Data Collection CLI (datactl)

Table of Contents

• Data Collection CLI (datactl)
• Installation
• Usage
• Getting started
• Exporting from DataService sources
• Exporting from IBM License Metric Tool sources
• Using the FIPS enabled datactl container

Installation

The tool is available via prebuilt executables on the latest release. To install the tool to your local system, download the targz file and extract it to a folder on your path.

Usage

As an OC Plugin

If the oc-datactl file is installed to your path. The oc command will recognize it as a plugin. You may call oc datactl

As a standalone tool

Datactl tool can be used standalone. Just move oc-datactl to your path and use oc-datactl directly.

Getting started

1. Get your Pull Secret.

2. Log in to your cluster.

3. Setup your configuration. When prompted, provide the pull secret token as the Upload API Secret.

   oc datactl config init

   This will create the default configuration on your home directory. ~/.datactl/config

4. Add the role-binding to the default service account on operator-namespace.

   Install the ClusterRole and create the ClusterRoleBinding for the default service account for the namespace the IBM Metrics Operator's DataService is installed to by default: ibm-software-central. The datactl tool will use this service account by default.

   oc apply -f resources/service-account-role.yaml // file found in release
   oc create clusterrolebinding rhm-files --clusterrole=rhm-files --serviceaccount=ibm-software-central:default

5. Now you're configured. You can start using the export commands.

Exporting from DataService sources

Recommended approach is to run the commands in this order:

// Must be logged in to the cluster

// Add the dataservice as a source, to which you are logged into with your current context
datactl sources add dataservice --use-default-context --insecure-skip-tls-verify=true --allow-self-signed=true --namespace=ibm-software-central

// Pull the data from dataservice sources
oc datactl export pull --source-type=dataservice

// If you're connected to the internet
oc datactl export push

// If no errors from push.
oc datactl export commit

Let's break down what each one is doing.

oc datactl sources add dataservice --use-default-context --insecure-skip-tls-verify=true --allow-self-signed=true --namespace=ibm-software-central

• Adds the default-context cluster's dataservice as a source for pulling
• Writes the source data-service-endpoint to ~/.datactl/config
• The DataService to connect to is in the ibm-software-central namespace

oc datactl export pull

• Pulls files from data service and stores them in a tar file under your ~/.datactl/data folder.
• Writes the status of the files found in ~/.datactl/config

oc datactl export push

• Files pulled by the previous command are pushed to IBM Software Central.
• If this process errors, do not commit. Retry the export push or open a support ticket.

oc datactl export commit

• Commits the files to the dataservice.
• At this point you're telling the data service that you've retrieved these files and have or will submit them to IBM Software Central.
• After some time, the files in dataservice will be cleaned up to save space.

If you want to transfer it somewhere else, you can find the tar file under your ~/.datactl/data/ directory.

Exporting from IBM License Metric Tool sources

Prerequisite: API Token is required to get data from IBM License Metric Tool (ILMT). Login to your ILMT environment, go to Profile and click Show token under API Token section.

First step is to configure ILMT data source. Execute following command

datactl sources add ilmt

and provide ILMT hostname, port number and token

To pull data from ILMT, execute command

datactl export pull --source-type=ilmt

First time you will be asked to provide start date. Next time last synchronization date is stored in config file and will be updated to pull data from last synchronization date.

To push data to IBM Software Central execute command

datactl export push

Using the FIPS enabled datactl container

A containerized FIPS enabled version of datactl is provided, built with Red Hat's go-toolset

1. Create the .datactl directory locally on the host

   mkdir -p $HOME/.datactl
   

2. Setup your configuration, binding the .datactl and .kube directories, and providing the api endpoint and pull secret token

   docker run --rm \
   --mount type=bind,source=$HOME/.datactl,target=/root/.datactl \
   --mount type=bind,source=$HOME/.kube,target=/root/.kube \
   quay.io/rh-marketplace/datactl:latest config init --api swc.saas.ibm.com --token ${TOKEN}
   

3. Add a data source, such as dataservice from your current OpenShift cluster context

   docker run --rm \
   --mount type=bind,source=$HOME/.datactl,target=/root/.datactl \
   --mount type=bind,source=$HOME/.kube,target=/root/.kube \
   quay.io/rh-marketplace/datactl:latest sources add dataservice --insecure-skip-tls-verify=true --use-default-context --allow-self-signed=true --namespace=ibm-software-central
   

4. Pull from the data source

   docker run --rm \
   --name datactl \
   --mount type=bind,source=$HOME/.datactl,target=/root/.datactl \
   --mount type=bind,source=$HOME/.kube,target=/root/.kube \
   quay.io/rh-marketplace/datactl:latest export pull
   

5. Push data to IBM Software Central

   docker run --rm \
   --name datactl \
   --mount type=bind,source=$HOME/.datactl,target=/root/.datactl \
   --mount type=bind,source=$HOME/.kube,target=/root/.kube \
   quay.io/rh-marketplace/datactl:latest export push
   

6. Commit

   docker run --rm \
   --name datactl \
   --mount type=bind,source=$HOME/.datactl,target=/root/.datactl \
   --mount type=bind,source=$HOME/.kube,target=/root/.kube \
   quay.io/rh-marketplace/datactl:latest export commit