<?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 Springup on Medium]]></title>
        <description><![CDATA[Stories by Springup on Medium]]></description>
        <link>https://medium.com/@gitfog?source=rss-1e12108b757------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*jP9wToTa4FBACfkETszQEQ.png</url>
            <title>Stories by Springup on Medium</title>
            <link>https://medium.com/@gitfog?source=rss-1e12108b757------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 11:18:13 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@gitfog/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[Should we make high availability at the application level or the file system level in Kubernetes?]]></title>
            <link>https://medium.com/@gitfog/should-we-make-high-availability-at-the-application-level-or-the-file-system-level-in-kubernetes-94616773891d?source=rss-1e12108b757------2</link>
            <guid isPermaLink="false">https://medium.com/p/94616773891d</guid>
            <dc:creator><![CDATA[Springup]]></dc:creator>
            <pubDate>Mon, 17 Apr 2023 05:21:04 GMT</pubDate>
            <atom:updated>2023-04-17T05:21:04.963Z</atom:updated>
            <content:encoded><![CDATA[<p>I think the answer should be file system level, cause if you make HA at application level, you have you setup HA solution one by one for database, redis, etc. every component in the cluster. Those solutions might even consume larger resource than file system level HA such as openebs, mayastor…</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=94616773891d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How to restrict accessing IP in k8s behind LoadBalancer with Proxy Protocol when using HTTPS?]]></title>
            <link>https://medium.com/@gitfog/how-to-enable-source-ip-whitelist-in-k8s-behind-loadbalancer-with-proxy-protocol-when-using-https-15d3a1d2ce26?source=rss-1e12108b757------2</link>
            <guid isPermaLink="false">https://medium.com/p/15d3a1d2ce26</guid>
            <category><![CDATA[kubernetes]]></category>
            <category><![CDATA[linode]]></category>
            <category><![CDATA[load-balancer]]></category>
            <dc:creator><![CDATA[Springup]]></dc:creator>
            <pubDate>Tue, 02 Nov 2021 05:35:44 GMT</pubDate>
            <atom:updated>2021-11-02T05:39:57.819Z</atom:updated>
            <content:encoded><![CDATA[<p>When you provide services/sites from k8s behind LoadBalancer through HTTPS, you may want to enable source IP whitelist to restrict accessing IPs, for this you can choose the traditional way `X-Forwarded-For` with Protocol HTTP or HTTPS in LoadBalancer, that will terminate the SSL connection at LoadBalancer then add `X-Forwarded-For` with source IP to the headers of consequent communications which are transparent with the backends in k8s. If you intend to use LetsEncrypt at the same time, you’ll suffer the problem that only one cert pair can be configured to the LoadBalancer, however you might want to support multiple domains at the same LoadBalancer. What can we do then?</p><p>We can use Proxy Protocol。</p><p>How：</p><ol><li>Select Proxy “TCP” and enable Proxy Protocol option at LoadBalancer settings page, ensure these two settings are correct, and are not accidentally reset by applying load balancer service yaml through k8s (It looks like the annotation `service.beta.kubernetes.io/do-loadbalancer-enable-proxy-protocol` should work for LoadBalancer, but it doesn’t work for Linode at least).</li><li>Add `nginx.ingress.kubernetes.io/whitelist-source-range`</li></ol><pre>apiVersion: networking.k8s.io/v1</pre><pre>kind: Ingress</pre><pre>metadata:</pre><pre>name: xxx</pre><pre>namespace: xxx</pre><pre>annotations:</pre><pre>kubernetes.io/ingress.class: nginx</pre><pre>kubernetes.io/tls-acme: &quot;true&quot;</pre><pre>nginx.ingress.kubernetes.io/whitelist-source-range: &quot;123.123.123.123/24&quot;</pre><p>3. Add `use-proxy-protocol`</p><pre>kind: ConfigMap</pre><pre>apiVersion: v1</pre><pre>metadata:</pre><pre>name: nginx-configuration</pre><pre>namespace: oftenpay</pre><pre>labels:</pre><pre>app.kubernetes.io/name: ingress-nginx</pre><pre>app.kubernetes.io/part-of: ingress-nginx</pre><pre>data:</pre><pre>use-proxy-protocol: &quot;true&quot;</pre><p>Then it should work!</p><p>References:</p><ul><li><a href="https://www.linode.com/community/questions/366/how-do-i-configure-my-nodebalancer-to-pass-through-ssl-connections-to-the-back-e">How do I configure my NodeBalancer to pass through SSL connections to the back-end nodes? | Linode Questions</a></li><li><a href="https://www.linode.com/docs/guides/nodebalancer-proxypass-configuration/">Using Proxy Protocol with NodeBalancers</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=15d3a1d2ce26" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>