HN user

BenElgar

159 karma

I'm the founder of Krucible (https://usekrucible.com) and former engineer at MongoDB. I love building tools for other developers, particularly in the infrastructure space.

[ my public key: https://keybase.io/benelgar; my proof: https://keybase.io/benelgar/sigs/m5aFJYK7Q9eEwo97rx9nX7lUC8La33bauZAD1wFJYwU ]

Email: me@benelgar.com | ben@usekrucible.com

meet.hn/city/us-New-York

Interests: Technology, Startups, Philosophy, Privacy, Programming, Networking, Investment

---

Posts5
Comments43
View on HN

Why do you include Social Security in the list of things that won't be eliminated? I'm not American but in my understanding UBI would almost completely eliminate Social Security, along with putting a big dent in the need for Medicare and Medicaid.

Author here. At the last two companies I've worked at we really needed—and didn't have—a solution for spinning up throwaway Kubernetes clusters that we could use for testing and development. Krucible is an attempt to solve that problem.

Krucible replaces minikube/kind[0] with faster, cloud-hosted Kubernetes clusters that are easier to spin up from inside of CI/CD environments and that don't consume a large amount of local resources.

Happy to answer any questions!

[0] Or whichever other Kubernetes provider you use for testing and development

Kubernetes supports resource requests and resources quotas to combat this. You should be protecting your production workloads this way anyway.

That's certainly good advice and would significantly reduce the likelihood of issues but it doesn't handle all cases. For instance it's not particularly easy to quota network bandwidth.

Ultimately all of these problems are likely solvable—we just think that Krucible is easier, simpler and safer.

There are definitely some use cases where having a local-to-your-machine development environment is more suitable but, as you say, Krucible is easier to set up and comes with features that just running k3d doesn't, like snapshots. Our experience is that ensuring that developers are all running in the same environment is pretty valuable by itself.

By having a dedicated Kuberntes cluster you reduce the blast radius. For instance, if you roll out a new version of your services that unexpected consumes a significant quantity of resources, if you were running that on your production cluster that could interfere with your production workload.

In a similar vein, if you have 10 or even 100 end-to-end test suites, with Krucible you could run them all in parallel, significantly reducing the time taken, without fear of them impacting each other. In your shared cluster scenario you would be limited by the size of your cluster.

I think Krucible could be very helpful for you here. Each student could sign up for a Krucible account and spin up up to 25 free clusters. Should be plenty for teaching purposes.

You could also prepare a cluster with a certain configuration in advance, snapshot it and then create new clusters from that snapshot so that all students could quickly get started with a Kubernetes cluster in a given state.

Get in touch and I'd be happy to help you get things set up. Email is ben@usekrucible.com.

Really it depends what you're testing and deploying. For instance, if you're developing a new microservice, it's helpful to be working with the same service discovery mechanism that is running in your production cluster. When you're testing, presumably you also want to ideally test any Kubernetes changes that you make before you deploy to your production servers. Krucible is ideal for that.

Krucible is hosted, so you're not running a VM that's consuming resources and that you have to manage—spinning up Minikube from within a CI environment isn't necessarily the easiest of tasks. With Krucible it's just a single API call to create a Kubernetes cluster. This also allows you to parallelise your test suite easily as you can spin up as many clusters as you need at the same time.

The snapshots feature is also a big differentiator: you can set up a cluster, take a snapshot and then share that with your team so that you're all running identical Kubernetes clusters.

Author here. At the last two companies I've worked at we really needed—and didn't have—a solution for spinning up throwaway Kubernetes clusters that we could use for testing and development. Krucible is an attempt to solve that problem.

We've just released a really cool feature called Snapshots that allows you to image a running Kubernetes cluster, including the state of all applications, and then create new clusters from that image. It's great for creating consistent development environments or quick starting test environments.

Happy to answer any questions people might have.

Author here. At the last two companies I've worked at we really needed—and didn't have—a solution for spinning up throwaway Kubernetes clusters that we could use for testing and development. Krucible is an attempt to solve that problem.

We've just released a really cool feature called Snapshots that allows you to image a running Kubernetes cluster, including the state of all applications, and then create new clusters from that image. It's great for creating consistent development environments or quick starting test environments.

Happy to answer any questions people might have.

Author here. At the last two companies I've worked at we really needed—and didn't have—a solution for spinning up throwaway Kubernetes clusters that we could, for example, easily and reliably test our deployment process in. Kind (https://github.com/kubernetes-sigs/kind) does a pretty good job of addressing that but requires setting up a bunch of infrastructure that needs to be maintained and doesn't easily work in a containerised CI environment. Krucible is an attempt at solving those problems by allowing you to spin up Kubernetes clusters via an API quickly and affordably.

Krucible is still in its early stages but I wanted to show it to the community and get feedback. Happy to answer any questions.

Tor Browser 8.5 7 years ago

Though it's important to note that using Tor directly rather than the browser bundle means that your browser fingerprint is even more distinct than it would otherwise be.

Speaking for at least some of the H1Bs and other non-immigrant visa holders, they'll leave—taking the savings from their Bay Area salary back to Europe, Asia, Central America or another place with reasonable living costs. I've always thought that that was one of the most compelling, pragmatic reasons to fast track the citizenship process.

For an example of the top of my head, take tax returns. In the UK the vast majority of people don't need to file one—your employer deducts the right amount each month from your salary and off you go. Moving to America I found I had to spend at least half a dozen hours trying to figure out what I had to do, eventually giving up and paying for a service to do it for me and am still unsure I filed correctly.

For another example, take getting a driver's license. In the UK you fill in a form online, they extract any other relevant details from your passport records and put in your credit card details and they send you a license within the week. It's a 15 minute process. In America I got to finding out I had to go to the DMV (and that I should go to the DMV in another city because it's somehow superior) and promptly gave up.

I know you said aside from healthcare, but even if you had the hypothetical best insurance available, with no co-pay or deductible, just keeping track of all of the arcane and opaque numbers, policies and restrictions is a huge administrative burden.

Now to be clear, I understand why some of these systems are the way they are and, in some cases, can see the necessity, but the fact remains that in almost any other developed nation there simply wouldn't be an equivalent administrative burden to just existing.

Finally! I've been making the point for years now that the reason new developers start with AWS is because of their much more useful 12 month free tier. People from GCP would come to our university every couple of years and ask us "Why don't you use GCP?" and be told by countless computer science undergraduates that they didn't NEED $300 for a couple of months, they needed at least a year of time with a cheap instance or two. It seems they finally listened.

In my mind, the problem with Approval Voting is that it's harder to decide how one should vote. What's the cut-off? Do I vote for everyone I really believe in? Or everyone I don't loathe? And it's exactly that sort of question which leads to tactical voting. With IRV it's easy. Just order them until you're indifferent to the rest of the candidates.