Archives

This is quite clever. DevOps’ish is brought to you by Accurics. Just announced: Terrascan extends Policy as Code to Kubernetes. People One lone hero in production is not sustainable-not for you, not for high-functioning teams, and not for customers who depend on your service. Collaborate well by instrumenting observability from the very beginning, and enable more resilient teams to build more reliable systems sustainably. In our guide, Developing a Culture of Observability, we lay out why o11y culture and tools go hand-in-hand. Learn how to build a culture of observability with Honeycomb. SPONSORED Chris Short: K8s Release Team - Cloud Native Computing Foundation - NVIDIA to buy Arm for $40B - Oracle and TikTok - OpenShift TV - Walmart Brings Back Gateway Computers by Tech Breakfast Podcast I sat down and talked the news with the folks over at Tech Breakfast Podcast. It was a fun conversation in which I shared my assessment of the Tik Tok situation (it is rather dire). ...

General 9-11, war, and mental health warnings This is hell week for me every year. The week before 9-11-2001, I lost my best friend. At the same time, mourning his loss in North Carolina still (I was stationed at Langley AFB, VA). I woke up early the morning after leaving the mountains and the funeral activities in Hendersonville. I was at my parents’ house. They’d taken time off the week before dealing with my dad’s mother’s (my grandmother’s) passing. While they were incredibly helpful, they’d missed all the work they could afford to, so we were leaving that day to be back home by nightfall. I flipped over to the news to see a weather update. I saw one smoking tower. Minutes later, I saw another plane smack into the second tower live. Scarred for life (I studied NYC architecture throughout my youth), I jumped in the shower after telling my ex-wife, “If they hit the Pentagon, we’re leaving, immediately.” She began frantically packing up our miraculously still sleeping daughter and her stuff. ...

Welcome! Your environment makes up so much of who you are in real life. I often wonder if your work environment reflects how your infrastructure will look. If you’re okay with satisfying that 80% use case, does your infra look like AWS with hundreds of services some of which are quite stable and others not so much? Or if you the perfect be the enemy of the good and be like a lot of companies and spend a lot of time spinning their wheels waiting for the right thing to happen. I’ve seen countless examples of this. But, what about when your environment is resilient, scalable, and cloud-enabled. What about when psychological safety is achieved AND the idea of cost per minute to the business during an outage is established? We see so many examples of bad ways and maybe a few good ways of building. I would love to see more use cases around the things that are possible when the business value is known down to the individual and the organization is a high-performer. Highlighting more of the good, I hope, would engrain “the why” behind the need for change way more than features and individual capabilities. ...

Welcome! It would appear the clock has run out for all the Heptio folks to get their payouts from the VMware acquisition. A few Kubernetes contributors are jumping ship from VMware. Four in a little over a week is nothing short of a sign, though. These aren’t the kinds of folks who like to hang out in big corporate orgs and feel like cogs in the machine. They want to feel like they’re making an impact and see the value in their work. They, like Deming, appreciate and take pride in their work. There is nothing at all wrong with this. I 100% understand, given that Red Hat is the biggest organization I’ve ever been a part of outside the Air Force. I can understand the appeal of the late-stage startup life too. But, there is comfort in working for a 25-year-old organization inside a much larger organization that probably won’t muck with us too much given the current global situation. VMware completely sucked all the Heptio out of Heptio the second they started renaming and rebranding Heptio projects to VMware products. I have to hand it to Red Hat. Realizing that the CoreOS name meant something and keeping some of its projects active and incorporating the CoreOS name into products shows that at the very least, Red Hat appreciated CoreOS’s work. I don’t know if the same can be said for VMware of Heptio. Perception is often reality in these instances because we’ll never know otherwise unless folks speak up (which they are often not allowed to). ...

Welcome! What a week it was. KubeCon EU 2020 was this past week. And amongst all the announcements there was a lot of activity and interaction too. Two things I realized this week were, first, y’all are really bubble gumming and duct taping clusters together into production like it’s no big deal just waiting to be bit by something. Second, the Kubernetes community is really an amazing bunch of people, in general. I asked what some folks thought were the highlights of the week’s festivities and I’d like to share those here: Alexis “Horgix” Chotard said, “The openness and transparency of Case Studies, especially the one from @milesbxf and @suhailpatel here Joshua Bezaleel said, “Holly Cummins keynote on climate. That was amazing.” There was a ton of praise for Ian Coldwater and Brad Geesaman’s talk, Advanced Persistence Threats: The Future of Kubernetes Attacks. Ian Coldwater was the talk of the conference pretty much. I have no idea when any of these talks are going to be publicly available. But, you can log back into the terrible platform and watch the talks and keynotes on-demand at the moment. ...