unRAID fork(), concepts, ideas and discussion, Running unRAID on another distro.

[grumpybutfun]

It does NO good to post about how much better you are than Tom ...

 

As far as product development, support, innovation, etc. as ONE person I am not... But a TEAM of us, we are. To argue that point is futile.

 

Would you have me lie?

 

Does anyone here not think that the mods haven't been sounding a fire alarm fire and he has probably received 10 - 20 emails to get on here?

 

Tom cannot take 5 minutes, respond to the Mods email, this thread, the 10 or so users who are in the support forum with red balled drives, missing drives, Kernel Panics, etc.

 

I have a job, GF, family, friends, responsibilities, commitments, hobbies, etc. and unlike Tom... I am not being PAID to SUPPORT a PRODUCT. I support 100s of users like the rest of you do on my own time.

 

How can you seriously say with a straight face and expect any of to believe that Tom supports his product when he is never on his own forum, doesn't police it, has 380 posts in 8 years in the support forum, has never written a plugin, never thanked anyone for writing a plugin, never thanked me or anyone else for the ESXi, Xen, KVM, Etc. guides, still hasn't added Xen / KVM guest support in the last several releases (even though it takes 2 seconds), still hasn't completed the new WebGUI for 5.0 release that took 2+ years in the make, etc?

 

Jump up and down and tell everyone and their lying eyes that Development Cycles for unRAID are fast?

 

We are still using software that is 4 years old?

 

Do you know how many holes / Security Bulletins have been release just for PHP alone? Most of the packages we are using have been end of life for YEARS.

 

Why are will still using NFS3?

 

Why are we still using Samba3?

 

Why are we still using a 32-Bit Kernel?

 

Why are we all experts in compiling Linux Kernels to add our TV, NIC, RAID Controllers?

 

Why are we all experts in Virtualization and spending $500 - $1000 on specific hardware and jumping through a milllion hoops just so we can install owncloud, Sickbeard, Couchpotato, etc. and have it work reliably?

 

although I suspect you'd find it's not as simple as you've implied to replicate all of the features.

 

I will let the 20 - 30 screenshots I am going to post shortly speak for themselves.

 

As for the threatened release being "torture tested" ... yet released next week ... I'd argue that those two comments are diametrically opposed.    You don't do thorough software testing in less than a week 

 

Regardless of what other features folks want, I think it's very safe to say that the #1 attribute folks want is for the underlying NAS to be "rock solid" reliable.    That has always been Tom's #1 focus -- to ensure that changes didn't impact that.  A one week release/test/call-it-torture-tested schedule doesn't give much confidence in that regard.

 

Are you talking about the fun we all had from 4.7 for 2+ years till 5.0 with no issues? Right....

 

I am several other users who will remain nameless have been running unRAID on openSUSE, Arch, CentOS, Debian for MONTHS and we have thrown EVERYTHING at the book at it.

 

Drive Failures, Power Failures, Countless Parity Checks, Adding Drives, Removing Drives, Copying TBs of Data back and forth, delete GBs and GBs of files and restoring with Resier fstack tools, running unRAID on the Host on one set of drives and unRAID in a VM on another set of drives simultaneously and transfer TBs of data for days on end, etc. It's been tested on AMDs, Intels, SAS / RAID / SATA Controllers, Linux Desktop, XMBC, Plex, TV Turners, Etc. all going, stopping, etc. Crashing VMs, resets, etc.

[AndroidCat]

grumpybutfun->

  **With your implementation would it be possible to use current unRAID array right away without any import/conversion?

If somebody already has working array of xxTB will it be recognized by your distribution? You understand that it would be a killer when we somehow have to transfer that much data from one place to another.

 

**Then ideally will I be able to swap ReiserFS for BTRFS (or other filesystem of choice) one disk at a time?

 

Really appreciate your work and speaking the truth about unRAID being behind.

[grumpybutfun]

Way to go Grumpy, quote half of what I said and ignore the rest.  I then deleted it before you even replied because I didn't want a pissing match.  Like I said stealing might not be the right word, I know Linux is free. If you are as good as you say you are why not start from scratch instead of using Limetech/Tom's code and call it something else?

 

Perhaps you should go look at the code yourself and see exactly how many lines of code that Tom wrote and then get back me. I think you will be surprised that 95% of it was written by two other people that have no association or being compensated by Tom. I bet they are completely unaware that unRAID even exists.

 

Yet... You want me to go write everything 100% from scratch? Please explain to me why that is okay for him but I am committing some sort of crime?

 

The very OS you want to use is CentOS for unRAID... It is a 100% clone of Red Hat (commercial product), why is that okay?

 

Why is an unRAID "fork" in a different distro, with a package manager, own emhttp, own Linux Kernel, and modifications to the md.c "stealing" but the others above are not?

 

Hypocritical if you ask me.

 

I wish Tom would post more but thats how it's been from day one.

 

Agreed. And a group of us and plenty of users here don't have to take it. We will take our forked version of unRAID and provide a much different experience if Tom does not want to step up.

[meep]

And for some light relief....

 

if/when we get access to Nirvana, it's going to need a name.

 

Some thought starters;

 

unRaid-ACE*

unPaid <--not a typo

forkOFF

nuRAID

 

(some of these may have been suggested already, please feel free to add more suggestions - or we can take it to a new thread if it's considered OT here)

 

*antichrist edition (my fav.)

 

[grumpybutfun]

grumpybutfun->

  **With your implementation would it be possible to use current unRAID array right away without any import/conversion?

 

Yes.

 

If somebody already has working array of xxTB will it be recognized by your distribution?

 

Your current unRAID will work in the Host or a VM and it would be unaware that anything has changed.

 

You understand that it would be a killer when we somehow have to transfer that much data from one place to another.

 

I am well aware. That would be something we would plan out / discuss as a community (much like Tom himself would do) before dropping that bomb.

 

Anyway... unRAID "AntiChrist Edition" has support for EXT2, EXT3, EXT4, Reiser, JFS, XFS (The default File System in Red Hat 7 and probably CentOS 7 too), GFS, BTRFS, MSDOS, vFAT, NTFS, HFS+, NFS3 and 4, Samba 3 and 4, AoE, iSCSI, VirtFS, etc.

 

**Then ideally will I be able to swap ReiserFS for BTRFS (or other filesystem of choice) one disk at a time?

 

That would be ideal. We are working on other things but eventually we will get to working on that.

 

I can tell you that I have 2 unRAID working on one box (one on the host and one VM) using different drives and have sent TBs and TBs of data between them both. Perhaps, that might be another solution for people... I dunno yet but eventually we all will figure out the best method.

 

Really appreciate your work and speaking the truth about unRAID being behind.

 

I guess I am suppose to drink the Kool-Aid and be iFollower (Apple Fan Boys) with unRAID and pretend what we experience isn't really happening. Maybe everyone else here owns stock in the LimeTech or something... It has to be the only explanation I can think of.

[grumpybutfun]

And for some light relief....

 

if/when we get access to Nirvana, it's going to need a name.

 

Some thought starters;

 

unRaid-ACE*

unPaid <--not a typo

forkOFF

nuRAID

 

(some of these may have been suggested already, please feel free to add more suggestions - or we can take it to a new thread if it's considered OT here)

 

*antichrist edition (my fav.)

 

Now THAT IS FUNNY!!!!!

[Kryspy]

I can concede that currently I have stopped using unRAID for many of the valid reasons raised.  Most reasoning lies behind the fact I cannot wait for plugins to be updated etc.  Currently I am using Ubuntu server 12.04 with SnapRAID  and aufs for pooling.

 

I receive daily e-mails about syncs, use webmin to administer and all of this is free.

 

I see the points being made and partially understand the whole GNU licensing but I think where the confusion lies is that unRAID is a package and the md driver you quoted is only part of the package. If you were to change a few colours, a font or two and rename it to something else and re-release it that I think would be considered theft?

 

Kryspy

[WeeboTech]

Does anyone here not think that the mods haven't been sounding a fire alarm fire and he has probably received 10 - 20 emails to get on here?

 

Tom cannot take 5 minutes, respond to the Mods email, this thread, the 10 or so users who are in the support forum with red balled drives, missing drives, Kernel Panics, etc.

 

 

I noted earlier Tom #1 has been ill all week. He usually logs onto the forum at least once a day.

He hasn't been around in a few days and Tom #2 confirmed that Tom #1 has been ill.

[WeeboTech]

Anyway... unRAID "AntiChrist Edition" has support for EXT2, EXT3, EXT4, Reiser, JFS, XFS (The default File System in Red Hat 7 and probably CentOS 7 too), GFS, BTRFS, MSDOS, vFAT, NTFS, HFS+, NFS3 and 4, Samba 3 and 4, AoE, iSCSI, VirtFS, etc.

 

I would really prefer you to choose a different descriptive 'edition' name.

 

Other then that, what modifications were done to the md.c driver to allow use of the other filesystems.

From what was said in the past, they do not work correctly. Tom said the md.c driver did not support directIO by design.

[jumperalex]

Anyway... unRAID "AntiChrist Edition" has support for EXT2, EXT3, EXT4, Reiser, JFS, XFS (The default File System in Red Hat 7 and probably CentOS 7 too), GFS, BTRFS, MSDOS, vFAT, NTFS, HFS+, NFS3 and 4, Samba 3 and 4, AoE, iSCSI, VirtFS, etc.

 

Grump, honest question of technical curiosity ... I seem to recall the reason Tom choose RFS way back when had something to do with the way it handles inodes.

 

Have you dealt with that a different way?  Do the newer filesystems have better support for whatever it was that made RFS the better choice?

 

Am I just clueless?  yeah yeah yeah I know

[ironicbadger]

Just checking in chaps. Had some personal stuff come up so won't be very available till next week.

 

Apologies if I ignore a PM or other stuff in the interim, that is not my intention.

 

This thread is breathless is it not?! Wowzer.

 

Sent from my Nexus 5 using Tapatalk

 

 

[ironicbadger]

Does anyone here not think that the mods haven't been sounding a fire alarm fire and he has probably received 10 - 20 emails to get on here?

 

Tom cannot take 5 minutes, respond to the Mods email, this thread, the 10 or so users who are in the support forum with red balled drives, missing drives, Kernel Panics, etc.

 

 

I noted earlier Tom #1 has been ill all week. He usually logs onto the forum at least once a day.

He hasn't been around in a few days and Tom #2 confirmed that Tom #1 has been ill.

 

This does at least explain a few things. A quick note would not have gone amiss as I have just done above perhaps?

 

Godspeed to a full recovery.

 

Sent from my Nexus 5 using Tapatalk

 

[lionelhutz]

I see the points being made and partially understand the whole GNU licensing but I think where the confusion lies is that unRAID is a package and the md driver you quoted is only part of the package. If you were to change a few colours, a font or two and rename it to something else and re-release it that I think would be considered theft?

 

Kryspy

 

It depends on what you are referring to.

 

Anyone can use the md driver as they see fit any way they want. It started with a GNU license which means the changes by Tom can't be copyrighted and also that he must release the source code. And the source code is in the distribution. Anyone can use it to create new storage OS with the same parity protected array capability as unRAID. At that point, it is no longer unRAID, it is something else.

 

The interface, emhttp, was created by Tom from scratch. He has every right to license and put conditions on that software. Being copyrighted software, no-one else has the right to re-package and distribute emhttp in any manner just because they feel like it.

 

 

[grumpybutfun]

It depends on what you are referring to.

 

Anyone can use the md driver as they see fit any way they want. It started with a GNU license which means the changes by Tom can't be copyrighted and also that he must release the source code. And the source code is in the distribution. Anyone can use it to create new storage OS with the same parity protected array capability as unRAID. At that point, it is no longer unRAID, it is something else.

 

The interface, emhttp, was created by Tom from scratch. He has every right to license and put conditions on that software. Being copyrighted software, no-one else has the right to re-package and distribute emhttp in any manner just because they feel like it.

 

I find it remarkable that many of you believe that Red Hat, VMWare (ESXi), Proxmox, Oracle, FreeNAS, NAS4Free, Nappit, EMC, Openfiler, Media Server Vault, etc. etc. etc. etc. all have to release their source code (even for their management tools / webguis / etc.) but somehow / someway Tom has better Lawyers and Billions of Dollars that those companies don't have to "get around" the GNU license.

 

It FLAT OUT STATES this in the GNU License...

 

You cannot incorporate GPL-covered software in a proprietary system. The goal of the GPL is to grant everyone the freedom to copy, redistribute, understand, and modify a program. If you could incorporate GPL-covered software into a non-free system, it would have the effect of making the GPL-covered software non-free too.

 

A system incorporating a GPL-covered program is an extended version of that program. The GPL says that any extended version of the program must be released under the GPL if it is released at all. This is for two reasons: to make sure that users who get the software get the freedom they should have, and to encourage people to give back improvements that they make.

 

Linux (the kernel in the GNU/Linux operating system) is distributed under GNU GPL version 2. Does distributing a nonfree driver meant to link with Linux violate the GPL?

 

Yes, this is a violation, because effectively this makes a larger combined work. The fact that the user is expected to put the pieces together does not really change anything.

 

Each contributor to Linux who holds copyright on a substantial part of the code can enforce the GPL and we encourage each of them to take action against those distributing nonfree Linux-drivers.

 

Linking ABC statically or dynamically with other modules is making a combined work based on ABC. Thus, the terms and conditions of the GNU General Public License cover the whole combination.

 

An “aggregate” consists of a number of separate programs, distributed together on the same CD-ROM or other media. The GPL permits you to create and distribute an aggregate, even when the licenses of the other software are non-free or GPL-incompatible. The only condition is that you cannot release the aggregate under a license that prohibits users from exercising rights that each program's individual license would grant them.

[cassiusdrow]

Anyway... unRAID "AntiChrist Edition" has support for EXT2, EXT3, EXT4, Reiser, JFS, XFS (The default File System in Red Hat 7 and probably CentOS 7 too), GFS, BTRFS, MSDOS, vFAT, NTFS, HFS+, NFS3 and 4, Samba 3 and 4, AoE, iSCSI, VirtFS, etc.

 

Grump, honest question of technical curiosity ... I seem to recall the reason Tom choose RFS way back when had something to do with the way it handles inodes.

 

Have you dealt with that a different way?  Do the newer filesystems have better support for whatever it was that made RFS the better choice?

 

Am I just clueless?  yeah yeah yeah I know

 

 

Post from Tom about reasons for choosing RFS: http://lime-technology.com/forum/index.php?topic=15099.msg142148#msg142148

[grumpybutfun]

What I find the most remarkable about this whole thing is that you people busting my balls or thinking I am some sort of evil genuis simply do not understand what is involved with implementing everything I said in this thread into "unRAID AntiChrist Edition".

 

1. Install CentOS

 

Download Linux Kernel

Copy .config file

Apply Kernel Patch for unRAID

Compile Kernel and Modules

 

Note: You wouldn't even need to do the the above if I provided the compiled kernel package for you.

 

2. Copy emhttp

 

3. Install I think 7 packages from the CentOS core repo.

 

4. Fix 2 symboloic links.

 

Yahtzee!

 

You are now running unRAID 64-Bit on CentOS using 100% of the very same unRAID Kernel module and emhttp since 2+ years ago.

 

I should just post a damn guide for doing the above so everyone will get off my ass and see how much "development" (cough, cough) Tom or I really did.

 

I bet the same a-holes who have been busting my balls will have a 64 Bit Version of unRAID installed on their system (with all the crap we have all talked about in this thread included) in 30 minutes or less and maybe then they would realize that the joke is on them.

 

Go look at the guide for adding your TV Tuner card in unRAID. It's 1,000 times more difficult and what 30 more steps?

 

[Kryspy]

lionelhutz,

 

I meant it as you stated.

 

Kryspy

[lionelhutz]

Geeze, I'm not going to pretend to be a lawyer. Tom calls it copyrighted and I'm have no intention of take any actions to dispute that, so I accept that's how it is. If someone else wants the source code for emhttp released because they believe it should be available then they can go chase LimeTech via email or laywers and get the source code released.

 

[BRiT]

As someone who admins multiple other discussion forums, this entire drama series has been quite the entertainment and diversion. It also reaffirms my decisions and beliefs in how highly effective discussion forums should and should not be managed. I thank each and every participant.

[WeeboTech]

I suppose one has to look at all the licenses of the shared libraries emhttp is dynamically linking to.

From what I read in another location,  if the shared library is dynamically linked, then depending on the libraries license, the proprietary application may not have the same license imposed.

 

If any library imposes, GPL and not LGPL, the source is supposed to be available.

A reasonable charge can be imposed and the method of how the source is presented is up to the author.

[grumpybutfun]

Geeze, I'm not going to pretend to be a lawyer. Tom calls it copyrighted and I'm have no intention of take any actions to dispute that, so I accept that's how it is. If someone else wants the source code for emhttp released because they believe it should be available then they can go chase LimeTech via email or laywers and get the source code released.

 

If only we had Millions of Dollars and the Lawyers Tom has...

[jsutter55]

Anyway... unRAID "AntiChrist Edition" has support for EXT2, EXT3, EXT4, Reiser, JFS, XFS (The default File System in Red Hat 7 and probably CentOS 7 too), GFS, BTRFS, MSDOS, vFAT, NTFS, HFS+, NFS3 and 4, Samba 3 and 4, AoE, iSCSI, VirtFS, etc.

 

I would really prefer you to choose a different descriptive 'edition' name.

 

Other then that, what modifications were done to the md.c driver to allow use of the other filesystems.

From what was said in the past, they do not work correctly. Tom said the md.c driver did not support directIO by design.

 

unRAID KSE? (KitchenSinkEdition) 

[garycase]

Anyway... unRAID "AntiChrist Edition" has support for EXT2, EXT3, EXT4, Reiser, JFS, XFS (The default File System in Red Hat 7 and probably CentOS 7 too), GFS, BTRFS, MSDOS, vFAT, NTFS, HFS+, NFS3 and 4, Samba 3 and 4, AoE, iSCSI, VirtFS, etc.

 

I would really prefer you to choose a different descriptive 'edition' name.

 

Other then that, what modifications were done to the md.c driver to allow use of the other filesystems.

From what was said in the past, they do not work correctly. Tom said the md.c driver did not support directIO by design.

 

unRAID KSE? (KitchenSinkEdition) 

 

That would be a major improvement.  I agree with WeeboTech that the name that's been used is inappropriate and very offensive.

 

[Kryspy]

Only to those who believe in christ.

 

Kryspy

[grumpybutfun]

As someone who admins multiple other discussion forums, this entire drama series has been quite the entertainment and diversion. It also reaffirms my decisions and beliefs in how highly effective discussion forums should and should not be managed. I thank each and every participant.

 

Lol! I was on conference calls all day.

 

Once this Ice Age is over... I will be able to head out to the client site and the party is over for me. I would rather be water-boarded than sleep in one those airport chairs all night.

 

There are some really knowledgeable people on here and they have great points of views and good debaters. I actually learned quite a bit today.