Probably not large scale, but definitely a spike. We’ve been seeing more signups lately due to the US talking shit about Canada.

We publish some graphs here https://fedecan.ca/en/guide/data/lemmy/users

As a media client, totally. Lots of people run jellyfin on Android TV sticks and things like that.

I’d also suggest throwing more cores at it than you think it needs, and increasing the nic buffer sizes. This is what finally solved the random 1-2 minute outages I was seeing on lemmy.ca, turns out we were somehow saturating the queues even at relatively low traffic levels. (these issues might be specific to our model nic)

Finally, make sure hardware offloading is disabled, and your CPU type is set to host.

I thought the % underneath was the number of users who couldn’t see that instance, but I guess it’s % of the active fediverse users?

Hmm, piefed.ca is supposed to follow the lemmy.ca defed list, I’d expect both to be the same.

Use of colors for software is confusing, since I expected it to be based on the % or something. A legend would be clutch for those of us that look at pictures and get confused, then read the text…

https://github.com/mdshack/shotshare maybe?

#2 with opnsense

Personally I’m fine with paid apps here, lots of people use tailscale for example. I think the larger issue is the drive-by spamming without contributing outside of their own promotion thread.

I like the comment elsewhere in this thread referencing a subreddit that requires X comments over Y days in the community first.

They should hang out a while first and not have only posts promoting their software, and not only have comments in those threads.

The lemmy attitude is very anti commercialization, and they don’t know any better. That doesn’t mean we should allow it.

I think new accounts that show up to shil their app should be banned. They’re not actively participating in the community, it’s just spam. There’s been a huge uptick recently.

piefed.ca is set to just follow lemmy.ca, not sure how often it refreshes…

Some instances operate on an allowlist rather than just blindly being open, but honestly the number of problem instances is much lower than you might expect. Most are just wide open. Spinning up an instance will cost you some time and a domain, so it’s not exactly a zero cost effort.

Honestly I have no idea what “chan” has to do with this, I only know it as the japanese honorific. Edit: Oh, chan like 4chan? That one didn’t occur to me.

Thanks, just defederated us

I don’t. Of my 120tb, I only care about the 4tb of personal data and I push that to a cloud backup. The rest can just be downloaded again.

Why are you relying on example.com as a health check? To be really blunt about it, if you’re using it then you’ve misconfigured your stuff.

From their docs:

These web services are provided as best effort, but are not designed to support production applications. While incidental traffic for incorrectly configured applications is expected, please do not design applications that require the example domains to have operating HTTP service.

Based on what? A quick peek and I didn’t see any of the stuff I expect from LLMs

At my house around 10-15. For lemmy.ca and our other sites, 35ish maybe. At work… hundreds.

Check lsmod and see if the kernel modules are loaded. Nvidia smi is probably loading them for you on first run.

Edit your bootloader config and turn off quiet / splash so you actually get a useful boot log.

Imagine a disk shelf of 24 drives being connected by a pair of sas loops. You’d want the faster speeds then. It’s not about individual drives.