I would consider using your Synology for what it’s good at - storage.

My homelab has a Synology DS1618 and servers are Lenovo M90q systems. They have enough compute to get the job done, and use the Synology NFS mount for storage.

I’ll come out with an anti-recommendation: Don’t do GitLab.

They used to be quite good, but lately (as in the past two years or so) they’ve been putting things behind a licensing paywall.

Now if your company wants to pay for GitLab, then maybe consider it? But I’d probably look at some of the other options people have mentioned in this thread.

Oh snap, are you the developer of Viewtube? If so, first off - great job. I do the infrastructure side of IT for my day job but aside from some basic go, I couldn’t code something like this to save my life.

I wish I had the chops to contribute to the project.

Heck, you could do a pre-stage play where you delegate to localhost an ansible.builtin.get_url to download the compose file before doing the rest.

It’s anonymous bulk text posting - great for sharing logs, but don’t discount the more grey side of the internet. If you browse recent public posts there’s often some fun things like scam links, credentials, etc.

It’s definitely fallen out of favor for password dumps though.

I agree, but it’s a hard pill to swallow that Meta is the best partner to grow the fediverse. There are real lessons to learn from Embrace, Extend, Extinguish (look at XMPP and Google), not to mention privacy concerns and content moderation issues that seem to be a “feature” for Meta products vs bugs.

I’m not sure what Zuck is up to, but for whatever it’s worth I think the best think the fediverse can do is be somewhat reactive to Meta’s movements.

If instances start getting overwhelmed with content, then block.

If Meta starts showing signs of EEE, then instances can block.

And us users can move to instances that we feel match our personal stances on things - hate Facebook like the plague? Look at one of the defederated/blocking instances. Do you miss interacting with a larger audience? Stay on instances that are embracing (or withholding judgment) the Meta federation.

It’s a complex topic to be sure, and the only way we’ll know the right way to deal with it is with the benefit of hindsight in a few years

Not the OP, but Eternal September references the massive culture impact on Usenet when ISPs started lowering the barrier to joining the then somewhat exclusive forum-esque part of the internet.

I didn’t intend to use it on the chest freezer - it was mostly for the modem, but since I had spare battery capacity and outlets I thought what the heck.

The power load is practically nothing until it cycles, and even then it’s fairly efficient - my current runtime is estimated to be about 18 hours, more than enough to come up with an alternative if we lose power in a storm.

While I appreciate the sentiment, most traditional VMs do not like to have their power killed (especially non-journaling file systems).

Even crash consistent applications can be impacted if the underlying host fs is affected by power loss.

I do think that backup are a valid suggestion here, provided that the backup is an interrupted by a power surge or loss.

I agree that 99.999% uptime is a pipedream for most home labs, but I personally think a UPS is worth it, if only to give yourself the option to gracefully shut down systems in the event of a power outage.

Eventually, I’ll get a working script that checks the battery backup for mains power loss and handle the graceful shutdown for me, but right now that extra 10-15 minutes of battery backup is enough for a manual effort.

This is why I have about five of these bad boys: CyberPower CP1500PFCLCD.

One is in my utility room for my cable modem and our chest freezer, three back up my homelab and wifi AP, and one is for my office.

They’ve been bulletproof through storms, and when we’ve lost power, but not Internet I can’t keep on working.

The big thing to look for is number of battery+surge outlets vs just surge outlets. Typically they top out at 1500VA - the more overhead for what you’re powering, the longer you can go without mains power.

A screen/display is helpful for at-a-glance information like expected runtime, current output, etc.

This is the right answer. I have dockerized Calibre and Calibre-Web for initial intake, then use Calibre-Web’s OPDS feed with my Moon+ Android app for reading on my tablet/phone.

Calibre handles type conversions, metadata sync, and file organization.

Calibre-Web works well for browser reading on my PC.

Adding to this - I have those adapters to, ans fyi they don’t support jumbo frames.

I’ve switched over to using a publicly resolvable domain name, but with a lan prefix (e.g. lan.mystuff.dev) so that I can do DNS challenge let’s encrypt certs.

Paired with PowerDNS that acts as an authority for the lan.mystuff.dev domain, I can go to a legit certificate/SSL protected https://sonarr.lan.mystuff.dev url. If I wanted to, I could add cloudflare records for the same services exposed through my router (Like for vpn.lan.mystuff.dev) so that both internal and external resolution is possible.

I’ve been converted to the ortholinear layout since my EOTW Preonic, now I have that plus a Planck and a second hand Drop Preonic - but I’m interested in going to a split kb next.

This looks awesome, but may I ask - what function does the dual rollers serve? I get rotary encoders mapping to volume, but I don’t know what two would be used for.

It really depends on what you want to run. I have a clusterHat (four Pi Zeros on a Pi4) that runs a Hashicorp Vault cluster with minimal usage.

The big thing about self hosting is what happens if you (or other people) start to depend on your service - what do you do about hardware failures? Maintenance windows for patching?

To start off, a Pi is fine, but you’ll probably start maxing out your compute and memory (again, depending on workload).

Dropbox keeps hounding me to upgrade to a paid level - back when they were incentivizing sign-ups with free storage for the inviter and invitee, I spin up several throwaway accounts and VMs to give myself bonus storage (I’m just over 21GB).

It’s plenty for some critical secondary backups of wedding photos, docs, and an encrypted volume or two - but I absolutely cannot see myself paying for even their cheapest tier.

After about a year of using Keycloak for some #dayjob side projects, I literally just stood it up in my homelab.

It does have a learning curve, but it supports OIDC and SAML - those two should get most internal services covered.

Back end can federate with AD or LDAP - for the real stinkers who refuse to support SSO. (Looking at you Netbox)

This is what I’ve been thinking lately. There’s no way to unring that bell after an influx of several million people “join” Mastodon through Threads.

Plus threads has an entire team of engineers who can be abused to get out better looking apps, sys admins who ensure the threads servers are running at minimal load - ensuring a top tier Fediverse experience. Already we’ve seen a burst of indie developers for Lemmy and Mastodon, but what if I’m not concerned that my microblog app needs to know health biometric data - threads is up when my niche instance goes down.

That’s how they get you. Come to threads for a “better user experience”! You can still follow the weird Bean memes, but with a better UI/UX! Don’t worry, we at Facebook won’t defederate with everyone else once we hit critical mass!