FreeNAS development - why are the processes still shaky?

[danb35]

Here's the problem:

FN10 was a fiasco, of course. And though iX ultimately decided to put it out of its (and our) misery, the elephant remains in the room: how did this happen (i.e., what was the series of business decisions that led to "ship this half-baked abominationproduct"), and what will iX do to keep it from happening again? As these are fairly internal business questions, it's understandable you might not want to discuss them publicly, but if you don't do so, you leave the community wondering. The best we have is "trust us, we've really changed." And then we see:

• Beta releases in the -STABLE update train--we saw this with 11.0, 11.1, and now 11.2. This makes people think the beta is the full release; we've seen that happen in all three of these releases.
• The Download link on the forum pointing to the Beta, which prompted this thread (and wasn't fixed until three days later, when I posted about it again).
• The download page steering people toward the beta, falsely suggesting that 11.1 only handles storage, while the beta adds virtualization and plugins. Between these three, it really seems like a concerted effort to paint the beta as a stable release, when you know it isn't.
  
• A series of broken updates. Every 11.1-Un broke iocage networking, and each of them broke it in a different way. We hear that iocage is the way forward with jails, but it's hard to follow that path when it just doesn't work. 11.1-U3, IIRC, very badly broke Samba. An earlier update had a memory leak that brought systems to their knees when trying to do a scrub. I won't catalog the remaining issues; they're discussed in the forums and on the bug tracker.

If I might briefly take the liberty of speaking for your user base, we want to trust you. We want to trust that the developers of software we use every day to store our important data have the wisdom, the motivation, and the skill to do it right. But history like the above makes it hard to do that.

Mode note: This thread was spun off from this thread about the download FreeNAS link pointing to a Beta version. This was the starting point of this discussion.

 

[Redcoat]

Possibly adding to potential confusion - I just looked at the Guide tab in my 11.1-U5 installs and was served up FreeNAS® 11.2-BETA1 User Guide ?

And the "current release" button on http://doc.freenas.org/ goes to FreeNAS® 11.1-U4 User Guide, not a version for U5.

 

[Ericloewe]

I just looked at the Guide tab in my 11.1-U5 installs and was served up FreeNAS® 11.2-BETA1 User Guide ?

That's weird. Definitely file a bug report for that, someone might have thrown the wrong switch.

goes to FreeNAS® 11.1-U4 User Guide, not a version for U5.

That's not too weird. -U5 shouldn't have any manual-relevant changes, IIRC.

 

[gpsguy]

As I recall, the commits for the docs went from 11.1-u4 to 11.2-BETA1.

That's why I couldn't make a PDF version of 11.1-U5.

U5 shouldn't have any manual-relevant changes

 

[Redcoat]

That's weird. Definitely file a bug report for that, someone might have thrown the wrong switch.

https://redmine.ixsystems.com/issues/39178

 

[Ericloewe]

@Bgadd, many of these points have been raised in the other megathread. Would it be possible for you to provide some insight into these?

 

[wblock]

As I recall, the commits for the docs went from 11.1-u4 to 11.2-BETA1.

That's why I couldn't make a PDF version of 11.1-U5.

The master branch of freenas-docs is for the old UI, and should have been used for 11.1-U5. The angulargui branch is the one for the new UI.

 

[appoli]

@danb35 I don’t want to throw any unnecessary shade since I’m still pretty new in the freenas world/community, but a some of the things you mentioned had a big impact on me while some other development related issues are still effecting me:

- I heard about freenas and after some time decided to take the plunge and build a custom NAS unit. I went to the freenas website and looked at the download menu; at the time there were two production builds available and I figured that I would go with the latest one since I had the necessary hardware and the new features sounded great. So I downloaded and installed FreeNAS Corral. I’m sure you can figure what the fallout was like. To be totally transparent I should have done more research on my own etc, but I assumed production was production (and to be honest the build worked great for me, in fact in many ways the performance was better than that of 11, except when using the GUI I believe. I only switched because I eventually hit a dead end where I couldn’t get the host docker machine to boot). The result was me having a lot of data and applications setup for docker & in particular that specific docker implementation (as well as my network having been tuned for that specific docker implementation).
I ended up spending a lot of time making a stop gap until future updates would be released & ended up spending more money because various developers stated that the problem was a lack of some resource or another (these are 3rd party developers, it was actually so to poor plugin designs is my guess).

-the current issue I’m experiencing right now is some issue with the way ram is managed, specifically when there are VMs that are running different OS’. It seems like this may be fixed in the new release (the issue is a constantly growing inactive bucket and growing use of Swap, where the laundry bucket is never used and thus ram isn’t cleaned). The fix seems to have stemmed from the bug requesting a ‘seatbelt’ for VM memory useage - this would address the growing inactive bucket probably regardless of how it’s implemented, but if implemented ‘correctly’ there are other things that would need tweaking, not just a restriction on how much ram is used by the VM. We would want to see a fix that results in the pagedaemon working correctly as per the newer FreeBSD releases; one that takes unused inactive memory and sends it to laundry for cleaning when memory pressure is present or cycles the inactive memory to active if applicable.

I don’t want to be overly harsh because I think this is a great platform with a great community and a large amount of development that comes at no cost to me. I’m complaining about some pretty specific stuff and to ‘make things worse’ I’m both new to the community and a relative noob. I’m bringing this up because it sounds like some valid points are being made - I did put a degree of trust into the developers hands when I downloaded what was labeled a production version & when there are serious bugs for a feature that is in the spotlight in general, but also for that release specifically - especially when this implementation seems to be worse in a number of ways than the previous one (but I don’t need to worry about the VM booting so there is that!).

I’m not sure what the answer or remedy is since this is a product of a private company that is being offered for free - so whatever transparency they provide is really at their own discretion and anything is more than they are required to do (unless, of course, these bugs also apply to their commercial clients and thus the manner in which they are resolved is of import to said clients).
More detail regarding bugs, how they are addressed, updates to builds, what’s coming up, better documentation, and more attention/care shown when making releases/linking to them are all great/very much appreciated. On the other hand these words can be said with little way of them being enforceable (in the short term at least) which is what danb35 is bringing up.
I would like for these things to be addressed, but I am also congnizant that addressing these issues costs money - money that we aren’t being charged.


I wrote the above saga to make the point that I am both very invested in this & I have been effected by it (both my time and my money). I would like a some progress to be made, but my experience tells me that when you identify an issue that rests with someone else you can’t just point it out and expect to see much/any progress made - especially when we aren’t paying for said progress. One of the approaches I have had better luck with is adding some suggestions/possible avenues of remediation along with identifying the issue (it’s easy to complain about an issue and then complain about the solution when you have put no skin in the game...).

So, to make things a bit more productive what do you think about making an addendum to your post and throwing out/asking for some ideas to address the identified gap?

You pointed out some items that are in the process of change. One is switching to iocage jails. We also have a new GUI that is coming out (which ix has been kind enough to open a thread for our feedback which I hope is being taken very seriously since the last time I tried out the new UI I found it VERY lacking & couldn’t even use it in 90% of the scenarios that I use the UI eg on my mobile), the matter of transitioning warden to iocage jails, the whole VM/Docker issue along with memory useage, and I’m sure there are some other big ticket items that are up for change.

So, does anyone have any suggestions for how the aforementioned issues could be addressed to the point at which the community would be more comfortable with the output?

Aside from taking threads & their comments for big changes (Eg new UI, a number of comments either make a good point or identify an issue that we would like to have resolved before the new UI takes over. Such as ensuring that the UI will have all the same functionality as the old one before it goes production & whatever ‘high level’ bugs/interface issues are addressed before it goes live - as two examples) and porting them over to the bug/feature tracker/JIRA using tags of some sort and a peer review of some sort for their review. And then having some info being posted once they are addressed with said info being ported/linked back to the forum for a community re-review; I can’t think of much that wouldn’t cost too much or take too long to implement (and I’m aware that this suggestion would take effort and money even though much of it is already in place).
The objective is to maybe identify a couple key issues (or whatever capacity IX says they are able to take on) using something like a community vote to determine what will receive this extra scrutiny/review followed by a method of uploading a degree of dialogue with some checks that need to be passed (so that the feedback is accepted and acted upon accordingly).

I’d be curious to see if anyone can come up with some other suggestions & it should be noted that Betas are exactly that, pre releases so that the community can find any issues. Maybe a checklist of some sort that covers areas of change that some volunteers offer to review and must be completed before the version goes live? (An attempt to stem issues such as the jails not working in the previous 11.1n releases) & it should be noted that accidents also happen. They shouldn’t in a real world, but one should also take the time to double check what version they ended up downloading (this is often common practice with hash files and such).

Apologies for the epic saga of a post - hope the feedback is helpful & once again thanks for developing a platform that so many people use and in such a wide manner of implementations. I sometimes forget that the software I use to store computer backups, important documents, various media files, host VMs running media/networking/cloud storage/home security/etc applications for my home/family and friends, and it’s community, is also used by professionals working for large and serious companies to handle their information needs. This is a piece of software that makes a lot of things possible and I do appreciate that

 

[gpsguy]

I don't recall seeing a 11.1-U5 on the master branch. As I recall, it jumped from 11.1-U4 to the 11.2 beta.

Yes, I know about the angulargui branch - I've generated PDF's for both the classic and angular versions of the 11.2-BETA1 documentation.

The master branch of freenas-docs is for the old UI, and should have been used for 11.1-U5. The angulargui branch is the one for the new UI.

 

[Bgadd]

@Bgadd, many of these points have been raised in the other megathread. Would it be possible for you to provide some insight into these?

Hey All,


Thanks for pinging me. I have recently tried to be more involved with the forums, and I plan to check for updates every couple of days as time allows. At some point in the future, I am hoping to set up an AMA for the community to discuss in more detail in case you would like to find out more details on ways iX has taken steps to improve our quality overall, but for now let me address the topics at hand at a high-level.


First off, we have not forgotten about Corral, I promise you that. We may not talk about it publicly, but we are very aware of the impact it had on our excellent community, but also it impacted our teams.


As stated, I will not go into specifics right now, but I will say quality is one of the areas that I am specifically concentrating my efforts in by working with leadership throughout iX. Quality is not just testing the software, but our processes. For example, how we collect information from users (community or TrueNAS). We are establishing a template to require certain information when submitting a bug or feature. We will have a triage process and then add it to our backlog to begin the screening of the issue. During the screening process, our engineers will assign a “severity” rating (which uses a “impact x likelihood” matrix) to allow the teams to prioritize the work. This simple change will help the teams to make sure they address high severity bugs over low-impact bugs.


Also, we understand the need for more stable releases, so in late Q2, we committed to begin an Agile approach to our development lifecycle and methodology. For instance, instead of iX depending solely on the community’s feedback or testing, we are taking a more aggressive testing stance. In the past year, our Quality Manager has created an Automation Team, HW Team and also a Manual Testing Team. This structure allows us to have people that concentrate on different aspects of the product and aligns us with industry standards.


Also, we have taken an iterative approach to our release schedule. Within those iterations, we have small testing phases. Each small release will have anywhere from 1-4 weeks of testing before the release is deployed. It is such a large range because it depends on which release we are currently in, for example, Alpha, Beta, RC, Proper, -Update Release. This approach also allows us to get releases to you quicker and also fix bugs users submit more quickly. After we do our first public release, in this case 11.2Beta1, we will have a new release about every 8 weeks until we get to the “proper” (11.2FN), then the cadence will be determined by the bugs and issues that are found. After that, the cadence will be determined by the number of bugs reported.


Again, this is all new to iX and to our community, especially in 11.2. I recognize it is quite a bit of change to introduce, but we knew it needed to be done to make an impact. These changes will not fix everything overnight, but we are taking the steps we need to make improvements. We are committed to continue to make an amazing product for all our users. We will never be content with one way of doing things, so we will continue to evaluate our processes through the lifecycle to make sure we making our products better, not worse.


Let me conclude by saying these forums are excellent places to find information and get assistance, but if you have issues, submit tickets. Put as much information in the ticket as possible. Let us know the environment you are in so we can better pinpoint the issue and fix it the first time. Some issues we can fix at the snap of a finger with no impact to the software. Others could require aredesign of a foundational part of the software and take months to fix.


Ben Gadd MBA,CSM
Program Manager
iXsystems, Inc.
To submit a bug, go to https://redmine.ixsystems.com/projects/freenas/issues/new

 

[danb35]

@Bgadd,

Thanks for the insights into what you're doing differently. But why then the apparent attempt to present beta software as though it were stable? Although the forums link has now been changed to point to a stable release, the other points I made remain--and there are probably at least a dozen threads here with some variation of "I installed 11.2 stable and...".

 

[kdragon75]

I imagine there is a fair bit of in-fighting on this topic. Testing cost money and iX is for profit. I trust there are many at iX that support making the best possible product by doing more internal testing but others that just see us as beta testers for TruNAS. Personally I'm on the edge of going to pure FreeBSD. The 11.2-RELEASE will be the deciding factor.

 

[KrisBee]

@Bgadd ,

At the moment I feel there are two main disincentives for users reporting bugs/issues. Firstly, I’m sure iXsystems would prefer that the same bugs are not reported multiple times by users, and I know I don’t want to waste time submitting bugs which are already logged. But finding if a bug/issue is already in the system is not straightforward.

For example, if wanted know all bugs/issues relating to virtual machine creation in FN11.2 Beta1, there seems to be no simple and reliable way to do this. Filtering issues by subject description is hit and miss, do you search on “vm” or “bhvye” or what? You can’t add more than one subject filter and AFAIK when using multiple filters, these are only ANDed. So you might end up trawling through the roadmap lists, which for FN11.2 Beta1 means 11.2-BETA2, 11.2-RC1 and even Master – FreeNAS Nigthlies.

One of the bugs I found in FN11.12 Beta1 ( see: https://forums.freenas.org/index.php?threads/multiple-vm-bugs-in-fn11-2-beta.68503/ ) turned out to have already been seen in the Master Nightlies and logged as #34975. It was reported a couple of weeks before Beta1 was even released and has ended up being assigned to 11.2-Beta2. So, I hope the template you intend to establish is going to make this kind of search easier?

Secondly, why would I bother to report bugs/issues in Beta and RC releases when I have no clear indication of the state of completeness of any given feature? For example, even a short cursory test of the creation and use of virtual machines in FN11.2 Beta1 would tell you a lot of the user input validation is missing. It’s doubtful this is not an area yet to be worked on, yet without some kind of matrix of area against completeness a user might think these are bugs that should be reported but could in fact easily be wasting their time doing so.

Unless and until there is a more straight forward way of knowing which bugs have already been reported and clearer idea of what “feature complete” really means, I don’t feel inclined to submit bug reports.

 

[Bgadd]

At the moment I feel there are two main disincentives for users reporting bugs/issues. Firstly, I’m sure iXsystems would prefer that the same bugs are not reported multiple times by users, and I know I don’t want to waste time submitting bugs which are already logged. But finding if a bug/issue is already in the system is not straightforward.


For example, if wanted know all bugs/issues relating to virtual machine creation in FN11.2 Beta1, there seems to be no simple and reliable way to do this. Filtering issues by subject description is hit and miss, do you search on “vm” or “bhvye” or what? You can’t add more than one subject filter and AFAIK when using multiple filters, these are only ANDed. So you might end up trawling through the roadmap lists, which for FN11.2 Beta1 means 11.2-BETA2, 11.2-RC1 and even Master – FreeNAS Nigthlies.


One of the bugs I found in FN11.12 Beta1 ( see: https://forums.freenas.org/index.php?threads/multiple-vm-bugs-in-fn11-2-beta.68503/ ) turned out to have already been seen in the Master Nightlies and logged as #34975. It was reported a couple of weeks before Beta1 was even released and has ended up being assigned to 11.2-Beta2. So, I hope the template you intend to establish is going to make this kind of search easier?


Secondly, why would I bother to report bugs/issues in Beta and RC releases when I have no clear indication of the state of completeness of any given feature? For example, even a short cursory test of the creation and use of virtual machines in FN11.2 Beta1 would tell you a lot of the user input validation is missing. It’s doubtful this is not an area yet to be worked on, yet without some kind of matrix of area against completeness a user might think these are bugs that should be reported but could in fact easily be wasting their time doing so.


Unless and until there is a more straightforward way of knowing which bugs have already been reported and clearer idea of what “feature complete” really means, I don’t feel inclined to submit bug reports.

----------

Regarding duplicate issues/bug, go ahead and submit them. Since we are doing the screening, we can link or close the bug if it is a duplicate. We do not mind it. In fact, it helps to validate the issue being reported and you may be able to provide more insight to the issue.


You bring up some of the same issues we deal with regarding Redmine, to help you and others out we can put together a guide to assist you out in searching or creating filters. Something I learned recently, there are actually other "skins or themes" you can apply to your account and they can make the UX a little better. To access the themes, sign in, go to My Account>Themes then find out that you prefer.

Searching for more than one subject is near impossible because it is a text box, so would you have to use the option “contains” in order to find a bug that contains the same work in the subject line. However, if you want to find bugs that are in the same category, there is a way to add in more than one category or other filters that allow it….Select the field you want to search, for this example I am going to choose “category.” Next to the list of categories’ drop down menu, there is a tiny (+), select that and then like other drop-downs, hold CTRL and then select the combo you would like, then apply. Do check out the different themes, maybe there is one that highlights the (+) better.

We will do what we can to make it easier for you to submit and search for issues. Myself and others at iX can always create better public queries if need be, so I may make a new thread to collect a few ideas on what will make your experience better.

I hope this helps and do feel free to PM me if you have any specific issues you need help with and I will do my best to help if I can.

 

[diskdiddler]

Here's the problem:

FN10 was a fiasco, of course. And though iX ultimately decided to put it out of its (and our) misery, the elephant remains in the room: how did this happen (i.e., what was the series of business decisions that led to "ship this half-baked abominationproduct"), and what will iX do to keep it from happening again? As these are fairly internal business questions, it's understandable you might not want to discuss them publicly, but if you don't do so, you leave the community wondering. The best we have is "trust us, we've really changed." And then we see:

• Beta releases in the -STABLE update train--we saw this with 11.0, 11.1, and now 11.2. This makes people think the beta is the full release; we've seen that happen in all three of these releases.
• The Download link on the forum pointing to the Beta, which prompted this thread (and wasn't fixed until three days later, when I posted about it again).
• The download page steering people toward the beta, falsely suggesting that 11.1 only handles storage, while the beta adds virtualization and plugins. Between these three, it really seems like a concerted effort to paint the beta as a stable release, when you know it isn't.
  
• A series of broken updates. Every 11.1-Un broke iocage networking, and each of them broke it in a different way. We hear that iocage is the way forward with jails, but it's hard to follow that path when it just doesn't work. 11.1-U3, IIRC, very badly broke Samba. An earlier update had a memory leak that brought systems to their knees when trying to do a scrub. I won't catalog the remaining issues; they're discussed in the forums and on the bug tracker.

If I might briefly take the liberty of speaking for your user base, we want to trust you. We want to trust that the developers of software we use every day to store our important data have the wisdom, the motivation, and the skill to do it right. But history like the above makes it hard to do that.

Mode note: This thread was spun off from this thread about the download FreeNAS link pointing to a Beta version. This was the starting point of this discussion.

To my knowledge, besides the download link pointing to the wrong location (which is fixed) at no point was FreeNAS 11.2 BETA1 pushed as an update.
The tweet and news article specify that it's a beta and in order to get it on an existing system via the update method, you need to actively not only check for updates on your existing install but then click the box to switch train.
Not that this is a get out of jail free card, but it seems unlikely many would have moved to 11.2 'accidentally' people did need to manually do this. (yeah, ok the name should've been 11.2-BETA, I agree)

I keep saying this and I keep hoping I'm right, but 11.0 -> 11.2 is actually a pretty darn large change. I am willing to give these guys a pass with all the issues, because the entire backend for plugins / jails is being re-done and the UI on top of that. Not only that, but the new UI is, surprisingly, slowly getting there.

In a weird way, I wish people were all 'stuck' on 11.1 or heck, even 9 and 11.2 became "the new 10" (shudder)
Assuming BETA2 and RC1 are tested well, I'm suspecting 11.2 will be genuinely pretty darn good.

FWIW, I can say I didn't really suffer any issues with the 11.1x series and the 11.2 BETA has been surprisingly reliable.
It is free software, so we're never going to get the 'full synology experience' - which I'd love, because I could recommend to dummy friends then, but it just won't be like that unfortunately.

 

[diskdiddler]

@Bgadd ,

At the moment I feel there are two main disincentives for users reporting bugs/issues. Firstly, I’m sure iXsystems would prefer that the same bugs are not reported multiple times by users, and I know I don’t want to waste time submitting bugs which are already logged. But finding if a bug/issue is already in the system is not straightforward.

For example, if wanted know all bugs/issues relating to virtual machine creation in FN11.2 Beta1, there seems to be no simple and reliable way to do this. Filtering issues by subject description is hit and miss, do you search on “vm” or “bhvye” or what? You can’t add more than one subject filter and AFAIK when using multiple filters, these are only ANDed. So you might end up trawling through the roadmap lists, which for FN11.2 Beta1 means 11.2-BETA2, 11.2-RC1 and even Master – FreeNAS Nigthlies.

One of the bugs I found in FN11.12 Beta1 ( see: https://forums.freenas.org/index.php?threads/multiple-vm-bugs-in-fn11-2-beta.68503/ ) turned out to have already been seen in the Master Nightlies and logged as #34975. It was reported a couple of weeks before Beta1 was even released and has ended up being assigned to 11.2-Beta2. So, I hope the template you intend to establish is going to make this kind of search easier?

Secondly, why would I bother to report bugs/issues in Beta and RC releases when I have no clear indication of the state of completeness of any given feature? For example, even a short cursory test of the creation and use of virtual machines in FN11.2 Beta1 would tell you a lot of the user input validation is missing. It’s doubtful this is not an area yet to be worked on, yet without some kind of matrix of area against completeness a user might think these are bugs that should be reported but could in fact easily be wasting their time doing so.

Unless and until there is a more straight forward way of knowing which bugs have already been reported and clearer idea of what “feature complete” really means, I don’t feel inclined to submit bug reports.

I really recommend that you log bugs regardless. I've found the responses on the bug tracker to be friendly/professional, despite me not being a qualified QA guy.
They take things seriously and they even seem to accept a fair few feature suggestions which is very nice.

It won't improve without us making them aware, that's what it's there for.
I do think the search could do with a little work somehow, I'd like to know if I'm logging a dupe, that being said, I'd rather log a dupe than not log at all.

It's not as difficult as you think to do one either, they aren't monsters with "WHY ISN'T THERE X" or "YOU DIDN'T PROVIDE Y" - it's all pretty decent on there. Stuff seems to get fixed fairly rapidly.
I've dropped in several and had them fixed fairly quickly.

Please, contribute to the community, it can only help it, not harm it.

 

[diskdiddler]

I imagine there is a fair bit of in-fighting on this topic. Testing cost money and iX is for profit. I trust there are many at iX that support making the best possible product by doing more internal testing but others that just see us as beta testers for TruNAS. Personally I'm on the edge of going to pure FreeBSD. The 11.2-RELEASE will be the deciding factor.

If you're honestly doing mission critical stuff on your machine, the smart move is to simply not update for say 2 to 4 weeks, have a quick glance at the forums before you do, then go for it.
The "if it ain't broke, don't fix it" should apply here. As much as it's difficult to be patient (I know!) I generally have found you'll get a great experience that way.

 

[diskdiddler]

how we collect information from users (community or TrueNAS).

I've been thinking about this, for the past couple of days and I was wondering if there would be a way to be a part of a database of hardware and software uses, for FreeNAS (opt-in, of course)
I'd gladly opt in for monthly reporting which sends you guys the following

Hardware specifications (in full)
FreeNAS version
Services enabled / disabled
Jail quantity / plugin quantity / VM quantity etc

That kind of thing, would that be of any use to the dev team, to have a rough idea of what customers are using 'out there' almost like a steam hardware survey?
Understandably, some members would never opt in, but if it helps in any capacity I'm more than fine with this.

 

[danb35]

at no point was FreeNAS 11.2 BETA1 pushed as an update.

Well, there was the download link here, which was fixed three days after I opened the thread. There's the update train that's called 11.2-STABLE when it isn't (and, again, a dozen or more threads here from people who think they've installed 11.2 Stable when there's no such thing--and the same thing happened with 11.1, and before that with 11.0, so iX really should know better by now). There's the download page that suggests, incorrectly, that 11.1 doesn't support plugins or VMs, and presents the beta as what sounds like the better option, with no warning that it's, well, beta, and shouldn't be used in production (though it does at least say "beta"). Betas shouldn't be encouraged for production use, but it looks like iX is doing just that.

I'd gladly opt in for monthly reporting which sends you guys the following

You're aware that FreeNAS already sends quite a bit of telemetry on an out-out basis, right?

 

[diskdiddler]

Well, there was the download link here, which was fixed three days after I opened the thread. There's the update train that's called 11.2-STABLE when it isn't (and, again, a dozen or more threads here from people who think they've installed 11.2 Stable when there's no such thing--and the same thing happened with 11.1, and before that with 11.0, so iX really should know better by now). There's the download page that suggests, incorrectly, that 11.1 doesn't support plugins or VMs, and presents the beta as what sounds like the better option, with no warning that it's, well, beta, and shouldn't be used in production (though it does at least say "beta"). Betas shouldn't be encouraged for production use, but it looks like iX is doing just that.

You're aware that FreeNAS already sends quite a bit of telemetry on an out-out basis, right?

The name was wrong - but no (existing) users, got switched without manually instigating this. Likely a tiny handful who happened to get curious and pulled down the dropdown box.
Anyhow, what is done, is sadly done - I'm sure based on this thread, they may name the 11.3 BETA branch differently? Or perhaps have a warning box if you try switching.
I do sympathise but it could've been much much worse (it's quite stable)

I didn't know about the telemetry or whatever the term is. I'm good with that, I hope it helps them. I want this thing to be good and continue to be good. In time it could eat Synologys lunch (especially based on their atrocious pricing)