cpw

joined 1 year ago
[–] cpw@lemmy.ca 13 points 5 hours ago

I took what came out of the box, very much factory default here. My offspring are figuring it out at the minute, Imma let them cook.

[–] cpw@lemmy.ca 48 points 3 days ago (2 children)

But the CEO's third luxury yacht? What about that?

[–] cpw@lemmy.ca 21 points 3 days ago (6 children)

Price decreases are actually negative inflation and have all sorts of whacked out effects on an economy. It was a concern during COVID due to the huge drop in consumer spending forcing some prices to start to decrease.

[–] cpw@lemmy.ca 1 points 3 days ago

No idea. It was working great until this morning.

7
submitted 3 days ago* (last edited 3 days ago) by cpw@lemmy.ca to c/lemmyconnect@lemmy.ca
 

This morning, connect is empty. All view modes are blank like the screenshot. Randomly ONE post shows up if I refresh a LOT, but immediately disappears. I can read and view my profile, but any posts I try and load from there are empty too.

Edit: I appear to be able to scroll the blank space for quite some time.

Editedit: other clients do not appear to be affected.

[–] cpw@lemmy.ca 2 points 5 days ago

Probably some sort of trolling effort sadly. Like an as yet unaired bit for a tv show.

[–] cpw@lemmy.ca 14 points 5 days ago (1 children)

And this is the fifth line of four..

[–] cpw@lemmy.ca 10 points 1 week ago

I'll bet the Intel management engine is just as "vulnerable". The only context this is likely a concern is large scale corpo deployments, without verified supply chains to the source. Love how the security researcher handwaves that there's "plenty of existing exploits" that can be used to install the exploit into the SMM, without giving any suggestions of how.

[–] cpw@lemmy.ca 21 points 1 week ago (1 children)

Bought a mach e instead of a model 3 or y because Elon is a dick. Glad to stick it to the wanker. The mach e is awesome btw 👍

[–] cpw@lemmy.ca 1 points 2 weeks ago (1 children)

Your second point is key. In an ideal world, open source could rival and even beat the best paid offerings (see: blender). But in most cases it just doesn't. There's not a dedicated team working on the open source products, working with HCI experts and designers on every detail of the product. It doesn't preclude the open source being better (see, again: blender), but it does push a LOT of workload onto a bunch of hobbyist developers working in their spare time. The resultant burnout is typically why you see these projects sputtering along for years and years. I don't know how to solve those problems either, but they're your real "roadblocks".

[–] cpw@lemmy.ca 4 points 2 weeks ago

I agree with your fundamental point, learning new shit is definitely fun for me. But there's lots of different people and some just don't. I can definitely sympathize with someone who's income depends on one of these workflows, and why they can't disrupt that for "fun learning sake". There's only so many hours in a day and some people have different priorities.

[–] cpw@lemmy.ca 33 points 2 weeks ago (8 children)

This guide is misleading. Sure, the product functionalities overlap, but if you have a mature workflow, you will not be able to switch without investing a LOT of effort in relearning your workflow on the new product stack. This is one of my MAIN reasons I hate the "I tried to switch to Linux and failed" genre of content. You're not going to find identical like-for-like replacements in Linux world that won't require significant effort to relearn. It's something us Linux users through and through need to bear in mind.

Also, we need to be cognisant that "just switching to Linux" narratives, fueled off infographics like this, will lead to frustration and dismissal.

No, I don't know how to change this - and morphing e.g. gimp to be a clone of Photoshop isn't the answer either.

[–] cpw@lemmy.ca 1 points 3 weeks ago (1 children)

Fair enough. But the fact I can't even use it to connect to my homelab proxmox cluster kinda has to be a dealbreaker for me. Even a trial period to allow me to try and experience everything would be sufficient in my opinion. On the fuzzy thing, I'm using gnome desktop, with latest gnome shell in debian sid, on an Nvidia 20280 using the proprietary driver (latest in debian experimental). It's connected to three 2k/1440p monitors running at 144/60/60hz. If that helps at all. The tooltips are most notably fuzzy. It looks like it's being antialiased multiple times or something?

 

Hi, so I have a very individual homelab. It's a collection of stuff accumulated over nearly 30 years of doing weird stuff.

For the past 9 years it's been running as a bunch of lxc containers (privileged because unprivileged did not exist, back then) but several of those containers are p2v conversions of physical hosts dating back to debian woody and earlier. They're all upgraded to at least buster, most are bookworm. Stuff like asterisk, email, home assistant, nextcloud, matrix synapse run there these days.

The server is a 15 year old HP gen6 thing, and is getting quite long in the tooth. There's also a dedicated cheapy microserver with an i4 running opnsense on bare metal as a firewall.

Trying to run stuff like local voice stuff for home assistant is showing the HP's age quite badly. Also, our area is getting fibre, and the opnsense box is maxed out at gigabit. More speed would be nice.

So, I'm in two minds. The homelab has been a lot of fun over the years, but I'm over 50 now, I want lower maintenance. This latest wave of upgrades is making me rethink the next 20 years of homelab. I don't want to leave something stupidly "only me" if I were to die tomorrow (diabetes is a fickle bastard). My wife might want to try and carry on this thing - it runs some useful stuff around the house (but it should be noted that nothing in this house requires a server or cloud) - and that's not going to happen with the current solution.

I think I might have a path, using proxmox, from where I am now, to something that can be deployed on e.g. a bunch of ms01 class devices. I'm thinking to convert the existing HP server to proxmox, to allow me to redeploy all my existing lxc containers into the proxmox world. As I acquire hardware over the next year, I can look at a k8s migration of the services onto a small, MUCH lower power cluster. One of the keys is that I don't want to have big outages of services for days or weeks while I migrate everything so it's gotta be a rolling upgrade as it were.

I'm here soliciting feedback. Has anyone ever migrated from a deeply legacy homebrew homelab into something like this? Does it reduce the workload long term? What's the practicality of this for someone rather less tech savvy?

Thanks!

view more: next ›