I am certain of it. Became very clear in for example how they handled bug reports and feature requests - many sorted as ”will only be fixed/implemented in Scale” - clearly according to internal decisions / policy which has been part of strategy for a long time.
3 Likes
Can I jump in? 
See here.
I hope I’m wrong, and that the latest version of SCALE does indeed create a default 2-GiB “buffer partition” at the beginning of every disk when a vdev/pool is created.
2 Likes
Current EE, at least, doesn’t. Dragonfish didn’t either. Haven’t tried the Fangtooth beta.
It is surprising they’d take this away, when the purpose of that partition was never to use it as swap. But these are the same folks who have left autoexpand broken through several major releases, despite several bugs being closed as “fixed” on the issue.
3 Likes
I can understand the gripe, but it assumes a Master Plan that we have stuck to. The reality is it is a series of decisions, based on technology challenges and customer feedback. Things we didn’t know included:
How well K8s and Docker would integrate with ZFS?
How well would ZFS and Linux integrate to deliver quality?
How Gluster and ZFS would integrate?
How would customers like a UI change for a regular maintenance release?
How would jails and iocage evolve and would Apps be stable?
Which OS would be easier to work with for new NVMe products?
How long would it take to get to a better quality level on HA products?
As you pointed out, we did release 13.0 and many updates + 13.3. We didn’t make a committment to develop CORE forever, but we have maintained it as a very good storage products. Apps/Plugins is where we could not maintain at a quality level.
If we had a master plan, we would not have done K8s and Gluster 
6 Likes
I must be seriously damaged by years in the corporate world, because when I see that and translate it from corp speech into English, I see:
“Doesn’t change the plans” - we plan to fully support our paying customers for as long as it takes. We may not add features, but we sure as hell don’t want to piss them off.
“Continued support” - paying support. That’s what support means. We will absolutely honor our support contracts. We will also make sure it doesn’t mean a faster refresh cycle for you, because that’d piss you off.
“Complimentary to each other” - SCALE is the new shiny and we hope it’ll take off. Certainly that’s where all our effort goes. In the meantime we do need to pay salaries so do continue to pay us for CORE contracts. We need to hedge our bets a little because SCALE is new, we don’t want to make you, our paying customers, nervous right now, and we don’t 100 percent know that SCALE is the future. But we will work to make it so. We promise we won’t piss paying customers off. If and when it’s time to migrate because SCALE works like we hope it will, we will help our valued (paying) customers with the migration. We know you have a refresh cycle and you won’t have to accelerate it because of this.
Again probably just my brain. Anything written in Corp English I translate to “how do we make money and not lose customers”. Maybe that’s a hair too cynical.
I should also probably add that I understood the reasoning and agreed with it then, and do now. It’s painful, but less so than being on an OS your customers’ admins have never touched, and that has serious issues with driver support.
I didn’t feel betrayed then and don’t do now. I also don’t fault them for using corp speech to say it all - it’s how the game is played. Being extremely forthright in the language, instead of hedging and weasel words, loses you customers.
A little bit like why politicians can utter a lot of words without saying anything, maybe. Extreme clarity and honesty is not a winning strategy. I can wish it were - doesn’t make it so.
3 Likes
Maybe one more note of appreciation: Enterprise customers probably couldn’t care less about apps. That this much effort goes into the hobby and SMB community to make apps shine is amazing.
Yes it should pay off in the form of mindshare and buzz … and, it’s still amazing.
As is raidz expansion. No self-respecting enterprise will touch that. But hobbyists and maybe even SMB love it. And developing that didn’t come cheap.
For all my snarking about corp speech, I see ix as one of the good ones.
5 Likes
Well, there’s enterprise, and there’s enterprise. The folks who have racks full of servers? I’m sure you’re right, they wouldn’t use either of those features. But I suspect there are a good number of paying customers who want and use one or both of them.
So I don’t think iX did this work solely from the goodness of their hearts–but I likewise appreciate it.
1 Like
As pointed out, at the time TrueNAS 12. And nothing in that post seems false with the benefit of hind sight.
Ie at the time the plan was as said.
I’ve talked to iX about it, at the time SCALE was a major punt. They had no idea how well it would be received. It was received well, and over the last 4 years you would expect plans to evolve.
And here we are.
CORE has had a good run. And you can still use its core functionality safely.
But guys, if you want new features and shinies, it’s time to get on the scale train… erm. CE train.
2 Likes
All it takes for an enterprise customer to use apps is one install of Minio, which is a very enterprisey thing to do on a storage appliance.
1 Like
I reckon frigate is a very appetising option for small businesses as well.
That’s correct, we really didn’t know how well SCALE would be received. We expected a larger market for Linux, but it was a new customer base.
In the 1st year: 50% of new users, opted for SCALE. Almost no existing users. Quality was not as good as CORE.
In 2nd year: Most of of the new users used SCALE, but only a small % of existing users migrated. Quality was getting closer to CORE.
In 3rd year: Virtually all new users used SCALE and combined growth was much higher. A larger % of users started to migrate from CORE to SCALE. SCALE users passed CORE users. Electric Eel (with Docker) accelerated growth significantly. Quality is now better than CORE (e. HA failover times are faster).
Did we know that would happen? No, we hoped for good adoption, but we could not anticipate the quality path and rapid acceptance.
Also, we found our paying customers did not want UI changes, unless the quality of software was better and the feature set was better. They have made the transition even slower.
4 Likes
Nope, spot on. Reality is iXsystems is in it to make money. They are not a charity. I’m not defending them, I’m just stating the obvious. Non-paying customers really are the lab rats (test subjects) that are virtually free to them. I say virtually free because big corporate does interact with us and someone’s time is money, but it is money well spent. Also the non-paying customers are free advertising/recommendations to draw in paying customers. It is smart business, and I like the free software, even if I am testing it. And no one needs to be a lab rat, you can use a solid version that has been deemed Stable (a few bug fixes into the version).
I would have to say that the speed of the acceptance and adoption was more in line with the fact that CORE was being touted as coming to EOL. At least that is what it felt like to me. But I like TrueNAS, and I will use it for years to come.
Agreed, that was part of it, especially for those who want Apps.
If a support contract would be needed for Core to continue to be maintained, i would get. Your point is rather irrelevant as this not about cost at all. It is about the switch from one technology to another where the latter is not ready yet for enterprise usage but the previous technology is pretty much abandoned or “maintained” which in this case, means exactly the same thing.
What makes it not ready for enterprise usage?
1 Like
I don’t understand, what do you mean it’s not ready for commercial use?
I didn’t know that you’re the one who dictates the rules of which technology is ready for commercial use.
Someone who didn’t pay anything for it, just complains, throws a tantrum like a child, and doesn’t even have the ability to develop something on their own.
I don’t think that’s how things work. I can agree with many points that need to be improved, like any other tool, but saying that it’s not ready for the commercial world is almost a joke.
Just a hint: “enterprise usage” doesn’t mean “commercial use” (by far).
Anything advertised on Shein, Temu or Alibaba is “commercial” … migrating those VMs over by just restarting on first reboot after nine years uptime is “enterprise”.
Well, that is what IX systems states, not me.
I find it interesting that you join the conversation out of nowhere, not even being able to differentiate between Enterprise and Commercial (IX systems only promotes Core for enterprise use, not Scale), missing the entire point (the point being that in my humble view, Scale should have been an enterprise-worthy, reliable replacement for Core before they stopped development on Core) and then throw around insults. Nice way as a first post after 10 months of inactivity on this forum. Keep it up 
1 Like
Now you also want to dictate who can or cannot join the conversation?
And my question was precisely about Enterprise, the broker that changed to Commercial.
And I didn’t insult anyone, I’m just stating facts, so far you’ve just been throwing a tantrum, as if you were a football fan, if it’s not my team, it’s no good.
And just because I haven’t posted in 10 months doesn’t mean I’m not always reading and following truenas, by the way I’ve been following FreeNAS since mid-2011, I saw FreeNAS become Truenas and now the change to Scale and CE.
1 Like