Move News


limetech

Recommended Posts

In general, I agree this has sort of devolved into a Lounge-worthy thread.  Personally, I think complaints about a lack of updates, in the thread within the announcements folder where the owner/developer suggested he would make updates, is relevant.  But, arguably the accompanying discussion is not.  Still, I wouldn't call this thread a "beta release" thread, nor would I call the announcements folder a folder just for beta releases.  I kind of like how SageTV used to have final release and beta release threads that people could subscribe to in order to get emails when updates came out.

 

I find it a little disturbing that some folks actually check in on how frequently people look at the forum and talk about what people are doing with their time.  Maybe Tom has another job; is that our business?  Did you get what you paid for?  I did, even though I have to admit I spent extra money upgrading to the pro version rather than buying it the first time. 

 

The reason to check to see if he's logged in is that that is the only sign some of us have that he hasn't walked away from the product. Now, I obviously haven't been concerned about that in the last few months, but there have been some "dry spells."  For other one-man development teams, a lack of forum posts and software updates are the only sign that the product has been abandoned (see DragonGlobal's ShowAnalyzer for a recent example).

 

And, while I don't mean to imply in any way Tom shouldn't have another job, if Lime-Tech is a second job, that suggests development and support could be only as time allows based on his other job.  That might not be a big deal to some.  But, I think I don't think its entirely unfair for prospective customers to consider.  If it is a second job, it also suggests he might not be dependent on the company, meaning he could drop it whenever he wanted (which, I think argues for more frequent updates).

 

You guys writing long letters criticizing Tom need to consider the possibility that he's spending time reading your whining instead of doing things that will help you.  Why not take a positive approach or say nothing at all.  Perhaps offer to help, if you are able to. 

 

I think this might come across a lot snottier than I intend, but I don't know how else to write this.  I think its fair to say, based on the comments in this thread, that there's a fair number of unRAID users that would like to see more frequent progress updates.  And this definitely isn't the first time this has been discussed on this forums.  If Tom wants to be responsive to this set of customers, then I don't think it's a waste of 10-15 minutes for him to read the messages in this thread and make a decision on how to proceed.  If he doesn't care about that, then he's simply not going to read the thread, in which case now time is wasted.

 

Now, I agree its a big waste of time for him to read these messages if he's already made up his mind on what he's going to do.  I basically lost hope a long time ago that we'd see more frequent updates.  But then he did post the message suggested he'd give weekly updates on the progress of v5 development.

 

Edit: Bolded existing text above for emphasis.

Link to comment
  • Replies 216
  • Created
  • Last Reply

Top Posters In This Topic

Let me post a quote of the Announcement board rules.

The Announcement Board is used to announce the availability of new unRAID OS releases, and the reporting and discussion of bugs and/or features specific to the announced release only.  Any off-topic posts may be deleted or moved without warning.

 

As far as "Has unRAID been abandon scenario"

 

The initial post is pretty clear that there is intention to continue unRAID.

Link to comment

Then arguably the thread started out in the wrong folder.  But, more generally, the announcements folder is regularly used for things other than what is described in the board rules.  See: "Forum Upgraded, and Other News," "Sorry about the recent website/forum outage," "Forum Theme Issues" for some examples from the first page.  In practice, the folder is used for general announcements from Lime Tech, and accompanying discussion around those announcements.  But, in general, I mostly agree with a decision to split/move the thread to the Lounge.

 

Also, as I said in my previous post (bolded now), I don't think unRAID has been abandoned (nor do I suspect it will be any time soon).  We do have a fairly recent update, and we know Tom is going through a major life change right now.  But this thread is discussion what has been an ongoing problem in some peoples' minds.  I don't think it would take too much effort to find the other threads where people have asked "what happened to Tom."  There's even been some limited discussion of back up plans if Tom does drop unRAID- I think you're one of the people to say you could reverse engineer the closed source portions of code.  So, it's something people have worried about in the past, to varying degrees.

 

I'm really not trying to be a jerk here.  I'm only trying to argue the advantages of more frequent updates.

Link to comment

 

  Up until recently I recommended UnRaid to all my friends that were looking for large storage servers for their media collection.

However with many now moving to 3TB drives, and the lack of confidence in UnRaid's stability, as well as not knowing if or when these deficiencies will be addressed I no longer tell them about UnRaid. I simply tell them, their guess is as good as mine. (Not a good answer, but I don't want to be the one responsible for them losing years and years worth of media collecting.)

 

  I am running two servers due to the fact I can not get Parity to run correctly on any of the beta 5 releases. I would like to sell my old server, but until I see a drastic improvement in 5.0 releases, I have to keep it as a back up. (And it is quickly running out of storage space.)

I am not going to abandon UnRaid, at least no time soon. But at the same time, I no longer have that wonderful sensation I once had knowing that my media collection was more secure with UnRaid!

 

  I will try not to complain so much, but that doesn't mean I am happy about things.

Link to comment

Part of the argument (I prefer debate) sort of throws fuel on the flame with comparison and notion of abandonment.

The first post clearly shows unRAID and limetech is intending to move forward albeit at a slower pace then others prefer.

 

The original post isn't even a month old.

 

I've been here 5 years, Joe L, 6, other members have been here a long time.

Let's not fuel the flame with notations and scenarios that do not exist.

 

This has been the pattern for a long time. We all (me too) want to see that pattern improve.

Speak your mind, but keep it on target and strictly relevant. (especially since any tangents or comparisons can be easily misinterpreted).

Link to comment

 

  Up until recently I recommended UnRaid to all my friends that were looking for large storage servers for their media collection.

However with many now moving to 3TB drives, and the lack of confidence in UnRaid's stability, as well as not knowing if or when these deficiencies will be addressed I no longer tell them about UnRaid. I simply tell them, their guess is as good as mine. (Not a good answer, but I don't want to be the one responsible for them losing years and years worth of media collecting.)

 

  I am running two servers due to the fact I can not get Parity to run correctly on any of the beta 5 releases. I would like to sell my old server, but until I see a drastic improvement in 5.0 releases, I have to keep it as a back up. (And it is quickly running out of storage space.)

I am not going to abandon UnRaid, at least no time soon. But at the same time, I no longer have that wonderful sensation I once had knowing that my media collection was more secure with UnRaid!

 

  I will try not to complain so much, but that doesn't mean I am happy about things.

 

It's hard for me to recommend it in a corporate environment also.

I would also like to point out, it's not,  "if the problem area's are going to be addressed." It's more when. We've seen other members compile more recent kernels and test with them. From what I can remember, some problems still exist.

 

if you are unhappy, you are communicating those points as I think you should.

Link to comment

Part of the argument (I prefer debate) sort of throws fuel on the flame with comparison and notion of abandonment.

The first post clearly shows unRAID and limetech is intending to move forward albeit at a slower pace then others prefer.

 

The original post isn't even a month old.

 

[snip]

 

Speak your mind, but keep it on target and strictly relevant. (especially since any tangents or comparisons can be easily misinterpreted).

 

Let me try to be a bit more clear.  I'm not very concerned about unRAID being "abandoned" right now.  I'm convinced, more than ever, that we'll see version 5 out the door relatively soon (i.e., in a few months).  But, in the long term, I'm much more concerned about it. There was even a time in the past where Tom didn't give any updates for a few months where I didn't expect to see another update.  You can probably strike that one up to being a newbie, and not seeing things like that in the past, but its probably something some of the other newbies thought too. 

 

I think the "fear of abandonment" is real and completely rational for customers of small companies/projects like LimeTech/unRAID.  I admit I'm probably overly sensitive because I'm a user SageTV, where I always said I was worried Sage could shut down overnight (after which people always assured me Sage seemed strong and healthy, which was probably true, but didn't change the result).  Still, trying to think objectively as I can, I think its a reasonable fear to have.  Turnover in IT jobs is fairly high, and people get sick of working on one project.  A big company can transition a project to another person- a one man project can't do that as easily.

 

I don't think that's a tangent, because I think that's very, very closely linked to my point about more frequent progress reports.  I don't want to read updates because I want to see that Tom is making adequate progress, or to see what the expected release dates are.  Tom will presumably work as quickly as his time allows.  I want to read updates because I want to see he's still actively developing the product and that it seems to be a priority in his work life.

Link to comment
I don't think that's a tangent, because I think that's very, very closely linked to my point about more frequent progress reports.  I don't want to read updates because I want to see that Tom is making adequate progress, or to see what the expected release dates are.  Tom will presumably work as quickly as his time allows.  I want to read updates because I want to see he's still actively developing the product and that it seems to be a priority in his work life.

 

It's a tangent. This part of the point, is flaming a fire with a fear based on another project's history.

It's obvious he's actively working with the product. He could be researching, coding, preparing other ideas, etc, etc.  He stated his goal less then a month ago.  The target wasn't met and unfortunately this makes people have expectations and become sensitive if they are not realized. 

 

  Now the core issue in what people are unhappy about is diluted with whispers of a perception that is not true, or immediately relevant, and a conversation surrounding such notion.

 

I'm going to ask you kindly to stop bringing up the abandonment subject matter.

If you absolutely must discuss this, then please do it directly with Tom.

Link to comment

Wait... what's this "huge bug" in 4.7 that we desperately need a 4.7.1 for?  I've been very stable on 4.7 the last while, is there anything I need to be aware of in the future if I ever get enough $ for another 8 port SATA card and some more drives?

 

Tom has stated that there is a bug when rebuilding the array and accessing it where the parity data could be written wrong.

 

This bug was discovered in 5.0beta and was subsequently also found in 4.7

Link to comment

Wait... what's this "huge bug" in 4.7 that we desperately need a 4.7.1 for?  I've been very stable on 4.7 the last while, is there anything I need to be aware of in the future if I ever get enough $ for another 8 port SATA card and some more drives?

 

Tom has stated that there is a bug when rebuilding the array and accessing it where the parity data could be written wrong.

 

This bug was discovered in 5.0beta and was subsequently also found in 4.7

 

The possibility can be reduced by not writing to the array during a rebuild. It can be corrected with a correcting party check after rebuild.

Link to comment

The possibility can be reduced by not writing to the array during a rebuild. It can be corrected with a correcting party check after rebuild.

 

This scares me.

 

But back on topic, as a user of unRAID of a few years, I've enjoyed using it and it's served its purpose.  It was better than WHS, and power efficient, and fairly hands off.  At the time, it was an excellent product.  I don't blame Tom for lack of updates.  It's his gig, and while I'd like more updates, it's not like you can expect them when it's just one guy.

 

Regardless, once Windows Server 8 hits RTM, I will be migrating to Storage Spaces.  Having unRAID-like features inside a Windows Server OS is a no brainer.

Link to comment

Having unRAID-like features inside a Windows Server OS is a no brainer.

hehe... the part above is what would keep me from using it... but then I avoid Windows whenever possible.  I have to use it for work but when I get home the only time I touch windows is when I need to put a new DVD/Blu Ray purchase on my server.

Link to comment

The possibility can be reduced by not writing to the array during a rebuild. It can be corrected with a correcting party check after rebuild.

 

This scares me.

 

And it should,  just mounting a filesystem causes the superblock to be updated, which in itself writes to the array.

Link to comment

The possibility can be reduced by not writing to the array during a rebuild. It can be corrected with a correcting party check after rebuild.

 

This scares me.

 

And it should,  just mounting a filesystem causes the superblock to be updated, which in itself writes to the array.

Just a note that it's not a problem with UNRAID itself, it's been in every previous version of Linux in all the MD drivers and has existed for years.

Link to comment

The possibility can be reduced by not writing to the array during a rebuild. It can be corrected with a correcting party check after rebuild.

 

This scares me.

 

And it should,  just mounting a filesystem causes the superblock to be updated, which in itself writes to the array.

Just a note that it's not a problem with UNRAID itself, it's been in every previous version of Linux in all the MD drivers and has existed for years.

 

It occurs very rarely. There have been less than ten reports on this forum.

Link to comment

The possibility can be reduced by not writing to the array during a rebuild. It can be corrected with a correcting party check after rebuild.

 

This scares me.

 

And it should,  just mounting a filesystem causes the superblock to be updated, which in itself writes to the array.

The issue only shows itself if a block is written to at the same time parity/re-construction is calculated for that block.    It is rare, and that is why it remained un-fixed in many generations of Linux, as it rarely showed itself.  The bigger bug in 4.7 is where it will re-write an MBR on data disks if the superblock on the flash drive is re-constructed. (It does not always point to the correct start block for the partition.  I think it uses the system default, but that might be wrong on older disks.  They then slow up as "unformatted")
Link to comment

I stumbled across this bug myself under a v5 beta, but I didn't realize it was present in the 4.7 release.

 

Regardless, I just don't allow any machine to access the array when rebuilding data or parity...

 

let me add, If you've crashed, chances are you have dirty filesystems.

A fsck that fixes anything or a mount of the filesystem in read/write mode, writes to the filesystem's superblock.

unfortunately, there's no way to prevent any writes to the filesystem. We've been fortunate that this bug has not bitten us more. 

Link to comment

Just a note that it's not a problem with UNRAID itself, it's been in every previous version of Linux in all the MD drivers and has existed for years.

 

That is absolutley not true.  The bug IS in Tom's code, which was derrived from the linux MD drive long ago.  No recent linux MD drivers ahve had this issue.

Link to comment

Just a note that it's not a problem with UNRAID itself, it's been in every previous version of Linux in all the MD drivers and has existed for years.

 

That is absolutley not true.  The bug IS in Tom's code, which was derrived from the linux MD drive long ago.  No recent linux MD drivers ahve had this issue.

 

That's akin to installing an engine you purchased into a chassis you built yourself and then someone blaming the chassis because the engine has a bad gasket.

 

Version 4.7 was built using the faulty md drivers - they ALL were faulty at that time. Current betas use newer md drivers that the issue was resolved.

Link to comment

Just a note that it's not a problem with UNRAID itself, it's been in every previous version of Linux in all the MD drivers and has existed for years.

 

That is absolutley not true.  The bug IS in Tom's code, which was derrived from the linux MD drive long ago.  No recent linux MD drivers ahve had this issue.

 

That's akin to installing an engine you purchased into a chassis you built yourself and then someone blaming the chassis because the engine has a bad gasket.

 

Version 4.7 was built using the faulty md drivers - they ALL were faulty at that time. Current betas use newer md drivers that the issue was resolved.

 

Tom forked the MD5 md driver way back tomake the unRAId MD driver and does NOT track the changed that are put in to the standard Linux MD code.  If he did track it, the "explicit block device plugging" bug that he references in the Beta-12 release notes would not have been there as the standard linux MD code was updated when explicit block device plugging was added.

 

If you want to check this for your self, the code for unRAID driver is in /usr/src/linux/drivers/md on your unRAID server and teh standard linux md code can be seen here:  http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=tree;f=drivers/md

Link to comment
  • 2 weeks later...

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.