unRAID Server release 4.5-beta6 available

[fireman]

I think it practical that users could continue to compile their own updated kernels with the md driver to support new hardware.

 

Many already have.  Detailed instructions are in the Wiki.

 

I have not built a replacement, even though I'm certain I could, as the current interface works just fine.

 

Agreed.... no need to do so, particularly as there are many ways to get extended functionality (add a web server, compile your own kernel, unMenu, etc)

 

I really want to support Tom and Lime-technology instead of undermining his business.   

 

So do I.  I would, however, like to know that Tom has escrowed the source code.... just in case.

 

+1

 

I, by no means, have the necessary skills to go beyond what I already have on my server so that's why I am hesitant to put more money in that project at this time. I know that a $100 isn't a whole lot of money for many and as a matter of fact, it isn't all that much money for me too but I want to make sure my money is spent on something durable. I want Lime-technology to grow so I can enjoy my unRaid server for a long time. I won't even mind spending more, if need be, for upgrades and add-ons because they're worth it to me. But I refuse to put money in a product that I can't figure if it's still gonna be around next year or even next month...

[Copey]

Any chance someone could sum-up the issues with this beta or is it wise to stick with 4.4.2 for the mo'?

 

Cheers

 

 

[mark-hc]

Stick with 4.4.2.

If 4.4.2 has problems.

Go with 4.3.3

 

I am running 4.3.3.

I am able to strema blu ray discs from unraid to xbmc just fine and dandy. What more does anyone want? 

[RobJ]

Any chance someone could sum-up the issues with this beta or is it wise to stick with 4.4.2 for the mo'?

 

I'll try, but welcome input from others.  Here are the issues as far as I can tell, and you can decide for yourself whether they apply to you.  This is of course completely unofficial and subjective, relying too much on my memory.

 

Unofficial Assessment of unRAID v4.5-beta6

 

* Active Directory support

** appears to be partially working, but buggy yet, and not well tested, very beta

** see the Release Notes and the Active Directory issues forum

 

* NFS support

** seems to be mostly working, but with some problems, still beta

** see the Release Notes and the UnRAID Topical Index, NFS section

 

* Cache disk

** For those using User Shares and a Cache disk, a performance problem has arisen in one of the recent betas.  Normally, the Cache disk provides full speed writing to User Shares, by bypassing the work of parity updating, postponing it to off-hours.  For some reason, files appear to be stored to the Cache drive now, but at the same speed as if parity is being updated.  This is perhaps the primary reason why a user might prefer staying with v4.4.2, until the speed is restored in a future version.  There are no reports of data loss or corruption, just the loss of the speed improvement that the Cache disk provides.

** It would be useful to hear from more users that save to a User Share and are using the Cache disk, in order to better define this problem, whether it applies to all or only some of this set of users.  If it only applies to some, then it needs to be determined what is different about their setup.

** I believe all relevant posts about this issue are in this very thread, this very long thread!

 

* Samba

** Very recently, a *possible* problem with the startup of the Samba modules has been reported.  I don't have a good feel for this one yet, as to what the issue even is.

** Only about 2 to 5 users may have seen this, most have not experienced it.

** There has not been any data loss reported, as far as I can recall, but there is a loss of access to the unRAID server, temporary.

** A possibly common thread is that most or all of those that may have experienced this were connecting to the unRAID server from a Windows 7 machine.  Obviously, we need more information and testing.

** See 4.5 Beta 6 Samba is stopped problem and 4.5 beta 6 - Can't connect to server

 

* 20 drives

** As of v4.5-beta4, the drive count was raised from 16 to 20, that's 19 data drives plus the parity drive.  However, it has been found that while you can add more drives, the array usually does not run successfully, so this gain should not be considered fully functional as yet.

** There is no problem supporting the same 16 drives as before, so there is no reason not to use these beta versions because of this issue.

** See this post and the subsequent posts, which include a link to an additional related thread

 

* Other, unknown

** I *think* there were a very few early adopters of v4.5-beta6, that reverted back to an earlier version, probably v4.4.2.  I don't know why.

** There are often a few users impacted by changes in the included Linux kernel.  A particular kernel may not work as well with a particular set of hardware, but this may or may not explain any of the early 'reverters'.

** You may find their posts in the early pages of this thread.

** There is a very minor issue involving the Spin Down button, see this post.  Does not affect normal operation, and carries no risk to data.

 

* General performance and stability

** Apart from the issues above, I believe most veteran users feel that v4.5-beta6 is as stable and fast and well-tested and widely used as any of the versions prior to it.

** I don't recall ANY issues involving data loss or corruption, related to either v4.5-beta6 or v4.4.2.

** Performance-wise, if you are impacted by the Cache disk problem, then v4.4.2 is a much faster choice, but otherwise v4.5-beta6 may be faster because of the disabling of NCQ.

** This is completely my own opinion, so don't quote me, but from the many posts and syslogs I have seen, my feel for current version usage is roughly:

     40%  v4.5-beta6

     30%  v4.4.2

     15%  v4.3.3

     15%  all others

** Please do check out the Release Notes, to compare the versions.  You may be surprised by how many things have been fixed since v4.4.2.

 

If I have missed anything, or made a mistake, or you disagree with something above, then please say so (gently and constructively, I'm a little sensitive).  I'll be happy to edit/update this post.

[limetech]

Any chance someone could sum-up the issues with this beta or is it wise to stick with 4.4.2 for the mo'?

 

I'll try, but welcome input from others.  Here are the issues as far as I can tell, and you can decide for yourself whether they apply to you.  This is of course completely unofficial and subjective, relying too much on my memory.

 

 

Nice summary.

[JarDo]

Very useful (and impressive) update.  I am running 4.5-beta6 and I can confirm that my experience is that the cache drive provides no increase in write speed.

[Guzzi]

Any chance someone could sum-up the issues with this beta or is it wise to stick with 4.4.2 for the mo'?

 

I'll try, but welcome input from others.  Here are the issues as far as I can tell, and you can decide for yourself whether they apply to you.  This is of course completely unofficial and subjective, relying too much on my memory.

 

 

You should also mention the 16+ disk problem - beta intrroduced the usage of 20 drives, but I do not know of any user being able to use it without breaking the whole set - exceeding 16 drives (incl. parity) throws exceptions.

 

btw - lt - welcome back ;-)

[Joe L.]

If I have missed anything, or made a mistake, or you disagree with something above, then please say so (gently and constructively, I'm a little sensitive).  I'll be happy to edit/update this post.

You did not mention the inability to grow past the original 15 data drives.  

 

First mentioned here with the "kernel: BUG: unable to handle kernel paging request " error from the syslog:

http://lime-technology.com/forum/index.php?topic=3764.msg33390#msg33390

 

Joe L.

[ClunkClunk]

Thanks for the summary. I was wondering why the cache drive I added provided no real benefits that I could detect or benchmark.

[PhilH]

Welcome back Tom. Are you back to stay for awhile? 

[RobJ]

Thank you Tom and all.

 

I did forget the 20 drives problem, and one other minor issue (a spin down oddity).  I should say that this drive count issue is not a reason to revert, but should be characterized as an announced feature that is not yet fully functional.  I have added both to the post above.

[fitbrit]

Thanks RobJ, and welcome back Tom. I'm really looking forward to 19 data drives!

[NLS]

Nice summary.

 

Pretty poor reply after all those months "Limetech".

 

I can't help not to find this funny:

 

So do I.  I would, however, like to know that Tom has escrowed the source code.... just in case.

 

Seems bubba is touched by all this after all.

 

Anyway, I hate that LimeTech seems to need food chewed by a few Linux wizards here, before swallowing. I certainly wish he pays you guys for it.

 

I would really wish there was a real viable alternative to unRAID right now, since this whole thing (and you've noticed I didn't take part at all) does show a few things clear enough (whatever bubba contradicts to those people ...with his "computer brain" with pure "if-then-else" logic - which is why he worked for an F500 company and he is not owning one, no insult meant bubba). ...And of course in the same sense it has clearly touched new (or wannabe) users.

 

I wouldn't be very surprised if Tom just appeared, exactly because he noticed his sales line dropping (imagine the "great" advertising to new users this thread gives).

 

In fact if there is the slightest chance Tom is even remotely thinking of charging new fees for 5.0 (as is common with major updates), I am jumping ship even if the second alternative is way back. Even if 5.0 is magic. A magic 5.0 with Tom's weird attitude towards his most loyal/active clients (the forum members) is not so magic. He clearly prefers clients that pay and disappear. We would all love to have such clients, but I don't think there are many on Planet Earth in this market.

 

 

[bubbaQ]

Seems bubba is touched by all this after all.

 

Not at all.  I have no problems whatsoever with Limetech's service or performance to me.  I'm just fine with no updates for the next 20 years, if that's Lime's choice.  unRAID does everything it did the same way it did as the day I bought it.... and I didn't buy vaporware or expect a single thing from Lime in the future.  Anything else above that is a free bonus.  If I go and buy a new laser printer, and get a free 16GB flash drive with it, and the flash drive dies, the printer company doesn't owe me squat.  20 drives, cache disk, user shares, etc, are all just gravy that I can and do live without.

 

Escrowing code is a protection against something that was not a "choice" of the developer, i.e. death or disability.

 

I have an application I have made no updates to since 1994.  It's used in the garment finishing industry, and still being used unmodified.  Still get paid for new licenses occasionally.  Code is escrowed.

 

In fact if there is the slightest chance Tom is even remotely thinking of charging new fees for 5.0 (as is common with major updates), I am jumping ship even if the second alternative is way back.

 

If jumping ship means you will desert us and leave the boards in peace, then Tom, PLEASE charge for the update. Somehow I doubt you will follow through on that promise though.  Perhaps I can shame you into it by talking about it repeatedly on the boards.... that seems to be how some people think they can get what they want here.

 

The posts bemoaning lack of something from Lime the poster things they are entitled to, bear an uncanny resemblance to a 4-year old stamping his feet and holding his breath.

 

 

[WeeboTech]

Well said BubbaQ.

People do have the choice to take their current ball and go home with it.

[NLS]

BubbaQ I will not answer in the same manner as I feel I owe you a few things (and thank you for them) and I am not a man that forgets such things (plus I realize that the style usually goes with the job... for example an Admin takes pride in being BOFH and the dev has to be the guy with no manners but full of analytical reasoning for everything that presumably lets the opposition having nothing to say ...well in the best case - done both for a while).

 

I can only say I didn't deserve that, esp. having NOT been a part of this whole parody of a thread (in turn being in a parody of an Official Company Forum).

I have whined from time to time but quite less than others.

 

My original points stand. Yours probably do too. There is no single truth here, we don't have a 1 and a 0.

 

I cannot say I understand the limitless brotherly spirit (as fellow devs) to Tom though.

 

I don't want to go "NAS-like" and start arguing point by point, I don't want to bother, I don't even have the time. Seems I have more time than Tom though.

 

Weebo people have the choice to take their current ball and go home with it, or stay here and whine like CUSTOMERS some times do (well, fact of life). With a reason too if you ask me (although I know a couple of you disagree).

 

As a person, I wish Tom the best with his life and his ventures.

As a customer, Limetech I have to say sucks in so many ways.

 

 

[WeeboTech]

Weebo people have the choice to take their current ball and go home with it, or stay here and whine like CUSTOMERS some times do (well, fact of life). With a reason too if you ask me (although I know a couple of you disagree).

 

It's a "beta". 

We usually get 1-2 updates a year.  It's not that customers have not been served with a reliable version or help when needed.

It's continued moaning about a beta and lack of progress with it.  What was paid for works. There is no SLA for future enhancements or future release dates.

 

I don't like the delay in forward motion, yet I do have the choice of writing my own interface to the MD driver and moving forward.

(Something that I'm doing on my own anyway).

[JarDo]

Oh, Boy!!

 

I don't quite know what to think.  I see the point to both of your positions.  unRAID still continues to do for me exactly what I bought it for and all the additional features added since the time of my original purchase were not benefits that I was anticipating at the time I made my purchase.

 

But since that time, my unRAID server has become a big part of my life.  Of course I've installed many of the great additions contributed by the members (unmenu, unraid_notify, unraid-web, etc...) and I use many of the user-contributed utilities (i.e. preclear_disk).  I found the period of time that all these contributions were being made in full-force exciting.  I love to tinker.

 

But, it seems to me that the user contributions have slowed in the last few months as the contributors are waiting for a future release that is supposed to include new hooks allowing these contributors to extend their creations (to the benefit of users like myself )  For that reason I'm hopeful for a v5.0 release.  And yes, I'd probably pay for it just to keep with the release that the contributors are using (Who I believe should not have to pay).

 

In the mean-time unRAID has encouraged me to spend hours every day working on my own personal utilities.  I've come a long way teaching myself shell scripting (BASH is quirky to say the least) and am working on a project to populate (sync) a database (MYSql) with information about my music collection (which is much too large now to manage without the ability to query a database)  I'm about finished with this part.  Next, I'm going to tackle PHP and put a UI on top of that database.  When I get that far, I too will share with the community if anyone is interested.

[WizADSL]

I'm just glad to see limetech back and hope that development continues.  If it helps any, I'd be willing to pay for 5.0.

[NLS]

Let me make something more clear: I would (possibly) be willing to pay for unRAID 5.0; I wouldn't be willing to pay Limetech for unRAID 5.0.

 

Unfortunately separating the two, is not possible.

Many people will have the same internal fight like me if the time comes and we need to think about it.

 

Too much talk about money, without having an indication about this happening though.

 

BTW, even if 5.0 would cost money, that wouldn't mean 4.X should stop being debugged. Just mentioning.

 

[lionelhutz]

Very good work as usual Rob, that sums it up very well.

 

The posts bemoaning lack of something from Lime the poster things they are entitled to, bear an uncanny resemblance to a 4-year old stamping his feet and holding his breath.

 

It really reminds me of the Satellite hacking boards of earlier years where the users would bitch and moan about no new fix so badly that the developers would just finally declare "Screw You" and leave. It's funny how so many people think they are entitled to get things for FREE. Getting something extra for free is a privilege, not an entitlement.

 

To me, any free software update after the point when I purchased it is a big bonus, not a requirement. "I purchased a product that does exactly what it was advertised as doing when I purchased it but the developer still owes me more!". Seriously? You won't get any major free updates from Microsoft, or many other software companies for that matter.

 

Having said that, I look forward to the possibilities that version 5 will bring and I do hope that it is produced.

 

Peter

 

[Kewjoe]

Tom / Limetech, I hope you realize the complainers are few in numbers (and seem to only contain 3 letters in their name). There are many more people very happy with your product and are patient for if and when you provide us with upgrades/updates. Thanks!

[BryantD]

Tom / Limetech, I hope you realize the complainers are few in numbers (and seem to only contain 3 letters in their name). There are many more people very happy with your product and are patient for if and when you provide us with upgrades/updates. Thanks!

 

Exactly!

[Kevlar21]

To all the people asking for the "complaints" to stop, please realize that unRAID is not perfect and problems do exist, and those with problems do often require assistance to rectify them. I don't understand this attitude of arrogance that permeates threads like this one.  When someone has a complaint or grievance that is not addressed, of course they will be vocal about it.  There's the attitude now that "hey, my unRAID installation works fine, it's worked fine since the day I bought it, so all other criticisms are null and void". I have this mental picture of unRAID being a car manufacturer with users complaining of poor gas mileage, acceleration performance etc.  They contact the company that manufactured the product for help and find out that all the factory's, dealerships, and repair bays have decided not to answer the phone for the past 6 months. Frustrated, they ask other unRAID car owners for help and are met with responses that our cars are perfect, and anybody who criticizes them should shut up.

 

What's going to happen if the lack of official support and communication continues, and currently happy users have to replace hardware, or implement a new server build and possibly run into problems? The current situation regarding the lack of fixes/supports/communication affects every single user of unRAID whether they realize it or not. Maybe dropping the fanboy attitude and working towards addressing everyone's problems on the forum just might BENEFIT THE ENTIRE COMMUNITY.

 

 

[WeeboTech]

Put the complaints about limetech support issues in another thread. Leave this one alone for beta bugs.