQueries
, which provide the actual SQL query to run, and ReportDataSources
, which are what define the data available to the ReportQueries and Reports.
Many use cases are addressed out-of-the-box with the predefined ReportQueries
and ReportDataSources
that come installed with metering, so you do not need to define your own unless you have a use-case not covered by what is predefined.
4.1.1. Reports
The Report custom resource is used to manage the execution and status of reports. Metering produces reports derived from usage data sources, which can be used in further analysis and filtering.
A single Report resource represents a job that manages a database table and updates it with new information according to a schedule. The Report exposes the data in that table via the reporting-operator HTTP API. Reports with a spec.schedule
field set are always running and track what time periods it has collected data for. This ensures that if metering is shutdown or unavailable for an extended period of time, it will backfill the data starting where it left off. If the schedule is unset, then the Report will run once for the time specified by the reportingStart
and reportingEnd
. By default, reports wait for ReportDataSources to have fully imported any data covered in the reporting period. If the Report has a schedule, it will wait to run until the data in the period currently being processed has finished importing.
4.1.1.1. Example Report with a Schedule
The following example Report will contain information on every Pod’s CPU requests, and will run every hour, adding the last hours worth of data each time it runs.
apiVersion: metering.openshift.io/v1 kind: Report metadata: name: pod-cpu-request-hourly spec: query: "pod-cpu-request" reportingStart: "2019-07-01T00:00:00Z" schedule: period: "hourly" hourly: minute: 0 second: 0
4.1.1.2. Example Report without a Schedule (Run-Once)
The following example Report will contain information on every Pod’s CPU requests for all of July. After completion, it does not run again.
apiVersion: metering.openshift.io/v1 kind: Report metadata: name: pod-cpu-request-hourly spec: query: "pod-cpu-request" reportingStart: "2019-07-01T00:00:00Z" reportingEnd: "2019-07-31T00:00:00Z"
4.1.1.3. query
Names the ReportQuery used to generate the report. The report query controls the schema of the report as well as how the results are processed.
query
is a required field.
Use the oc
CLI to obtain a list of available ReportQuery objects:
$ oc -n openshift-metering get reportqueries NAME AGE cluster-cpu-capacity 23m cluster-cpu-capacity-raw 23m cluster-cpu-usage 23m cluster-cpu-usage-raw 23m cluster-cpu-utilization 23m cluster-memory-capacity 23m cluster-memory-capacity-raw 23m cluster-memory-usage 23m cluster-memory-usage-raw 23m cluster-memory-utilization 23m cluster-persistentvolumeclaim-request 23m namespace-cpu-request 23m namespace-cpu-usage 23m namespace-cpu-utilization 23m namespace-memory-request 23m namespace-memory-usage 23m namespace-memory-utilization 23m namespace-persistentvolumeclaim-request 23m namespace-persistentvolumeclaim-usage 23m node-cpu-allocatable 23m node-cpu-allocatable-raw 23m node-cpu-capacity 23m node-cpu-capacity-raw 23m node-cpu-utilization 23m node-memory-allocatable 23m node-memory-allocatable-raw 23m node-memory-capacity 23m node-memory-capacity-raw 23m node-memory-utilization 23m persistentvolumeclaim-capacity 23m persistentvolumeclaim-capacity-raw 23m persistentvolumeclaim-phase-raw 23m persistentvolumeclaim-request 23m persistentvolumeclaim-request-raw 23m persistentvolumeclaim-usage 23m persistentvolumeclaim-usage-raw 23m persistentvolumeclaim-usage-with-phase-raw 23m pod-cpu-request 23m pod-cpu-request-raw 23m pod-cpu-usage 23m pod-cpu-usage-raw 23m pod-memory-request 23m pod-memory-request-raw 23m pod-memory-usage 23m pod-memory-usage-raw 23m
ReportQueries with the -raw
suffix are used by other ReportQueries to build more complex queries, and should not be used directly for reports.
namespace-
prefixed queries aggregate Pod CPU/memory requests by namespace, providing a list of namespaces and their overall usage based on resource requests.
pod-
prefixed queries are similar to namespace-
prefixed queries but aggregate information by Pod rather than namespace. These queries include the Pod’s namespace and node.
node-
prefixed queries return information about each node’s total available resources.
aws-
prefixed queries are specific to AWS. Queries suffixed with -aws
return the same data as queries of the same name without the suffix, and correlate usage with the EC2 billing data.
The aws-ec2-billing-data
report is used by other queries, and should not be used as a standalone report. The aws-ec2-cluster-cost
report provides a total cost based on the nodes included in the cluster, and the sum of their costs for the time period being reported on.
For a complete list of fields, use the oc
CLI to get the ReportQuery as YAML, and check the spec.columns
field:
For example, run:
$ oc -n openshift-metering get reportqueries namespace-memory-request -o yaml
You should see output like:
apiVersion: metering.openshift.io/v1 kind: ReportQuery metadata: name: namespace-memory-request labels: operator-metering: "true" spec: columns: - name: period_start type: timestamp unit: date - name: period_end type: timestamp unit: date - name: namespace type: varchar unit: kubernetes_namespace - name: pod_request_memory_byte_seconds type: double unit: byte_seconds
4.1.1.4. schedule
The spec.schedule
configuration block defines when the report runs. The main fields in the schedule
section are period
, and then depending on the value of period
, the fields hourly
, daily
, weekly
, and monthly
allow you to fine-tune when the report runs.
For example, if period
is set to weekly
, you can add a weekly
field to the spec.schedule
block. The following example will run once a week on Wednesday, at 1 PM (hour 13 in the day).
... schedule: period: "weekly" weekly: dayOfWeek: "wednesday" hour: 13 ...
4.1.1.4.1. period
Valid values of schedule.period
are listed below, and the options available to set for a given period are also listed.
hourly
-
minute
-
second
-
daily
-
hour
-
minute
-
second
-
weekly
-
dayOfWeek
-
hour
-
minute
-
second
-
monthly
-
dayOfMonth
-
hour
-
minute
-
second
-
cron
-
expression
-
Generally, the hour
, minute
, second
fields control when in the day the report runs, and dayOfWeek
/dayOfMonth
control what day of the week, or day of month the report runs on, if it is a weekly or monthly report period.
For each of these fields, there is a range of valid values:
-
hour
is an integer value between 0-23. -
minute
is an integer value between 0-59. -
second
is an integer value between 0-59. -
dayOfWeek
is a string value that expects the day of the week (spelled out). -
dayOfMonth
is an integer value between 1-31.
For cron periods, normal cron expressions are valid:
-
expression: "*/5 * * * *"
4.1.1.5. reportingStart
To support running a Report against existing data, you can set the spec.reportingStart
field to a RFC3339 timestamp to tell the Report to run according to its schedule
starting from reportingStart
rather than the current time. One important thing to understand is that this will result in the reporting-operator running many queries in succession for each interval in the schedule that is between the reportingStart
time and the current time. This could be thousands of queries if the period is less than daily and the reportingStart
is more than a few months back. If reportingStart
is left unset, the Report will run at the next full reportingPeriod after the time the report is created.
As an example of how to use this field, if you had data already collected dating back to January 1st, 2019, which you wanted to be included in your Report, you could create a report with the following values:
apiVersion: metering.openshift.io/v1 kind: Report metadata: name: pod-cpu-request-hourly spec: query: "pod-cpu-request" schedule: period: "hourly" reportingStart: "2019-01-01T00:00:00Z"
4.1.1.6. reportingEnd
To configure a Report to only run until a specified time, you can set the spec.reportingEnd
field to an RFC3339 timestamp. The value of this field will cause the Report to stop running on its schedule after it has finished generating reporting data for the period covered from its start time until reportingEnd
. Because a schedule will most likely not align with reportingEnd, the last period in the schedule will be shortened to end at the specified reportingEnd time. If left unset, then the Report will run forever, or until a reportingEnd
is set on the Report.
For example, if you wanted to create a report that runs once a week for the month of July:
apiVersion: metering.openshift.io/v1 kind: Report metadata: name: pod-cpu-request-hourly spec: query: "pod-cpu-request" schedule: period: "weekly" reportingStart: "2019-07-01T00:00:00Z" reportingEnd: "2019-07-31T00:00:00Z"
4.1.1.7. runImmediately
When runImmediately
is set to true
, the report will be run immediately. This behavior ensures that the report is immediately processed and queued without requiring additional scheduling parameters.
When runImmediately
is set to true
you must set a reportingEnd
and reportingStart
value.
4.1.1.8. inputs
The spec.inputs
field of a Report can be used to override or set values defined in a ReportQuery’s spec.inputs
field.
It is a list of name-value pairs:
spec: inputs: - name: "NamespaceCPUUsageReportName" value: "namespace-cpu-usage-hourly"
The name
of an input must exist in the ReportQuery’s inputs
list. The value
of the input must be the correct type for the input’s type
.
4.1.1.9. Roll-up Reports
Report data is stored in the database much like metrics themselves, and therefore, can be used in aggregated or roll-up reports. A simple use case for a roll-up report is to spread the time required to produce a report over a longer period of time; instead of: requiring a monthly report to query and add all data over an entire month, the task can be split into daily reports that each run over a thirtieth of the data.
A custom roll-up report requires a custom report query. The ReportQuery template processor provides a function: reportTableName
that can get the necessary table name from a Report’s metadata.name
.
Below is a snippet taken from a built-in query:
# Taken from pod-cpu.yaml spec: ... inputs: - name: ReportingStart type: time - name: ReportingEnd type: time - name: NamespaceCPUUsageReportName type: Report - name: PodCpuUsageRawDataSourceName type: ReportDataSource default: pod-cpu-usage-raw ... query: | ... {|- if .Report.Inputs.NamespaceCPUUsageReportName |} namespace, sum(pod_usage_cpu_core_seconds) as pod_usage_cpu_core_seconds FROM {| .Report.Inputs.NamespaceCPUUsageReportName | reportTableName |} ...
# aggregated-report.yaml spec: query: "namespace-cpu-usage" inputs: - name: "NamespaceCPUUsageReportName" value: "namespace-cpu-usage-hourly"
4.1.1.9.1. Report Status
The execution of a scheduled report can be tracked using its status field. Any errors occurring during the preparation of a report will be recorded here.
The status
field of a Report currently has two fields:
-
conditions
: Conditions is a list of conditions, each of which have atype
,status
,reason
, andmessage
field. Possible values of a condition’stype
field areRunning
andFailure
, indicating the current state of the scheduled report. Thereason
indicates why itscondition
is in its current state with thestatus
being eithertrue
,false
or,unknown
. Themessage
provides a human readable indicating why the condition is in the current state. For detailed information on thereason
values seepkg/apis/metering/v1/util/report_util.go
. -
lastReportTime
: Indicates the time Metering has collected data up to.
4.2. Storage Locations
A StorageLocation is a custom resource that configures where data will be stored by the reporting-operator. This includes the data collected from Prometheus, and the results produced by generating a Report custom resource.
You only need to configure a StorageLocation if you want to store data in multiple locations, like multiple S3 buckets or both S3 and HDFS, or if you wish to access a database in Hive/Presto that was not created by metering. For most users this is not a requirement, and the documentation on configuring metering is sufficent to configure all necessary storage components.
4.2.1. StorageLocation examples
This first example is what the built-in local storage option looks like. It is configured to use Hive, and by default data is stored wherever Hive is configured to use storage (HDFS, S3, or a ReadWriteMany PVC).
Local storage example
apiVersion: metering.openshift.io/v1 kind: StorageLocation metadata: name: hive labels: operator-metering: "true" spec: hive: 1 databaseName: metering 2 unmanagedDatabase: false 3
- 1
- If the
hive
section is present, then the StorageLocation will be configured to store data in Presto by creating the table using Hive server. Only databaseName and unmanagedDatabase are required fields. - 2
- The name of the database within hive.
- 3
- If
true
, then this StorageLocation will not be actively managed, and the databaseName is expected to already exist in Hive. Iffalse
, this will cause the reporting-operator to create the database in Hive.
The next example uses an AWS S3 bucket for storage. The prefix is appended to the bucket name when constructing the path to use.
Remote storage example
apiVersion: metering.openshift.io/v1
kind: StorageLocation
metadata:
name: example-s3-storage
labels:
operator-metering: "true"
spec:
hive:
databaseName: example_s3_storage
unmanagedDatabase: false
location: "s3a://bucket-name/path/within/bucket" 1
- 1
- (optional) The filesystem URL for Presto and Hive to use for the database. This can be an
hdfs://
ors3a://
filesystem URL.
There are some additional optional fields that can be specified in the hive
section:
- (optional) defaultTableProperties: Contains configuration options for creating tables using Hive.
- (optional) fileFormat: The file format used for storing files in the filesystem. See the Hive Documentation on File Storage Format for a list of options and more details.
- (optional) rowFormat: Controls the Hive row format. This controls how Hive serializes and deserializes rows. See the Hive Documentation on Row Formats and SerDe for more details.
4.2.2. Default StorageLocation
If an annotation storagelocation.metering.openshift.io/is-default
exists and is set to true
on a StorageLocation resource, then that resource becomes the default storage resource. Any components with a storage configuration option where StorageLocation is not specified will use the default storage resource. There can be only one default storage resource. If more than one resource with the annotation exists, an error is logged because the Operator cannot determine the default.
Default storage example
apiVersion: metering.openshift.io/v1 kind: StorageLocation metadata: name: example-s3-storage labels: operator-metering: "true" annotations: storagelocation.metering.openshift.io/is-default: "true" spec: hive: databaseName: example_s3_storage unmanagedDatabase: false location: "s3a://bucket-name/path/within/bucket"
Chapter 5. Using Metering
5.1. Prerequisites
- Install Metering
- Review the details about the available options that can be configured for a Report and how they function.
5.2. Writing Reports
Writing a Report is the way to process and analyze data using Metering.
To write a Report, you must define a Report resource in a YAML file, specify the required parameters, and create it in the openshift-metering
namespace by using oc
.
Prerequisites
- Metering is installed.
Procedure
Change to the
openshift-metering
project:$ oc project openshift-metering
Create a Report resource as a YAML file:
Create a YAML file with the following content:
apiVersion: metering.openshift.io/v1 kind: Report metadata: name: namespace-cpu-request-2019 1 namespace: openshift-metering spec: reportingStart: '2019-01-01T00:00:00Z' reportingEnd: '2019-12-30T23:59:59Z' query: namespace-cpu-request 2 runImmediately: true 3
- 2
- The
query
specifies ReportQuery used to generate the Report. Change this based on what you want to report on. For a list of options, runoc get reportqueries | grep -v raw
. - 1
- Use a descriptive name about what the Report does for
metadata.name
. A good name is the query, and the schedule or period you used. - 3
- Set
runImmediately
totrue
for it to run with whatever data is available, or set it tofalse
if you want it to wait forreportingEnd
to pass.
Run the following command to create the Report:
$ oc create -f <file-name>.yaml report.metering.openshift.io/namespace-cpu-request-2019 created
You can list Reports and their
Running
status with the following command:$ oc get reports NAME QUERY SCHEDULE RUNNING FAILED LAST REPORT TIME AGE namespace-cpu-request-2019 namespace-cpu-request Finished 2019-12-30T23:59:59Z 26s
5.3. Viewing Report results
Viewing a Report’s results involves querying the reporting-api Route
and authenticating to the API using your OpenShift Container Platform credentials. Reports can be retrieved as JSON
, CSV
, or Tabular
formats.
Prerequisites
- Metering is installed.
-
To access Report results, you must either be a cluster administrator, or you need to be granted access using the
report-exporter
role in theopenshift-metering
namespace.
Procedure
Change to the
openshift-metering
project:$ oc project openshift-metering
Query the reporting API for results:
Get the route to the
reporting-api
:$ meteringRoute="$(oc get routes metering -o jsonpath='{.spec.host}')" $ echo "$meteringRoute"
Get the token of your current user to be used in the request:
$ token="$(oc whoami -t)"
To get the results, use
curl
to make a request to the reporting API for your report:$ reportName=namespace-cpu-request-2019 1 $ reportFormat=csv 2 $ curl --insecure -H "Authorization: Bearer ${token}" "https://${meteringRoute}/api/v1/reports/get?name=${reportName}&namespace=openshift-metering&format=$reportFormat"
The response should look similar to the following (example output is with
reportName=namespace-cpu-request-2019
andreportFormat=csv
):period_start,period_end,namespace,pod_request_cpu_core_seconds 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-apiserver,11745.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-apiserver-operator,261.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-authentication,522.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-authentication-operator,261.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-cloud-credential-operator,261.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-cluster-machine-approver,261.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-cluster-node-tuning-operator,3385.800000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-cluster-samples-operator,261.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-cluster-version,522.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-console,522.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-console-operator,261.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-controller-manager,7830.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-controller-manager-operator,261.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-dns,34372.800000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-dns-operator,261.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-etcd,23490.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-image-registry,5993.400000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-ingress,5220.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-ingress-operator,261.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-kube-apiserver,12528.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-kube-apiserver-operator,261.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-kube-controller-manager,8613.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-kube-controller-manager-operator,261.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-machine-api,1305.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-machine-config-operator,9637.800000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-metering,19575.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-monitoring,6256.800000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-network-operator,261.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-sdn,94503.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-service-ca,783.000000 2019-01-01 00:00:00 +0000 UTC,2019-12-30 23:59:59 +0000 UTC,openshift-service-ca-operator,261.000000
Chapter 6. Examples of using metering
Use the following example Reports to get started measuring capacity, usage, and utilization in your cluster. These examples showcase the various types of reports metering offers, along with a selection of the predefined queries.
6.1. Prerequisites
- Install Metering
- Review the details about writing and viewing reports.
6.2. Measure cluster capacity hourly and daily
The following Report demonstrates how to measure cluster capacity both hourly and daily. The daily Report works by aggregating the hourly Report’s results.
The following report measures cluster CPU capacity every hour.
Hourly CPU capacity by cluster example
apiVersion: metering.openshift.io/v1
kind: Report
metadata:
name: cluster-cpu-capacity-hourly
spec:
query: "cluster-cpu-capacity"
schedule:
period: "hourly" 1
- 1
- You could change this period to
daily
to get a daily report, but with larger data sets it is more efficient to use an hourly report, then aggregate your hourly data into a daily report.
The following report aggregates the hourly data into a daily report.
Daily CPU capacity by cluster example
apiVersion: metering.openshift.io/v1 kind: Report metadata: name: cluster-cpu-capacity-daily 1 spec: query: "cluster-cpu-capacity" 2 inputs: 3 - name: ClusterCpuCapacityReportName value: cluster-cpu-capacity-hourly schedule: period: "daily"
- 1
- To stay organized, remember to change the name of your Report if you change any of the other values.
- 2
- You can also measure
cluster-memory-capacity
. Remember to update the query in the associated hourly report as well. - 3
- The
inputs
section configures this report to aggregate the hourly report. Specifically,value: cluster-cpu-capacity-hourly
is the name of the hourly report that gets aggregated.
6.3. Measure cluster usage with a one-time Report
The following Reports to measure cluster usage from a specific starting date forward. The Report only runs once, after you save it and apply it.
CPU usage by cluster example
apiVersion: metering.openshift.io/v1 kind: Report metadata: name: cluster-cpu-usage-2019 1 spec: reportingStart: '2019-01-01T00:00:00Z' 2 reportingEnd: '2019-12-30T23:59:59Z' query: cluster-cpu-usage 3 runImmediately: true 4
- 1
- To stay organized, remember to change the name of your Report if you change any of the other values.
- 2
- Configures the Reports to start using data from the
reportingStart
timestamp until thereportingEnd
timestamp. - 3
- Adjust your query here. You can also measure cluster usage with the
cluster-memory-usage
query. - 4
- This tells the Report to run immediately after saving it and applying it.
6.4. Measure cluster utilization using cron expressions
You can also use cron expressions when configuring the period of your reports. The following report measures cluster utilization by looking at CPU utilization from 9am-5pm every weekday.
Weekday CPU utilization by cluster example
apiVersion: metering.openshift.io/v1 kind: Report metadata: name: cluster-cpu-utilization-weekdays 1 spec: query: "cluster-cpu-utilization" 2 schedule: period: "cron" expression: 0 0 * * 1-5 3
Chapter 7. Troubleshooting and debugging metering
Use the following sections to help troubleshoot and debug specific issues with metering.
In addition to the information in this section, be sure to review the following topics:
7.1. Troubleshooting metering
A common issue with metering is Pods failing to start. Pods might fail to start due to lack of resources or if they have a dependency on a resource that does not exist, such as a StorageClass or Secret.
7.1.1. Not enough compute resources
A common issue when installing or running metering is lack of compute resources. Ensure that metering is allocated the minimum resource requirements described in the installation prerequisites.
To determine if the issue is with resources or scheduling, follow the troubleshooting instructions included in the Kubernetes document Managing Compute Resources for Containers.
7.1.2. StorageClass not configured
Metering requires that a default StorageClass be configured for dynamic provisioning.
See the documentation on configuring metering for information on how to check if there are any StorageClasses configured for the cluster, how to set the default, and how to configure metering to use a StorageClass other than the default.
7.1.3. Secret not configured correctly
A common issue with metering is providing the incorrect secret when configuring your persistent storage. Be sure to review the example configuration files and create you secret according to the guidelines for your storage provider.
7.2. Debugging metering
Debugging metering is much easier when you interact directly with the various components. The sections below detail how you can connect and query Presto and Hive as well as view the dashboards of the Presto and HDFS components.
All of the commands in this section assume you have installed metering through OperatorHub in the openshift-metering
namespace.
7.2.1. Get reporting operator logs
The command below will follow the logs of the reporting-operator
.
$ oc -n openshift-metering logs -f "$(oc -n openshift-metering get pods -l app=reporting-operator -o name | cut -c 5-)" -c reporting-operator
7.2.2. Query Presto using presto-cli
The following command opens an interactive presto-cli session where you can query Presto. This session runs in the same container as Presto and launches an additional Java instance, which can create memory limits for the Pod. If this occurs, you should increase the memory request and limits of the Presto Pod.
By default, Presto is configured to communicate using TLS. You must to run the following command to run Presto queries:
$ oc -n openshift-metering exec -it "$(oc -n openshift-metering get pods -l app=presto,presto=coordinator -o name | cut -d/ -f2)" -- /usr/local/bin/presto-cli --server https://presto:8080 --catalog hive --schema default --user root --keystore-path /opt/presto/tls/keystore.pem
Once you run this command, a prompt appears where you can run queries. Use the show tables from metering;
query to view the list of tables:
$ presto:default> show tables from metering; Table datasource_your_namespace_cluster_cpu_capacity_raw datasource_your_namespace_cluster_cpu_usage_raw datasource_your_namespace_cluster_memory_capacity_raw datasource_your_namespace_cluster_memory_usage_raw datasource_your_namespace_node_allocatable_cpu_cores datasource_your_namespace_node_allocatable_memory_bytes datasource_your_namespace_node_capacity_cpu_cores datasource_your_namespace_node_capacity_memory_bytes datasource_your_namespace_node_cpu_allocatable_raw datasource_your_namespace_node_cpu_capacity_raw datasource_your_namespace_node_memory_allocatable_raw datasource_your_namespace_node_memory_capacity_raw datasource_your_namespace_persistentvolumeclaim_capacity_bytes datasource_your_namespace_persistentvolumeclaim_capacity_raw datasource_your_namespace_persistentvolumeclaim_phase datasource_your_namespace_persistentvolumeclaim_phase_raw datasource_your_namespace_persistentvolumeclaim_request_bytes datasource_your_namespace_persistentvolumeclaim_request_raw datasource_your_namespace_persistentvolumeclaim_usage_bytes datasource_your_namespace_persistentvolumeclaim_usage_raw datasource_your_namespace_persistentvolumeclaim_usage_with_phase_raw datasource_your_namespace_pod_cpu_request_raw datasource_your_namespace_pod_cpu_usage_raw datasource_your_namespace_pod_limit_cpu_cores datasource_your_namespace_pod_limit_memory_bytes datasource_your_namespace_pod_memory_request_raw datasource_your_namespace_pod_memory_usage_raw datasource_your_namespace_pod_persistentvolumeclaim_request_info datasource_your_namespace_pod_request_cpu_cores datasource_your_namespace_pod_request_memory_bytes datasource_your_namespace_pod_usage_cpu_cores datasource_your_namespace_pod_usage_memory_bytes (32 rows) Query 20190503_175727_00107_3venm, FINISHED, 1 node Splits: 19 total, 19 done (100.00%) 0:02 [32 rows, 2.23KB] [19 rows/s, 1.37KB/s] presto:default>
7.2.3. Query Hive using beeline
The following opens an interactive beeline session where you can query Hive. This session runs in the same container as Hive and launches an additional Java instance, which can create memory limits for the Pod. If this occurs, you should increase the memory request and limits of the Hive Pod.
$ oc -n openshift-metering exec -it $(oc -n openshift-metering get pods -l app=hive,hive=server -o name | cut -d/ -f2) -c hiveserver2 -- beeline -u 'jdbc:hive2://127.0.0.1:10000/default;auth=noSasl'
Once you run this command, a prompt appears where you can run queries. Use the show tables;
query to view the list of tables:
$ 0: jdbc:hive2://127.0.0.1:10000/default> show tables from metering; +----------------------------------------------------+ | tab_name | +----------------------------------------------------+ | datasource_your_namespace_cluster_cpu_capacity_raw | | datasource_your_namespace_cluster_cpu_usage_raw | | datasource_your_namespace_cluster_memory_capacity_raw | | datasource_your_namespace_cluster_memory_usage_raw | | datasource_your_namespace_node_allocatable_cpu_cores | | datasource_your_namespace_node_allocatable_memory_bytes | | datasource_your_namespace_node_capacity_cpu_cores | | datasource_your_namespace_node_capacity_memory_bytes | | datasource_your_namespace_node_cpu_allocatable_raw | | datasource_your_namespace_node_cpu_capacity_raw | | datasource_your_namespace_node_memory_allocatable_raw | | datasource_your_namespace_node_memory_capacity_raw | | datasource_your_namespace_persistentvolumeclaim_capacity_bytes | | datasource_your_namespace_persistentvolumeclaim_capacity_raw | | datasource_your_namespace_persistentvolumeclaim_phase | | datasource_your_namespace_persistentvolumeclaim_phase_raw | | datasource_your_namespace_persistentvolumeclaim_request_bytes | | datasource_your_namespace_persistentvolumeclaim_request_raw | | datasource_your_namespace_persistentvolumeclaim_usage_bytes | | datasource_your_namespace_persistentvolumeclaim_usage_raw | | datasource_your_namespace_persistentvolumeclaim_usage_with_phase_raw | | datasource_your_namespace_pod_cpu_request_raw | | datasource_your_namespace_pod_cpu_usage_raw | | datasource_your_namespace_pod_limit_cpu_cores | | datasource_your_namespace_pod_limit_memory_bytes | | datasource_your_namespace_pod_memory_request_raw | | datasource_your_namespace_pod_memory_usage_raw | | datasource_your_namespace_pod_persistentvolumeclaim_request_info | | datasource_your_namespace_pod_request_cpu_cores | | datasource_your_namespace_pod_request_memory_bytes | | datasource_your_namespace_pod_usage_cpu_cores | | datasource_your_namespace_pod_usage_memory_bytes | +----------------------------------------------------+ 32 rows selected (13.101 seconds) 0: jdbc:hive2://127.0.0.1:10000/default>
7.2.4. Port-forward to the Hive web UI
Run the following command:
$ oc -n openshift-metering port-forward hive-server-0 10002
You can now open http://127.0.0.1:10002 in your browser window to view the Hive web interface.
7.2.5. Port-forward to hdfs
To the namenode:
$ oc -n openshift-metering port-forward hdfs-namenode-0 9870
You can now open http://127.0.0.1:9870 in your browser window to view the HDFS web interface.
To the first datanode:
$ oc -n openshift-metering port-forward hdfs-datanode-0 9864
To check other datanodes, run the above command, replacing hdfs-datanode-0
with the Pod you want to view information on.
7.2.6. Metering Ansible Operator
Metering uses the Ansible Operator to watch and reconcile resources in a cluster environment. When debugging a failed metering installation, it can be helpful to view the Ansible logs or status of your MeteringConfig custom resource.
7.2.6.1. Accessing ansible logs
In the default installation, the metering Operator is deployed as a Pod. In this case, we can check the logs of the ansible container within this Pod:
$ oc -n openshift-metering logs $(oc -n openshift-metering get pods -l app=metering-operator -o name | cut -d/ -f2) -c ansible
Alternatively, you can view the logs of the Operator container (replace -c ansible
with -c operator
) for condensed output.
7.2.6.2. Checking the MeteringConfig Status
It can be helpful to view the .status
field of your MeteringConfig custom resource to debug any recent failures. The following command shows status messages with type Invalid
:
$ oc -n openshift-metering get meteringconfig operator-metering -o=jsonpath='{.status.conditions[?(@.type=="Invalid")].message}'
Chapter 8. Uninstalling metering
You can remove metering from your OpenShift Container Platform cluster.
Metering does not manage or delete Amazon S3 bucket data. After uninstalling metering, you must manually clean up S3 buckets that were used to store metering data.
8.1. Removing the Metering Operator from your cluster
Remove the Metering Operator from your cluster by following the documentation on deleting Operators from a cluster.
Removing the Metering Operator from your cluster does not remove its CustomResourceDefinitions or managed resources. See the following sections on Uninstalling a metering namespace and Uninstalling metering CustomResourceDefinitions for steps to remove any remaining metering components.
8.2. Uninstalling a metering namespace
Uninstall your metering namespace, for example the openshift-metering
namespace, by removing the MeteringConfig resource and deleting the openshift-metering
namespace.
Prerequisites
- The Metering Operator is removed from your cluster.
Procedure
Remove all resources created by the Metering Operator:
$ oc --namespace openshift-metering delete meteringconfig --all
After the previous step is complete, verify that all Pods in the
openshift-metering
namespace are deleted or are reporting a terminating state:$ oc --namespace openshift-metering get pods
Delete the
openshift-metering
namespace:$ oc delete namespace openshift-metering
8.3. Uninstalling metering CustomResourceDefinitions
The metering CustomResourceDefinitions (CRDs) remain in the cluster after the Metering Operator is uninstalled and the openshift-metering
namespace is deleted.
Deleting the metering CRDs disrupts any additional metering installations in other namespaces in your cluster. Ensure that there are no other metering installations before proceeding.
Prerequisites
-
The MeteringConfig custom resource in the
openshift-metering
namespace is deleted. -
The
openshift-metering
namespace is deleted.
Procedure
Delete the remaining metering CRDs:
$ oc get crd -o name | grep "metering.openshift.io" | xargs oc delete
Legal Notice
Copyright © 2024 Red Hat, Inc.
OpenShift documentation is licensed under the Apache License 2.0 (https://www.apache.org/licenses/LICENSE-2.0).
Modified versions must remove all Red Hat trademarks.
Portions adapted from https://github.com/kubernetes-incubator/service-catalog/ with modifications by Red Hat.
Red Hat, Red Hat Enterprise Linux, the Red Hat logo, the Shadowman logo, JBoss, OpenShift, Fedora, the Infinity logo, and RHCE are trademarks of Red Hat, Inc., registered in the United States and other countries.
Linux® is the registered trademark of Linus Torvalds in the United States and other countries.
Java® is a registered trademark of Oracle and/or its affiliates.
XFS® is a trademark of Silicon Graphics International Corp. or its subsidiaries in the United States and/or other countries.
MySQL® is a registered trademark of MySQL AB in the United States, the European Union and other countries.
Node.js® is an official trademark of Joyent. Red Hat Software Collections is not formally related to or endorsed by the official Joyent Node.js open source or commercial project.
The OpenStack® Word Mark and OpenStack logo are either registered trademarks/service marks or trademarks/service marks of the OpenStack Foundation, in the United States and other countries and are used with the OpenStack Foundation’s permission. We are not affiliated with, endorsed or sponsored by the OpenStack Foundation, or the OpenStack community.
All other trademarks are the property of their respective owners.