a local filesystem remote
a cloud provider remote
rclone serve s3 to serve a local s3 compatible API for the local remote
something orchestating an rclone copy, rclone sync or rclone bisync command to sync the local and cloud remotes. Could be triggered on a timer, or using another tool to watch the local directory for changes and trigger a sync.
I think you can run rclone as a service, they recommend you use bisync on crontab, for example.
What's bisync? I don't really know, as it's new to me, but it's an experimental beta feature from rclone and sounds neat.
https://rclone.org/bisync/#sharing-an-encrypted-folder-tree-...
> Sharing an encrypted folder tree between hosts
> bisync can keep a local folder in sync with a cloud service, but what if you have some highly sensitive files to be synched?
> Usage of a cloud service is for exchanging both routine and sensitive personal files between one's home network, one's personal notebook when on the road, and with one's work computer. The routine data is not sensitive. For the sensitive data, configure an rclone crypt remote to point to a subdirectory within the local disk tree that is bisync'd to Dropbox, and then set up an bisync for this local crypt directory to a directory outside of the main sync tree.
This is quite an amazing man page, at no point does it say what the command does.
It has a lot of details upfront about it being beta, and where its working directory is, and what the limitations are. But what does it do?
https://github.com/rclone/rclone
Specifically, check out this new beta feature called bisync which has use cases that sound a lot like yours:
https://rclone.org/bisync/#sharing-an-encrypted-folder-tree-...
Before investigating it for eventual production use, I heard quite a bit about how complicated it was to use, but for my use case (storage for enterprise and academic k8s clusters) it's actually been quite simple to deploy and use. cephadm (one of many ceph management tools) can handle nearly all our bootstrapping and management needs. Little to no tweaking or configuration needed. Fairly low overhead. Very reliable and resilient to adverse conditions. Easy to handle different storage types and data retention needs.
One thing I will warn you, if you go the ceph docs site right now and just start browsing, it is in fact quite overwhelming because ceph has the capacity to handle a ton of edge cases and unusual environments. I'd recommend taking 15 minutes and build this [1], which gives you a fully functional toy-sized ceph cluster on a single node. Then, hit the docs to fill in the gaps of what you need to know for your deployment.
I went into the office and met the two founders. Both were nice and welcoming, but it felt like there was absolutely zero process or structure in the company at all. They didn't have a product or vision. It was more or less "we're building cool stuff".
I wasn't asked a single technical or business question. I was asked what I would like to work on and they suggested I come into the office and do some open source work on whatever I want and if I enjoyed it they would hire me.
Not a bad experience, but very bizarre and out of the norm.
Fly does something similar, except they pay you for the evaluation work.
It's sad because this was a great product that grew to be what it was as a result of community interest, and now they're destroying the community version to try to force revenue. Maybe they'll even triple the price again for no reason other than that they can.
goals change man. who of us knew what we'd be doing today 9 years ago when minio was founded
fetch("https://dl.min.io/server/minio/agplv3-ack", {
mode: "no-cors",
})
https://en.wikipedia.org/wiki/Neo4j#Criticisms
> Neo4j sued PureThink, a small business that had used a power created under the terms of the GNU AGPL, to remove a restrictive Commons clause that Neo4j had added to the AGPL license. The United States District Court for the Northern District of California made a decision on 2024-07-22 to impose $597,000 in actual damages on PureThink, having previously decided that PureThink had violated the DMCA by removing the Commons Clause from Neo4j's AGPL license, and that it had violated trademark law by continuing to use the name Neo4j in selling to government agencies.
https://jamesoclaire.com/2025/05/27/how-to-self-host-your-ow...
The deleted console UI helped you understand the data in the system, get started with it and do some basic admin. But for any real application built on it, you'd need to build some of your own tooling and get acquainted with the mc command line tool. It was a very shallow UI - it didn't help with ACLs where you just had a big box to paste your JSON into.
It also had a lot of asterisks to show you which tabs were premium - at least they were visible.
Now it lets you browse buckets, but not create users? All the premium tabs are gone. If I hadn't used it before this I'd have assumed it was just raw and unfinished.
It's solid software, but the documentation really only covers the happy paths, and there's no community around it.
Annual pricing for the complete package & support is about 75% of Backblaze's, but you have to bring your own hardware and network. I guess it compares much more favorably against S3, or maybe better still if you're going to exabytes.
But yeah this just seems like damage to the on-ramp for people who want to grow into it.
Ceph seems like the only battle-tested alternative. It has a community around it, but the setup is more demanding than minio.
That's one of my favorite obscure projects
It's also perhaps worth pointing out that they once had a very strange interpretation of AGPL that caused quite the drama some years ago: https://github.com/minio/minio/issues/13308#issuecomment-929...
> Also, just want to mention that the AGPL license requires that all software connecting with MinIO be 100% open source for you/your users not to be in violation of the license.
That's the type of people we are dealing with here, probably best to stay away from it. If AGPL itself didn't already deter you.
S3 is pretty well established and commoditized and there are alternatives.
There is literally no moat here, let alone one that would hold up for charging nearly 100 grand. Seems like a pretty bizarre play.
- Ceph
- SeaweedFS
- Garage
- CubeFS
- Apache Ozone
Do any of them fill the niche that MinIO sits in? (Relatively easy to set up and operate and with all the typical enterprise bells & whistles)
Ozone/Ceph -> Way more complicated
Among others, an inevitable torrent of forks of MinIO by people who want to keep the UI but backport any other security or architectural fixes.
Or, as the author points out and if it works for you, a 33% price increase to go to B2 from Backblaze and not have to manage your own hardware, network, infrastructure, etc.
Or, for enterprise shops, another storage appliance with an S3-compatible interface (e.g. Dell has some options)
But for those wanting other options and self-host something it's not clear cut.
There’s probably a better more original source…
anotherhue•1d ago
If a developer needs a pay day, are we surprised if they take it from an investor who wants returns?
phoronixrly•1d ago
nine_k•1d ago
The saddest thing here is that MinIO is apparently not important enough for the big names that use it to receive donations / sponsorship funds sufficient to continue as is :-/
overfeed•1d ago
pjmlp•13h ago
Downstream only gets to do whatever they are willing to contribute upstream.
j1elo•1d ago
mikepurvis•1d ago
I think in that scenario, it would be much more clear-cut that the responses would be:
1. Well that's a bummer that MinIO wasn't sustainable as it was; I'll just stick with 1.x forever, or
2. Well thank goodness this new company stepped up and is taking over maintenance and development; I'm willing to pay them to get ongoing support and having continued access to the web console feature in new versions seems like a fair price to pay.
jchw•1d ago
I have some unsolicited advice: please don't open source something that you are actively planning to sell. Nobody will take this advice of course, because delivering your product as an open source project is an easy way to generate some leads and even get some free labor here and there. But if you don't want people to feel deceived, start with your expectations out front and magically people won't be angry at you.
On the other hand, I agree people wouldn't be angry at company B in this scenario, because they're not the ones who set the incorrect expectations to begin with. OTOH though I really do doubt "we just removed some features so we could charge you for them" would be a successful pitch as a brand new vendor for something.
I'm struggling to understand how there's even still a lot of debate on this subject because it's very simple, if you can pull off a sustainable FOSS project then the world will be grateful but in many cases it becomes very clear over time that there was effectively no plan, just a lot of hope that maybe things would work out somehow. Whatever good that is.
mikepurvis•1d ago
I think the pitch would be more like "we want to focus our more limited resources toward ensuring the core of the product is as good as possible, and there isn't a way to continue shipping an unmaintained version of the UI, so we haven't brought that part of MinIO into the community version of <new product name XX>. By offering it as a paid option, we hope to be able to sustainably maintain both the open and closed parts of XX over the long term."
jchw•1d ago
With open source, I think what companies really want is an open source project that is maintained by people with stakes in the project, but complementary ones rather than primary. Helping pay for a new business seems like a worse proposition than just getting some major stakeholders to donate a headcount or two; yeah, it's expensive, but it's probably a better status quo and should nearly indefinitely cut out any concerns about rent seeking behavior.
turtlebits•1d ago
jchw•1d ago
linsomniac•1d ago
Also, github sponsors seems to work some as well. I enabled it last year and I've had a few gracious souls throw up to $50 at a time into my tip jar, though I've sent out way more than has come in (but that's ok).
I sure wish I could get my company to start contributing some significant funds though. We get a huge benefit from open source, but never seem to have the money to spend towards it.
lenerdenator•1d ago
What, and pay the people actually creating the value?!
robertlagrant•1d ago
lenerdenator•1d ago
Everyone's happy to get software in exchange for nothing from a project with only a handful of people working on it, until there's a problem in the code that is a real bear and causes massive vulnerabilities, and on a long enough horizon, that will happen to any sort of code running on machines connected to the internet.
If your entire business model relies on that code, it'd be wise to back the project, even with token support. I'm talking about less than cost of one team building outing.
lenerdenator•1d ago
We need more people looking to Jimmy Wales as an example of how FLOSS can work instead of Silicon Valley.
ujkhsjkdhf234•1d ago
lenerdenator•1d ago
He's led the effort in a way that has enabled to project to stick to the core values.
ujkhsjkdhf234•1d ago
overfeed•1d ago
ujkhsjkdhf234•1d ago
overfeed•23h ago
ujkhsjkdhf234•22h ago
overfeed•18h ago
jsiepkes•1d ago
mystified5016•1d ago
As a society we should probably consider such behavior unacceptable and deserving of strong governmental protection. But alas, this behavior makes line go up.
riku_iki•1d ago
they do not burn it to the ground, community has open source high quality product for absolutely free developed on investors money. Anyone can fork it, and do whatever he wants. I imagine if product is very good, and parent corp will decide to close sources, there could be some smaller business/consultancy supporting OSS version of this, or even MS/Google take over it as it happened many times e.g. ElasticSearch vs OpenSearch.
pjmlp•13h ago
weinzierl•1d ago
Only contribute to projects that:
1. Have a Copyleft license
2. Have no CLA and do not require copyright assignment
3. Have enough independent contributors to make relicensing practically impossible
Bonus points if the project is governed by a 501(3) (not a 501(6)).
MinIO fails point 2. Redis failed point 1. Happy to hear a good example for failing point 3. Basically a (A)GPL product that was only developed by employees.
pas•1d ago
(Unfortunately I never had good experiences with MinIO, because it was just a bit less complicated than Ceph and the docs and admin tools much less mature and stable, and due to this I simply never got to performance evaluation.)
pjmlp•13h ago
No wonder that we are slowly coming back to the shareware, public domain and trial demos, used to distribute software on 8 and 16 bit home computing.
JimBlackwood•12h ago
If MinIO were to require a license to run, I’m sure I could convince a boss at work to pay several thousands a month for it.
Currently, we are paying about 5000€/month for the hardware to run our MinIO cluster and 0€/month for the MinIO devs. If we now want to keep using their UI, the license costs would be €20.000,-/month. That is an insane gap