Replaces Python with new Kata guide in swupd-search; revises dev workstation.

Signed-off-by: Michael Vincerra <michael.vincerra@intel.com>
This commit is contained in:
Michael Vincerra
2018-08-15 15:27:06 -07:00
parent 1f95d68954
commit a244881bac
2 changed files with 59 additions and 58 deletions
@@ -6,26 +6,26 @@ Developer Workstation
Overview
********
*Developer Workstation* helps you find the bundles you need to
*Developer Workstation* helps you find the :ref:`bundles-about` you need to
start your |CL| development project.
First, we recommend learning how to use :ref:`swupd <swupd-guide>`.
Before continuing, we recommend that you learn how to use :ref:`swupd <swupd-guide>`. Visit our :ref:`swupd-about` page to understand how |CL|
simplifies software versioning in contrast to other Linux\* distributions.
How to use this document
========================
Workstation Setup
=================
This guide helps you understand the minimum bundles required to get started.
After installing them, you can add more bundles relevant to your use case.
To run any process required for Clear Linux developement, you may want to
To run any process required for Clear Linux development, you may choose to
add the large bundle :ref:`*os-clr-on-clr* <enable-user-space>`. However,
given how many packages this bundle contains, you may prefer instead to
deploy a leaner OS with only bundles relevant to your project. Developer
given how many packages this bundle contains, you may instead want to deploy
a leaner OS with only those bundles relevant to your project. Developer
Workstation responds to this need.
Use Table 1, *Clear Linux Developer Profiles*, to identify the *minimum
required bundles* to get started developing based on your role or project.
While your role may not fit neatly into one of these categories, use Table 1
as a starting point.
While your role may not neatly fit in one of these categories, consider using Table 1 as a starting point.
.. list-table:: **Table 1. Clear Linux Developer Profiles**
:widths: 20, 20, 20, 20
@@ -75,9 +75,11 @@ as a starting point.
-
-
`swupd` search
==============
Second, try out :ref:`swupd search <swupd-search>`, to learn the commands to
search for and add bundles relevant to your project.
We recommend trying out :ref:`swupd search <swupd-search>`, to learn the
commands to search for and add bundles relevant to your project.
:ref:`swupd-search` shows you how to:
@@ -12,104 +12,103 @@ a bundle.
Assumptions
***********
You have:
This guide assumes you:
* A basic knowledge of using :ref:`swupd <swupd-guide>`
* An Understanding how :ref:`swupd <swupd-about>` differs from
* Possess a basic knowledge of :ref:`swupd <swupd-guide>`
* Understand :ref:`how swupd differs <swupd-about>` from
other Linux\* distributions
* A plan to use :ref:`mixer` to build your own |CLOSIA|for a specific
use case
* Plan to use :ref:`mixer` to build your own |CL| OS
How do I search for a bundle?
*****************************
Example: Python Data Science
============================
Example: Kata Containers
========================
We're developing a custom Clear Linux OS for data science with Python. We'll
develop our own mix, from which we'll create a release image. That image
will be distributed to data center (DC) clients across the United States
who need to determine DC workload balancing during peak energy hours.
We will create a tool to analyze energy consumption based on population
statistics and consumption data. Our tool's dashboard will show heat-maps
of where and charts of when energy consumption peaks in large
metropolitan areas.
Containers have revolutionized the way we manage cloud infrastructure. Whereas traditional containers share the same OS kernel, raising security concerns, with Kata Containers, each container has its own kernel instance and runs on its own :abbr:`Virtual Machine (VM)`. Whether you're running 3 or 300 nodes on your cluster, Kata Containers provide a lightweight, fast, and secure option for app/container management.
So far, we know we need a Python *data science* bundle.
In |CL|, you only need to add one bundle to use `Kata Containers`_:
`containers virt`_. We also recommend our tutorial: :ref:`kata`
First, use :command:`swupd search` with a general term like *Python*.
So far, we need a *kata* containers bundle. So how do we search for it? First, use :command:`swupd search` with **only one term** like *kata*.
#. Enter this command, followed by 'Python' as the search term:
#. Enter this command, followed by 'kata' as the search term:
.. code-block:: bash
sudo swupd search -b Python
sudo swupd search -b kata
.. note::
`swupd search` searches for matching paths in the manifest data.
Enter only one term, or one hyphenated term, at a time.
Use the command :command:`man swupd` to learn more.
`swupd search` downloads |CL| manifest data and searches for matching paths. Enter only one term, or hyphenated term, per search. Use the command :command:`man swupd` to learn more.
`-b` flag, or `--binary`, means: Restrict search to program binary paths.
#. Viewing `swupd search` result, we choose the best match for our use case.
#. The `swupd search` results show a match for our use case.
.. code-block:: console
Bundle python-data-science (923 MB to install)
/usr/bin/ipython3
/usr/bin/ipython
Bundle containers-virt (834 MB to install)
/usr/bin/kata-virtfs-lite-proxy-helper
/usr/bin/kata-runtime
/usr/bin/kata-qemu-lite-system-x86_64
/usr/bin/kata-qemu-lite-pr-helper
/usr/bin/kata-qemu-lite-ga
/usr/bin/kata-collect-data.sh
.. note::
The results *Python* is one of several shown.
If the bundle is already installed, *[installed]* appears in search results. If this doesn't apppear, the bundle needs to be installed.
If the bundle is already installed, *[installed]* appears in search results. If that doesn't apppear, the bundle needs to be installed.
#. Add the bundle `python-data-science`:
#. Add the bundle `containers-virt`:
.. code-block:: bash
sudo swupd bundle-add python-data-science
sudo swupd bundle-add containers-virt
.. note::
To add other bundles, replace `python-data-science` with your selected bundle.
To add other bundles, replace `containers-virt` with your selected bundle.
#. When prompted, enter your password.
.. note::
You should see console data similar to the following:
#. Upon successful installation, your console should show similar data:
.. code-block:: console
Password:
Downloading packs...
Extracting python-data-science pack for version 23710
...50%
Extracting python-extras pack for version 23830
...100%
Extracting containers-virt pack for version 24430
...50%
Extracting kernel-container pack for version 24430
...100%
Starting download of remaining update content. This may take a while...
...100%
...100%
Finishing download of update content...
Installing bundle(s) files...
...100%
...100%
Calling post-update helper scripts.
Successfully installed 1 bundle
.. note::
For developers who do not wish to adopt the |CL| Common Tooling Framework (e.g., Autospec, etc.), select the complementary :file:`-dev` bundle in order to successfully build each bundle.
FAQ
===
Find answers to these common questions:
* How do I show all :ref:`bundles available<swupd-guide>`?
* How do I install and *use* :ref:`Kata Containers <kata>` on |CL|?
* How do I :ref:`kata_migration`?
* How do I show all :ref:`bundles available <swupd-guide>`?
* How do I :ref:`update swupd<swupd-guide>`?
* How do I :ref:`remove bundles<swupd-guide>`?
.. note::
For developers who do not wish to adopt the |CL| Common Tooling Framework (e.g., Autospec, etc.), select the complementary :file:`-dev` bundle in order to successfully build each bundle.
.. _Kata Containers: https://clearlinux.org/blogs/clear-linux-os-announces-support-kata-containers
.. _containers virt: https://github.com/clearlinux/clr-bundles/blob/master/bundles/containers-virt