unRAID Server release 4.5-beta6 available

[WizADSL]

Addressing your speed issue....

 

Reading and rewriting the SAME sector on rotating magnetic media is a performance disaster.... that's the laws of physics.

 

Back in the day write speed could not keep up with rotational speed, so we had interleaving, and an improper interleave setting would send your disk performance in the toilet.  Having to read and rewrite the same sector is similar to having a too-aggressive interleaving setting -- you spend a LOT of time waiting for the disk to spin back around (as opposed to waiting for the head to seek).  This is why high rotational speed drives as parity affect writes as much as they do.

 

The only way to improve this is some sort of aggregator/elevator method of reading and rewriting, but that itself is fraught with peril in this type of environment.... not to mention would be a complete ground-up rewrite of the core.

 

If you want proof that the bottleneck is the disk and not unRAID, replace parity with a PCIe SSD.  I would almost be willing to rent one just to prove it to the doubters.

 

 

I hadn't really considered the fact that the parity drive is having its data area UPDATED rather than appended when you update a data disk, so you are absolutely right, write latency would be an issue.  I realize that a system crash during my next question would be a very bad thing (as well as a power outage, but I have a UPS), but could write buffering on the parity disk be used to mitigate the problem?

 

The other thing I had forgotten to mention is that I am using a cache disk in my system (not a 1.5 barracuda).  I just did a quick test,  the cache drive has a read speed of 54 MB/s, I'll guess that the write speed should be around 43 MB/s (80% of read).  I got 13.6 MB/s copying an 8.46GB file over gigabit (to a user share w/ cache drive).  In my case I am using a newer model "consumer grade" netgear 8 port gigabit switch.  The cables I am using are cat5e (I know cat6 is supposed to be faster, but even on higher-grade gigabit switches I have seen little difference on short runs).  Does this sound inline with what the speed should be?

 

Although I have only used the 4.4 final and 4.5 beta releases of unraid, I have seen several people post that the previous releases yielded much better performance, although I have not tested that for myself.  Have you noticed what those users have claimed yourself (assuming you are running one of the later versions)?

 

I appreciate your input on this, thanks!

[Kewjoe]

I'm running a WD 200gig IDE drive as my cache and i get 40+ MB/s speeds. This is from a Windows 7 box (all network tweaks, disabling IPv6, and others found on the wiki) over a Dlink DIR-655 Router with Cat6 cables.

[Two Wire]

 

unRAID is a product that is worth many times what it sells for to some people... and not worth 1/10 of the price to others.  I think most complainers should simply find a different product.  Perhaps if they tried that, they'd discover what a deal unRIAD is. How many "alternatives" get mentioned in the forums, and are shot down as inferior or too expensive?

 

We should stop treating these forums like some official 24/7 support forum and realize they are here for our added benefit. Tom could easily just close the forum down and just support his product via email. Many products and companies have forums that don't have employees responding at all or in any constant basis.

 

 

+ 1 

[Kevlar21]

unRAID is a product that is worth many times what it sells for to some people... and not worth 1/10 of the price to others.  I think most complainers should simply find a different product.

 

What!!!!!? First off I think the majority of us here believe that unRAID is a great product. So your solution is that if we have problems or support issues we should just find something different? UnRAID is a product that we love, and unfortunately even the best products have issues and all we want is a more reliable outlet for support. The Veteran users do an amazing job of helping out in the forums, but it just seems that more and more issues are cropping up and some issues are not getting the attention they may need in Limetech's absence.

[NAS]

 

unRAID is a product that is worth many times what it sells for to some people... and not worth 1/10 of the price to others.  I think most complainers should simply find a different product.  Perhaps if they tried that, they'd discover what a deal unRIAD is. How many "alternatives" get mentioned in the forums, and are shot down as inferior or too expensive?

 

We should stop treating these forums like some official 24/7 support forum and realize they are here for our added benefit. Tom could easily just close the forum down and just support his product via email. Many products and companies have forums that don't have employees responding at all or in any constant basis.

 

 

+ 1 

 

 

lol no one asking for 24/7 support... did you guys even read the thread?

 

Were asking for the odd post every now and again as opposed to the current rate of 1 post per every half year. Even one post once a week would be vastly superior. Come on lets put what being asked for in perspective before making silly fanboy "unRAIDs fantastic" responses.

 

We know unRAID is fantastic thats why were here every day.

 

 

[boof]

There seem to be a few interesting statements and viewpoints in this thread :

 

- you paid for unraid based on what it was at the time. Any expectation of future updates is flawed as it is not guaranteed.

 

- unraid 5 will be awesome.

 

- For many users its $ value in features and data handling is much higher than the actual cost of a license.

 

I have to take issue with them all.

 

Buying unraid on its own merits at the time and nothing else is, of course, entirely correct and logical. My problem is that for myself and I suspect others, storage and the purchase of unraid *is* a long term plan. I can't justify the outlay on buying 20*1.5TB or 2TB drives in one go. And one of the key selling points of unraid is its easy future expandability. This isn't inherently spelt out as 'we will thus provide updates' but based on the history of the program and that for unraid bought today to be competitive in 2 years time then this would need to occur it seems reasonable to factor it in.

 

The problem is that this future expandability entirely relies on unraid being updated. This isn't from the core unraid md driver perspective as 'it just works' at the moment - but the fact that unraid is distributed as it's own linux distribution. Which means we rely on updates to include new drivers for new motherboards and controller chipsets all of which factor into future expansion. If we must treat unraid as a point in time snapshot never to be compatible with future hardware due to completely unreliable updates this should at least be heavily highlighted in the user wiki as this will fundamentally change many peoples purchasing decisions. An 'easy' solution to reduce support overheard in this respect would be to split unraid from the OS and provide packages for common linux distributions (RHEL/Centos, ubuntu, slackware) and let the normal mainstream kernels handle driver upgrades. I realise there is work going on at the moment to do this in the community but not all of unraid is open source so what we will end up with will be subtly different.

 

Tom has hinted at the next version of unraid and potential features and I'm going to single out bubbaq's comment that it's going to be good. This surely is the entire crux of this conversation. We have nothing to say it's actually being developed or any time frame for a release if there ever is one. The fact bubbaq has made this statement lends some very credible weight to it's accuracy. But I then have to wonder if there's a 'backchannel' of information here we lesser forum contributors are missing out on.

 

If the perceived value of unraid is higher than the current cost and the current cost model is causing problems in term of revenue stream and / or support...raise the cost?

 

I've touched on this in a previous post that I don't understand the end game of the current licensing model - although I obviously do not have all the data available to me - and I half expect significant changes to any future releases licensing model. But if raising the cost or charging on a per release basis means more development and support then so be it.

 

As some final thoughts as I said above storage is a long term strategy for me and this was one of the large factors in going with unraid. I'm very happy with what it does and how it works, my servers sits tucked away in a corner just doing what it does with no issues. However I only have 5 disks in it at the moment. My future plans relied on adding more disks which requires at least one new disk controller and upgrading my plus license to pro. I'm now very very hesitant to do that as unless I buy enough controller capacity in one go to support all my potential future disks (which is what I wanted to avoid  - unraid should allow incremental hardware additions) I have no guarantee that future controllers will work or at the very least the ones at the best price / performance will be supported. At present when my current capacity is exhausted I'll be going back to my 'storage feature matrix' which I painstakingly went through before deciding on unraid. At this moment in time unraid will be someway from the top, whereas before it was very clear cut.

 

I've recommended unraid to maybe a dozen people over the past 6 months - all have now come to me concerned about the current state of updates. Many of them have either moved away from unraid altogether or have at least postponed purchase of a plus/pro license (upgrade or otherwise) after evaluating the free version. This, I'm sure, is but a splash in the plan revenue wise but scale up my (not so uncommon?) experience to a few more forum members and it starts taking a toll. Both on revenue and unraid's reputation which has, until now, been A+ wherever I've looked.

 

So I don't really want to come on here and moan - I hope the above is seen as constructive. As I say if things don't move on then that's life, as has been pointed out I got what I paid for and no more. My next storage evolution will come much much sooner than I'd planned or hoped but it won't involve unraid.

 

I would be very dissapointed though.

 

All of this speculation / conversation could be fixed by a 2 minute post update from Lime Tech which makes it all the more baffling.

 

Can I ask how many peoples concerns over this are :

 

- because of core unraid features that they have problems with or are looking for new features to be added.

 

- due to the OS like nature of unraid and potential future hardware support problems.

 

It might provide some interesting data.

 

 

[olympia]

Although I have only used the 4.4 final and 4.5 beta releases of unraid, I have seen several people post that the previous releases yielded much better performance, although I have not tested that for myself.  Have you noticed what those users have claimed yourself (assuming you are running one of the later versions)?

I appreciate your input on this, thanks!

 

What version are you using now?

Cache drive speed is "broken" in the latest 4.5 betas. You can only achieve about 1/4 of the write speeds comparing to 4.4 final.

Which is in my case about 13-14MB/sec too, instead of the 70+MB/sec I have with 4.4 final.

[WizADSL]

Although I have only used the 4.4 final and 4.5 beta releases of unraid, I have seen several people post that the previous releases yielded much better performance, although I have not tested that for myself.  Have you noticed what those users have claimed yourself (assuming you are running one of the later versions)?

I appreciate your input on this, thanks!

 

What version are you using now?

Cache drive speed is "broken" in the latest 4.5 betas. You can only achieve about 1/4 of the write speeds comparing to 4.4 final.

Which is in my case about 13-14MB/sec too, instead of the 70+MB/sec I have with 4.4 final.

 

Using 4.5b6 right now.

[lionelhutz]

Come on guys, is the beta release thread the best place to post and post and post about Tom not having a presence on the forums?

 

This wasn't even called a support thread for the release, just an announcement or release thread. There wasn't anything put in the initial about posting your issues so that Tom would get them addressed.  Try not to go on for pages commenting about this either...

 

Peter

 

[olympia]

Although I have only used the 4.4 final and 4.5 beta releases of unraid, I have seen several people post that the previous releases yielded much better performance, although I have not tested that for myself.  Have you noticed what those users have claimed yourself (assuming you are running one of the later versions)?

I appreciate your input on this, thanks!

 

What version are you using now?

Cache drive speed is "broken" in the latest 4.5 betas. You can only achieve about 1/4 of the write speeds comparing to 4.4 final.

Which is in my case about 13-14MB/sec too, instead of the 70+MB/sec I have with 4.4 final.

 

Using 4.5b6 right now.

 

Try to go back to 4.4 final, if you can. You will immedietly notice the huge difference in writings to cache drive.

[NAS]

Come on guys, is the beta release thread the best place to post and post and post about Tom not having a presence on the forums?

 

This wasn't even called a support thread for the release, just an announcement or release thread. There wasn't anything put in the initial about posting your issues so that Tom would get them addressed.  Try not to go on for pages commenting about this either...

 

On the contrary this is the best place as it is the most likely place for it to get noticed.

[WeeboTech]

Come on guys, is the beta release thread the best place to post and post and post about Tom not having a presence on the forums?

 

This wasn't even called a support thread for the release, just an announcement or release thread. There wasn't anything put in the initial about posting your issues so that Tom would get them addressed.  Try not to go on for pages commenting about this either...

 

On the contrary this is the best place as it is the most likely place for it to get noticed.

 

ENOUGH!!! I'm with lionelhutz.

A response was forwarded, if you want to continue off the topic of 4.5-beta6 please start another thread.

[NAS]

It will be enough when Limetech LLC reposnds. Until then the more posts the better both for and agains. This is too serious a subject for it to be ignored for ANY reason.

 

Sorry but Limetech has know about this conversation for several days now. If it takes 100 more post to get a 5 minute response then that's fine in my book.

 

3 months is long enough to be polite.

[WeeboTech]

It will be enough when Limetech LLC reposnds. Until then the more posts the better both for and agains. This is too serious a subject for it to be ignored for ANY reason.

 

Sorry but Limetech has know about this conversation for several days now. If it takes 100 more post to get a 5 minute response then that's fine in my book.

 

6 months is long enough to be polite.

 

What about wasting our time too?

It does not matter how many posts there are if limetech has not logged on in over a week.

Please keep this thread related to the Beta.

 

Please post,  issues with production,  support of production and expectations of such,  in it's own related thread.

[NAS]

OK sure.

 

Limetech can we have an update on the beta please

[lionelhutz]

Yes, enough is enough! Do you guys not realize that when there are pages and pages of complaining mixed in with a few reports of problems that it's unlikely Tom will spend the time to go through the pages of off topic posts that have been generated to find the legitate problems? Going on and on for days and weeks when the person at the heart of the discussion is not listening is a waste of everyone's time and clutters this thread with useless garbage.

 

Honestly, we're lucky Tom has even allowed responses to be post responses here because this section could be used to only post the announcement and then the lock the thread. So, show him some respect and discuss issues with the beta here so he has some useful info about this release to use at a later time if he does future development.

 

Use the Forum Feedback or Lounge sections if you must complain in public.

 

Peter

 

[cwr]

Use the Forum Feedback or Lounge sections if you must complain in public.

[WizADSL]

I think for the moment those that are calling for a stop to the requests for limetech to respond are probably right, surely the posts have been read if they are going to be read and the choice to ignore them has already been made.  The reason I am making THIS post is just to mention to those complaining about the posts, that your anger is misdirected when you reprimand the posters for asking for a reply from limetech, it is limetech that has allowed this to continue for as long as it has, if a short reply is beyond limetech's abilities then perhaps the situation is truly grave.  On the other hand, it is understandable that your anger be directed at us since we ARE here to be written to.

 

Did I mention how much I hate that things have gotten to this point?

[Jon]

I think for the moment those that are calling for a stop to the requests for limetech to respond are probably right, surely the posts have been read if they are going to be read and the choice to ignore them has already been made.  The reason I am making THIS post is just to mention to those complaining about the posts, that your anger is misdirected when you reprimand the posters for asking for a reply from limetech, it is limetech that has allowed this to continue for as long as it has, if a short reply is beyond limetech's abilities then perhaps the situation is truly grave.  On the other hand, it is understandable that your anger be directed at us since we ARE here to be written to.

 

Did I mention how much I hate that things have gotten to this point?

 

I'm a nobody here, but I agree.  Tom chooses not to reply to this thread or he would have already.  Anyone can look up his profile and see that he was here just yesterday.

 

Last Active:  September 15, 2009, 12:24:59 PM

 

We all pretty much use it as a media server, so step away from the forums and go watch a movie or listen to some music

 

 

[kapperz]

WOW, starting with page ~15 the gloves start to come off. I've already start to invest into my first unRAID and will complete the project (before I saw this thread). I've been a developer for nearly 10yrs and can understand both the developer's and user's points. Maybe interest has been lost on the project, maybe there was a family emergency, maybe the dev needed a vacation (or all three). As long as critical bugs are address to maintain a stable product, I'll be happy. From what I gather, unRAID works very well on cheap hardware and that's why I've decided to go with it. I was almost ready to buy a Drobo until a read countless angry reviews about data being loss and terrible customer support (lots of other issues too).

 

I will start with a stable version at first. And because I'm a Linux/unRAID n00b.

 

I do want to thank all the active posters for quickly answering my questions. My stress is reduced knowing when I actually start building (next few weeks), I have support.

[lionelhutz]

The reason I am making THIS post is just to mention to those complaining about the posts, that your anger is misdirected when you reprimand the posters for asking for a reply from limetech, it is limetech that has allowed this to continue for as long as it has, if a short reply is beyond limetech's abilities then perhaps the situation is truly grave.  On the other hand, it is understandable that your anger be directed at us since we ARE here to be written to.

 

No, no misdirection at all. Tom didn't post 4 or 5 pages of posts that didn't belong here. I'm not angry, I just find what has been going on VERY rude towards Tom.

 

Peter

[mark-hc]

I HAVE A BURNING CROSS IN ONE HAND AND GOD HELP ME A PITCHFORK IN THE OTHER! I'M COMING FOR YOU TOM! I'M COOOOOOOOMIIIIIIIIIIIIIIIINNNNNNNGGG!!!!!!! GOD HEEEEELP MEEEEE!!!! TOOOOOOOOOMMMM!!!!!

 

Man that felt good... 

[eeking]

- due to the OS like nature of unraid and potential future hardware support problems.

I'll admit I'm not entirely certain about the internals, but isn't the md driver part of unraid open source? Only emhttp is closed source. So I think it practical that users could continue to compile their own updated kernels with the md driver to support new hardware. That's not what we want, certainly, but even if unraid development halted tomorrow, the functionality as it exists today to be carried forward almost indefinitely.

[Joe L.]

- due to the OS like nature of unraid and potential future hardware support problems.

I'll admit I'm not entirely certain about the internals, but isn't the md driver part of unraid open source? Only emhttp is closed source. So I think it practical that users could continue to compile their own updated kernels with the md driver to support new hardware. That's not what we want, certainly, but even if unraid development halted tomorrow, the functionality as it exists today to be carried forward almost indefinitely.

There are a number of us with the skills to build a new "emhttp" equivalent, if unRAID were ever to be abandoned by lime-technology and there was a need to write a replacement.   I have not built a replacement, even though I'm certain I could, as the current interface works just fine.  It is not perfect, but it is a media server... and, it has not been abandoned by lime-technology... 

 

I really just want to use it so I can watch movies, see photos, listen to music, etc.   I can work around the speed limitations, remaining bugs, and operational quirks (don't press restore when you want to "restore" data... as it does not, but might erase it instead).   I really want to support Tom and Lime-technology instead of undermining his business.   

 

As you said, the "md" driver is GPL, and source code would allow us to continue to evolve unRAID to work with any foreseeable hardware using the existing emhttp executable to manage the array.

 

I purposely chose unRAID for this exact reason. I can make it do whatever I need...  but so far, other than Tom taking time to focus on issues in his life other than this forum for the past few months, there is no need to start a re-write.

 

Joe L.

[bubbaQ]

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.