Hacker Newsnew | past | comments | ask | show | jobs | submit | jacksgt's commentslogin

Which software are folks using to run self-hosted / on-premises S3? I'm looking to replace MinIO in my homelab and would like to keep a S3 management UI.


Some that I've seen:

SeaweedFS: https://github.com/seaweedfs/seaweedfs (Apache 2.0)

Garage: https://git.deuxfleurs.fr/Deuxfleurs/garage (AGPL)

Zenko: https://github.com/scality/Zenko (Apache 2.0)


But which ones have a management UI?


None of those, if a quick search of "ui/gui/user experience" is of any indication.

Also looking for an alternative to MinIO since I also rely on the GUI for some things. But it feels like these things should probably be separate? Since they all expose a S3 compatible API interface, couldn't we have one GUI that can be used regardless of the backend?


It's great when building software and testing locally, e.g. in a docker setup.


diffoci is a tool that compares Docker/OCI container images for helping debugging and creating reproducible builds. Similar to container-diff, which has unfortunately archived in March: https://github.com/GoogleContainerTools/container-diff/commi...


Location: Switzerland / EU

Remote: Yes

Willing to relocate: No

Technologies: Kubernetes, OpenShift, Containers, CI/CD, Helm, Prometheus, Grafana, Linux, Go, Postgres, Python

Résumé/CV: https://blog.cubieserver.de

Email: jacksgt at mailbox dot org

Description:

Hi HN, I'm currently looking for my next step as a Cloud/DevOps/Platform Engineer. Out of the technologies listed above, I particularly consider myself a expert in OpenShift and Kubernetes with several Yeats of professional experience. I like building internal platforms, private clouds and focusing on developer experience. More generally I have solid CS fundamentals and am good at problem solving, brainstorming, rapid prototyping and knowledge sharing.


Adam Gordon Bell interviews Adam Jacob - the original developer of the Configuration Management Tool "Chef" - and his journey of leading the company.


I think Loki is pretty much the easiest thing you can find (if you want it to be multi server, at least). Loki whole approach comes down to avoiding expensive indexing (compared to Elastic search et all.), and really on "grep" for searching instead.


What I hate about Elastic is it special grep syntax I can never get right...

I tried loki around v1.0 and it didn't seem to offer much back then...


I would really like to start converting some of my personal media and websites to use JPEG XL, but the momentum doesn't seem there yet - despite clear technical and practical benefits.


I just went ahead and did it, then included fallbacks to supported formats. What I really want is a standard way to let the browser request lossy or lossless images.


"CoRecursive: Coding Stories" (from Adam Gordon Bell) is one of the Podcasts I look forward to the most. Incredible high production value, yet very personal and down to earth.

Probably one of the best episodes is the interview with Andreas Kling, the author of Serenity OS

https://corecursive.com/serenity-os-with-andreas-kling/


That was a great episode. I also really enjoyed this one: https://corecursive.com/prison-programming-with-rick-wolter/


Came to suggest the same! Was nice to be able to upvote another users suggestion instead!


What do you mean by error correction? In which sense does restic not do error correction?


If you flip a bit in some file in a restic or Borg repository, you will lose data (the amount of which depends on the location, possibly the repository is corrupted). With error correction, that bit flip is corrected by the software from parities.


  Location: Switzerland / EU

  Remote: Yes

  Willing to relocate: No

  Technologies: Kubernetes, OpenShift, Containers, CI/CD, Helm, Prometheus, Grafana, Linux, Go, Postgres, Python

  Résumé/CV: https://blog.cubieserver.de/about/ 

  Email: jacksgt at mailbox dot org

  Description:
Hi HN, I'm currently looking for interesting consulting / freelancing positions.

Out of the technologies listed above, I particularly consider myself a specialist in OpenShift and Kubernetes - as I'm a currently an OpenShift cluster admin, I have several years of professional experience with these.

I like building internal platforms, private clouds and focusing on developer experience.

More generally I have solid CS fundamentals and am good at problem solving, brainstorming, rapid prototyping and knowledge sharing.


Off topic: it's such a breath of fresh air to read this content without 1) having to close half a dozen popups and 2) all in a single post and not painfully spread out across multiple messages.


Doesn't do targeted ads, so doesn't have to have an EU cookie popup.

Doesn't have an app, so doesn't have to try to make you install the app.

Doesn't do registered users growth hacking, doesn't have to have sign-up dialog.

The problem is, if and when they decide to monetise this thing they will have to have all of these because the money people and the analytics will tell them they have to.

Everything is much more fun when it's paid by someone else, that's why the old web was so nice. The content was produced for free and the distribution was handled by VCs. Today, these VCs are recouping their investments.


No, all the bloggers I was reading on the old web were paying for their hosting a few bucks a month and no VC were involved.

Some of them included ads but they were mostly in a sidebar.


They were paying for the costs? Then they were the VC. Some of them actually grew to be huge and wealthy.


You equate VC with tracking and then equate VC with anyone that provides money whatever the intention. By that logic, you imply that anyone providing money (including a mom for his 14yo son) will track everyone and gather a lot of stats and push an app ? Remember WhatsApp did cost a few bucks a year and was self sufficient, it was bought by fb not because of cost of operation but to ensure market share.


me paying $5 to wordpress: i'm a venture capitalist now mom

(seriously, this is a very silly way of using terms to cover extremely small amounts of capital)


Did you mean paying for PHP site hosting instead, or paying to remove ads on the WordPress.com SaaS?


The size of the capital doesn't change the reality of the financial structure.


I don’t think I understand the argument that you’re making.

Are you saying that anyone that pays for anything is a venture capitalist?

That doesn’t seem like a very useful definition, so I think I’ve misunderstood you ’:-)


> «if you use money — you are a capitalist»


> Doesn't have an app

It does have an app. See https://joinmastodon.org/apps for not only the official app, but also literally dozens of third-party apps.


Distributing a static HTML page content does not need a VC. Nginx on an RPI on my home connection does provide sufficient level of performance and availability. If I need more because my content is way popular, I guess a monetization scheme (asking for a tip) might cover it ?


> If I need more because my content is way popular, I guess a monetization scheme (asking for a tip) might cover it ?

Hetzner will give you a powerful server for ~30 bucks/month and includes 20TB of bandwidth for free (and overages are charged at ~1$/TB, almost 90x less than AWS). That's enough to host and serve a lot of content.


I really hope you're right but I don't think it will play out that simply for mastodon. I have high hopes for it, hopefully these problems are solved but with this rate of growth, what will happen when there's a big political event that used to cripple Twitter back in the day? I don't think tips will cover it.


Mastodon (and the general "fediverse") is an inefficient disaster by design, but if you ignore that and go back to old-school forum software, your Raspberry Pi will be just fine for a few hundred concurrent users (and way more for read-only traffic).


If you are not getting paid for your servers you are the VC/Angel.


A bit of a difference in motivation between a hobbyist and a VC.


Agreed! Here you are more a patron or sponsor who's disinterestedly giving spare money for something you think is good for the rest of the people.

Although the Angel investor expects a few shares of a company, I like the analogy of an 'angel sponsor'.


How does that make them the VC/Angel?


They provide the capital required for that venture.


So if I order lunch, am I providing the capital required for the venture of ordering lunch? Is there anything in the universe that's not venture capital?

When all you know is a hammer...


No. If you are s 14 years who is into hobbits and you are writing a fan fiction, your mom is a venture capitalist when feeds you and gives you pocket money which you used to buy a domain. She can be a venture capitalist or angel investor depending on her expectations. I guess it also can be a case of racketeering if she does that you just shut up, then you are the VC investing in your own stuff with the money you extorted by being a really annoying kid :)

Anyway, it doesn't matter. The gist is, someone else than the user paid for the experience without immediate expectation for profit from side channels and that's how we had free and awesome things.


Most of the time, it wasn't other people. People were paying for their own hosting for their own purposes, much like how I order lunch.


No, you miss the point. Because of those who were paying their own hosting the rest was able to have free and awesome stuff. The paying out of pocket to show your own stuff to other people was the act of funding a venture. Some ended up turning these into profitable businesses.


I just think it's weird calling hobbies "venture capital". But if that's your thing, at least now I understand it.


> The content was produced for free and the distribution was handled by VCs

And even that, content distribution was handled by a volunteer happy to chip in a few bucks to pay for shared hosting.

Content distribution (and infrastructure in general) nowadays is cheaper than ever thanks to technological advances (today's entry-level MacBook is more powerful than a lot of servers from 10 years ago).

There is absolutely a way to distribute content for very cheap nowadays if you know how to - you just have to avoid the rent-seekers like cloud providers.


On that vein, I recently joined home-barista, an old school web forum for coffee geeks.

That site is seemingly frozen in time from the early 2000s. There are no trackers - there's no need, since it is already filled with a self selected group. The ads are just simple banners. And best of all it filled with a group of passionate, kind and helpful folks. A simpler site from a simpler time. One of my favorite haunts on the web.


> Doesn't have an app, so doesn't have to try to make you install the app.

Well, it does, but does nothing user hostile with it, so doesn't gain anything from coercing them to install it.


> Everything is much more fun when it's paid by someone else, that's why the old web was so nice.

What are you talking about? The old web was filled with ads. They just weren't tracking you every nanosecond of your life.

Google's AdWords was launched in 2000. DoubleClick that Google acquired in 2008 was launched in 1995. Ad exchanges are from 1998. And so on https://en.wikipedia.org/wiki/Timeline_of_online_advertising


> They just weren't tracking you every nanosecond of your life.

Exactly.

But only some of that was paid by ads.


> But only some of that was paid by ads.

Ah yes. They were paid by altruists who had the foresight to know that 30 years later they will be able to track everyone and collect their dues.


Those who relied on making profits were decimated as VC funded user experience was much superior. It was an era of land-grab.


Those who relied on anything were decimated in the bubble of 2001.

Also, you're pretending that all of those sites were making a profit, or operated under the assumption of making a profit. There was a lot of money thrown at any and all internet companies by the end of 1990s. It's just that the market was much smaller.


Once upon a time this was called a “blog post”.


You might like it even more when reading through elk.zone

https://elk.zone/infosec.exchange/@marcan@treehouse.systems

There is popup about the preview, but that might be acceptable. There is zen mode to remove all disturbances on right bottom.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: