Hacker News Reader

Comment by placatedmayhem | Hacker News Reader

I'm curious whether the community will trust Redis-the-company again after this, or if they'll choose to stick with Valkey. The other concern is at least some big company legal departments are wary of AGPL software, which makes Valkey, still BSD, more attractive to them.

Edit: Regardless, thank you and the rest of the folks inside Redis for pushing to bring this back to OSS!

View on HN

We kept using redis, the license change never affected us. We had no reason to switch.

I imagine there is quite a large, quiet fraction (majority) of users who were the same way.

Not to say it’s not an important discussion!

Many people switched to Valkey and didn't even know it. A lot like how many users are using MariaDB but think they are using MySQL.

Several major linux distros transparently switched to Valkey and the users are none-the-wiser. On Fedora, for example, doing `sudo dnf install redis` just installs Valkey.

Well that's a reason not to use a distro right? If I type `sudo dnf install redis`, I want to install redis not valkey.

I disagree probably because I'm just used to it. Most of your major distros do this, including Fedora and Debian (the two largest). Usually this is due to license issues.

"I'm used to it" is a terrible reason for disagreement.

If I update my packages and suddenly an open-source software is replaced with a closed-source one, I would blame my distro. It's totally normal for distros to replace packages like this with the best replacement they have.

This would suck if you wrote scripts that looked for a process called 'redis' running on your machine.

I'd imagine the package would have a symlink for redis that points to valkey

Even with a symlink, the running process is going to be called valkey, not redis

Only if valkey explicitly changes the program name after it starts, or Does something like calls exec on it's program directly. At least on linux.

I think that is not normal at all, and absolutely should not be normalized.

It is much worse for my package manager to install a totally different software, than for my package manager to install a new version of the software I asked for that now has a different license. Also as an aside, SSPL is not closed-source.

If the distro wants to do something, they can throw a warning up saying "this package is now licensed with the SSPL, would you still like to install? Try installing valkey for a BSD-licensed alternative". But installing software I didn't ask for is bad, actually.

You're a bit late to the party. Its been normalized for almost as long as distros have existed.

No, you don't just randomly have a different package installed one day, at least on major distros. The next distro release will include the new package. If for any reason you care, you can always go install the other one you want instead as well, it just won't be part of the default package repos.

Generally, the replacement packages are 1:1 with the one they are replacing, and/or compatibility shims are included during the install. Its seamless. Also, generally the package manager does tell you what it's installing.

The major Linux distros are very careful about this stuff. The two largest have huge enterprise user bases, and it's never been a problem.

Many of the Linux distros are extremely opinionated on what goes into their default package repositories - it's a major reason why you choose certain distros. You are delegating all of this concern about packages, compatibility, bug/security fixes, and licenses and whatever to the maintainers of the distro. They are very careful not to break existing systems, and aren't going to surprise you one day with a major disruptive change. For them to replace Redis, for instance, with Valkey, it's going to be on the next major os release, it'll be a drop in replacement (all Redis commands continue to work, etc), and you'll have an opportunity to see this change while installing packages. This isn't "shoot from the hip" npm style stuff...

On Arch Linux, you are explicitly asked whether to replace a package with the (distro-)designated successor.

> But installing software I didn't ask for is bad, actually.

Except it's a fork, so it is what you asked for. The name changed, sure, and there's a different governance arrangement. But pretty much the entire point of using a distro is offloading decisions such as which developers and repositories to trust onto the maintainers.

If you want to make those decisions for yourself then you should obviously be cloning and building from source. I'm not just saying that - I myself do exactly that in cases where it matters to me.

If you don't care about license purity then perhaps don't use a distro that explicitly filters on that?

> Also as an aside, SSPL is not closed-source.

It's also not a free and open source software license. This is disqualifies software using it from being in the main repos for various distros.

Seems like some sort of user prompting could be acceptable, though.

I understand your point, but that's how distros handles mostly licence issues. And I do believe that's the right way, we should strive for OSS projects in a distro that literally focuses on it.

Not really. Post-closing the source, the thing called “Redis” was the actual fork, and Valkey was the original community-built product. Users who want existing configurations to continue working on the same terms need “install redis” command not to break their licensing expectations.

First, they didn't "close the source". The new license is not closed source. You can argue why you think the license is bad, but it is not closed source.

Second, I don't know about you, but continuing to function in the same way is my primary need for systems I am managing. When my provisioning system installs a package by name, i expect it to work in the same way as before. Switching binary names breaks that promise.

My setup has scripts that do things like check that a process named "redis" is running... this will break if the process is now called "valkey"

I feel like all the commenters live in some kind of crazy alternate world where purity of license matters more than stability of systems.

> When my provisioning system installs a package by name, i expect it to work in the same way as before.

But where do you put the blame?

The distro for making that change, or the redis company for breaking your software stack?

Look, I know licensing decisions are important and a lot of people care a lot about them.

For me and my company, though, it just doesn't matter. We don't use redis in a way that would ever come into conflict with the license, so it really doesn't affect me. Redis didn't break my software stack with the license change. I am sorry, but I just can't get up the energy to care that much about which license they choose. If it helps them make money, fine go for it. I can't root that hard for the side Amazon is on.

> Well that's a reason not to use a distro right? If I type `sudo dnf install redis`, I want to install redis not valkey.

Using a distro that handles things your way is your privilege. I assume most people who install packages care about the functionality they provide, not the brand name - so it seems like a fair default for distros that aim to appeal to broad user bases imo.

> Using a distro that handles things your way is your privilege. I assume most people who install packages care about the functionality they provide, not the brand name

This is an amusing example of "the duality of man," having just finished reading a bunch of comments to the effect of "the user should have the ultimate say as to what apps they can install from the App Store" in the Apple thread.

But you do have the ultimate say in this case. You can reconfigure the mainstream package managers in all sorts of ways, set up your own repos, anything you'd like.

I use a distro because I don't have the time or energy to do all of that for myself. It's a purely voluntary arrangement unlike the iOS/Android duopoly that modern society is increasingly built on top of.

It's not just the brand name, it's also the binary name. All the support code I have expects the binary to be named 'redis'

Also, what happens if functionality drifts?

So the redis package renames it to redis. No big deal.

You're thinking of it as if upstream is primary. Actually, when you use a distribution package, the distribution has the final say over the contents of the package.

100% if they wanna stop shipping redis they should just remove the redis package

Problem is, redis threw their users under the bus by relicensing [1], and user would end up with an outdated redis version.

1: https://github.com/redis/redis/commit/0b34396924eca4edc52446...

I LIKE this. Why would you not use a distro if it just removes packages? Or force closed source on you if it's legal to install the binaries? I do not know it's the same on Ubuntu, but I consciously made the decision and typed out the valkey package. If I would not know now about all the BS and would just want Redis I would LOVE for my distro to just install a replacement without me noticing anything. Maybe with a little hint and conformation message during install why this is happening and that is it. Hats off to Fedora maintainers, this is how you make the end user happy!

Same here. The response from the community was valid, but basically didn't affect us.

From the blog post it seems like existing users kept using Redis but new users adopted alternatives instead.

> it seems like existing users kept using Redis

Redis user since it appeared and I switched my servers (~15) to Valkey - partially because of the shenanigans, partially because Arch is moving Redis to archive.

In my current compagny (multibillions dollars kind of), most if not all redis has been replaced by valkey

Redis is a threat to the compagny and the licence change was taken very seriously, as all legal-related threat.

All of this aside, Redis-the-company has some of the least tactful salespeople I've come across in my long stint in this industry. Used car sales level tactics.

Between that and the licensing, I would never consider dealing with them.

The team of sales was, AFAIK, rebuilt from scratch recently. Please if this happened recently tell me, and I'll make sure to report back. Thanks.

We switched to Valkey on our Elasticache instances and immediately noticed a performance improvement in our usecase that allowed us to reduce number of instances. Not really interested in moving back to Redis at this point.

I am thinking the same that going to AGPL may actually push more people to Valkey.

Although I haven't checked if ValKey any substantial development since the fork.

Yeah, there has been a lot of stuff like performance [1] and efficiency improvements [2]. A lot of the contributors, that didn't work for Redis labs but worked on Redis OSS before the fork, moved to Valkey and they continued to contribute.

[1] https://valkey.io/blog/unlock-one-million-rps-part2/ [2] https://valkey.io/blog/new-hash-table/

Valkey has RDMA support, which offers significant performance improvements.

Well Valkey has more commits to their repository then Redis does, and more contributors. So it appears to be active.

https://github.com/valkey-io/valkey

https://github.com/redis/redis

In fact, some of Redis 8’s new features were taken from Valkey source code.

Lol feels a little hypocritical to complain about proprietary clouds, then take open source software from a competitor and embed it into your own

Is that why they're AGPL now?

Statistically nobody is using valkey.

Are there usage stats available? How do you know this?

My guess is most people are using Redis via cloud providers. Did any cloud providers switch away from Redis?

AWS supports Valkey for Elasticache, and they actually bill it 33% cheaper. We use it and it works well.

AWS Elasticache gave a nice discount to switch to valkey w/ 1 button click no downtime migration path

ValKey is cheaper in AWS than Reddis

Wasn’t the whole point (or one of the major ones) of Valkey so that AWS could use it?

My guess is they are making it up. AWS has no public information, but there are some high profile customers that have migrated https://aws.amazon.com/elasticache/customers/.

Without sources, it's a "statistically worthless" comment :)

Based on Internet-accessible services the number of Valkey servers is low (~120):

https://trends.shodan.io/search?query=valkey_version+port%3A...

Here's a chart of all Redis-compatible services (~55,000):

https://trends.shodan.io/search?query=port%3A6379+redis_vers...

And how representative are publicly accessible redis/valkey instances for redis/valkey usage in general? And can shodan even differentiate Redis from a Valkey instance setup in a backwards-compatible way without being able to authenticate?

In absolute numbers probably not highly representative but the relative numbers are meaningful to measure adoption. And no, it requires the user to disable authentication in order to get the service details to differentiate between Redis and Valkey. But again, you can compare unauthenticated Redis to unauthenticated Valkey to see how the percentages are changing over time.

I don't know about valkey but I got word Nvidia was switching away from Redis.

Amazon really encourages valkey in the elasticache dashboard. There's a banner advertising lower prices and it's listed first in the dropdown when you go to create one. Default settings do have power.

Sure, but the impact of new customers and their decisions take a long time before they impact net statistics. All evidence I can find, regardless of domain or context, suggests Redis vs. valkey marketshare is something around a 99%/1% difference.

> All evidence I can find, regardless of domain or context, suggests Redis vs. valkey marketshare is something around a 99%/1% difference.

What evidence did you find?

[flagged]

Why/how would you expect an LLM to know that info?

LLMs are not a source.

but what if they confirm my priors?

[flagged]

I've been using Valkey simply because after I updated to the latest Fedora version, it dropped redis and pointed me to Valkey instead. I assume as more distros do this and more people update their systems, the Valkey user base will grow. But perhaps with the AGPL redis that will no longer be the case.

Yeah, all the open source distributions and most open source projects switching to valkey must be "nobody".

If you use the latest versions of Redis, you are benefiting from the continued efforts of the Valkey development community. [1]

This is Open Source working well.

Unfortunately, the reverse flow does not work.

[1] https://github.com/redis/redis/pull/13638

I wonder how that works legally with CLA. If the person who originally wrote the code is not the one who signs off the PR. I assume the lawyers have signed off on it.

Did they maintain the author's copyright notice as required by BSD-3?

Well, now that Redis is once again Open Source and even Free Software, that should change.

That kind of assertion really needs some backup or it's just noise. I'll be honest and say that I have no idea what the usage stats for Valkey are -- and it may be that it's a drop in the bucket compared to Redis. But I don't know. Can you back this up or is this just your gut feeling?

Yet. It's a drop-in replacement, and both faster and cheaper.

what do you mean? i work at a FAANG-adjacent company and our entire engineering org was told to switch to valkey, with an internal deadline from ops. My team supports a public facing service and we made the switch 2 months ago.

It was pretty easy, a small config change and some performance testing to make sure it worked well at scale.

Maybe nobody is talking about it online but some people have definitely switched.

I very much doubt that anyone will stick with valkey after the PaaS providers switch back to just offering Redis proper.

Why would PaaS providers switch back to offering Redis? They've clearly all already invested a lot in Valkey (AWS, GCP, Heroku).

AWS, GCP, surely are invested: they paid for ValKey, they forked to avoid doing revenue sharing with Redis in any way :D IMHO it's a matter of what the community does, and it, in turn, this depends on how well we are able to develop Redis.

It's not just licensing and hyper-scalers, it's also a matter of development quality and direction. For instance, now in Redis you can find substantial more stuff not available in ValKey, including hash items expires, Vector Sets that are very useful for a number of things, the probabilistic data structures just introduced with Redis 8, and so forth.

If Redis is superior then sticking with Valkey would just be throwing good money after bad. Hopefully those companies are competent enough to understand the concept of sunk costs.

Maybe Valkey has served its purpose in pressuring Redis into playing ball.

Just answering "why would". Whether or not Redis is better then Valkey or if it would be worth it to switch back is not something I know.

AWS and GCP offer valkey-based versions of products that are typically based on Redis, but those versions are currently, generally, preview-grade, and statistically zero customers are using them. They still offer the original, Redis-based versions of those products, which, statistically, 100% of their customers are using.

Do you have data to back up your claims? I see a lot of customer claims for Valkey here, https://aws.amazon.com/elasticache/customers/. Neither of the AWS or GCP offerings are in preview.

>and statistically zero customers are using them

You've said this twice now, but not provided any data or even a hand-wave to a possible source so that others could go get the data and look at it.

If it's statistically something, where are the stats?

[flagged]

>My claim that statistically zero (cloud provider) customers are using valkey should, I sincerely hope, be self-evident.

I have no idea what the actual stats are. But no, I don't find your "statistically 0%" to be self-evident, especially in light of the other comments and links in this thread, and what I've heard elsewhere.

I was hoping, since you presented it so confidently, that you had something more than "trust me". In another comment you say you have evidence of marketshare, maybe you could post that?

Their stats are "I asked an LLM", no, that's not joke, see this thread: https://news.ycombinator.com/item?id=43860256

>Nobody has performed any kind of automatic or default transition from Redis to valkey for existing customers.

>My claim that statistically zero (cloud provider) customers are using valkey should, I sincerely hope, be self-evident.

This is simply not true. For example, Aiven (a cloud provider) completely ended support for Redis at the end of March and migrated existing users to Valkey. https://aiven.io/docs/platform/reference/end-of-life#aiven-f...

I recently moved on to a new company, but my prior company had a pretty large scale Elasticache Redis deployment in production (over 50 large clusters in us-east-1), and were in the middle of a complete migration to Valkey due cost savings, improved performance, and reduction in memory usage.

We've already completed migrating several large production clusters and I can confidently say that the migration had been pretty smooth and seamless.

Valkey is certainly production ready (at least on AWS it is). The team is looking forward to expedite and complete the migration

Reply on news.ycombinator.com