mjg235

6 ms·

1. 1.▲
   by mjg235 4y ago

   Hey I'm Michael, the author of the article, and I did consider this but didn't go into detail for fear of bloating the read. I think there's two points here. 1. Not all software has viable OSS alternatives. The most obviou
2. 2.▲
   by mjg235 4y ago

   Yeah it's wild that it's been a pain point for so long without a properly supported solve.
3. 3.▲
   by mjg235 4y ago

   A basic stateless web app behind a load balancer can tolerate unstable network identifiers as long as you automate registering/deregistering the new names, which kubernetes services do very well. Stateful apps can be a bit more excepti
4. 4.▲
   by mjg235 4y ago

   People are definitely sleeping on k8s' capability in managing stateful workloads. We run various datastores on it all the time, and it's quite reliable for a lot of use cases (largely because the underlying cloud block stores it&
5. 5.▲
   by mjg235 4y ago

   I actually think k8s is (or can be) a decent solve for non-critical, low cost databases if you're confident about navigating the toolchain. The one big gap is storage resizing, which is part of why we're pointing it out.
6. 6.▲
   by mjg235 4y ago

   The flaw is definitely most acute when it is combined with operators, but it would make storage resizes vastly more accessible for less capable devs if it were able to be done declaratively via the spec instead of via a painful workaround.
7. 7.▲
   Kubernetes StatefulSets are Broken(plural.sh)

   107 points by mjg235 4y ago|96 comments