chucky_z

I will second this. They get a ton of shit for not properly maintaining projects but damn do they try to constantly be moving forwards even if it’s not always super innovating.

I second this. I went all in on the dream machine and some of its parts and it failed in an unrecoverable way, I sent it in for repairs; and it failed again in a different unrecoverable way. I just use the crappy Comcast router with a ubiquiti AP now.

fzf!!!

I used to ask it for a pre-pre screen. It had a maybe 20% fail rate.

Run it yourself if you’re concerned =)

https://github.com/charmbracelet/charm

how long ago did you use nomad? nomad integrates with consul but isn't backed by it. it's also pretty trivial to run consul in quite a shitty network environment by bumping up some of the settings (they should probably change their 'production' suggestions).

Looks like this is for a consistent workflow for developers where managing a VM or learning nerdctl is too much. I say this as truthful and neither sarcastically or maliciously.

Considering Twitters global status $8/mo to post would kill it outside more wealthy countries basically instantly.

I think there's some "cheat codes" now involving some of these things though. For instance, you can use some of AWS but keep your actual compute/networking out and have things like security patching and server inventory be completely solved by using AWS SSM. There's also options like EKS Anywhere to have managed on-prem k8s for way, way cheaper than running it in AWS proper. These kinds of services I think are the future for hybrid/on-prem folks.

I used Phabricator a long time ago before we eventually moved to Github due to a handful of blocking issues with the `arc` tool, and CircleCI only supporting Github (this was even pre-circle 2.0 days).

I really, really liked Phabricator and would happily have replaced Jira + Confluence + GitHub with Phorge if I was starting somewhere from scratch.

I’m at a point in my career and as an individual where if someone is on mat/pat leave all my complaints go out the window. It’s just about understanding the situation, and maybe all of us can explain to those individuals complaining that even if they never have children, that them giving the parents space to raise a well-balanced individual is a net benefit for all of society.

Yo ho yo ho…

Keep those backups of your Nintendo games archived somewhere.

Figure out how to revive https://github.com/coreos/fleet as something native in systemd?

You could run Nomad and use parameterized jobs with raw_exec to get something like this. I know that systemd has some lifecycle stuff where you can create/run systemd units directly through the cli instead of using unit files which should allow you to hook the lifecycle into a nomad style service invocation but I can't recall exactly how to do this.

This doesn't quite roll into the 'only systemd,' but it allows you to build exactly what you're looking for.

I believe the OP is talking about LoadBalancer (https://kubernetes.io/docs/tasks/access-application-cluster/...), and I believe you're talking about Ingress (https://kubernetes.io/docs/concepts/services-networking/ingr...).

Yea consul connect is a good choice for mTLS, it's enforced by default with no way to turn it off.

You can also go DIY with something like Traefik and just configure everything to require TLS with client TLS pass-through for enforced mutual TLS, that puts you into 'require a set of central load balancers' but you still get all the benefits.

One point around Nomad is that the point on it requiring low latency.... isn't really true. I run a cluster with some servers physically on different coasts of the United States with clients as far as Singapore. It all just works :). (This does need some tuning but once it works, it just works)

https://learn.hashicorp.com/vault ok then hire some random joe shmoe sysadmin and teach them.

Hashicorp Vault is always my goto even for small companies. It seems too much but it’s really not. A single instance is scalable enough to handle quite a bit of traffic.

Another good alternative if you need something more SAASy is the 1pass API product

Interesting reasoning! I had a similar thought about why raft doesn’t allow observers the other day.

I used observers with Gluster previously and went from annoying split brain scenarios to flawless clusters just by adding a few, and their resource usage was basically nothing.

From my perspective, I think the biggest issue with gRPC is it using HTTP/2. I understand that there’s a lot of reasons to say “No, HTTP/2 is far superior to HTTP/1.1.” However, in terms of proxying _outside Google_ HTTP/2 has lagged, and continues to lags at the L7 proxy layer. I recently performed a lot of high-throughput proxying comparing HAProxy, Traefik, and Envoy. HTTP/1.1 outperformed HTTP/2 (even H2C) by a pretty fair margin. Enough that if gRPC used HTTP/1.1 we could use noticeably less hardware. I could see this holding true even with a service mesh.

never. double down on them. if https://twitter.com/LucyLavend/status/1543166082223063041 was what every app gave me i'd have a much better time in life.

They're also adding in a basic secrets k/v store in the next version. Their intention (I believe) is to target small use-cases and IOT use-cases; while allowing folks to scale to gigantic levels when mixing in Consul/Vault.

Traefik and Nomad play very nice together if ingress is your main concern.

This is not true anymore. They released service discovery a little bit ago.

When the feature doesn't exist! For existing workflows/functionality; I think using a new tool for a feature that doesn't (and won't for some time) exist is a _perfectly_ good reason to introduce the tool. On top of that you're looking at using something that is mature and well-used, Lucene/Solr.

If you wanted to be cheeky you could go ask your CTO if they would foot the bill for someone like 2nd Quadrant to write CJK full-text search into pgsql. ;)

I've done this for some features/issues in random OSS projects and if you feel someone else has the same problem I always feel paying money to have it implemented is well worth it.

AWS Outposts are tragically limiting in so many ways. If you are a small shop they are great but if you don’t fit their model exactly you are SOL.

Same problem with Anthos.

You can work around the limitations by having a million points of presence with them and splitting up your workloads but there’s real costs associated with that model. Running 1 rack in 100 DCs is a lot harder than 10 racks in 10 DCs.

If you had one early on consider it again. The first 6 months was abysmal for me, but ever since https://github.com/lima-vm/lima has been locked down I have ways to compile everything and run any container I want. `nerdctl` has taken a tiny bit of effort to learn but it's been well worth it.

Alibris is really great, and I've gotten a few hand written notes from bookstores I've bought from. Even if they're fake I'm a sucker for these kinds of things.

Principle CJ