Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
mdaniel
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
27 ms
·
481.
▲
by
mdaniel
1y ago
I think you missed the error message; I wasn't whining about for_each syntax masquerading as if it was a property of the resource, I was pointing out the moronic behavior of attempting to iterate over a list of strings and being given
482.
▲
by
mdaniel
1y ago
> At some point you won't need a fully dedicated ops team I think programmers are more likely to go extinct before that version of reality materializes. That's my secret plan on how to survive the alleged AI apocalypse: AI ain&
483.
▲
by
mdaniel
1y ago
I'll bite: what is the systemd primitive for secret injection? for declaring which EBS volume it needs mounted where? (I don't mean Mount= I mean a conceptual MountVolumeByTag[Name=my-pgdata-0]=/var/lib/postgres
484.
▲
by
mdaniel
1y ago
In addition to what the sibling comment pointed out about that being an EKS-ism, yes, I know for 100% certainty VPC-CNI will allocate Pod IP addresses from the Subnet's definition, which includes public IP addresses. We used that to ci
485.
▲
by
mdaniel
1y ago
To the best of my knowledge, nothing is stopping you from doing any of those things right now. Including, ironically, authentication for pod-to-pod calls, since that's how Service Accounts work today. That even crosses the Kubernetes A
486.
▲
by
mdaniel
1y ago
The comment you replied to cited 12TB of NVMe, and I can assure you that 12TB of ECC RAM costs way more than NVMe
487.
▲
by
mdaniel
1y ago
> I don't get the etcd hate. I'm going out on a limb to say you've only ever used hosted Kubernetes, then. A sibling comment mentioned their need for vanity tooling to babysit etcd and my experience was similar. If you are
488.
▲
by
mdaniel
1y ago
Secure from what, friend? It's a credential leak waiting to happen, to say nothing of the need to now manage IAM Users in AWS. That is the 2000s way of authenticating with AWS, and reminds me of people who still use passwords for ssh.
489.
▲
by
mdaniel
1y ago
I'd guess management access, or the old school way of doing vLANs. Kubernetes offers Network Policies to solve the risk of untrusted workloads in the cluster accessing both pods and ports on pods that they shouldn't https:/&
490.
▲
by
mdaniel
1y ago
You stopped typing; what does Kubernetes do a bad job at with relation to scheduling workloads that declare they need at least 1 GPU resource but should be limited to no more than 4 GPU resources on a given Node? https://kubernet
491.
▲
by
mdaniel
1y ago
It was mentioned pretty deep in another thread, but this is just straight up user hostile variable "my_list" { default = ["a", "b"] } resource whatever something { for_each = var.my_list } The gi
492.
▲
by
mdaniel
1y ago
Fun fact, while digging into the sibling comment's complaint about the OpenAPI spec, I learned that it actually advertises multiple content-types: application/json application/json;stream=watch application/vnd.ku
493.
▲
by
mdaniel
1y ago
So, what I'm hearing is that it should tie itself to a commercial company, who now have a private equity master to answer to, versus an open source technology run by a foundation Besides, easily half of this thread is whining about hel
494.
▲
by
mdaniel
1y ago
Ah, so that explains it: https://github.com/skateco/skate#:~:text=leverages%20podman%...
495.
▲
by
mdaniel
1y ago
and then the rest of the owl is working out the merge conflicts :-D I don't know how to have a cute git analogy for "but first, git deletes your production database, and then recreates it, because some attribute changed that made
496.
▲
by
mdaniel
1y ago
I recognize that I'm biased, but you'll want to strongly consider whether https://rivet.gg/docs/config is getting your audience where they can be successful, as compared to (e.g.) https://kubernete
497.
▲
by
mdaniel
1y ago
IMHO, the rest of that sentence is "be too complex for some metric within some audience" I can assure you that trying to reproduce kubernetes with a shitload of shell scripts, autoscaling groups, cloudwatch metrics, and hopes-and-
498.
▲
by
mdaniel
1y ago
Loading it in what? I just slurped the 1.8MB openapi.json for v1.31 into Mockroon and it fired right up instantly
499.
▲
by
mdaniel
1y ago
For my curiosity, what threat model is mTLS and encryption between pods driving down? Do you run untrusted workloads in your cluster and you're afraid they're going to exfil your ... I dunno, SQL login to the in-cluster Postgres?
500.
▲
by
mdaniel
1y ago
I repeatedly see this "yaml isn't typesafe" claim but have no idea where it's coming from since all the Kubernetes APIs are OpenAPI, and thus JSON Schema, and since YAML is a superset of JSON it is necessarily typesafe E
501.
▲
by
mdaniel
1y ago
As the sibling comment points out, I think that would be a perfectly fine helm replacement, but I would never ever want to feed starlark into k8s apis directly
502.
▲
by
mdaniel
1y ago
orly, what structure does this represent? outer { example { foo = "bar" } example { foo = "baz" } } it reminds me of the insanity of toml [lol] [[whut]] foo = "bar"
503.
▲
by
mdaniel
1y ago
> a one-shot reconciler that wires that all together each time you plan and apply? You skipped the { while true; do tofu plan; tofu apply; echo "well shit"; patch; done; } part since the providers do fuck-all about actually, no
504.
▲
by
mdaniel
1y ago
heh, I think you didn't read the room given this directory https://github.com/agentsea/nebulous/blob/v0.1.88/deploy/cha... Also, ohgawd please never ever do this ever ohgawd https://
505.
▲
by
mdaniel
1y ago
Apologies, but what influence does Kubernetes have over the way AWS deals with attach and detach behavior of EBS? Or is your assertion that Kubernetes should be its own storage provider and EBS can eat dirt?
506.
▲
by
mdaniel
1y ago
Relevant: Show HN: Canine – A Heroku alternative built on Kubernetes - https://news.ycombinator.com/item?id=44292103 - June, 2025 (125 comments)
507.
▲
by
mdaniel
1y ago
As far as I know one can do that right now, since wasmedge (Apache 2) exposes a CRI interface https://wasmedge.org/docs/develop/deploy/oci-runtime/crun#pr... (et al)
508.
▲
by
mdaniel
1y ago
Sounds like the kustomize mental model: take code you potentially don't control, apply patches to it until it behaves like you wish, apply If the documentation and IDE story for kustomize was better, I'd be its biggest champion
509.
▲
by
mdaniel
1y ago
CRDs are not tied to golang in any way whatsoever; < https://www.redhat.com/en/blog/writing-custom-controller-pyt... > and < https://metacontroller.github.io/metacontroller/guide/cr
510.
▲
by
mdaniel
1y ago
Crash Loop Back Off
More ›