fduran

With servers, if they have a huge uptime and sysadmins etc are afraid to restart them, then they have "pets" with high liability, requiring Infra as Code https://docs.sadservers.com/blog/destroy-and-deploy-the-joys...

Fedora may be becoming the default for desktops, not for servers (Debian possibly the default for servers).

Yes, the website is for prospective (and current) clients.

A small annoyance in startup circles is getting feedback about my website front page along the lines of "I didn't understand your hero, everybody should understand in one sentence what you do". Well, no, my clients will self-select as in not everybody needs to understand what "troubleshooting servers" or "devops" is :-)

There is def some luck involved, as in you don't know beforehand what's going to be successful.

"You find a problem in a niche, say accounting for plumbers, and build for that, then you just go and market to these people". It's way better to work on something you are familiar with and you like.

They have saved _more_ than two hours per dev and week. There's a compound factor and now code can be more reliable (less outages or emergencies fixing bugs) etc. Also having a sane working environment helps engineers not quitting, which is very expensive if they are replaced.

Nice, this is like SadServers with a twist, excellent :-)

Suerte! Unrelated, growing up in Spain it always baffled me that identification was based on a photo on your DNI. Stories of siblings or even friends that had a passing resemblance to each other sharing DNIs was a common story.

This is not a new or novel idea

Yep https://x.com/fduran/status/1134283398594387969

hello, thanks for the feedback. Just deployed a new image that only checks for the objective, not at what docker network somebody uses.

It is hard to have a checker that eliminates both false positives and false negatives in general, but we always try to minimize false negatives and we failed initially here.

Hello, SadServers guy here.

We have scenarios running on k8s, both on single VMs (the ones you can see in the scenario list) and we also have a beta/PoC k8s cluster where we currently run a couple of scenarios as single pod (a docker container) or as a full system (the "kubernetes playgrounds", which is kind of hidden while we test it).

Is this what you were wondering? we do have pending to introduce podman scenarios as well

Hello, creator here. Have you checked your dashboard? otherwise please contact us (email or form in the website) and we'll be happy to help

(Creator here) thanks! and we may at some point get into Windows :-)

Hello, SadServers guy here. Free VMs are sandboxed (no way in or out other than coming in through the proxy) for security reasons. Paid accounts have VMs with internet access and SSH access (and your pub key is added to all VMs for convenience)

Checking out how the platform works was two clicks away: home -> give me a server.

I don't know of any other SaaS which gives you a VM with one click without any registration but we do it.

In any case thanks for the feedback, I've put a button on this /advent page for clarity, cheers

hello, creator here, sorry about that. In this case you can click again on the "Open the Server Terminal in a New Window" button

https://sadservers.com/advent

I haven't set up an advent event (maybe I should) but you can do yourself a challenge a day from SadServers.com

This is similar to our experience migrating a SadServers k8s service; basically Hetzner is way cheaper and more performant than the big cloud providers BUT you have to take into account man-hours for the migration and be careful with missing services or ancillary tooling: https://docs.sadservers.com/blog/migrating-k8s-out-of-cloud-...

Big cloud provider (AWS, Azure, GP) is great for all the managed ecosystem; if you mostly only need raw computing (CPU, memory, bandwidth), then a provider like Hertzner makes a lot of sense (plus they have API and basic services like LB/firewall and object storage).

We at SadServers moved from big cloud managed K8s to Hetzner + Edka and it's an order of magnitude cheaper (obv some perks are missing).

I love Asciinema and have made a small contribution, support the project with a donation: https://docs.asciinema.org/donations/#individuals

If you want to see how Asciinema looks when people are troubleshooting Linux systems: https://replay.sadservers.com/

Hola, I may try this out (once the HN hug of death slows down). At SadServers we ran a GKE cluster with 0.2 vCPUs workload for a PoC at it was costing us CAD $50/day until we figured out thing like GCP using a pod not doing anything just keeping a node up using 1 CPU, hidden Prometheus metrics etc.

shameless plug: you can practice this in a free VM https://docs.sadservers.com/docs/scenario-guides/practical-l... (there's a typo there to keep you on your feet)

At SadServers we use Asciinema to record some scenario sessions, this is how we did it: https://github.com/sadservers/sadservers?tab=readme-ov-file#... (we still have some issues to iron out on our side)

The author Marcin is a nice fellow and as a reminder asciinema development relies on donations and sponsorships https://github.com/sponsors/ku1ik

Good food for thought but the article assumes how currently DevOps and Dev teams (and there's also an Ops team?) operate and there's a lot of different ways in how "DevOps" is done at different organizations.

Take any combination of dev, platform engineering, devops, SRE, ops teams (existing or not) and combine with different responsibilities and how they work together and there's a company out there using that model.

Heck, we don't even have a common understanding of what a "DevOps" team does https://docs.sadservers.com/blog/what-the-f-is-devops/

You area a Site Reliability Engineer and you get a page notifying your service is down and you rush to see your logs, metrics and other dashboards, start communicating in Slack and entering checking commands in a shell prompt.

In the same way pilots get put in emergency situations in flight simulators, I'm building an "SRE incident simulator" , a generalization of SadServers.

(SadServers guy here) Good article and although I don't see the author here, thanks for the mention :-)

(SadServers guy here) happy to help!

I've called that "Organization as Code" some years back :-)

Hello, SadServers author here. Zero shame in adapting an idea you found somewhere; I'm super happy you built something useful.

I've thought of adding programming debug scenarios (I even got sadbugs.com lol), may implement in the future.

I'd love to see what you've done, please feel free to connect :-)

Thanks, pot capacity being scheduled differently would explain the behavior.

Almost all my ec2 instances are spot, and actually I can compare the distribution with the on-demand ones.

My spot instances are very short lived (15-30 mins max) and AFAIK I've never seen a spot instance force-terminated (this would be hard to find I think).