A somewhat harsh blog post about NAS OS's / FreeNAS

[anodos]

Read this blog post and feel it's a bit unfair towards FreeNAS. Of course, it's hard to argue with something that doesn't give concrete examples. http://www.smbitjournal.com/2015/07/the-jurassic-park-effect/

If we take a common example and look at FreeNAS we can see how this is a poor alignment of “difficulties.” FreeNAS is FreeBSD with an additional interface on top. Anything that FreeNAS can do, FreeBSD an do. There is no loss of functionality by going to FreeBSD. When something fails, in either case, the system administrator must have a good working knowledge of FreeBSD in order to exact repairs. There is no escaping this. FreeBSD knowledge is common in the industry and getting outside help is relatively easy. Using FreeNAS adds several complications, the biggest being that any and all customizations made by the FreeNAS GUI are special knowledge needed for troubleshooting on top of the knowledge already needed to operate FreeBSD. So this is a large knowledge set as well as more things to fail. It is also a relatively uncommon knowledge set as FreeNAS is a niche storage product from a small vendor and FreeBSD is a major enterprise IT platform (plus all use of FreeNAS is FreeBSD use but only a tiny percentage of FreeBSD use is FreeNAS.) So we can see that using a NAS OS just adds risk over and over again.

This same issue carries over into the communities that grow up around these products. If you look to communities around FreeBSD, Linux or Windows for guidance and assistance you deal with large numbers of IT professionals, skilled system admins and those with business and enterprise experience. Of course, hobbyists, the uninformed and others participate too, but these are the enterprise IT platforms and all the knowledge of the industry is available to you when implementing these products. Compare this to the community of a NAS OS. By its very nature, only people struggling with the administration of a standard operating system and/or storage basics would look at a NAS OS package and so this naturally filters the membership in their communities to include only the people from whom we would be best to avoid getting advice. This creates an isolated culture of misinformation and misunderstandings around storage and storage products. Myths abound, guidance often becomes reckless and dangerous and industry best practices are ignored as if decades of accumulated experience had never happened.

A NAS OS also, commonly, introduces lags in patching and updates. A NAS OS will almost always and almost necessarily trail its parent OS on security and stability updates and will very often follow months or years behind on major features. In one very well known scenario, OpenFiler, the product was built on an upstream non-enterprise base (RPath Linux) which lacked community and vendor support, failed and was abandoned leaving downstream users, included everyone on OpenFiler, abandoned without the ecosystem needed to support them. Using a NAS OS means trusting not just the large, enterprise and well known primary OS vendor that makes the base OS but trusting the NAS OS vendor as well. And the NAS OS vendor is orders of magnitude more likely to fail if they are basing their products on enterprise class base OSes.

I think the statement that the community only consists of those struggling with the OS is also not quite fair. There are quite a few very knowledgeable members of the freenas community who write prolifically and help the forums from being an 'isolated culture of misinformation'.

 

[nightshade00013]

I agree with the statement that the NAS os versions lag behind the base os. That is true with a lot of things and isn't really a problem unless you think the bleeding edge is the best place to live. Look at the Android ecosystem for instance. Google devices are generally on the bleeding edge and while some added features and functionality is great sometimes things come out that break the device functionality. The OEM's that use android are ALWAYS a step behind a google device, then again they often times have less issues since they are not on the newest version. Would it be nice if a phone is at least a little newer version of the OS, sure but then again a lot of the problem there comes from the carrier and not the manufacturer.

FreeNAS 9.3 has had some lag behind at this point which caused a few issues like driver support for certain things like USB 3.0 and some of the newest drive controllers but overall it has still done very well and it sounds to me that some decisions have been made that will begin to allow the upgrade schedule to work just a little faster to stay closer to the FreeBSD versions.

Also by the author's reasoning any version of linux other than a parent version (think along the lines of ubuntu vs debian the parent) are bad because they most recent version of one os will lag behind the other. The same would be in effect for say, RHEL and CentOS. CentOS is built off of the RHEL source but is always going to lag behind at times, does that effectively make it a worse version of the OS?

I think it's also funny that in the comments they are talking about people using extremely underpowered hardware like a raspberry pi and comparing it to a full server. Could it be better to use the latest and greatest version of FreeBSD to host a NAS, sure a lot of features would be implemented early on. But I think it has a whole lot more to do with people being able to dedicate their day to running a server that can do everything (FreeBSD) vs something that is easy to work with and does what you need it to (FreeNAS). I personally would rather have a life beyond configuring a server and honestly doubt at this point in my life could even get everything working as well on my own. FreeNAS may be a little harder to support than a straight FreeBSD install by someone who works with FreeBSD all the time but I can guarantee that someone who is coming from a windows only background would look at setting up a FreeBSD version that can do the things that FreeNAS can do would throw their hands up in the air and go back to a hobbled windows computer as their server.

 

[Ericloewe]

Maybe he should write a blog post on how Windows for Embedded Devices (or whatever the hell they're calling it out now) lags behind regular desktop Windows. Or how Windows Server lags behind desktop Widows (*gasp*!).

 

[anodos]

Maybe he should write a blog post on how Windows for Embedded Devices (or whatever the hell they're calling it out now) lags behind regular desktop Windows. Or how Windows Server lags behind desktop Widows (*gasp*!).

Or how xenserver lags behind centos. ;)

 

[Ericloewe]

This same issue carries over into the communities that grow up around these products. If you look to communities around FreeBSD, Linux or Windows for guidance and assistance you deal with large numbers of IT professionals, skilled system admins and those with business and enterprise experience.

Someone has never been to the Ubuntu forums or any forum centered on Windows...

 

[Mirfster]

Yeah, the simple fact that the article got a whopping 4 comments in 15 days (before it was locked) speaks volumes to me... :D

Why post something and then not be willing to converse about it?

Especially when you consider 1/4th of the comments is made up of:

because BACON!

 

[Jailer]

Made it about half way through before I closed the tab. I'm not an IT guy or even in "the biz" and it didn't take me long to figure out that the post was complete garbage.

Ifin you don't like it, don't use it but leave us happy end users to our own devices. I think the community here is doing just fine. ;)

 

[Yatti420]

This creates an isolated culture of misinformation and misunderstandings around storage and storage products. Myths abound, guidance often becomes reckless and dangerous and industry best practices are ignored as if decades of accumulated experience had never happened.

Maybe in like like 2005 with FreeNAS 0.7x etc..




** #BecauseIts2015 - Does the USA want to keep Justin? We don't want him anymore ;) **

 

[Arwen]

One thing I may have missed, (orginal article was a bit LONG), is that FreeNAS allows
OS disk replacement EASILY.

It would be nice if Linux, (and FreeBSD, Solaris, etc...), had a mechanisim in place to
store the OS configuration files without extra crap. In some ways "/etc/" used to be the
place. But, what happens today is there are default files in "/etc/" that a user may not
need to change. Thus, no need to back that up as configuration.

So a new scheme might be;

/etc/default/hosts
/etc/default/nsswitch.conf
/etc/default/ssh/...
/etc/hosts
/etc/nsswitch.conf
/etc/ssh/...
etc...

When an application needs to know if NIS is supported is "nsswitch.conf, it's library checks
"/etc/" first. If not available, then "/etc/default".

Certainly not perfect.

 

[NAS-Plus]

Someone just sent links to the blog posts that this thread was based on (different links though then were provided previously):
https://www.storagecraft.com/blog/the-jurassic-park-effect-part-1/
https://www.storagecraft.com/blog/the-jurassic-park-effect-part-2/

I had a few questions after reviewing the blogs if anyone is able to weigh in:
1. Does FreeNAS add features that are not available out-of-the-box to a FreeBSD user (either via GUI or Console/command line)? In other words, does FreeNAS add a "NAS" GUI around FreeBSD or does it add the GUI and add unique features not available in FreeBSD?
2. I know that iXsystems provides support to customers that purchase iXsystems hardware. Are there others on the forum, or elsewhere, that provide professional (paid) support for FreeNAS and TrueNAS users regardless of whether iXsystems hardware is involved?
3. Does FreeNAS get security and other stability updates applicable to FreeBSD in a timely manner?
4. Is there anything that can be accomplished using a FreeBSD server that cannot be accomplished with FreeNAS? In other words, does FreeNAS dumb down FreeBSD?
5. Can a volume (i.e. mirror or raidz2) be removed from a FreeNAS server and imported into a FreeBSD server and be used? At a minimum, can data be copied from this volume onto other media?

 

[nightshade00013]

1 - The GUI and middleware are big additions in and of themselves and you will not find anything to do the job on top of FreeBSD that you can add yourself. And FreeBSD can be configured to do the the things FreeNAS can but you will spend HOURS upon HOURS doing it. There has been some suggestions about developing FreeNAS as a pkg to be installed on top of a regular FreeBSD installation but there are some points of integration that from my understanding just would not play very nice with certain updates changing a function.

2 - I am sure there are people out there who can offer paid professional support. But most of those places will also do other networking or pc repair services. I believe a few members on the forums will also do this when someone is having major problems and need the extra help but we try and work through the issues on the forums as it's better for everyone involved to learn from mistakes and issues.

3 - Yes, when new versions of FreeBSD are rolled out the FreeNAS team rolls them into the next release that is based on the same version. It did take awhile to get up to FreeBSD 11 due to some poor decisions with previous editions but now that we are on 11 updates seem to come out about every 2 to 3 weeks, these updates are rolled in with the FreeNAS modifications and they do in house testing as well as RC's to make sure everything meshes well.

4 - With a regular FreeBSD server you will be able to do a few things that you can not do with FreeNAS. But in all honesty if you are doing those types of things you probably do not want to have them on the same install as your NAS anyway. Plus the average joe who wants a good solution for data storage will have little need and a major pain trying to make FreeBSD do what FreeNAS does.

5 - Yep, may need the same version as what your underlying FreeNAS install was in case the pools feature flags were changed but should be able to import the pool and just run with it.

As far as the article's you referenced is that it is more geared towards the IT pro who deals with things all day and would have no issue setting FreeBSD to do what needs to be done. I know a few of our members are in that caliber of doing things. But there is also a cost vs benefit that needs to be done before you just default to FreeBSD. Figure a small one person IT department that needs to deal with 100 people along with all the servers and networking hardware involved. How long should that person take away from their support work to deal with the file server? You could go with a windows only environment and have a gui centric setup but in the same light you have shifted the cost of management from the IT person's time to the cost of the OS and the user licensing.

To me FreeNAS is a happy medium since it is free to use, doesn't have the per user licensing scheme and also is able to be managed with a gui which will make a lot of things easier to work with. It is also stable as well as able to run on a decent subset of server grade hardware.

It also looks at using off the shelf proprietary NAS boxes which IMHO are only good for one thing (file storage) and home use. Putting them into a commercial environment is honestly just asking for trouble.

 

[NAS-Plus]

Thank you for the excellent, through, and very clear responses! These answers are largely what I guessed, and many of the reasons that I like FreeNAS, but I did not want to assume.

 

[anodos]

I had a few questions after reviewing the blogs if anyone is able to weigh in:
1. Does FreeNAS add features that are not available out-of-the-box to a FreeBSD user (either via GUI or Console/command line)? In other words, does FreeNAS add a "NAS" GUI around FreeBSD or does it add the GUI and add unique features not available in FreeBSD?

In addition to the things @nightshade00013 mentioned, there are a few iXsystems-developed Samba VFS modules that are not available in vanilla FreeBSD (winmsa, zfs_space), an idmap backend (idmap_fruit), and Samba modifications to make it properly provision a Samba4 AD DC on FreeBSD ZFS. FreeNAS also has a REST API that lets you automate many administrative activities. Those are just a few things that I can think of off the top of my head.

2. I know that iXsystems provides support to customers that purchase iXsystems hardware. Are there others on the forum, or elsewhere, that provide professional (paid) support for FreeNAS and TrueNAS users regardless of whether iXsystems hardware is involved?

The support that iXsystems provides for TrueNAS is very different from what you get from other sources. For instance, recently one of their customers noticed a regression in Samba speed, and they figured out the exact commit to the FreeBSD kernel that caused the problem, fixed it upstream, and merged the fix into FreeNAS. I'm not aware of anyone else who does this, and if someone did, they would have a hard time doing it cheaper than iX.

3. Does FreeNAS get security and other stability updates applicable to FreeBSD in a timely manner?

Yes.

4. Is there anything that can be accomplished using a FreeBSD server that cannot be accomplished with FreeNAS? In other words, does FreeNAS dumb down FreeBSD?

Of course. For instance, you can install a desktop environment on a FreeBSD server, listen to music from it, and browse for pictures of unicorns on the internet. Now, that doesn't mean it's a good idea to do this. FreeNAS is focused, which is a good thing.

5. Can a volume (i.e. mirror or raidz2) be removed from a FreeNAS server and imported into a FreeBSD server and be used? At a minimum, can data be copied from this volume onto other media?

Yes

 

[jgreco]

As far as the article's you referenced is that it is more geared towards the IT pro who deals with things all day and would have no issue setting FreeBSD to do what needs to be done. I know a few of our members are in that caliber of doing things. But there is also a cost vs benefit that needs to be done before you just default to FreeBSD.

I've been doing this stuff since the days of 386BSD, which puts me solidly in the "old farts" club, and been doing things like service provider level Usenet news and Squid caching on FreeBSD, which stresses the system at unbelievable levels that no normal deployment would ever approach. There's tuning stuff that I've had to do that only a handful of other people on the planet have ever done. I do enough system builds that I've got automated tools I've written to handle the hundreds of systems under management here.

The issue isn't whether or not you *could* make it work. Over the years, I've deployed numerous Samba-on-FreeBSD fileservers, or NFS servers, or AFP servers, etc. The question is how much effort you want to put into it. Are you really interested in exploring all the latest FreeBSD issues and determining which version is best for interop with Samba? Do you want to be on the bleeding edge of having to discover those things? Do you really want to spend several hours installing and configuring it? Or would you rather just spend 30 minutes deploying a FreeNAS image, one where software developers have spent significant time creating all the nifty features like graphing and monitoring? One where the community has beat the $#!+*#!*# out of it and it is known to work well? One that, when you upgrade it, just continues to work?

I have other fires I'd prefer to fight. I don't feel a driving urge to build the software stack for large complex fileservers.

 

[gpsguy]

You can't purchase TrueNAS without buying their supported hardware.

If you buy a FreeNAS system (like the FreeNAS Mini) all you receive is hardware support from iXsystems.

If you need paid support for FreeNAS, they can put you in touch with a "3rd party consultant". See - https://www.ixsystems.com/freenas-commercial-support/

Are there others on the forum, or elsewhere, that provide professional (paid) support for FreeNAS and TrueNAS users regardless of whether iXsystems hardware is involved?

 

[NAS-Plus]

Excellent responses! My primary interest in FreeNAS/TrueNAS is from a business use point of view. These are largely the answers that I had hoped that I would hear.