Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Old school Linux administration – my next homelab generation

scholz.ruhr

Post reply on HN
Print view View original

71–80 of 115 posts

Re: Old school Linux administration – my next homelab generation

#71

Post by ksbrooksjr »

bayindirh wrote: post #31

As a person who manages a big fleet of servers containing both pets and cattle, the upkeep of the pets is nowhere near the cloud-lovers drum-up. A server installed with half-decent care can run uninterrupted for a long long time, given minimal maintenance and usual care (update, and reboot if you change the kernel). Also, not installing a n+3 Kubernetes cluster with an external storage backend reduces overheads and n…

Even changing the kernel can often be done without rebooting the system depending on which distro you're using. Quite a few distros now include some sort of kernel livepatching service including Ubuntu, Amazon Linux, and RHEL.

Re: Old school Linux administration – my next homelab generation

#72

Post by bayindirh »

ksbrooksjr wrote: post #71
bayindirh wrote: post #31

As a person who manages a big fleet of servers containing both pets and cattle, the upkeep of the pets is nowhere near the cloud-lovers drum-up. A server installed with half-decent care can run uninterrupted for a long long time, given minimal maintenance and usual care (update, and reboot if you change the kernel). Also, not installing a n+3 Kubernetes cluster with an external storage backend reduces overheads and n…

Even changing the kernel can often be done without rebooting the system depending on which distro you're using. Quite a few distros now include some sort of kernel livepatching service including Ubuntu, Amazon Linux, and RHEL.

Yeah, We're aware of that, but none of our pets is mission critical enough to prevent 5 minutes of downtime, at most, in most cases.

Re: Old school Linux administration – my next homelab generation

#73

Post by ksbrooksjr »

bayindirh wrote: post #72
ksbrooksjr wrote: post #71

Earlier quoted context omitted.

Even changing the kernel can often be done without rebooting the system depending on which distro you're using. Quite a few distros now include some sort of kernel livepatching service including Ubuntu, Amazon Linux, and RHEL.

Yeah, We're aware of that, but none of our pets is mission critical enough to prevent 5 minutes of downtime, at most, in most cases.

That's understandable. Not every service needs a perfect uptime record.

Re: Old school Linux administration – my next homelab generation

#74

Post by wernerb »

jethro_tell wrote: post #67
blinkingled wrote: post #56

Earlier quoted context omitted.

But then there's the problem that now you have two teams - ops team doesn't understand the app and app team doesn't understand the ops/infra side. I of course agree with your point that there should be two teams but you need a few guys who understand both app dev and ops/infra/OSes/k8s internals etc. And finding these people has been nearly impossible.

You can just add an ops guy with minimal coding to a dev team and it works wonders. Ops guy does the ops stuff, writes the shell scrips and consults with the devs on how to build what they need to build. The ops guy has an opportunity to level up his coding since the devs are doing code reviews for him and devs have an opportunity level up their ops knowledge because they are working with someone who understands what…

I think it's great to "bring" knowledge to devs in this model. But not commit to individual wishes for features and implement them willy illy. The "ops guy" needs a team too, preferably to build a company shared platform that suits the large majority. This only goes of course when you either hit scale or are a large company. When smaller it's perfectly fine to unload unwanted things to one or two people. I do very strongly believe in "you build it you run it". Keeping servers up is different than keeping your application up. I think a dev should know what health checks/probes are, what a good period time they need for their application. It's the "ops guy"'s job to make the dev guys job as easy as possible and bring expertise and knowledge when needed

Re: Old school Linux administration – my next homelab generation

#75

Post by GordonS »

bayindirh wrote: post #31

As a person who manages a big fleet of servers containing both pets and cattle, the upkeep of the pets is nowhere near the cloud-lovers drum-up. A server installed with half-decent care can run uninterrupted for a long long time, given minimal maintenance and usual care (update, and reboot if you change the kernel). Also, not installing a n+3 Kubernetes cluster with an external storage backend reduces overheads and n…

> A server installed with half-decent care can run uninterrupted for a long long time, given minimal maintenance and usual care (update, and reboot if you change the kernel).

For my earlier home setups, this was actually part of the problem! My servers and apps were so zero-touch, that by the time I needed to do anything, I'd forgotten everything about them!

Now, I could have meticulously documented everything, but... I find that pretty boring. The thing with Docker is that, to some extent, Dockerfiles are a kind of documentation. They also mean I can run my workloads on any server - I don't need a special snowflake server that I'm scared to touch.

Re: Old school Linux administration – my next homelab generation

#77

Post by j45 »

woopwoop24 wrote: post #63

HA is overrated, i'd much rather go for a low mean time to repair. Backups, reinstall, ansible playbook is my way to go if hardware fails, which is quite rare to be honest. HA goes beyond hardware in terms of "electricity, internet connection, storage, location etc.., IMO people quite often underestimate what it means to have real HA --> second location with identical setup to shift workload or even have active-activ…

HA is built into platforms like proxmox, what is the stress?

Re: Old school Linux administration – my next homelab generation

#80

Post by hereyougo2 »

davidwritesbugs wrote: post #78

Can someome explain "Pets" and "Cattle"? Not seen this useage before, think I get it but want to be sure. Pets are servers for love and fun, Cattle are servers exploited for loveless meat & milk?

https://cloudscaling.com/blog/cloud-computing/the-history-of...
Post reply on HN