I want to upgrade some of my older machines with some new, high(er) capacity SSDs (SATA and nvme). I don't need super high speeds, just something in the TB range in terms of storage.

Problem is, there's so much garbage out there, I can't really tell, which SSD is inexpensive and reliable and which is just utter garbage.

I thought about buying new, but last gen Samsung/WD SSDs.

Intenso and Fanxiang both seem to have been around for a few years, but reviews seem to be mixed.

Basically the title. I have Nixos running on a server that runs completely headless and while playing around today, I noticed that the rebuild takes longer than expected and apparently that's because firefox gets compiled.

Now, I don't have any GUI installed, and even if I had, I don't see a reason to compile FF from source.

My packages are just Jellyfin, Samba, Gitea, Nextcloud, virt-manager. None of these should depend on FF.

I have a QCOW2 image (Homeassistant VM), that I ran for several months without problems.

A few days ago, I reinstalled the VM host,so I copied the image to a backup drive and now wanted to start a VM from this image.

However, it always end up hanging at "booting from hard disk" and takes up 100% load on one core.

On the VM host, I imported the image like this:

# copied from HAOS wiki
sudo virt-install --name hass --description "Home Assistant OS" --os-variant=generic --ram=2048 --vcpus=2 --disk /var/vm/hass.qcow2,bus=sata --import --graphics none 

To ensure that my host wasn't broken, I tried the same image on another machine, that I know can run VMs (virtual machine manager, using the GUI), but same result. One core at 100% and no change at all.

I even let it run over night, but it was still at this point.

One machine runs NixOS, the other Debian 12.

What could cause this? There are no errors in journalctl or /var/log/qemu.

I'm currently a senior developer, but relatively new in the role of a "lead". In my current project, I'm having a kind of co-lead and we have two devs working in our team. So a rather small enterprise.

Now my boss told me, that going forward, I will probably be leading larger and more complex projects (possible rather soon).

Since I'm constantly doubting myself, I would really like to learn more about how to be an effective/likeable lead. I've had too many "leads" who were just dogshit, professionally and as a person. I don't want to be that (at least the professional part).

So, I guess my question is: what helped you? Books, articles, just random hints or strategies? I'll take everything.

Ich lebe in einer größeren Stadt, in der es aber de facto nur zwei Lokalzeitungen gibt, die beide auf mehreren Ebenen verwerflich sind und dann auch noch journalistisch ausgesprochen fragwürdig.

Leider sind genauso diese beiden aber auch die einzige Quelle für "echte" lokale Nachrichten. Gefühlt bekomme ich kaum etwas von meiner eigenen Stadt mit - und das ist eigentlich ziemlich schade.

Deswegen mal die Frage an euch: beißt ihr in den sauren Apfel und kauft die Zeitung? Ist die Lage bei euch besser?

In Germany there's an app called "Jodel", which is essentially like a localized reddit/lemmy. That means, you only see posts from people near you (the default is something like 10km, I think).

This is of course awesome for localized events, Craigslist style posts, or just discussions about local stuff.

I wondered, despite creating local communities on Lemmy or tags for your city, is there anything like it on the fediverse?

I have an HAOS-VM running on a machine attached to my router via cable, and another (wifi attached) machine that I want to WoL from HA.

Apparently, my router does not forward WoL packets from wired ethernet to wifi, so I just plugged in a usb-wifi dongle and connected HAOS via the dongle - wifi works, no problem.

WoL in general also works, I can wake the machine up from my MacBook without a problem. I can also use the very same dongle on a Debian desktop to send WoL packets.

However, doing it from HA fails. My configuration.yaml looks like this:

wake_on_lan:
switch:
  - platform: wake_on_lan
    name: wake_<machine>
    mac: <mac>

I tried deactivating the KVM brigde, so that the VM really only connects via wifi, but no changes here either.

There are no errors in any logs, so it doesn't look like it's failing per se.

To me it looks almost like the packet is send to nirvana, but there's no way to configure the network interface in the WoL platform either.

Update

I kind of got everything to work, albeit in a rather weird way.

There already is a built-in WoL Service, accessible via the GUI, however, I have to use the Advanced Mode to access it, otherwise it simply doesn't show up. Using that service and the proper broadcast address, I'm now able to wake the target machine via a button. What's a bit concerning to me is that the "advanced mode" toggle seems to be lost after a reboot. I get a warning, that the WoL service couldn't be found, that only resolves after setting the advanced mode off and on again - that's rather stupid.

I'm thinking about buying a small 3D printer for the odd project once in a while.

Problem is, I will not use it very often and I don't have much desk space for it to sit around.

Ideally (and I know this is utopia), I would like a device that I can pull out of a closet, fasten four screws, plug it in and be ready to go.

Is there something even remotely like that available? Every review I've seen just seems to assume that printers are basically static.

I have a Dell Optiplex 3060 here, that I used as a backup desktop with Linux, but now I'm trying to use it essentially as a streaming host for games (Fallout, GTA...), unfortunately that means Windows.

And even less fortunate: Windows seems to think, fan speeds only know one direction: up.

Essentially, the machine starts nice and reasonably quite, but after some load (e.g. a game), the fans never spin down again. Even if the temps are fine (all cores at <30°C, GPU at 48°C), it keeps running in turbine mode.

The only "fix" is a sleep or power cycle.

Since this machine is supposed to run relatively long hours and sit in my room, this is quite annoying and I'm kind of out of ideas.

Newest BIOS and all the Dell Magic™ are installed.

I just browsed eBay a bit and saw that older, used SAS drives can be had pretty cheap - 30€ for 4TB, but of course rather old drives, sometimes 10 years old.

Now, I wouldn't expect ultra reliable, ultra fast, super cheap drives here. But this offer seems compelling, even buying a spare drive for higher redundancy would still be pretty cheap.

Question is: am I too optimistic here? Are these drives bound to fail within 3 months?

I'm using Feedly (google reader clone) to keep track of my news. However, there are tons of duplicates (same event/topic different sources).

I was just thinking about using text summaries + similarity analysis (possible AI driven) to cluster groups of articles. Are there already solutions for that? I could build it myself, but I'm not exactly the best web dev.

I'm trying to use an RPi Pico W as a temp/humidity sensor using a DHT20.

It kind of works - at least sometimes, but I keep "losing" sensors more or less randomly.

I connected everything up like here (using MicroPython): https://github.com/flrrth/pico-dht20 There are currently 4 sensor-boards, 3 soldered, one on a breadboard.

The error modes I could observe are:

1. DHT20 fails to init - sometimes after the first read, sometimes after days. Resetting the machine works sometimes, if not, power cycling usually does the trick

2. The board just "stops" after about 5min - the serial console just says "device disconnected". Power cycling is the only option.

My measurement work by having a timer fire every minute, connect to wifi, read from the sensor, and then send an mqtt message (either the values or an error message) and shutdown wifi again.

My current ideas why it could fail (but I'm not an electronics guy at all):

• There is some kind of "rogue current" messing with some IC.
• Some component is broken
• Maybe the power draw is too low or issuing sleep() messes with the USB-power connection somehow?

For me the problem is, I don't really know where to look for errors. The software works in principle, the soldering seems to be good enough to sometimes work for days, and looking too deep into the whole electronics side is beyond my capabilities.