<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Leticia Wanderley on Medium]]></title>
        <description><![CDATA[Stories by Leticia Wanderley on Medium]]></description>
        <link>https://medium.com/@lfwanderley?source=rss-cfaaec1c4dfe------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*E9jOxC9BPs_NWTuIuHIyjw.jpeg</url>
            <title>Stories by Leticia Wanderley on Medium</title>
            <link>https://medium.com/@lfwanderley?source=rss-cfaaec1c4dfe------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 17:18:06 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@lfwanderley/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Bye, Outreachy]]></title>
            <link>https://code.likeagirl.io/bye-outreachy-1db686a5301d?source=rss-cfaaec1c4dfe------2</link>
            <guid isPermaLink="false">https://medium.com/p/1db686a5301d</guid>
            <category><![CDATA[open-source]]></category>
            <category><![CDATA[codelikeagirl]]></category>
            <category><![CDATA[internships]]></category>
            <category><![CDATA[openstack]]></category>
            <category><![CDATA[outreachy]]></category>
            <dc:creator><![CDATA[Leticia Wanderley]]></dc:creator>
            <pubDate>Thu, 31 Aug 2017 00:54:54 GMT</pubDate>
            <atom:updated>2017-09-13T00:21:34.842Z</atom:updated>
            <content:encoded><![CDATA[<h4>My last post as an Outreachy intern</h4><p>This past few months were filled with novelty and learning. I have been able to grasp how it is to work with and to belong to an open source community. I have altogether learned a lot about cloud computing and OpenStack. And, hopefully, made meaningful contributions to this great project.</p><p>More specifically, during the internship I have been able to understand how Keystone uses LDAP and how it works together with the other OpenStack services. From the first task, which was to fix the DevStack LDAP plugin, to the addition of a continuous integration job to execute tests against an LDAP enabled scenario, I was able to understand several OpenStack concepts.</p><h4>There is still work to be done</h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/432/1*AvrnfFwUAOfFYXT_mZO5ig.png" /></figure><p>There are a couple of patches that still need merging. They update some identity tests on Tempest and the CI job structure, to include the tests’ update. This only goes to show how the work in open source is never done, there is always room for improvement. Even though the officially internship agenda is over, I will continue to contribute and to look after those patches.</p><h4>Ack</h4><p>Above all, I have to thank <a href="https://medium.com/u/1e0bff50067d">Raildo Mascena</a> and <a href="https://medium.com/u/3942f85e3a5f">Rodrigo Duarte</a>, my mentors. They’ve been awfully patient and supportive. Always available to answer the most rookie questions I could come up with. I have said this before, but they were pivotal for the fulfillment of the project. I wouldn’t have done anything without them. I am thankful for all the knowledge they shared with me.</p><h4>How to apply</h4><p>To anyone who is reading and thinking about applying to Outreachy,<strong> do it!</strong> You won’t regret it. The applications will open on <a href="https://twitter.com/outreachy/status/900061095666241536">September 7</a>. The participating organizations are not out yet, but you can take a look on the ones the <a href="https://wiki.gnome.org/Outreachy/History#Past_Participating_Organizations">have participated</a> before and start interacting with their communities. It does feel daunting to go ahead and introduce yourself to a lot of people you don’t know, but always keep in mind that those people, and projects, need extra hands and minds to help them. There will always be someone willing to help out (including myself).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/702/1*KJLOHBefMUduHQ2SRK4Acw.png" /><figcaption>Check out their <a href="https://twitter.com/outreachy">twitter feed</a> for more info</figcaption></figure><p>After the participating organizations are announced you can scroll through the projects and find one that looks interesting. Try emailing the mentors talking about how you want to engage with the project. A great thing about Outreachy is that it provides an application template, a set of questions/topics that an applicant needs to complete. You can take a look at my application <a href="https://docs.google.com/document/d/184xBdDi1ng77H5G4D-uBMeJ6OzX-pic3MKHXJOfZRTc/edit?usp=sharing">here</a>.</p><p>In order to be accepted in Outreachy you’ll need to have some code merged on the organization’s open source software. Don’t worry if you don’t get any code in until the application deadline, you can still update your application with the link to your contribution after the deadline has passed. A good tip to start contributing is to go through the project’s bug list and search for the ones marked as “easy” or “beginner”. Another good idea is to contribute by correcting the documentation, that way you can get some lines in whilst learning more about the project by going through its documentation.</p><p>After landing the internship, get in touch with your mentor, schedule a meeting and further discussed what she/he expects you accomplish on the project. You may be tormented with a lot questions, don’t feel afraid of asking them, you are there to learn, after all. When contributing always pay attention to the projects guidelines, how to format a commit message, how write your code… No need to fret if you get something wrong, a good soul will come and lead you in the right direction.</p><p>Last but not least, try and have fun. You are being paid to learn something new, contribute to a piece of software that hundreds of people use everyday and interact with different people from different backgrounds. Your contributions can have major impact on the project and the things you’ll learn have the potential to change the way you interact with software in general.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1db686a5301d" width="1" height="1" alt=""><hr><p><a href="https://code.likeagirl.io/bye-outreachy-1db686a5301d">Bye, Outreachy</a> was originally published in <a href="https://code.likeagirl.io/">Code Like A Girl</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[OpenStack CI Jobs]]></title>
            <link>https://medium.com/@lfwanderley/openstack-ci-jobs-e599f66422b3?source=rss-cfaaec1c4dfe------2</link>
            <guid isPermaLink="false">https://medium.com/p/e599f66422b3</guid>
            <category><![CDATA[continuous-integration]]></category>
            <category><![CDATA[openstack]]></category>
            <category><![CDATA[outreachy]]></category>
            <dc:creator><![CDATA[Leticia Wanderley]]></dc:creator>
            <pubDate>Fri, 18 Aug 2017 14:15:23 GMT</pubDate>
            <atom:updated>2017-08-18T14:15:23.619Z</atom:updated>
            <content:encoded><![CDATA[<h4>In order to have a maintainable, trustable project, all changes proposed have to be tested to provide assurance that the change won’t harm the system instead of improving it. With that in mind, you’ll read about my experience with OpenStack’s continuous integration jobs.</h4><h4><strong>Continuous Integration</strong></h4><p>Continuous integration is the development practice where the contributors’ code is merged into the main repository as often as possible. This assures that all contributors have access to the most up-to-date version of the code. In most cases, including OpenStack’s, having a CI system means having automated tests that run on every change (patch) submitted by a contributor. The result of these tests determines whether the change should/can be added to the main repository.</p><h4><strong>CI on OpenStack</strong></h4><p>OpenStack has a great test suite, it includes unit, style and integration tests, and a platform that deals with the commit-test-merge, aka continuous integration testing, process/infrastructure. Depending on the service in which a contributor proposed a change a distinct part of that suite is executed.</p><p>The definition of when to run a certain suite is done on via the <a href="https://github.com/openstack-infra/project-config">project-config</a> files. These files store the job definitions for every OpenStack service. They know which tests to run and how to configure the environment in which those tests should run.</p><p>For most of the unit and style tests, there’s not a lot of configuration needed, mostly just a Python tox virtualenv running on the service repository in which the change was proposed. However, the integration tests need a somewhat more advanced environment. There needs to be an OpenStack installation up and running, so the integration between the services can be tested. This installation is made using <a href="https://github.com/openstack-dev/devstack">DevStack</a>, a collection of scripts and functions that downloads and deploys the OpenStack services source codes, creating an OpenStack environment. The details about each particular DevStack installation and the tests that should execute on top of it can be found on the project-config jenkins folder, this details are called jobs.</p><h4><strong>My patches</strong></h4><p>My Outreachy project’s goal is to add a new CI job that executes tests on a scenario in which LDAP is enabled as identity back end. To do that there had to be a working DevStack configuration where LDAP was enabled and a job on project-config/jenkins/jobs that not only created a DevStack instance with the aforementioned configuration, but executed tempest test suite on it.</p><p>The DevStack configuration problem was resolved by <a href="https://review.openstack.org/#/c/472482/">fixing</a> the LDAP plugin and thus allowing an LDAP specific domain driver to be enabled and hold an LDAP domain.</p><p>The CI job has to apply the configuration variables that enable LDAP on its DevStack instance. This is done by adding the variables to the job builder definition.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/764/0*ra948kTKVemjdnph." /></figure><p>Then, the job definition needs to specify which tests should run on the environment that is going to be built. This specific case demands that all the test suite is executed because Keystone’s identity service is used throughout OpenStack services and all the changes proposed on it should not get in the way of any of its uses in other services.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*B44x6JlbfA3WH6Tt." /></figure><p>By setting the DEVSTACK_GATE_TEMPEST_FULL variable to “1”, jenkins job builder is told that it should run tempest entire integration test suite.</p><p>With the job definition done the infrastructure system has to know when it should be executed. The OpenStack project which deals with LDAP is Keystone, hence, any change made on Keystone could potentially affect the LDAP domain driver operation. For every patch proposed on Keystone the infrastructure system should execute it on the LDAP enabled scenario.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/0*czvjxexmkIkyYMEC." /></figure><p>Whenever a contributor proposes a change on Keystone, <a href="https://review.openstack.org/">Gerrit</a> (the review system) adds a new event on its stream. When <a href="http://status.openstack.org/zuul/">Zuul</a> (a pipeline used to automate test and merge patches to the source repository) acknowledges the new event it sends a message to jenkins for it to execute the jobs defined for the project. The jobs are defined on project-config’s <a href="https://github.com/openstack-infra/project-config/blob/master/zuul/layout.yaml#L10934">zuul/layout.yaml</a>. They can be of two types, voting and non-voting, if a voting job fails the patch proposed won’t merge until the contributor fix the failure and the job runs again with a pass, a non-voting job failure won’t get in the way of the patch getting merged.</p><p>As far as my contributions go, there’s now a non-voting CI <a href="https://review.openstack.org/#/c/483576/">job</a> that tests the LDAP domain enabled scenario when a change is proposed on Keystone. As this is a fairly new job it’s classified as experimental and only runs when someone adds an “experimental check” on a Gerrit review. After it gets a little bit older and more trustable it can be moved to the “check” pipeline and then become a voting job.</p><h4>References:</h4><ul><li><a href="https://docs.openstack.org/infra/manual/developers.html">https://docs.openstack.org/infra/manual/developers.html</a></li><li><a href="https://docs.openstack.org/infra/manual/creators.html">https://docs.openstack.org/infra/manual/creators.html</a></li><li><a href="http://www.joinfu.com/2014/01/understanding-the-openstack-ci-system/">http://www.joinfu.com/2014/01/understanding-the-openstack-ci-system/</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e599f66422b3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Outreachy project — Where do we stand?]]></title>
            <link>https://medium.com/@lfwanderley/outreachy-project-where-do-we-stand-c18cb36ce5a?source=rss-cfaaec1c4dfe------2</link>
            <guid isPermaLink="false">https://medium.com/p/c18cb36ce5a</guid>
            <category><![CDATA[software-development]]></category>
            <dc:creator><![CDATA[Leticia Wanderley]]></dc:creator>
            <pubDate>Fri, 11 Aug 2017 20:37:28 GMT</pubDate>
            <atom:updated>2017-08-11T20:37:28.306Z</atom:updated>
            <content:encoded><![CDATA[<h4>Updates on the patches and current status of Outreachy project</h4><h4>Some updates on the last post:</h4><ul><li>The CI job path was merged. This means that there’s a non-voting job that runs tests on Keystone’s LDAP enabled scenario.</li><li>The Tempest run returned a few failures on the Users and Groups identity tests (these failures will be discussed on this post).</li></ul><h4>Testing the scenario</h4><p>When the entire Tempest test suite was run, three API identity tests failed. Two of them on the list user test suite and one on the list groups test suite. Those tests are directly related to Keystone’s domains and domains drivers, hence, their failures were directly related to the changes made on the scenario.</p><p>After a little bit of research, the reason for the failures came up. They were not happening just because LDAP domain specific drivers were enabled. They were happening because when any kind of domain specific driver is enabled Keystone’s API only returns user and groups objects from the default domain. That causes the tests which previously expected the API to return all objects available to receive only objects from the default domain, and fail.</p><p>In order to fix that, our plan is to add a feature flag on <a href="https://review.openstack.org/#/c/490739/">DevStack</a> and <a href="https://review.openstack.org/#/c/490735/">Tempest</a> to, when LDAP is enabled, retrieve objects from all the domains on those tests.</p><h4>References</h4><p><a href="https://github.com/openstack/keystone/blob/master/doc/source/configuration.rst">https://github.com/openstack/keystone/blob/master/doc/source/configuration.rst</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c18cb36ce5a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Outreachy update]]></title>
            <link>https://medium.com/@lfwanderley/outreachy-update-d46aac44e50e?source=rss-cfaaec1c4dfe------2</link>
            <guid isPermaLink="false">https://medium.com/p/d46aac44e50e</guid>
            <category><![CDATA[openstack]]></category>
            <category><![CDATA[outreachy]]></category>
            <category><![CDATA[continuous-integration]]></category>
            <category><![CDATA[jenkins]]></category>
            <dc:creator><![CDATA[Leticia Wanderley]]></dc:creator>
            <pubDate>Thu, 03 Aug 2017 13:25:05 GMT</pubDate>
            <atom:updated>2017-08-09T19:21:31.039Z</atom:updated>
            <content:encoded><![CDATA[<h4>First of all, good news! The <a href="https://review.openstack.org/#/c/472482/">patch</a> that fixed DevStack LDAP plugin was merged.</h4><p>This past few weeks I have been focused on testing the new multi-domain with LDAP domain specific driver scenario. I have submitted a <a href="https://review.openstack.org/#/c/483576/">patch</a> adding a CI job on Jenkins that runs tests on a DevStack instance with that configuration. This patch runs Keystone Tempest plugin tests only, which is not the ideal test suite for the scenario. The ideal thing would be to run the entire Tempest test suite, this suite contains all the Users and Groups tests.</p><p>So my job now is to run the entire Tempest test suite, try and fix potential failures, that are bound to happen, using the new scenario and add theses tests to the CI job.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d46aac44e50e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[DevStack LDAP plugin — Part 2]]></title>
            <link>https://medium.com/@lfwanderley/devstack-ldap-plugin-part-2-d9359f8e14df?source=rss-cfaaec1c4dfe------2</link>
            <guid isPermaLink="false">https://medium.com/p/d9359f8e14df</guid>
            <category><![CDATA[outreachy]]></category>
            <category><![CDATA[authentication]]></category>
            <category><![CDATA[openstack]]></category>
            <category><![CDATA[ldap]]></category>
            <dc:creator><![CDATA[Leticia Wanderley]]></dc:creator>
            <pubDate>Mon, 17 Jul 2017 17:07:56 GMT</pubDate>
            <atom:updated>2017-08-09T19:38:19.701Z</atom:updated>
            <content:encoded><![CDATA[<h4>OpenStack Keystone domain with LDAP as identity back end</h4><p>Following up with my previous <a href="https://medium.com/@wanderleylf/devstack-ldap-plugin-part-1-baf3792e7681">post</a>, about LDAP plugin for DevStack, both Keystone and LDAP server services should be running. Now the challenge is to integrate those services and make Keystone access LDAP to retrieve information about users of a specific domain. The goal here is to add another domain to Keystone, a domain that uses LDAP as identity back end.</p><h4>Enabling multiple domains</h4><p>First of all, let’s enable domain specific drives and multiple domains on both Keystone and Horizon. To do this add these two new lines under identity on <em>etc/keystone.conf</em></p><pre>[identity]<br>domain_specific_drivers_enabled = True<br>domain_config_dir = /etc/keystone/domains</pre><p>And, to enable multiple domains on Horizon, add the following line to <em>horizon/openstack_dashboard/local/local_settings.py</em>.</p><pre>OPENSTACK_KEYSTONE_MULTIDOMAIN_SUPPORT=True</pre><h4>Configuring LDAP domain</h4><p>Using OpenStackClient create a new Keystone domain by running the command</p><pre>openstack — os-identity-api-version=3 domain create &lt;your-domain-name&gt;</pre><p>Now to configure your new domain add a new folder to<em> etc/keystone</em> called “domains”. Inside this folder add a conf file with your LDAP domain configuration. The file should be named <em>keystone.&lt;your-domain-name&gt;.conf </em>and contain the following lines</p><pre>[identity]<br>driver = ldap</pre><pre>[ldap]<br>url = ldap://localhost<br>user = cn=Manager,dc=openstack,dc=org<br>password = &lt;password&gt;<br>suffix = dc=openstack,dc=org<br>user_id_attribute = uid<br>user_mail_attribute = mail<br>user_name_attribute = cn<br>user_objectclass = inetOrgPerson<br>user_tree_dn = ou=Users,dc=openstack,dc=org<br>group_id_attribute = cn<br>group_name_attribute = cn<br>group_objectclass = groupOfNames<br>group_tree_dn = ou=Groups,dc=openstack,dc=org<br>user_allow_create = False<br>user_allow_update = False<br>user_allow_delete = False<br>group_allow_create = False<br>group_allow_update = False<br>group_allow_delete = False</pre><p>The domain component (dc), organization unit (ou) and common name (cn) configured should match the ones added to slapd by the <a href="https://github.com/openstack-dev/devstack/blob/master/lib/ldap#L104">ldap plugin</a>. (You can check those values running an <a href="http://www.openldap.org/software/man.cgi?query=ldapsearch&amp;apropos=0&amp;sektion=0&amp;manpath=OpenLDAP+2.0-Release&amp;format=html">ldapsearch</a>). As a LDAP domain should be read-only, we should disallow all user and group write actions (creation, update and deletion).</p><p>So far we have an extra domain on keystone and a domain configuration file inside <em>etc/keystone/domains </em>which sets the variables of a LDAP domain. It is important that the name of the domain created and the middle part of the file name are the same.</p><h4>Binding Keystone and LDAP</h4><p>Now we need to tell Keystone about the configuration file for its newly created domain. To do that we restart apache and keystone services, this should be enough to bind the domain and the LDAP server (slapd).</p><pre>sudo service apache2 reload<br>sudo systemctl restart devstack@keystone</pre><p>To test if everything went fine let’s add a new user to the LDAP service and check if keystone can retrieve it as one of its users. The process of adding a user will be done entirely on the LDAP server because, as mentioned before, the LDAP identity back end on Keystone is read-only.</p><p>Create a ldif file to hold your new user’s information, it can be called <em>user.ldif</em>, for example. Inside that file add these lines</p><pre>dn: cn=demo,ou=Users,dc=openstack,dc=org<br>cn: demo<br>displayName: demo<br>givenName: demo<br>mail: <a href="/service/mailto:demo@openstack.org">demo@openstack.org</a><br>objectClass: inetOrgPerson<br>objectClass: top<br>sn: demo<br>uid: demo<br>userPassword: demo</pre><p>Then run the following command</p><pre>ldapadd -x -w &lt;password&gt; -D “cn=Manager,dc=openstack,dc=org” -H ldap://localhost -c -f user.ldif</pre><p>This adds a new user called “demo” to LDAP. Now let’s check if Keystone can see the new user by running the command</p><pre>openstack user list --domain &lt;your-domain-name&gt;</pre><p>After trying to run this you might be prompt an error message about the need of authentication. If that’s the case, try and login to Horizon, download the <em>OpenStack RC File v3 </em>file and load it using source on your current shell window. Then, rerun the command above.</p><p>If you can see your user listed that means the integration was successful and you have got a domain with LDAP as identity back end.</p><h4>Using Horizon</h4><p>The steps above allow you to add users to your Keystone LDAP domain. In order to interact with those users on Horizon they need to be assigned to a project and, hence, have a specific role.</p><p>Projects and roles should not be stored on LDAP, that’s only for users and groups. So we will assign the users on LDAP to SQL projects. (Note that the project should be created on or retrieved from the default SQL domain.)</p><pre>project=openstack project create &lt;project_name&gt; --domain=default --or-show -f value -c id</pre><pre>user=openstack user show --domain=&lt;your-domain-name&gt; demo -f value -c id </pre><pre>role=openstack role create &lt;role_name&gt; --or-show -f value -c id</pre><pre>openstack role add $role --user $user --project $project</pre><p>After running these commands you should be able to login as the LDAP demo user on your domain on Horizon.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/481/1*R0ShcMSBhSAvYtHNYwrB1g.png" /></figure><h4>References:</h4><ul><li><a href="https://www.ibm.com/developerworks/cloud/library/cl-ldap-keystone/index.html">Enable the Keystone LDAP back end in OpenStack</a></li><li><a href="http://adam.younglogic.com/2015/02/adding-an-ldap-backed-domain-to-a-packstack-install/">Adding LDAP backed domain</a></li><li><a href="https://docs.openstack.org/tripleo-docs/latest/install/advanced_deployment/domain_specific_ldap_backends.html">Domain specific LDAP Backends</a></li><li><a href="https://www.slideshare.net/JessePretorius/openstack-keystone-with-ldap">Keystone with LDAP</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d9359f8e14df" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[DevStack LDAP plugin — Part 1]]></title>
            <link>https://medium.com/@lfwanderley/devstack-ldap-plugin-part-1-baf3792e7681?source=rss-cfaaec1c4dfe------2</link>
            <guid isPermaLink="false">https://medium.com/p/baf3792e7681</guid>
            <category><![CDATA[openstack]]></category>
            <category><![CDATA[ldap]]></category>
            <category><![CDATA[ubuntu]]></category>
            <category><![CDATA[outreachy]]></category>
            <dc:creator><![CDATA[Leticia Wanderley]]></dc:creator>
            <pubDate>Mon, 10 Jul 2017 18:01:08 GMT</pubDate>
            <atom:updated>2017-08-09T19:31:38.063Z</atom:updated>
            <content:encoded><![CDATA[<p>As promised my second post will be about my first task as an Outreachy intern. The project I am part of is about adding functional tests for a scenario where Keystone, OpenStack Identity Service, is running with LDAP as identity backend. On that scenario LDAP is the backend for one of the domains storing Keystone users.</p><blockquote>LDAP stands for Lightweight Directory Access Control, it is a data access protocol to interact with data contained in directory servers.</blockquote><p>In order to get familiar with OpenStack, keystone and LDAP, the usual initial step is to use LDAP as identity backend on DevStack. DevStack is a script that creates an OpenStack development environment and it is used to demonstrate interaction with OpenStack services. The only problem was that the LDAP plugin was broken, no one could automatically use an LDAP backend on DevStack. And there it was, my first task.</p><h3><strong>Fixing the LDAP plugin in DevStack</strong></h3><p>After lots of research &amp; re-runs of DevStack, it became clear that the main issue was that when using Ubuntu the directory server was not being started. So all of the interactions with the server that were scheduled on the script could not happen and exited execution with an error.</p><pre>SASL/EXTERNAL authentication started<br>SASL username: gidNumber=0+uidNumber=0,cn=peercred,cn=external,cn=auth<br>SASL SSF: 0<br>ldap_modify: No such object (32)<br>matched DN: cn=config<br>modifying entry &quot;olcDatabase={1}hdb,cn=config&quot;<br><br>+lib/ldap:install_ldap:1                   exit_trap<br>Error on exit</pre><p>DevStack uses <strong>OpenLDAP slapd</strong> as directory server, and the LDAP plugin script assumed that slapd was already running on Ubuntu, so it didn’t try to start it again.</p><pre>function install_ldap {</pre><pre>    ...</pre><pre>    printf &quot;installing OpenLDAP&quot;</pre><pre>    if is_ubuntu; then</pre><pre>        <strong># Ubuntu automatically starts LDAP so no need to call start_ldap()</strong></pre><pre>        :</pre><pre>    ...<br>}</pre><p>As it turned out that the slapd service was <strong>not</strong> running and it also needed to be configured. The configuration looks something like this:</p><pre>sudo debconf-set-selections &lt;&lt;EOF</pre><pre>    slapd slapd/internal/generated_adminpw password &lt;password&gt;</pre><pre>    slapd slapd/internal/adminpw password &lt;password&gt;</pre><pre>    slapd slapd/password2 password &lt;password&gt;</pre><pre>    slapd slapd/password1 password &lt;password&gt;</pre><pre>    slapd slapd/dump_database_destdir string /var/backups/slapd-VERSION</pre><pre>    slapd slapd/domain string &lt;keystone_domain&gt;</pre><pre>    slapd shared/organization string &lt;ldap_organization_domain&gt;</pre><pre>    slapd slapd/backend string HDB</pre><pre>    slapd slapd/purge_database boolean true</pre><pre>    slapd slapd/move_old_database boolean true</pre><pre>    slapd slapd/allow_ldap_v2 boolean false</pre><pre>    slapd slapd/no_configuration boolean false</pre><pre>    slapd slapd/dump_database select when needed</pre><pre>EOF</pre><pre>    sudo apt-get install -y slapd ldap-utils</pre><pre>    sudo dpkg-reconfigure -f noninteractive slapd</pre><p>Setting those variables, installing and configuring those services should be enough to get slapd up and running. To check the status of the slapd service execute the following command on the command line.</p><pre>sudo service slapd status</pre><p>After configuring the slapd service on Ubuntu the DevStack LDAP plugin was able to access the directory server and add entries to it. This means that by that point there should be an LDAP service and a Keystone service running. The next thing to do is to integrate those two services and have a Keystone domain accessing LDAP to retrieve its user’s information.</p><h4>A few useful links:</h4><ul><li><a href="https://nishayadavsite.wordpress.com/2016/06/29/how-to-set-up-work-environment-to-become-an-openstack-developer/">On setting up DevStack</a></li><li><a href="https://stackoverflow.com/questions/28937889/how-to-do-an-initial-setup-of-slapd-olc-with-ldapmodify">On configuring OpenLDAP slapd</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=baf3792e7681" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Welcome to Outreachy]]></title>
            <link>https://medium.com/@lfwanderley/welcome-to-outreachy-5283e79b7961?source=rss-cfaaec1c4dfe------2</link>
            <guid isPermaLink="false">https://medium.com/p/5283e79b7961</guid>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[openstack]]></category>
            <category><![CDATA[women-in-tech]]></category>
            <category><![CDATA[internships]]></category>
            <category><![CDATA[outreachy]]></category>
            <dc:creator><![CDATA[Leticia Wanderley]]></dc:creator>
            <pubDate>Mon, 12 Jun 2017 19:12:41 GMT</pubDate>
            <atom:updated>2017-08-09T19:38:47.828Z</atom:updated>
            <content:encoded><![CDATA[<p>Hi everyone,</p><p>On my very first post I’ll talk a bit about the Free and Open Source Software Outreachy internship program and my Openstack Outreachy project.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/702/1*KJLOHBefMUduHQ2SRK4Acw.png" /></figure><h3>First of all: <a href="https://gnome.org/outreachy/">Outreachy</a></h3><p>This program is all about getting people involved with Open Source Software, especially folks that are underrepresented on the whole technology field. However, the projects available are not restricted to coding, one can help the community by working with documentation, translation, marketing or design, not a single line of code involved.</p><p>A parallel can be draw between <a href="https://developers.google.com/open-source/gsoc/">Google Summer of Code</a> and Outreachy, the latter was actually inspired by the small number of female and minorities that applied for the previous.</p><p>An Outreachy intern is given the opportunity of being paid for contributing on an open source software. She/he applies to work on a project, submitted by Outreachy mentors, and, if accepted, works remotely for 3 months to develop that project. In order to complete their projects, interns receive guidance from the mentors that submitted the projects (and from the whole FOSS community!). The intern is expected to work full-time on the project. Outreachy projects are submitted by mentors that work on the participating organizations and are directly involved with the FOSS community.</p><p>Currently, there are two internship entries, one between May and August and another between December and March.</p><h3>Landing an internship</h3><p>To land an internship the prospective intern has to choose one or more projects from the participating organizations and send in applications for those projects, one application per project. The application template is already defined, which helps a lot when you are getting started. And the intern can edit the application after it is sent, so new things can be added if needed be.</p><p>Another requirement is that the candidate has to have merged code into the open source software she/he wants to work with, this merged patch has to be disclosed on the application. And it is important, but not mandatory, that the candidate reaches out to hers/his project mentor. It can be just to say ‘Hello’, but the mentor can provide useful information about the project and help with the application.</p><p>When I was applying I was lucky enough to have guidance from my now-mentors and I also found another Outreachy blog in which there was a link to an application that was accepted, it was crucial to my own application writing. For that reason, here’s a <a href="https://docs.google.com/document/d/184xBdDi1ng77H5G4D-uBMeJ6OzX-pic3MKHXJOfZRTc/edit?usp=sharing">link</a> to my application, I hope it helps some future applicant (good luck!).</p><h3>My Outreachy Project</h3><p>My project’s title is “Add continuous integration and tests for OpenStack Identity Service (keystone) LDAP support”, I’ll be working with <a href="https://www.openstack.org/software/releases/ocata/components/keystone">Keystone</a>, OpenStack’s identity service.</p><h4>For starters, a few definitions</h4><ul><li>“OpenStack is a cloud operating system that controls large pools of compute, storage, and networking resources throughout a datacenter, all managed through a dashboard that gives administrators control while empowering their users to provision resources through a web interface.”</li><li>“Keystone is an OpenStack service that provides API client authentication, service discovery, and distributed multi-tenant authorization by implementing OpenStack’s Identity API. It supports LDAP, OAuth, OpenID Connect, SAML and SQL.”</li><li>“The Lightweight Directory Access Protocol (LDAP) is an open, standards-based, mature, efficient, extensible, and popular means of interacting with the data contained in directory servers.”</li></ul><p>The projects aims to add functional tests for the scenario in which Keystone is used with LDAP as identity back end in order to add that scenario, along the tests, to a continuous integration job. The identity back end is responsible for the authentication, it validates credentials to provide or deny access to certain resources.</p><p>My first task within the project is to find out that’s wrong with, and hopefully fix, <a href="https://wiki.openstack.org/wiki/DevStack">devstack</a> ldap plugin, but hold on tight, this will be my next post subject.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5283e79b7961" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>