beto

Hey, friends!

In the last couple weeks I upgraded our instance to run the latest version of Lemmy, and because it was consuming more RAM today I upgraded the VPS to one with 4GB.

This means that the cost for running the instance have gone up. Historically I've been paying around $20 per month out of my pocket to run the instance (after Ko-fi donations), and this will increase to $35 after the upgrade.

Our cost per user is around $0.10/month, so it's not a lot. If you can contribute money, even $1/month, please do, it really helps.

Thanks!

Let's wait for the reviews to see if the transport is any good. It would be nice to have a decent quality new walkman on the market!

No major changes, just bug fixes.

Let me know if something's not working.

There has been a shitty incident where someone intentionally posted CSAM to a Lemmy community that unfortunately someone in our instance was subscribed to. Due to federation, the images might have been copied over to our instance storage.

Because of that I've taken the following measures until we understand the problem better:

1. I've defederated from lemmy.world and sh.itjust.works.
2. I've deleted all images uploaded since 2023-08-27 00:00:00 GMT.
3. I've disabled the image server.

I'll post updates when I have them. Thanks for understanding.

Link is also on the sidebar. We'll use it to keep people updated on releases and give support when people can't reach the Lemmy instance.

Also, I'm an IRC kind of person, so forgive my noobness with Matrix. 🙃

Hey, hey!

This morning (Pacific time) I changed the nameservers on lemmy.studio, switching to a service that provides a CDN and a WAF. It's a cautionary measure to protect us from DDOS attacks. That caused some instabilities, sorry for that.

I also noticed that the CDN was too aggressive, caching URLs regardless of the Content-type, and it seems that Lemmy uses content negotiation (the same endpoint can return HTML or JSON depending on the headers). I've fixed the caching to take the headers in consideration, so hopefully you won't be seeing JSON responses when accessing the website on your desktop. If you do, let me know!

On a related note, I created a spreadsheet showing the monthly expenses on the instance, and how much we're making from user donations on Ko-fi. The sheets was shared with the current supporters. If you can spare a dollar (or five) every month to help the instance run smoothly please do, anything helps.

Thanks!

cross-posted from: https://lemmy.studio/post/488779

Out of all the legendary albums on this list, I doubt many of them had their origins as an abandoned rock opera. Many arrangements and scraps of Pete Townshend’s abandoned Lifehouse project became the basis for Who’s Next, an album that has no underlying theme or storyline. This sense of freedom allowed The Who to focus on making great individual songs rather than an overarching story.

The result is The Who growing up in public. The songs combine the hard-hitting energy of the band in their youth with the more experimental elements explored on Tommy. The most noticeable improvement is Roger Daltrey’s voice, reaching heights that were only hinted at in the past. Keith Moon’s drum solo followed by Daltrey’s scream at the end of “Won’t Get Fooled Again” remains as one of rock’s greatest moments.

While everyone knows about the singles, from the opening keyboard of “Baba O’Riley” to the building acoustics of “Behind Blue Eyes,” every song on this record is a potential hit. Listen to the explosive chorus of “Bargain.” Check out a rare lead vocal from bassist John Entwistle on “My Wife.” With tracks like these, it’s easy to see why Who’s Next moved The Who from a great band of the ’60s to a rock superpower in the ’70s. — Joe Marvilli (2010)

Listen to it here.

Hi, folks!

Since I started lemmy.studio a lot of people have reached out offering help and financial support for our instance. While right now the hosting costs and administration efforts are mostly negligible, I want to make sure the instance can grown in a sustainable way, and I have some ideas on how to do that.

For financial support, I created a Ko-fi page. If you're not familiar with Ko-fi, they're very similar to Patreon, but the fees are optional and they pay creators immediately (not monthly). They also support one-off donations, which is nice. Right now there's a single monthly tier of $1 (though of course people can pay more, if they want).

The current monthly costs for the server, domain, and backups are around $25/month (edit: $45), and we have almost 200 users, so I'm hoping it will be relatively easier to get the costs covered. Any excess income would go towards a reserve to ensure we have a good runway and funds for upgrading the server as we grow. I'm going to create a spreadsheet and share with everyone donating, to keep this transparent.

Second, so far I've been the only administrator in our instance. The volume of reports is still small, usually a handful of spam posts per week, but as we grow we're going to need more people removing spam, banning bad actors, and ensuring we're building a supportive and encouraging culture across our communities. I'm planning to reach out to some "long-term" users that have offered help, but feel free to DM me if you want to help.

(An intermission about culture. This is a music-focused instance, and we're here to share the music we love and the music we make, and also help each other by sharing our process and tools. We're all united by the belief that music makes the world better. So let's put our differences asides, focus on what we have in common, and try to always be kind and supportive!)

Third, I've been also the only sys admin of the instance, responsible for backups and upgrades. Right now the instance is running on a personal team on Digital Ocean, together with other personal projects. My plan is to move the instance VPS to a separate team, so I can invite 1-2 people to share those responsibilities with me. Initially these would have to be voluntary, but I'm hoping that if we get enough support through Ko-fi we can compensate people doing this work.

Finally, for the long run I'd also love to set up a governance model. I'm not sure how this would work, but I'd rather have something where we can make decisions collectively.

Let me know your thoughts!

Edit: I just got my first AWS bill after switching the instance storage to S3, and while it's not bank breaking it caught me by surprise. My monthly bill is around $3-5, last month it was $28.64.

Edit 2: people are migrating to our instance from waveform.social because of malicious activity (DDOS and spurious file uploads), so if you can support the instance please do. I'm planning to move the database to a separate VPS, and maybe look into using CloudFlare to prevent us from being affected, and that will cost more money.

Hi, folks!

Sorry for the latest instabilities in the instance. For some reason we were getting spikes of 100% every now and then, and that would make the instance unstable.

I just finished upgrading to 0.18.3, which was released today. There should be a lot of CPU improvements since 0.18.0, so hopefully the instance will be more stable from now on.

Let me know if you encounter any problems! Since we're usually in the dark during instance upgrades I created a Mastodon account for the instance, so I can post status updates. You can find it here: https://mastodon.social/@lemmystudio

Meta just released Threads, a Twitter clone that is promised to be compatible with ActivityPub in the future. You sign up with your Instagram account and have the option of bringing in the people you follow there, even if they haven't a Threads account yet.

I'm planning to upgrade to 0.18 today and move the storage to S3, so I'll bring the instance down between 5-6pm PDT.

Update: this took longer than I expected. I started at 6pm because of work. I shut down the instance and took a snapshot, then brought it back up and upgraded to 0.18. The upgrade worked fine. I brought the services down and configured the instance to use S3 for object storage. That required migrating existing objects to an S3 bucket, which took quite some time. Then, after bringing the services back up again I got a "Server error" message. Unfortunately I had to stop at 8pm since I had something scheduled. At 9:15pm I was able to bring the services up successfully, and we're back!

Update 2: turns out we hit this bug, and the instance was down around 11pm. I changed the hostname of the instance and now it seems to be working fine.