<?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 Steve Kriss on Medium]]></title>
        <description><![CDATA[Stories by Steve Kriss on Medium]]></description>
        <link>https://medium.com/@stevekriss?source=rss-b82e21f7efd6------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*XKxCE7ezAAWXPJ01.</url>
            <title>Stories by Steve Kriss on Medium</title>
            <link>https://medium.com/@stevekriss?source=rss-b82e21f7efd6------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 13:12:44 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@stevekriss/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[Ark v0.9 alpha: Now With Restic!]]></title>
            <link>https://medium.com/heptio-blog/ark-v0-9-alpha-now-with-restic-14ad6b402ab3?source=rss-b82e21f7efd6------2</link>
            <guid isPermaLink="false">https://medium.com/p/14ad6b402ab3</guid>
            <category><![CDATA[cloud-computing]]></category>
            <category><![CDATA[kubernetes]]></category>
            <dc:creator><![CDATA[Steve Kriss]]></dc:creator>
            <pubDate>Thu, 14 Jun 2018 15:01:01 GMT</pubDate>
            <atom:updated>2018-06-14T15:01:01.082Z</atom:updated>
            <content:encoded><![CDATA[<p>We’re happy to announce that <a href="https://github.com/heptio/ark">Heptio Ark</a> now includes support for backing up and restoring Kubernetes volumes using a free open-source backup tool called <a href="https://restic.net/">restic</a>. We’ve just released a v0.9 alpha with this functionality — please try it out and let us know what you think!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/1*OP8jMAYHqIT8F9eYU_qY7Q.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Ggh-nvb4rZzCWq4UPIS2fw.png" /></figure><p>Ark has always allowed you to take snapshots of persistent volumes as part of your backups, provided you’re using one of the supported cloud providers’ block storage offerings (<a href="https://aws.amazon.com/ebs/">Amazon EBS Volumes</a>, <a href="https://azure.microsoft.com/en-us/services/managed-disks/">Azure Managed Disks</a>, <a href="https://cloud.google.com/persistent-disk/">Google Persistent Disks</a>). And, since v0.6.0, we’ve had a plugin model which enables anyone to easily implement additional object and block storage backends, outside of the main Ark repository.</p><p>With the integration of restic, Ark now natively supports backing up and restoring data from <em>any</em> Kubernetes volume or persistent volume. Restic takes file-level backups of your data, and has several key features that make it a great fit for Ark’s needs:</p><ul><li>Supports multiple storage backends, including Amazon S3, Google Cloud Storage, Azure Blob Storage, and Minio</li><li>Fully encrypts backup data at rest and in transit with AES-256 in counter mode</li><li>Only backs up data that has changed since the prior backup, using <a href="https://restic.net/blog/2015-09-12/restic-foundation1-cdc">content-defined chunking</a></li><li>De-duplicates data within a single backup for efficient use of storage</li></ul><p>Now, Ark can now be used for backup/restore across public cloud and bare-metal/on-premise environments, including:</p><ul><li>Backing up &amp; restoring emptyDir, NFS, and iSCSI volumes in an on-prem cluster</li><li>Backing up and restoring AzureFile volumes</li><li>Backing up emptyDir volumes from a cluster on Cloud A, and restoring them to a cluster on Cloud B</li></ul><p>For an Ark administrator/user, getting started with restic only requires a few steps:</p><ol><li>You set up a location for restic to store its data, and tell Ark where that is</li><li>You create encryption keys for restic to use for backups, one per namespace</li><li>You annotate the pods containing volumes to be backed up with restic</li><li>You take an Ark backup using the existing `ark backup create` command</li><li>You restore your backup using the existing `ark restore create` command</li></ol><p><strong>Alpha Testing</strong></p><p>The Ark v0.9 alpha is available for testing now! Instructions on getting started with the restic integration can be found <a href="https://github.com/heptio/ark/blob/master/docs/restic.md">here</a>. Please take this alpha release for a spin, and provide us with any and all feedback through <a href="https://github.com/heptio/ark/issues/new">GitHub issues</a>, in the <a href="https://kubernetes.slack.com/messages/ark-dr">#ark-dr channel</a> on Kubernetes Slack, or through our <a href="http://j.hept.io/ark-list">mailing list</a>. Note that there are still a couple of development tasks outstanding. Before the v0.9.0 GA release, we plan to:</p><ul><li>Enable easy restic encryption key rotation using the Ark CLI</li><li>Display restic volume backup and restore status in the `ark backup describe` and `ark restore describe` commands, respectively.</li></ul><p>DISCLAIMER: Ark’s integration with restic is still in alpha — things may change, bugs will be encountered, and you should not yet rely on this functionality for critical backups!</p><p><strong>More Information</strong></p><p>Additional detail on the integration with restic can be found <a href="https://github.com/heptio/ark/blob/master/docs/restic.md">here</a>.</p><p>For those interested in even more detail, the main PR can be found <a href="https://github.com/heptio/ark/pull/508">here</a>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=14ad6b402ab3" width="1" height="1" alt=""><hr><p><a href="https://medium.com/heptio-blog/ark-v0-9-alpha-now-with-restic-14ad6b402ab3">Ark v0.9 alpha: Now With Restic!</a> was originally published in <a href="https://medium.com/heptio-blog">Heptio</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Heptio KubeCon Recap: Disaster Recovery with Steve Kriss and Andy Goldstein]]></title>
            <link>https://medium.com/heptio-blog/heptio-kubecon-recap-disaster-recovery-with-steve-kriss-and-andy-goldstein-5b46fc91d7b?source=rss-b82e21f7efd6------2</link>
            <guid isPermaLink="false">https://medium.com/p/5b46fc91d7b</guid>
            <category><![CDATA[disaster-recovery]]></category>
            <category><![CDATA[heptio]]></category>
            <category><![CDATA[heptio-ark]]></category>
            <category><![CDATA[kubernetes]]></category>
            <dc:creator><![CDATA[Steve Kriss]]></dc:creator>
            <pubDate>Thu, 25 Jan 2018 16:51:43 GMT</pubDate>
            <atom:updated>2018-01-25T18:43:23.919Z</atom:updated>
            <content:encoded><![CDATA[<p>Today we’re continuing our KubeCon recap with a talk I presented along with <a href="https://medium.com/@andy.goldstein">Andy Goldstein</a> titled <strong>“Disaster Recovery for Your Kubernetes Clusters”</strong>. Read on for a summary and to watch the video.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Gkul8361xvd6oXLQ_MSw3w.jpeg" /></figure><p>It’s 3am. Your pager is beeping. Your Kubernetes cluster is down. What do you do? In this talk, we covered a variety of disaster scenarios you may encounter, and provided strategies for how to overcome them.</p><h4>Target audience:</h4><p>Kubernetes cluster administrators</p><h4>Why you should care:</h4><p>No one ever wants to lose their persistent data. Losses can be prevented, if you take the appropriate steps up front.</p><iframe src="https://cdn.embedly.com/widgets/media.html?src=https%3A%2F%2Fwww.youtube.com%2Fembed%2FqRPNuT080Hk%3Ffeature%3Doembed&amp;display_name=YouTube&amp;url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DqRPNuT080Hk&amp;image=https%3A%2F%2Fi.ytimg.com%2Fvi%2FqRPNuT080Hk%2Fhqdefault.jpg&amp;key=a19fcc184b9711e1b4764040d3dc5c07&amp;type=text%2Fhtml&amp;schema=youtube" width="854" height="480" frameborder="0" scrolling="no"><a href="https://medium.com/media/686866a14cf99122933fb46f2b155e56/href">https://medium.com/media/686866a14cf99122933fb46f2b155e56/href</a></iframe><h4>If you don’t remember anything else, make sure you:</h4><ul><li>Spend the time and effort to make it as easy as possible to deploy and re-deploy a new cluster, new node, etc. through automation with your tool of choice.</li><li>Have a solution in place for backing up the state within your clusters (i.e. the data in etcd and your persistent volumes).</li><li>Back up persistent volumes, not just etcd. Backing up just etcd isn’t sufficient, nor is it always accessible to you (e.g. on GKE or other managed Kubernetes services).</li><li>Use the Kubernetes discovery API to walk through all the resources in a cluster so you can back them up.</li></ul><h4>Best questions from the audience:</h4><p><strong>Q:</strong> What happens when you restore into a namespace that already has some content in it?</p><p><strong>A:</strong> Currently, we record any conflicts due to preexisting content as warnings. In the future, we plan to make conflict resolution configurable and pluggable, so you’ll be able to add your own logic for dealing with pre-existing items.</p><p><strong>Q:</strong> Are there any resources that don’t restore well?</p><p><strong>A:</strong> Services with Type = LoadBalancer. These are services that result in load balancers being provisioned in your cloud provider. The name of the load balancer is derived from the Service’s UID, which is randomly generated and assigned by the Kubernetes apiserver when the Service is created. Unfortunately, there is no way for Ark to influence the UID’s value. The end result is that when a LoadBalancer Service is restored, it generates a new cloud load balancer resource, instead of reusing a preexisting one.</p><h4>What’s next for Ark:</h4><ul><li>Possibly adding concept of PV snapshots to Kubernetes itself</li><li>Additional Ark features (backup targets, snapshotting &amp; backing up any file system, run in whatever namespace you want, backup &amp; restore across multiple regions/zones, custom conflict handlers, backup templates, etc.)</li></ul><h4>Where to learn more:</h4><p><a href="https://github.com/heptio/ark/">https://github.com/heptio/ark/</a></p><p>Kubernetes Slack — #ark-dr</p><p><a href="https://groups.google.com/forum/#!forum/heptio-ark">https://groups.google.com/forum/#!forum/heptio-ark</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5b46fc91d7b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/heptio-blog/heptio-kubecon-recap-disaster-recovery-with-steve-kriss-and-andy-goldstein-5b46fc91d7b">Heptio KubeCon Recap: Disaster Recovery with Steve Kriss and Andy Goldstein</a> was originally published in <a href="https://medium.com/heptio-blog">Heptio</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>