<?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[Hepta Analytics Blog - Medium]]></title>
        <description><![CDATA[Stories about going beyond information. - Medium]]></description>
        <link>https://medium.com/heptanalytics?source=rss----a336fb0c6921---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>Hepta Analytics Blog - Medium</title>
            <link>https://medium.com/heptanalytics?source=rss----a336fb0c6921---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 21:26:13 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/heptanalytics" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Cybersecurity Foundations: Application Security — My 2nd LinkedIn Learning Course]]></title>
            <link>https://medium.com/heptanalytics/cybersecurity-foundations-application-security-my-2nd-linkedin-learning-course-b95d50e352b7?source=rss----a336fb0c6921---4</link>
            <guid isPermaLink="false">https://medium.com/p/b95d50e352b7</guid>
            <category><![CDATA[free-course]]></category>
            <category><![CDATA[application-security]]></category>
            <category><![CDATA[linkedin-learning]]></category>
            <category><![CDATA[online-courses]]></category>
            <category><![CDATA[appsec]]></category>
            <dc:creator><![CDATA[Emmanuel Chebukati]]></dc:creator>
            <pubDate>Fri, 07 Jul 2023 20:47:17 GMT</pubDate>
            <atom:updated>2023-07-07T20:46:32.698Z</atom:updated>
            <content:encoded><![CDATA[<h3>Cybersecurity Foundations: Application Security — My 2nd LinkedIn Learning Course</h3><p>I am excited to announce the release of my 2nd LinkedIn Learning course titled Cybersecurity Foundations: Application Security. Check it out the link below:</p><p><a href="https://www.linkedin.com/learning/cybersecurity-foundations-application-security">Cybersecurity Foundations: Application Security Online Class | LinkedIn Learning, formerly Lynda.com</a></p><p>This demonstrative beginner course covers the essentials that those entering the cybersecurity field ought to know to have a solid understanding of application security.</p><p>In designing the course, I wanted it to be fresh and current so I used the <a href="https://www.hackerone.com/resources/i/1487910-2022-hacker-powered-security-report-q4fy23/0?">top 10 most reported vulnerabilities of 2022</a> as the baseline for the core chapters. This way, technical learners who successfully completed the course would be able to fit right into a role that required them to identify, triage, and resolve common security incidents. I also added a chapter on current trends in the appsec space featuring demonstrations on WAFs, DevSecOps, and training platforms.</p><p>To keep the course somewhat entertaining, I tried not to get too technical with the explanations. At every step of the scripting phase I pondered, “if this video was dubbed onto TikTok, would it stand-alone, be engaging, and make sense to my parents?”. If not, I worked harder on the demos, and softned the use of technical terms. In the end, I believe that I arrived at a good balance.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ef6l3WQBK8xhVCIighvV3w.jpeg" /></figure><p>Unlike my <a href="https://medium.com/heptanalytics/check-out-my-first-course-on-linkedin-learning-security-in-fintech-essential-training-4b56bc379b68">previous LinkedIn Learning course</a>, this one was fairly easy to record because I already knew how all the equipment and software worked. The only hiccup was a laptop switch mid-recording which delayed the anticipated course release. Happy retirement, my legendary Microsoft Surface Book!</p><p>I hope all cybersecurity entrants — and even those looking to refresh their knowledge of current vulnerabilities — enjoy taking the course. I look forward to any comments, questions or feedback you may have about my new course.</p><p>It’s free on LinkedIn Learning with the link above. Spread the word to teach others about Cybersecurity Foundations in Application Security!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b95d50e352b7" width="1" height="1" alt=""><hr><p><a href="https://medium.com/heptanalytics/cybersecurity-foundations-application-security-my-2nd-linkedin-learning-course-b95d50e352b7">Cybersecurity Foundations: Application Security — My 2nd LinkedIn Learning Course</a> was originally published in <a href="https://medium.com/heptanalytics">Hepta Analytics Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Running an NGINX Ingress Controller for each Kubernetes Namespace]]></title>
            <link>https://medium.com/heptanalytics/running-an-nginx-ingress-controller-for-each-kubernetes-namespace-b006a4326467?source=rss----a336fb0c6921---4</link>
            <guid isPermaLink="false">https://medium.com/p/b006a4326467</guid>
            <category><![CDATA[azure]]></category>
            <category><![CDATA[kubernetes]]></category>
            <category><![CDATA[cloud]]></category>
            <category><![CDATA[infrastructure]]></category>
            <dc:creator><![CDATA[Emmanuel Chebukati]]></dc:creator>
            <pubDate>Fri, 28 Jan 2022 12:43:29 GMT</pubDate>
            <atom:updated>2022-01-28T12:43:29.454Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*S2A-hY0Bcl_OiWz8.png" /></figure><p>You may find yourself needing to deploy multiple NGINX Ingress Controllers to serve each namespace on your Kubernetes cluster. This may be useful in a scenario where you have multiple client deployments on the same K8S cluster; and you want to assign a public load balancer IP address for each client to achieve logical separation. This blogpost explores how to do that.</p><p>We shall be using the Azure Kubernetes Service (AKS) environment to demonstrate the solution, although these commands should work across the major cloud providers. For the NGINX Ingress controller’s deployment, we shall use Helm 3: The package manager for Kubernetes. Note that we shall be using the NGINX <a href="https://github.com/kubernetes/ingress-nginx">community version</a> (<strong>ingress-nginx</strong>) which is maintained by the Kubernetes community, and not the <a href="https://github.com/nginxinc/kubernetes-ingress">NGINX version</a> (<strong>nginx-stable</strong>) which is maintained by F5 NGINX. You can only use helm3 for the community version deployment, and Kubernetes &gt;= 1.19 is also required.</p><p>Currently on my cluster are two deployments called “echello” in two different namespaces: <strong>main</strong> and <strong>client1</strong>. There is a main ingress controller in the “main-ingress” namespace with an external public IP already assigned.</p><figure><img alt="AKS Services and Ingress currently deployed." src="https://cdn-images-1.medium.com/max/1024/0*eFsfTUbvLvn_TKEs.png" /></figure><p>I now want to deploy another ingress controller to serve the resources of client1 in the client1 namespace. The following basic deployment creates the new ingress controller for <strong>client1</strong>:</p><pre>RELEASE_NAME=<strong>client1</strong>-ingress<br><br>helm repo add ingress-nginx https://kubernetes.github.io/ingress-nginx<br>helm repo update<br><br>helm install $RELEASE_NAME ingress-nginx/ingress-nginx  \<br>  --namespace $RELEASE_NAME --create-namespace \<br>  --set controller.ingressClassByName=true \<br>  --set controller.ingressClassResource.name=$RELEASE_NAME \<br>  --set controller.ingressClassResource.controllerValue=&quot;k8s.io/$RELEASE_NAME&quot; \<br>  --set controller.ingressClassResource.enabled=true \<br>  --set controller.replicaCount=2</pre><p>The above deployment takes about three minutes to run. In the meantime, let’s explore each block of the above command:</p><ul><li>Give the deployment unique RELEASE_NAME.</li><li>Add the correct chart repository, then update information of available charts locally.</li><li>Install the ingress-nginx chart and give it a unique RELEASE_NAME (i.e. client1-ingress)</li><li>Set the deployment in its own client-specific namespace for ingress controllers, and create the K8S namespace if it doesn’t exist.</li><li>Set the controller to process IngressClass per name to true. The default is “false”.</li><li>Set the name of the ingressClass. The default is “nginx”, but because this has to be unique per-controller, we set it as our $RELEASE_NAME here.</li><li>Set the controller-value of the controller that is processing this ingressClass. The default is “k8s.io/ingress-nginx”, but this ought to be unique per-deployment so we use $RELEASE_NAME.</li><li>Enable this ingressClass. The default is “true”, but let’s set it just to be sure.</li><li>Set the controller pods to 2. The default is 1, but it is advisable to run at least two ingress controllers for higher availability.</li></ul><p>When you run the above, you get a shiny new ingress controller in the <strong>client1-ingress</strong> namespace ready to serve clients:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Mt18XyuD2oeZMqCr.png" /></figure><p>To deploy a third ingress controller to serve another namespace (e.g. <strong>client2</strong>), simply modify the $RELEASE_NAME variable, and then run the helm install command again.</p><p>Now let’s proceed to look at how the ingress route file will look like. Remember to apply this in the appropriate client’s namespace (e.g. client1) so that it can “see” the chosen service:</p><pre>apiVersion: networking.k8s.io/v1<br>kind: Ingress<br>metadata:<br>  name: <strong>client1-ingress</strong><br>  annotations:<br>    nginx.ingress.kubernetes.io/ssl-redirect: &quot;false&quot;<br>    nginx.ingress.kubernetes.io/use-regex: &quot;true&quot;<br>spec:<br>  <strong>ingressClassName: client1-ingress</strong><br>  rules:<br>  - host: <strong>client1.domain.co</strong><br>    http:<br>      paths:<br>      - pathType: Prefix<br>        path: /<br>        backend:<br>          service:<br>            name: echello<br>            port:<br>              number: 80</pre><p>The key thing to note from the above is the spec of ingressClassName is set to our RELEASE_NAME (i.e. client1-ingress). This is a <a href="https://kubernetes.io/blog/2020/04/02/improvements-to-the-ingress-api-in-kubernetes-1.18/">newer way</a> of doing things as opposed to using the “kubernetes.io/ingress.class: client1-ingress” annotation in the past.</p><p>When you apply the file, you will now notice two Ingresses on the Azure portal; each pointing a different domain to a different IP address.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*NBIwFYqdiM58t2xz.png" /></figure><p>Finally, set the DNS A-record to point to the new IP, and you should be good to go!</p><p>Note: If you make a mistake with a setting on the helm chart, you may end up with a scenario where you have two different ingress controllers deployed on two different public IP addresses, but both the Ingresses on the same public IP as shown below:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*I-rGWYPGrRBMGCYp.png" /></figure><p>This does not achieve the desired logical separation. To fix this, run:</p><pre>helm uninstall $RELEASE_NAME --namespace=$RELEASE_NAME</pre><p>Afterwards, re-install the ingress controller with all the correct parameters as specified above.</p><p>This concludes this tutorial on running multiple NGINX Ingress Controllers, one for each Kubernetes Namespace. While there are other settings you can include, this basic setup should get you started. I’ve since released the IP addresses associated with this tutorial back to Azure.</p><p>I spent a lot longer than I should’ve setting this up given all the variations and changes, so I wrote this blogpost to save you from suffering as I did.</p><p>All the best with your deployments!</p><p>References:</p><ol><li><a href="https://kubernetes.github.io/ingress-nginx/">https://kubernetes.github.io/ingress-nginx/</a></li><li><a href="https://github.com/kubernetes/ingress-nginx/tree/main/charts/ingress-nginx">https://github.com/kubernetes/ingress-nginx/tree/main/charts/ingress-nginx</a></li></ol><p><em>Originally published at </em><a href="https://blog.heptanalytics.com/running-an-nginx-ingress-controller-for-each-kubernetes-namespace/"><em>Hepta Analytics Blog</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b006a4326467" width="1" height="1" alt=""><hr><p><a href="https://medium.com/heptanalytics/running-an-nginx-ingress-controller-for-each-kubernetes-namespace-b006a4326467">Running an NGINX Ingress Controller for each Kubernetes Namespace</a> was originally published in <a href="https://medium.com/heptanalytics">Hepta Analytics Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Check out my first course on LinkedIn Learning: Security in Fintech — Essential Training]]></title>
            <link>https://medium.com/heptanalytics/check-out-my-first-course-on-linkedin-learning-security-in-fintech-essential-training-4b56bc379b68?source=rss----a336fb0c6921---4</link>
            <guid isPermaLink="false">https://medium.com/p/4b56bc379b68</guid>
            <category><![CDATA[cloud]]></category>
            <category><![CDATA[infrastructure]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[cloud-computing]]></category>
            <dc:creator><![CDATA[Emmanuel Chebukati]]></dc:creator>
            <pubDate>Tue, 04 Jan 2022 14:19:48 GMT</pubDate>
            <atom:updated>2022-01-04T14:27:09.268Z</atom:updated>
            <content:encoded><![CDATA[<p>Today my first <a href="https://www.linkedin.com/learning/">LinkedIn Learning</a> course on securing fintech solutions went live! Check out the link below:</p><p><a href="https://www.linkedin.com/learning/embed/security-in-fintech-essential-training/securing-fintech-solutions?autoplay=false&amp;claim=AQE3YJ5v7P9qwAAAAX4lX6polJueo6ordzCgeScowkrQsFbBhWWCSa1cJhFWKswnth4sTgkkWklEIRy50EnohpGeqm9X3cLNBRVZauOC5Y3BeHrxwm5qLCxzuNbj6nKCFTczWDSiaVdYMn3k0XX_5_ITVnVpPoSCgwfcBSwg65oHcGF-7MKKdy4ThBrbz4FFc3SymWTwoyc-jI1Q94P8leZJELApb-V3tb97Ca0E72AT4dHrzj80cd6ObLGk1Fexs6unGk2PzR2DZcYfU1Yna0Wgb3ctmk-EOStjwzQEIRhzINxl1pxZ6vZMBch9YRTx0VXan1U6jtevHGJCWpLM38Z3KOknVmFGd6adMglT7LBMdR9MNOmoVK0pG3dZkL_DMssFd6_zMhpenwCmFWAdqMYPhSydJaGlog8ULilSEdcwP-Pfnwu2UQSY3GypdT2pm6zaaWq4DiGi8imR2tMfxX_qO5rTTIzISZ5mfU_HITRhsCvkAZfN_8HqYYpwSeYjggj48L4kSeX5-3Uozov--0tI5jHc-NdXP6sfZl9zh8iUrLAhvDM3e8_ZGkDLyvOdMc2DXgT_arJm9EP_R2zVSfoEybV6LTAdVm7c0OvW77r9HLxoQQqoYdBzjVz78Or6dyKygJcXvtUYDgHKT25bclZfYCupmUrn0JrL_rQAXPorhSSe-I3MyAokXjjjF-q1dDzC3qN5BzrhnaZyDnEkX17PpRMrl4RgMlVlCRYTvu0YaDajSUc&amp;lipi=urn%3Ali%3Apage%3Ad_learning_content%3B0EqycSJqQV%2BUI12stXS9qw%3D%3D&amp;licu">Securing fintech solutions - Amazon Web Services Video Tutorial | LinkedIn Learning, formerly Lynda.com</a></p><p><a href="https://www.linkedin.com/learning/security-in-fintech-essential-training/securing-fintech-solutions?trk=embed_lil"><strong>Securing fintech solutions</strong></a> from <a href="https://www.linkedin.com/learning/security-in-fintech-essential-training?trk=embed_lil"><strong>Security in Fintech Essential Training</strong></a> by <a href="https://www.linkedin.com/learning/instructors/emmanuel-chebukati?trk=embed_lil"><strong>Emmanuel Chebukati</strong></a></p><p>It was an exciting surprise to wake up to the notifications of the course’s release, and to see the initial reactions it elicited.</p><p>This demonstrative course covers the essentials that fintech providers and professionals in the industry ought to implement to arrive at a baseline security posture.</p><p>In designing the course, I leveraged my experience working with our fintech clients at <a href="https://www.heptanalytics.com/">Hepta Analytics</a>, replenished my knowledge on current best-practices and trends, and borrowed lessons from the experience of building and securing our own fintech solution <a href="https://heptapay.com/">HeptaPay</a>.</p><p>The result was <strong>17 nascent slides</strong>, <strong>49 media files</strong>, and over <strong>4 hours</strong> of raw footage, which were all beautifully packaged into a professional <strong>2h</strong> <strong>12m </strong>course by the fantastic post-production team at LinkedIn.</p><p>I hope you enjoy learning; and I look forward to any comments, questions or feedback you may have about my course.</p><p><em>Originally published at </em><a href="https://blog.heptanalytics.com/check-out-my-first-course-on-linkedin-learning-security-in-fintech-essential-training/"><em>Hepta Analytics Blog</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4b56bc379b68" width="1" height="1" alt=""><hr><p><a href="https://medium.com/heptanalytics/check-out-my-first-course-on-linkedin-learning-security-in-fintech-essential-training-4b56bc379b68">Check out my first course on LinkedIn Learning: Security in Fintech — Essential Training</a> was originally published in <a href="https://medium.com/heptanalytics">Hepta Analytics Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Connecting a Linux VPS to an AWS VPC using a S2S VPN with Static Routing]]></title>
            <link>https://medium.com/heptanalytics/connecting-a-linux-vps-to-an-aws-vpc-using-a-s2s-vpn-with-static-routing-f45eaf3f1191?source=rss----a336fb0c6921---4</link>
            <guid isPermaLink="false">https://medium.com/p/f45eaf3f1191</guid>
            <category><![CDATA[s2s-vpn]]></category>
            <category><![CDATA[aws-vpc]]></category>
            <category><![CDATA[racoon]]></category>
            <category><![CDATA[aws]]></category>
            <dc:creator><![CDATA[Emmanuel Chebukati]]></dc:creator>
            <pubDate>Mon, 27 Sep 2021 16:33:56 GMT</pubDate>
            <atom:updated>2021-09-27T16:33:56.089Z</atom:updated>
            <content:encoded><![CDATA[<p>This blogpost will cover how to connect a standalone Virtual Private Server (VPS) running Linux (specifically Debian) to AWS’ Virtual Private Cloud (VPC) using a site-to-site VPN with Static Routing.</p><p>This blogpost is relevant for those who find themselves having to integrate their AWS infrastructure with external sites where they do not own, or do not have permission to configure the gateway device (e.g. a Cisco ASA appliance).</p><p>In a <a href="https://blog.heptanalytics.com/connecting-a-linux-server-vps-to-aws-secure-virtual-private-cloud/">previous blogpost</a>, I broke down how to do this with dynamic routing (BGP). In this blogpost, we’ll explore the same with static routing which is even easier.</p><p>This blogpost assumes you have an AWS <a href="https://docs.aws.amazon.com/vpc/latest/userguide/VPC_Scenario2.html">VPC with public and private subnets (NAT)</a> which comes with the following elements in place:</p><ul><li>A VPC with a size /16 IPv4 CIDR block (10.0.0.0/16).</li><li>A public subnet with a size /24 IPv4 CIDR block (10.0.0.0/24).</li><li>A private subnet with a size /24 IPv4 CIDR block (10.0.1.0/24).</li><li>Two instances with private IPv4 addresses in the public and private subnets (10.0.0.10 and 10.0.1.10 respectively). The public instance also has a public IP.</li><li>A custom route table associated with the public subnet. This route table contains an entry that enables instances in the subnet to communicate with other instances in the VPC over IPv4.</li><li>The main route table associated with the private subnet. The route table contains an entry that enables instances in the subnet to communicate with other instances in the VPC over IPv4.</li><li>A instance / Virtual Private Server (VPS) in a different VPC (or even a different cloud provider) with a private IP range that doesn’t clash with 10.0.0.0/16. You should opt for an “Elastic”/Fixed Public IP address so that the gateway IP doesn’t change often.</li></ul><p>With those elements in place, let’s proceed to create a new customer gateway for our integration. A customer gateway is a resource that you create in AWS that represents the customer gateway device in your on-premises network. When you create a customer gateway, you provide information about your device to AWS.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/694/0*r53EKEV1U3fxQURI.png" /></figure><p>Click the “Customer Gateways” link under the “Virtual Private Network” tab of the VPC Console then click “Create Customer Gateway”.</p><figure><img alt="Create customer gateway AWS" src="https://cdn-images-1.medium.com/max/1024/0*SbB6F5dfsmtwtmgv.png" /></figure><p>Name the customer gateway appropriately (e.g. MyVPS_CG). Set the routing as static. Specify the IP address of the device (this is the public IP address of your VPS). Accept the rest of the defaults, then “Create Customer Gateway”. The operation takes about one minute.</p><p>Now we can create a Virtual Private Network (VPN) between the two sites for secure communication over private addressing. Click “Site-to-site VPNs” under the “Virtual Private Network” tab on the left pane then click “Create VPN Connection”.</p><figure><img alt="Creating a site-to-site VPN on AWS with static routing" src="https://cdn-images-1.medium.com/max/1024/0*iV1XejFJJIWZPztB.png" /></figure><p>Name the VPN appropriately (e.g. MyVPS_VPN), select your VPG, select your existing Customer Gateway, and set the routing as static. Then enter the internal IP prefixes that you will be communicating with on your secondary site (e.g. 172.31.93.1/32). Keep this range as small as is necessary for security purposes. You may set other options such as PSKs, but we’ll go with the defaults. Then click “Create VPN Connection”. The deployment will take about 5 minutes, and it will create two tunnels for you. One primary, and one secondary, both connecting to the same infrastructure.</p><p>Once the deployment is done, we can now download the configuration file which we will use to configure the VPN between our local site, and AWS. Click “Download Configuration”. We’ll go with the generic vendor, a generic platform, and a vendor agnostic software option.</p><figure><img alt="Download AWS Site to site VPN configuration" src="https://cdn-images-1.medium.com/max/1024/0*bz4e8kcl8jG-5x6y.png" /></figure><p>Here are the configs you need to take note of when the download is complete. Copy these configurations into your text editor and replace the addresses according to your download file:</p><pre>Tunnel 1<br>- Pre-Shared Key : AAAA<br>Outside IP Addresses:<br> - Customer Gateway : 1.1.1.1 <br> - Virtual Private Gateway : 2.2.2.2<br>Inside IP Addresses<br> - Customer Gateway : your.vps.private.ip/32<br> - Virtual Private Gateway : 10.0.0.0/16<br><br>Tunnel 2<br>- Pre-Shared Key : BBBB<br>Outside IP Addresses:<br> - Customer Gateway : 1.1.1.1 <br> - Virtual Private Gateway : 3.3.3.3<br>Inside IP Addresses<br> - Customer Gateway : your.vps.private.ip/32<br> - Virtual Private Gateway : 10.0.0.0/16</pre><h4>The Host Setup</h4><p>Next, log on to your <strong>Linux VPS Instance</strong> (i.e. the customer gateway) as <strong>root</strong> and begin by installing a few packages:</p><pre>apt-get update<br>apt-get install ipsec-tools racoon</pre><p>Next, we shall edit /etc/racoon/psk.txt and add our secret/pre-shared key to the file.</p><pre>2.2.2.2 AAAA<br>3.3.3.3 BBBB</pre><p>Next we shall modify /etc/racoon/racoon.conf as below. These configs match the current default AWS configurations for a site to site VPN. Make sure to indent the configurations properly:</p><pre>log notify;<br>path pre_shared_key &quot;/etc/racoon/psk.txt&quot;;<br>path certificate &quot;/etc/racoon/certs&quot;;<br>remote 2.2.2.2 {<br> exchange_mode main;<br> lifetime time 28800 seconds;<br> proposal {<br> encryption_algorithm aes128;<br> hash_algorithm sha1;<br> authentication_method pre_shared_key;<br> dh_group 2;<br> }<br> generate_policy off;<br>}<br><br>remote 3.3.3.3 {<br> exchange_mode main;<br> lifetime time 28800 seconds;<br> proposal {<br> encryption_algorithm aes128;<br> hash_algorithm sha1;<br> authentication_method pre_shared_key;<br> dh_group 2;<br> }<br> generate_policy off;<br>}<br><br>sainfo address your.vps.private.ip/32 any address 10.0.0.0/16 any {<br> pfs_group 2;<br> lifetime time 3600 seconds;<br> encryption_algorithm aes128;<br> authentication_algorithm hmac_sha1;<br> compression_algorithm deflate;<br>}</pre><p>The next step is to edit /etc/ipsec-tools.conf as follows:</p><pre>flush;<br>spdflush;<br><br>spdadd your.vps.private.ip/32 10.0.0.0/16 any -P out ipsec<br>esp/tunnel/your.vps.private.ip-2.2.2.2/require;<br>spdadd 10.0.0.0/16 your.vps.private.ip/32 any -P in ipsec<br>esp/tunnel/2.2.2.2-1.1.1.1/require;<br><br>spdadd your.vps.private.ip/32 10.0.0.0/16 any -P out ipsec<br>esp/tunnel/your.vps.private.ip-3.3.3.3/require;<br>spdadd 10.0.0.0/16 your.vps.private.ip/32 any -P in ipsec<br>esp/tunnel/3.3.3.3-1.1.1.1/require;</pre><p>Finally, we can degub racoon by running it in the foreground and checking for syntax errors.</p><pre>racoon -d -v -F -f /etc/racoon/racoon.conf</pre><figure><img alt="Run Racoon in the foreground" src="https://cdn-images-1.medium.com/max/1024/0*uvk0ZTyYqmmNDFSs.png" /></figure><p>If the output looks like above, it means it’s good to go. If not, refer to the error messages for potential errors such as syntax errors in one of the configuration files.</p><p>Finally restart setkey and racoon using these commands:</p><pre>/etc/init.d/setkey restart<br>/etc/init.d/racoon restart</pre><p>You may opt to start the two services on startup to make the VPN persistent after a reboot:</p><pre>sudo update-rc.d setkey enable<br>sudo update-rc.d racoon enable</pre><p>Now we need to simulate “interesting traffic” to bring the tunnel up. Let’s set up a ping to the private IP of a resource on the AWS Side. Make sure ICMP traffic is allowed through the security group of the Private Instance on the AWS side. You may opt to set this up as a per-minute crontab in order to keep the tunnel alive even when no traffic is passing through it.</p><pre>* * * * * ping 10.0.1.10</pre><p>In the meantime, we also need to edit our route tables on the VPC to allow traffic from our external host to go through the Virtual Private Gateway. We need to edit the route tables to enable traffic into the private subnet. Click “Route Tables” under the “Virtual Private Cloud” tab. Select the “Main” route table (which happens to be the one that controls the routes for the private subnet).</p><figure><img alt="Edit AWS VPC Route table for S2S VPN Static Routing" src="https://cdn-images-1.medium.com/max/1024/0*I6D-MYPLr1LDvcsq.png" /></figure><p>Click on the “Routes” tab, then click “Edit routes”.</p><figure><img alt="Edit AWS VPC Private Subnet Route table" src="https://cdn-images-1.medium.com/max/1024/0*asVksVvtuqP36HDh.png" /></figure><p>Then click “Add route” and enter the internal IP CIDR of your target server(s) on the partner side (E.g. 172.31.93.1/32). Keep this block as small as possible. Select the target as “Virtual Private Gateway” then select your VPG. Save the changes.</p><p>If you want to be able to enable communication between your VPS and your public resources, repeat the same process for the route table associated with the public subnet.</p><p>In a few moments, traffic from the ping you set earlier will start to flow, and your tunnel status will change to “Up”!</p><p>To add another partner, follow the same process of adding a customer gateway, creating the VPN, performing the configuration on the VPS, and editing the route tables accordingly.</p><p>This concludes this tutorial on securely connecting a standalone Virtual Private Server (VPS) running Linux (specifically Debian) to AWS’ Virtual Private Cloud (VPC) using a site-to-site VPN with Static Routing.</p><p><em>Originally published at </em><a href="https://blog.heptanalytics.com/connecting-a-linux-vps-to-an-aws-vpc-using-a-s2s-vpn-with-static-routing/"><em>Hepta Analytics Blog</em></a><em>.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f45eaf3f1191" width="1" height="1" alt=""><hr><p><a href="https://medium.com/heptanalytics/connecting-a-linux-vps-to-an-aws-vpc-using-a-s2s-vpn-with-static-routing-f45eaf3f1191">Connecting a Linux VPS to an AWS VPC using a S2S VPN with Static Routing</a> was originally published in <a href="https://medium.com/heptanalytics">Hepta Analytics Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Lessons Gleaned from Attending and Speaking at the World Economic Forum for Africa 2019 Gathering]]></title>
            <link>https://medium.com/heptanalytics/lessons-gleaned-from-attending-and-speaking-at-the-world-economic-forum-for-africa-2019-gathering-d569a62d90c1?source=rss----a336fb0c6921---4</link>
            <guid isPermaLink="false">https://medium.com/p/d569a62d90c1</guid>
            <dc:creator><![CDATA[Sylvia Makario]]></dc:creator>
            <pubDate>Sun, 28 Mar 2021 17:16:44 GMT</pubDate>
            <atom:updated>2021-03-28T17:16:44.254Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*hf26uazb-BZzR0Dz.jpg" /></figure><p>Last year in September, Hepta Analytics was amongst the few startup companies invited to participate in the World Economic Forum for Africa in Cape Town, South Africa. Such a rare opportunity for a young company like ours given the high profile individuals, such as heads of states invited to attend these types of events. It also included a great mix of local, regional and international companies execs, academic and civil society leaders, all coming together to discuss one thing: Shaping inclusive growth and shared futures in the Fourth Industrial Revolution.</p><p>Even though it has taken a while to write this reflective piece, I have been pondering on what to present it as, on the outcomes of that event where I represented Hepta Analytics in full capacity as an attendee and speaker.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*PswhX21tN0c6vgp9.jpg" /></figure><p>The major topic of the panel discussion I was slotted was tackling the role of Artificial intelligence and its contribution to the global economy as predicted to reach almost $16 trillion by 2030, and how Africans can seize the opportunity of utilizing it to improve their well being. The link to the conversation can be found below.</p><p><a href="https://newsvideo.su/video/11449553">Link on the WEF page</a></p><p><strong><em>or DW news website below</em></strong></p><p>I was also able to ask a panel of a few African presidents later on at the plenary session on their investment strategy in the space technology sector in the African Continent.</p><p>Question captured below :</p><p><em>Space technology, while capital intensive is becoming increasingly important to countries truly leapfrogging and realizing new economic dividends. How can we as young people help our countries and governments also compete in the global space race without losing focus on the more immediate challenges that our countries face, such as food insecurity.</em></p><h3><strong>Here are the outcomes and lessons learnt during and after the conference</strong></h3><p>The text will be organized in terms of <strong>concerns and issues</strong> and then outlining potential solutions to those problems. It feels literally effective to not only point out issues, but to also provide suggestions for potential solutions. I am looking forward to your comments and potential challenges shared and solutions for those in this field as well.</p><p>Based on discussions held during the event by various organizations from the African continent, both multinational and local, it was very evident that their major cry was the “lack of data science and other kinds of technology infrastructure” experts in the African continent.</p><p>According to most individuals I spoke to, their go to strategy was outsourcing this expertise outside of the African continent. Their excuse or as everyone says it “We are reducing our overheads such as spending tons of dollars in resources, training experts in-house”. However, this same individuals forgot how risky and careless that sounded, given the potential cybersecurity risks you might introduce to your systems when outsourcing critical infrastructure work or software that for example monitors traffic from sensitive sources such, as financial data and any activities that breaching your system would mean the end of your business.</p><p>In addition to that, you risk losing your intellectual property due to lack of very solid legal structures missing in the African continent on intellectual property.</p><p>But this is beside the point because I might be preaching to an already converted crowd, that has water tight systems for the mentioned issues. But stay with me here.</p><p>What is plaguing the Technology and Engineering industry in the African continent is issues running the gamut from having individuals masquerading as experts in various fields, operating and not delivering the level of quality they front as offering to their clients; not having systems in place facilitating enough and quality post high school training, leading to what they call half baked graduates ; lack of well-paying opportunities for individuals who have actually studied this disciplines, but later drop out to pursue whatever venture that pays them well; underpaid college professors that don’t deliver training to students to the level of quality needed by industry etc. There are several angles that can be investigated but we obviously don’t just want to blame the system. However, as most of us might be aware, these issues have led to diminished faith, especially for genuine and legit experts in the African continent. This becomes exacerbated even further by the anecdotes and rhetoric that continuously keeps getting fronted about that.</p><p>Furthermore, this has led to some governments across the african continent hiring “experts” outside of the African continent, flying them in to fix the issues on the ground. A process that incurs a hefty price in terms of tax payers money as well as missed opportunity on betting on companies like Hepta analytics and various technology companies across the continent, offering quality work and capable in all realms to offer solutions that fit the local or regional context.</p><p>We also can’t forget the issue of lack of self-education for organizations seeking these kinds of technological expertise. There is some sort of entitlement that comes with wanting to be educated on how these technologies such as machine learning, cloud computing and geospatial technologies can contribute towards improving your work flows, while being adamant to pay for the services offered. Basically wanting free consultancy from a local firm and using their information to hire a “more tactical team abroad.”</p><p>This puts most young local firms at crossroads. As much as young technology companies have to give some level of their information away in exchange of a contract for early days, it becomes burdensome when then this becomes an expectation, even more so when you truly have and believe that you have a highly trained and talented team to deliver quality work above and beyond. These moments or times make you learn early on and very quickly, how to avoid time suckers, by turning down some contracts no matter how painful that might feel. This also includes individuals peddling donation money to just have a piece of software that is just going to sit in their vault and not making the required impact. Yup, we have been there done that. It is part of the <strong><em>#Heptaway</em></strong> on making sure our integrity is water tight as an organization.</p><p><strong>Possible solution</strong></p><p>Collectively, these issues don’t have any straight jacket answer to them. Every company and individual in the African continent in the heavily specialized fields, has to analyze fully what they are worth in terms of compensation for their work.</p><p>However, Compensation becomes such a thorny issue especially when people minimize what you are worth based on your setting and throwing phrases around like “surely, you are in the African continent, you can’t be that expensive!”. Maybe we should collectively, as technology and Engineering experts decide then on what minimum we cannot go beyond when charging for service and product delivery. This would then enable provisions on setting standards beyond which, one should not accept as compensation, as an employer or employee for the industries sake. This would however be a whole other discussion and debate that would lead to nitpicking and various other ugly anecdotes.</p><p>On the other hand, the issue of individuals continuously peddling the lack of expertise in the African continent would probably be answered by only one question. “Why are all these individuals doing a very poor job in headhunting for the talent in the African continent?”. Is it the fact that the top quality talent refuses to be paid meagerly resulting to such conclusions, portraying a whole continent of 54 countries of not having Engineering and technology talent?. Such careless phrases and the enablers that is us, is causing more harm than good. Maybe it is time we looked inward and investigate what we internalize as truth.</p><p>There is also fundamentally a need, for more technology companies and experts in the African continent, embracing the act of speaking ,writing and owning their success stories and how good they are in what they do and how whatever they do, adds value to whatever work flows, organizations in the African continent are focused on optimizing. Do it with your chest Na! If you are good in your craft, own that with pride. That might change the game for those who come after you. Say that to the world and mountain tops :).</p><p>Clearly, humility certainly doesn’t pay bills but on the other hand, be proud backed with facts and real quality work. That is just how capitalism is set up and those who make a lot of noise sometimes are purported to be the experts in what they do even when it might not be true.</p><p>To conclude, African governments need to start investing in their own citizens and giving them opportunities before seeking that outside their countries or even the African continent in general. There are several other political tropes that come with this topic but the main message today is to give opportunities to local entrepreneurs and experts.</p><p>There are obviously a lot more that could be analyzed and discussed from the various pointers mentioned, but today we keep it short and simple. Looking forward to your thoughts on this. #askhepta #heptaAnalytics #machinelearning #DataScience #cloudinfrastructure #geospatialTechnologies</p><p>To access the report on the entire event on the WEF19 Africa, <a href="https://toplink.weforum.org/events/a0P0X00000MSE7TUAX/default/library#preview=9894">follow this link</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d569a62d90c1" width="1" height="1" alt=""><hr><p><a href="https://medium.com/heptanalytics/lessons-gleaned-from-attending-and-speaking-at-the-world-economic-forum-for-africa-2019-gathering-d569a62d90c1">Lessons Gleaned from Attending and Speaking at the World Economic Forum for Africa 2019 Gathering</a> was originally published in <a href="https://medium.com/heptanalytics">Hepta Analytics Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Animated Bar Graphs]]></title>
            <link>https://medium.com/heptanalytics/animated-bar-graphs-258ff39051fc?source=rss----a336fb0c6921---4</link>
            <guid isPermaLink="false">https://medium.com/p/258ff39051fc</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[football]]></category>
            <dc:creator><![CDATA[Hepta Analytics]]></dc:creator>
            <pubDate>Sun, 28 Mar 2021 17:16:43 GMT</pubDate>
            <atom:updated>2021-03-28T17:16:43.085Z</atom:updated>
            <content:encoded><![CDATA[<p>In recent times there has been a surge in animated bar graphs visualization. They are mainly used to show how certain events changed overtime. At Hepta Analytics, we used them to showcase how the premier league table changed over time during the just completed season. I will work you through how you can do it on your own.</p><h4>Prerequisites</h4><ul><li>D3.js</li></ul><p>First you will create the page using HTML and CSS.</p><p>NB<strong>: I am not good with design. So anyone looking for something pretty, this is where you lower your expectation. The goal here is to get something that works.</strong></p><pre>&lt;!doctype html&gt;<br>&lt;html lang=&quot;en&quot;&gt;<br>&lt;head&gt;<br>  &lt;meta charset=&quot;utf-8&quot;&gt;<br>  &lt;title&gt;bars&lt;/title&gt;<br>&lt;/head&gt;<br><br>&lt;body&gt;<br>        &lt;div class=&quot;wrapper&quot;&gt;<br>                &lt;center&gt;&lt;p&gt;&lt;h4 class=&quot;chart-title&quot;&gt;Week  &lt;span class=&quot;week&quot;&gt;&lt;/span&gt; &lt;/h4&gt;&lt;/p&gt;<br>                  &lt;p&gt;&lt;h4&gt;&lt;span class=&quot;ole&quot;&gt;&lt;/span&gt;&lt;/h4&gt;&lt;/p&gt;<br>                &lt;/center&gt;<br>                &lt;div class=&quot;chart&quot;&gt;&lt;/div&gt;<br>                <br>                &lt;p class=&quot;source&quot;&gt;<br>                  &lt;span &gt;Title Battle Premier League 2018/2019&lt;/span&gt;<br>                  &lt;div class=&quot;item&quot; &gt;<br>                      &lt;img src=&quot;favicon.ico&quot;/&gt;<br>                      &lt;span class=&quot;caption&quot;&gt;@heptanalytics&lt;/span&gt;<br>                  &lt;/div&gt;<br>                &lt;/p&gt;<br>                <br>              &lt;/div&gt;<br>              &lt;script src=&quot;https://d3js.org/d3.v4.min.js&quot;&gt;&lt;/script&gt;<br>              &lt;script src=&quot;https://d3js.org/d3-scale.v3.min.js&quot;&gt;&lt;/script&gt;<br>    &lt;/body&gt;<br>&lt;/html&gt;</pre><p>Now we are going to load the data which is in csv format.</p><pre>d3.csv(&quot;datafull.csv&quot;, function(d) {<br>      return {<br>        games : d.games,<br>        position : d.position,<br>        teamname : d.teamname,<br>        points : d.points<br>      };<br>    },function(data) {<br>// all the logic goes here<br>});</pre><p>The above code loads data with particular columns. The resulting selection is what will be parsed to the function for logic. You can change name of the column to whatever you want. Next we are going to select all the matches in the first matchday. Then we sort the teams .</p><pre>function sortData(data) {<br>      data.sort(function(a,b) {<br>        return b.points - a.points;<br>      });<br>    }<br><br>const lastMatch = 38;<br>let startMatch = 1;<br>var filtereddata = data.filter(function(d) { return d.games == startMatch.toString(); });<br>var selectedData = filtereddata.sort(function(a,b) {  return b.points - a.points;});</pre><p>After that, we need to draw the bar graph for the first match day. The x-axis will have points, y axis will have the team names. The drawBars() is where you will need to make changes if you want to alter the look and feel of the bars. You can change the color with the fill function. You can also change the width and height of the bars.</p><pre>function drawBars(el, data, t) {<br>      let barsG = el.select(&#39;.bars-g&#39;);<br>      if (barsG.empty()) {<br>        barsG = el.append(&#39;g&#39;)<br>        .attr(&#39;class&#39;, &#39;bars-g&#39;);<br>      }<br>      <br>      const bars = barsG<br>      .selectAll(&#39;.bar&#39;)<br>      .data(data, yAccessor);<br>      bars.exit()<br>      .remove();<br>      bars.enter()<br>      .append(&#39;rect&#39;)<br>      <br>      <br>      .attr(&#39;x&#39;, leftPadding)<br>      .merge(bars).transition(t)<br>      .attr(&#39;y&#39;, d =&gt; yScale(yAccessor(d)))<br>      .attr(&quot;fill&quot;, function(d) {<br>              <br>        return &quot;#000&quot;;<br>        <br>        <br>      })<br>      .attr(&#39;width&#39;, d =&gt; xScale(xAccessor(d)))<br>      .attr(&#39;height&#39;, yScale.bandwidth())<br>      <br>      .delay(delay);<br>    <br>    }</pre><pre>function drawXAxis(el) {<br>      el.append(&#39;g&#39;)<br>      .attr(&#39;class&#39;, &#39;axis axis--x&#39;)<br>      .attr(&#39;transform&#39;, `translate(${leftPadding},${height})`)<br>      .call(d3.axisBottom(xScale));<br>    }<br>    <br> function drawYAxis(el, data, t) {<br>      let axis = el.select(&#39;.axis--y&#39;);<br>      if (axis.empty()) {<br>        axis = el.append(&#39;g&#39;)<br>        .attr(&#39;class&#39;, &#39;axis axis--y&#39;);<br>      }<br><br>const yScale = d3.scaleBand()<br>    .rangeRound([0, height], 0.1)<br>    .padding(0.1);<br><br>yScale.domain(geoAreas);<br>drawXAxis(svg, selectedData.points);  <br>drawYAxis(svg, selectedData.teamname);<br>drawBars(svg, selectedData);</pre><p>This will work for only on match. If you want to animate throughout the season, we need to have a loop that will keep changing the match day. This will continue until the last day of the season. The code below does that.</p><pre>const interval = d3.interval(() =&gt; {<br>          const t = d3.transition().duration(400);<br>          <br>          startMatch += 1;<br>          filtereddata = data.filter(function(d) { return d.games == startYear.toString(); });<br>          selectedData = filtereddata.sort(function(a,b) {  return b.points - a.points;});<br>          <br>          d3.select(&#39;.week&#39;).text(startMatch.toString());<br><br>          <br>          <br>          yScale.domain(selectedData.map(yAccessor));<br>          drawYAxis(svg, selectedData, t);<br>          drawBars(svg, selectedData, t);<br>          <br>          if (startMatch === lastMatch) {<br>            interval.stop();<br>          }<br>        }, 1000);</pre><p>This will give you the animation of the teams throughout the season. If you want to make certain events notable, you will alter the drawBars() by adding a class attribute. If you want to change the color of the This can be seen it the fill attribute. Below is the code to do that.</p><pre>.attr(&#39;class&#39;, d =&gt; d.games === &#39;18&#39; ? &#39;bar wld&#39; : &#39;bar&#39;) //when the matchday is 18<br><br>.attr(&quot;fill&quot;, function(d) {<br>        if (d.teamname==&#39;Liverpool&#39; ){<br>          <br>          return &#39;Red&#39;;<br>        }else if (d.teamname==&#39;Man City&#39;){<br>          return &#39;#3182bd&#39;;<br>        }<br>        <br>        else  {<br>          return &quot;#000&quot;;<br>        } <br>        <br>      })</pre><p>The above code will customize the bar colors of the Liverpool and Man City.</p><p>After completing the code, the final result should be as below:</p><h3>Hepta Analytics on Twitter: &quot;As we head into the first of many long weekends without #PremierLeague football, let&#39;s take a look back at the 2018/19 season with a particular focus on the title race in this #dataviz. pic.twitter.com/wiCRctnB3j / Twitter&quot;</h3><p>As we head into the first of many long weekends without #PremierLeague football, let&#39;s take a look back at the 2018/19 season with a particular focus on the title race in this #dataviz. pic.twitter.com/wiCRctnB3j</p><p>To improve this, one can use logos instead of names and add the total points next to the bars. Cannot wait to see you all being creative. The full code and data can be found on my github account <a href="https://github.com/yvonnegitau/Animated-Bar-Graphs">https://github.com/yvonnegitau/Animated-Bar-Graphs</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=258ff39051fc" width="1" height="1" alt=""><hr><p><a href="https://medium.com/heptanalytics/animated-bar-graphs-258ff39051fc">Animated Bar Graphs</a> was originally published in <a href="https://medium.com/heptanalytics">Hepta Analytics Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Unpacking approaches for achieving digital inclusion for all]]></title>
            <link>https://medium.com/heptanalytics/unpacking-approaches-for-achieving-digital-inclusion-for-all-a2d43f9ed6c1?source=rss----a336fb0c6921---4</link>
            <guid isPermaLink="false">https://medium.com/p/a2d43f9ed6c1</guid>
            <dc:creator><![CDATA[Hepta Analytics]]></dc:creator>
            <pubDate>Sun, 28 Mar 2021 17:16:40 GMT</pubDate>
            <atom:updated>2021-03-28T17:16:40.494Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/510/0*nU2EQut_QkLYRvgl.jpg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/510/0*FOxsyIc1zX-q3Ddb.jpg" /></figure><p>The International Telecommunication Union (ITU) recent <a href="https://www.itu.int/en/ITU-D/Statistics/Pages/facts/default.aspx">report</a> shows that great strides are being made in expanding access to internet through increased availability of broadband networks. However, there is a substantial divide of unconnected and under connectedness — that still need to be addressed within communities in both developed and developing markets. A lot has been written to demonstrate how cost is remain one of major barriers to internet adoption.</p><p>Communities all over the world have a segment of their population that is marginalized because of economic headwinds that make it difficult for them to be full participant and contributor in social and economic development.</p><p>Access to internet creates an enabling environment for everyone to benefit from the opportunities of an increasingly digital economy that we live in today. It also fosters digital literacy and it is an empowerment multiplier to achieving broader goals not only in educational and health outcomes but also employability, entrepreneurship including small business support and a path to middle class. Therefore, closing the digital divide is both an economic and moral imperative that need persistent discussion and sharing of various approached with stronger potential for results.</p><p>Giving everyone access to the internet, otherwise known as digital inclusion can be achieved in the following ways;</p><p>First approach is by improving and cultivating a competitive environment in the market place. The commission needs to perform a diagnostic of existing and potential barriers for any new entrant into mobile service providers space.</p><p>This includes provisioning of data services, wireless and fiber platforms. This diagnosis must take to account impediments to two kinds of entrants; 1) Those which might want to install their own infrastructure and 2) Those considering an option of leasing unused capacity from the incumbent operators also known as Mobile Virtual Network Operator (MVNO).</p><p>To facilitate MVNO participation, the regulatory framework may mandate all the incumbent mobile operators to report how much of their installed capacity goes unused every month as well as addressing existing market failure in business transactions by acting as an enabler of the negotiations to protect the MVNO from being unfairly priced out and excluded.</p><p>The idea is that when barriers to entry are lowered, new entrants will want to come in with innovative business models that target the excluded and economically marginalized market. This has been done elsewhere like in the United States where the incumbents ignored and priced out a segment of the market but MVNO like <a href="https://www.cricketwireless.com/">CricketWireless</a>, <a href="https://www.boostmobile.com/">BootsMobile</a> and others joined in with innovative business models that suites and cater for the neglected low-income including those subscribers with-not so good credit history.</p><p>These MVNO begun taking up the market share of the very incumbent from whom they were leasing the infrastructure. The incumbents’ profitability was being affected and as a matter of survival the competitive environment resulted into low prices that benefited everyone. This is a sustainable way to regulate the market and achieve the desired outcome of digital inclusion instead of arbitrarily price fixing, which will distort the market and taking away the natural creativity incentive of a business enterprise.</p><p>The second approach is by intentionally creating a dedicated Digital Inclusive Fund. The regulatory framework can facilitate putting into the fund through part of the mobile operator spectrum auctioning revenue or as a percentage of valuation of mergers &amp; acquisition deals. Sustainment of the fund may come from <a href="https://www.isc.hbs.edu/about-michael-porter/affiliated-organizations-institutions/Pages/shared-value-initiative.aspx">shared value</a> contribution of private sector participants including mobile operators and other companies operating in localities as well as philanthropists.</p><p>A separate and independent organization should be given the mandate to work with local governments, Community Based Organizations (CBOs) through stakeholder engagements and local trusted brokers to implement Digital Inclusive Initiatives and programs using this fund.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a2d43f9ed6c1" width="1" height="1" alt=""><hr><p><a href="https://medium.com/heptanalytics/unpacking-approaches-for-achieving-digital-inclusion-for-all-a2d43f9ed6c1">Unpacking approaches for achieving digital inclusion for all</a> was originally published in <a href="https://medium.com/heptanalytics">Hepta Analytics Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Save Kubernetes Deployment Cost Using Nginx Ingress]]></title>
            <link>https://medium.com/heptanalytics/save-kubernetes-deployment-cost-using-nginx-ingress-77337b54595b?source=rss----a336fb0c6921---4</link>
            <guid isPermaLink="false">https://medium.com/p/77337b54595b</guid>
            <category><![CDATA[ingress]]></category>
            <category><![CDATA[kubernetes]]></category>
            <category><![CDATA[infrastructure]]></category>
            <category><![CDATA[cloud-computing]]></category>
            <dc:creator><![CDATA[Hepta Analytics]]></dc:creator>
            <pubDate>Sun, 28 Mar 2021 17:16:37 GMT</pubDate>
            <atom:updated>2021-03-28T17:16:37.551Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/300/0*Byn_pT28GNG8hQub.png" /></figure><p>Previously I have written about creating a Kubernetes cluster on <a href="https://www.bensonmurimi.com/diary/read/9/how-to-setup-kubernetes-cluster-on-aws-using-kops-full-guide">AWS</a> as well as on <a href="https://blog.heptanalytics.com/2019/02/27/574/">bare metal servers</a><br> Both of these setups require you to expose your application to the public and that comes with cost implications. Let me start with a little background on how application deployment work.<br> A <strong>Pod</strong> is the lowest compute unit in <em>Kubernetes</em> running one or more containers. Pods have their own cluster-private IP address, which means that containers within a Pod can all reach each other’s ports on localhost, and all pods in a cluster can see each other. However, Pods can die, whether crashing, rolling update or being removed due to down scaling, in which cases, their IP addresses change. Therefore directly accessing pods is not a reliable way to access the application.<br> The most common way of exposing pods/ services publicly is using a <strong>loadbalancer</strong>. When you are using the cloud (GCP, AWS, Azure, Digital Ocean etc) when you run a deployment with type loadbalancer, an external access link is automatically created. For GCP and Digital Ocean, this will be a public IP eg: <em>80.34.53.23</em> while for AWS it will be something like <em>acf86ld3bff1fd611e9as7580a43b4esb515d-792533560.us-east-1.elb.amazonaws.com</em> .<br> These end points can be used to configure DNS record in cloudflare or any other DNS management service you are using. However the loadbalancers come at a cost that increases as you deploy more applications since each application requires it’s dedicated loadbalancer. The figure below shows on high level how the deployment looks like.</p><figure><img alt="Service deployment with a loadbalancer" src="https://cdn-images-1.medium.com/max/564/0*QFXprJRSXi9d_khk.png" /></figure><p>This is where Ingress controllers becomes a savior. There are a couple of options to choose from Traefik, HAProxy and Kubernetes Ingress. In this blog I will focus on Kubernetes Nginx Ingress. The diagram below shows a complete setup of nginx ingress controller on high level.<br> Notice we are using one loadbalancer to configure the domain names irrespective the number of apps deployed. This loadbalancer points to the ingress controller which then decides where to direct traffic to based on the configured ingress rules and annotations.</p><p>The possibility of deploying multiple applications using a single loadbalancer makes nginx ingress very suitable and a cost savior unlike using a single loadbalancer for every application you deploy. I know now you are wondering what happens when traffic spikes up and the loadbalancer is overwhelmed?. In kubernetes you can scale the ingress controller and have multiple pods running, lets say 10 instances. This means kubernetes will distribute incoming traffic in the different pods in a round robin manner hence handling the traffic appropriately.</p><figure><img alt="Service deployment using nginx ingress" src="https://cdn-images-1.medium.com/max/553/0*AIqAsLoG8DfOxQYl.png" /></figure><p><strong>Implementation</strong><br> Configuring nginx ingress controller is easy and fast thanks to their awesome documentation which is very elaborate especially for cloud deployment.<br> Extra tweaking might be necessary for a bare metal setup since a loadbalancer option is not available. Nodeport is the only viable and stable option to expose your nginx ingress on a public ip.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=77337b54595b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/heptanalytics/save-kubernetes-deployment-cost-using-nginx-ingress-77337b54595b">Save Kubernetes Deployment Cost Using Nginx Ingress</a> was originally published in <a href="https://medium.com/heptanalytics">Hepta Analytics Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Privacy by Design — A Developers Perspective]]></title>
            <link>https://medium.com/heptanalytics/privacy-by-design-a-developers-perspective-2d003e4e743d?source=rss----a336fb0c6921---4</link>
            <guid isPermaLink="false">https://medium.com/p/2d003e4e743d</guid>
            <category><![CDATA[privacy]]></category>
            <dc:creator><![CDATA[Hepta Analytics]]></dc:creator>
            <pubDate>Sun, 28 Mar 2021 17:16:36 GMT</pubDate>
            <atom:updated>2021-03-28T17:16:36.575Z</atom:updated>
            <content:encoded><![CDATA[<p><em>by Benjamin Munyoki</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/0*F6sagN7uP9QmnoZJ.png" /></figure><p>Facebook’s annual developer conference over last week has had lots of talking points. Key among them is Mark Zuckerbergs bold statement that the future is private. Many have questioned if Facebook has any moral authority left to talk about privacy. For a company that thrives on advertisement served through scanning user data, it becomes hard to believe that Facebook is really serious about privacy of its user data. To backup the claim the social network giant has announced a major overhaul of its core platform including its messaging services. But why overhaul? Where did Facebook go wrong in the first place?</p><p>While many think privacy is more about who you give access to data (who you share the data with) or how you store data, it’s imperative to note most companies including Facebook miss the point from the start — during design. While how you transmit data, how you store data and who you share the data with are equally important parts in guarding user privacy, a key point in privacy lies in how you collect the data itself. Data collection is a broad subject and with the advance in technology, there are multiple ways applications collect data. In every step (collection, transit, storage and sharing), privacy should be one of the main design decisions that application developers should consider.</p><p>What is Privacy by Design? Privacy by Design is an approach used when developing new systems and applications. Simply put privacy should by default be factored in, in every step of system development. When Privacy by Design is followed, privacy becomes an integral part of project objectives, design process and planning. Privacy by Design has been touted for sometime but its major highlight has been the advent of GDPR.</p><h4>How can Developers Apply Privacy by Design:</h4><h4>Collect only what is needed</h4><p>Sometime last month I downloaded a torch application for my Android phone. Normally when installing an application for the first time, you have to go through the required permissions. I couldn’t believe it when location permission popped up! Why would a torch application need to know my location? Worse still the app crashed when I declined! Why does a torch application need location data to function?</p><h4>Limit what you share</h4><p>Often times, application need to share data with third parties. Most application do this through application programming interfaces (APIs). An API is a way to allow systems to communicate often times sharing data between them. When developing APIs developers should ensure that data is not shared indiscriminately. Differently API keys should have different access permission and only access the data permitted by their access.</p><h4>Factor in Privacy in Sprint Planning</h4><p>With agile, normally every sprint starts with a planning session. It imperative to discuss with the team how the new feature will be implemented with the users privacy factored in.</p><p>When followed, privacy by design can bring lots of benefits key among them customer trust, increased app users and application stability.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2d003e4e743d" width="1" height="1" alt=""><hr><p><a href="https://medium.com/heptanalytics/privacy-by-design-a-developers-perspective-2d003e4e743d">Privacy by Design — A Developers Perspective</a> was originally published in <a href="https://medium.com/heptanalytics">Hepta Analytics Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why Satellite Technology is Extremely Important in Food Security]]></title>
            <link>https://medium.com/heptanalytics/why-satellite-technology-is-extremely-important-in-food-security-e702d5d05b01?source=rss----a336fb0c6921---4</link>
            <guid isPermaLink="false">https://medium.com/p/e702d5d05b01</guid>
            <dc:creator><![CDATA[Sylvia Makario]]></dc:creator>
            <pubDate>Sun, 28 Mar 2021 17:16:35 GMT</pubDate>
            <atom:updated>2021-03-28T17:16:35.646Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/678/0*tnowWGJz81m2hIKI.png" /></figure><p>In my last post on <a href="https://blog.heptanalytics.com/2019/03/18/why-remote-sensing-is-fundamental-in-soil-mapping/">why remote sensing is important for soil moisture mapping</a>, I tackled fundamentals quite familiar, yet overlooked in agricultural activities and mostly for small scale farmers in the rural areas. This is therefore a reiteration that without quality data on the levels of climate change as it is happening on real time, there is no possible ways to prevent disasters such as as hunger and starvation in a country. Case in point, the recent catastrophe in northern Kenya where several Kenyans living in there faced and continuously face the threat of hunger due to delayed rains, which has culminated to having cases of death and malnutrition.</p><p>If we had quintessential ways of practicing agriculture, it being the largest economic activity practiced in the continent, every farmer would have access to real time information on any slight changes in the climate or even their soil moisture. This would happen early on before any disasters such as hunger hit, facilitating the coordination of efforts, on finding solutions for a nations survival in such a season.</p><p>There have been several efforts to achieve that, but most of it has been hampered by silo mentality. Most organizations that are invested in agriculture either do not share coordination efforts in making sure that food security is maintained in a country or there is literally a ton of bureaucratic processes involved. Streamlined processes and coordination, are fundamental to ensuring a successful dissemination ideas to some organizations if not all, that are the enablers of farmers getting the required information at the right time.</p><p>On that note, I reiterate the fundamental aspect of having quality real time information/ data that would benefit the farmers and everyone else in general in achieving optimal yield during the planting seasons, compensating for shocks such as hunger due to climate change. In this case, satellite data has fundamentally played a significant role in ensuring that in countries that have been quick to adapt to the capabilities of the technologies.</p><p>As much as i am advocating for African countries, investing in satellite technology, I am mostly advocating for a collective effort for all African governments. This is due to the fact that space technology in itself, is a very costly undertaking for a single African Nation to take on. This then calls for cohesion and agreements to have a robust space ecosystems fr the purpose of serving the greater african region in all aspects of required data.</p><p>With satellite technology in agriculture, there are several agricultural processes that can be monitored to the letter, enabling optimal yield and optimal use of resources available to countries to cushions themselves against disasters.</p><p>Food security as you might be aware, is fundamental in maintaining national security in a country. Therefore having it as a priority on the list of areas to tackle using satellite technology is important in supporting those efforts. With real time data, and a combination of technologies in agriculture such as precision farming technologies, so much such as disease/pest control would be achieved in very quick and easy ways, saving cost in the process. Machine learning models can be built to enable deterministic prediction methods for crop yields, informing government on the best decision to make when necessary.</p><p>In conclusion I would argue that African countries have to regroup and take measures especially on considering geospatial data as gateway to achieving food security in their individual countries.</p><p>The next post would tackle the framework that can looked into and steps that can be put in place to achieving this goal.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e702d5d05b01" width="1" height="1" alt=""><hr><p><a href="https://medium.com/heptanalytics/why-satellite-technology-is-extremely-important-in-food-security-e702d5d05b01">Why Satellite Technology is Extremely Important in Food Security</a> was originally published in <a href="https://medium.com/heptanalytics">Hepta Analytics Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>