The State of MineOS

Showing 21-79 of 79 messages
The State of MineOS Will 2/23/12 2:07 PM
MineOS CRUX is--I feel--nearly as final as it will get.  While there are countless ideas I have come up with that have gone undeveloped and would still be of great use, it's at the point where I believe the overall usability/educational experience has balanced out exactly where I want it.
  • Ideas such as the auto-refreshing AJAX-driven web-ui I am technically competent in pulling off, but aesthetically incompetent for.
  • The log parser/MySQL plus visual graph would be nice,  but overall is more fun than function. 
  • My share-your-world-on-the-cloud / s3 backup idea I still think would be amazing, but  would probably be unpopular
  • My MySQL filesystem (database held chunks, rather than file-based) would be amazing, but may be impractical to spend time developing if API modding changes code drastically in the next few months.
  • A couple of other great ideas, etc.
That said, while I'm at no shortage of ideas and ways to dump hundreds man-hours programming, the returns might not be as great as I'd expect and 'if it isnt broken, dont break it'.

So where does this leave me now?

Right now, with no new programming challenges (or at least, none being developed...I spend a lot of time theorizing, but not prototyping), I'd probably be happy with simply providing support for issues. However, the direction of the questions lead me to believe that MineOS it self does not need much more updating, but instead there could be use to a more general-education of Linux. I have a few ideas that I'd like to feel out to see if it interests any of you:
  1. Creating a collaborative fork
    Right now, MineOS CRUX is pretty much exclusively handled by me. Some code contributions come in here and there and dozens of feature requests, but I stand as the single decider and implementer of new code and functionality.  This sort of centralized management means updates come at my timeline, and in this case, last ISO update in October.

    Would any be interested in creating a non-crux based MineOS?  That is, I'm asking if anybody would be interested in learning the steps of transforming an OS into a MineOS while also being themselves decision makers.  Things like mineos.py are transferable, but getting a distributable ISO, automating installation and deployment tasks, as well as documenting are things that are both immensely educational (you bet it was for me!) as well as intensely rewarding. Realistically, one could consider making a MineOS to the likes of:
    * MineOS FreeBSD (oh, unix!)
    * Turnkey MineOS (http://www.turnkeylinux.org/)
    * MineOS EC2
    * Virtualbox Appliances
  2. Creating an unrelated Desktop OS
    While having a graphical GUI on a server has been largely requested, it has never been interesting to me to make a MineOS + GUI version.  On the other hand, if we wanted to create a Desktop OS, learn the process, maintain and distribute, this would be a huge learning opportunity for us all.  That said, I see no reason why existing materials couldn't be made to work with MineOS script, but the primary focus here is MineOS-independent: creating a usable, pretty, and customized desktop distro.
Keep in mind, the primary focus of these options is the learning. Neither of these will be options that give some perfect, immediate product. It will, however, give you a chance to learn the process that I have gone through--actively, and with my guidance and accumulated knowledge--to create something you may find useful. For me, it will give me the chance to finally coordinate a team-oriented project rather than a solo-work, employing things such as git, buildboxes, and--most significantly--it is an opportunity for me to see how a project grows organically.  I do not expect people to make unrealistic 'commitments' to the project, dedicating time you simply don't have, but if enough people are interested, hopefully the distribution will be thinned quite nicely and progress can be made even if often untouched. (think of CRUX the distro... they release a minor version (.1) every 3-4 months, and a major version (1.0) every 2-8 years).

So what do you all think?

Will


Re: The State of MineOS Matthew Smollinger 3/2/12 1:48 PM
Hey Will,
  Wanted to throw in my $0.02 here. I'd love to work on a fork of MineOS, however my life at the moment is busy with a house purchase. In a month or two, I'll be much more available once I'm moved in and settled. However I had a couple of ideas for areas I'd like to help improve:

- I started my programming life as a web dev (PHP) and have always wanted to look in python. I'd like to see the GUI revamped to handle things like logging, server reboots, and effectively make it so the average user can do what they need.

- My biggest issue with CRUX is the lack of documentation and a somewhat active but small community. For a fork, I'd either like to move to something like Gentoo, which has a great deal of documentation and support, or sure FreeBSD would be good too as that's equally active and documented (however my knowledge of linux vastly outweighs my BSD knowledge). Moving to a more maintained distro would also mean if people wanted to install 3rd party packages, there's less support work for us as we can point them to the official docs. 

- Virtualization is the way of the future, IMHO, so yeah I'd love to work on Appliancizing (new word!) the install. 

- Easier updating of external packages, possibly thru an automated git process, similar to homebrew on OS X. 

All these things I see as attainable with my knowledge-base, so let me know your thoughts, and where you'd see a fork going. 

--Matt--
Re: The State of MineOS Andrew Kallmeyer 3/2/12 2:24 PM
I have already done small modifications to the python scripts that run the webui, if there were a github project or something I would have forked it already. It could also be fun to implement a feature when I find time for it. Is the source already available in full anywhere?


On Thursday, February 23, 2012 5:07:33 PM UTC-5, Will wrote:
Re: The State of MineOS Zone 3/3/12 4:20 PM
I would love to to see these come the to the light of day!
Re: The State of MineOS Zone 3/4/12 5:32 AM
On a second note, I would like to see a option number one. Although I still don't have the technical skills to pull it off personally.
Re: The State of MineOS Will 3/4/12 1:10 PM

On Friday, March 2, 2012 3:24:41 PM UTC-7, Andrew Kallmeyer wrote:
I have already done small modifications to the python scripts that run the webui, if there were a github project or something I would have forked it already. It could also be fun to implement a feature when I find time for it. Is the source already available in full anywhere?


Which sources in particular are you looking for?  Currently, all the python scripts are fully open and readable, both on the ISO as well as the installed setup; for the most part, there are zero things in MineOS that arent fully transparent to the user.  All the minecraft specific files can be found online as well, at http://minecraft.codeemo.com/crux/rsync/ 
Re: The State of MineOS blha303 3/4/12 2:39 PM
I made a batch file that uses wget to pull all the files from
http://minecraft.codeemo.com/crux (minus the ISO files), put them in a
directory, then add them to a GitHub repo. Hopefully at some point I
can get instructions up there on how to configure MineOS on another
distribution.

https://github.com/blha303/crux
Re: The State of MineOS SBHouse 3/5/12 2:20 PM
I would love to see the MineOS scripts as an installable package for
Ubuntu.  Nope, I'm not a die hard Ubuntu nut - though I do like it.
Let me explain...

A year ago, I did not have the knowledge or experience to feel
confident working with a command line only Linux.  But thanks to my
kids getting into Minecraft, my desire to play WITH them on a server,
and Will's MineOS, I now have a fair amount of experience.  So
recently, I began building a home fileserver using the Turnkey Linux
ISO for their fileserver appliance.  But I added some stuff (like a
normal FTP so I can automate off-box backups of Minecraft server
archives from my Minecraft server) beyond that.  I've updated
packages.  I've updated Webmin.  I really like Webmin - even better
now since I understand what it's doing under the hood a lot of the
time.

I would love to see Minecraft be able to run on a TurnkeyLinux.org
(minimal overhead Ubuntu server distribution) with a Webmin module for
some additional management beyond what the MineOS scripts might do.
Why Ubuntu?  Because if I want to add any other functions to the
server, it's really easy to do that because there are so many packages
out there for Ubuntu or Debian.

Alas, I do not have programming knowledge or skills to help make a new
Linux distro around this idea.  But I'm putting my wish out there in
case anyone (Will??) would choose to undertake the challenge.

Until the, I'm still ecstatically happy with MineOS CRUX, especially
since it installs 64-bit kernel and 64-bit Java out of the box so I
can assign more than the 32-bit Java RAM limitation.  Unless something
else evolves from this project (or Mojang does something conflicting),
I don't see my platform for my Minecraft server changing anytime soon.

Thanks all!

- SBHouse
unk...@googlegroups.com 5/22/12 5:06 PM <This message has been deleted.>
Re: The State of MineOS Markus Geuken 8/7/12 10:44 PM
I would also like to get a package working for unbuntu actually. Since ive spent a bit of time there. and it has great hardware support. partly becuase i also want to learn it so i can manage our home computers more effciently. i know unbuntu might have high resource reqs, but even that would be interesting to see how much you can safely turn off. and how to manage it without the gui, while still having an option of loading up the gui if you feel you need to for some reason. 

I have a quad core with 8 gig of ram thats just waiting for me to install mineos on it. but the ISO would not install for some reason, and i havent really looked into setting up XEN or something (which would be lovely to learn as well. setting up a super lightweight Virtual environment. (although im sure there are plenty of guides for that stuff already). 

but yes the API might change things. 1.3.1 of minecraft has gone over to some kind of multicore. also ive started running into mods that work fine on windows but server wont start in mineoscrux at all. or the jar wont. perhaps due to the installed copy of java? 
unk...@googlegroups.com 8/8/12 12:16 PM <This message has been deleted.>
Re: The State of MineOS Will 8/8/12 12:28 PM
On Tuesday, August 7, 2012 10:44:44 PM UTC-7, Olga wrote:
I would also like to get a package working for unbuntu actually. Since ive spent a bit of time there. and it has great hardware support. partly becuase i also want to learn it so i can manage our home computers more effciently. i know unbuntu might have high resource reqs, but even that would be interesting to see how much you can safely turn off. and how to manage it without the gui, while still having an option of loading up the gui if you feel you need to for some reason. 

I have a quad core with 8 gig of ram thats just waiting for me to install mineos on it. but the ISO would not install for some reason, and i havent really looked into setting up XEN or something (which would be lovely to learn as well. setting up a super lightweight Virtual environment. (although im sure there are plenty of guides for that stuff already). 

but yes the API might change things. 1.3.1 of minecraft has gone over to some kind of multicore. also ive started running into mods that work fine on windows but server wont start in mineoscrux at all. or the jar wont. perhaps due to the installed copy of java? 

Ubuntu isn't really high resource requirements--it's actually quite low--it just feels that way since I aimed for such *low* resource requirements for MineOS.  As this forum has shown, there is a cost for that, which often manifests as MineOS not containing all the drivers for all the hardware out there, or making it easy to do things like...use flash drives.

Especially when considering that all the services are running (hiawatha, etc) when I count under ~30 mb, it could be that an equivalent Ubuntu installation with all services running could be...80mb.  This isn't exactly a footprint to fear, either...its relatively common for servers to have footprints this large and in modern days, 100mb is becoming a smaller and smaller slice of the pie, comparatively.

Ubuntu is a great server distro, and I recommend it to anybody with the know-how to turn it into the best minecraft server they can--MineOS exists, of course, because not everybody has that know-how...and CRUX to me is a more educational tool in learning the innards of linux rather than simply "apt-get some-precompiled-binary".

The problem with getting MineOS stuff on ubuntu is that theres a lot of interconnected software...iptables, hiawatha, crontabs.  In order to install MineOS onto ubuntu (like MineOS-ubuntu did, a community created project on github https://github.com/nfearnley/MineOS-Ubuntu), you have to make a lot of assumptions about what software the user is going to WANT to use (since he has the choice of much).

In other words, I can't create a package that just starts changing user's iptables and set up hiawatha...thats not what packages do--theyre self-contained units that don't touch other things....they depend on other things, but they don't modify.  Think of how sucky it'd be if installing "apt-get MineOS" suddenly changed your apache webserver config to work with minecraft mapping...and how that would be annoying if you already had something on your webserver.

I think the MineOS-ubuntu project required it was a completely brand new server, so this wasn't the case that you can ruin a good config...but that--too--is outside the scope of what one package should do, themeatically.  So to make a MineOS-ubuntu that was truly maintainable and safe would require a lot of scripting to assure proper installation without treading on other configs.  Unfortunately, it's also why I've never gone this route.   By offchance, such a thing is IDEAL for self-contained distros such as Turnkey (ubuntu)...but I have yet to see any real interest in that manifest either. If anything, I'm more inclined to go to 'LESS-EASY' distros (harder than crux, that is) because of the notion that 'the harder it is, the more you learn, the more competent you become'...and freebsd calls to me.  But the direction of MineOS still is undecided, awaiting that community push for where it should go (since I have mostly ruled out simply re-re-re-doing a new iso with newer scripts).
Re: The State of MineOS roguethu...@gmail.com 8/12/12 5:03 PM
My only warning on FreeBSD. It is *very* picky. (And in preticular vmware tools have to be partially recompiled for it atm) That said... I *DID* totally just set up freebsd 9 for minecraft hosting uses with LXDE because I'm lasy (and it totally doesnt use much ram)
Re: The State of MineOS Will 9/14/12 2:37 PM
Enough time has passed that my frustration with javascript has subsided.  Today I worked on building a 0.5.0 web-ui which seems to be working quite nicely, utilizing jquery and json to its full potential.

Admittedly, the UI isn't getting all that much prettier, but operation should be greatly improved, e.g., hitting buttons grays them out and refreshes itself, etc.  Unsure about a timeframe but it's actually quite exciting to me again so I might hack away at it quite speedily, who knows
Re: The State of MineOS Dragoon209 9/17/12 8:45 AM
Thanks for all your hard work on this.  I am tweaking my server daily, and none of it would be as fast or easy if I wasn't using MineOS Crux.  Thanks!!
Re: The State of MineOS Will 9/17/12 4:19 PM
Alrighty.  The web-ui has been making progress.  At this time, I have most of the underlying javascript done and done.   That is, almost every single function that 0.5.0 offers can be successfully executed via a JSON request.  All that remains now is making the pages usable in a way that is both pretty and functional.

I apologize, this new web-ui again lacks a aesthetic element.  Here's hoping I have everything I need to get the job done, though.


I'll end up making a video describing how the new profiles system works, and how update/propogate works, but in the meantime, the general idea is that you can template out the files used for each server, ensuring that you truly can clone servers or shard them  (for testing).  The buttons de-activate upon push and re-activate after confirmation of the action (success/failure).  The web-ui will now allow users to switch between the stable and beta scripts -- 'stable' and 'beta' being much scarier words for 'the version people have not found bugs in' and 'the version to test my fixes'

profiles will do more than just tekkit/canary/etc.  It'll also do mapping software, chunk fixing software...all of it will be updated/managed via what is called profiles, which should make it much more straightforward to add minecraft related software (linux related software still will be using root and CLI).

this isn't yet usable for testing (nor is it available online) because most of the pages that people need most are not yet implemented (such as the create page form). I can create a server with the UI, for example, but I can't change how much ram it uses from the web-ui yet). Once the 'create server' functionality is included, I'll be asking for help to make sure this release actually improves your guys lives...and I really think it does work nicely.

I think by the end of the workweek I'll have something that is really ready to be used full-time (by those who still have the comfort and confidence to use CLI). Hopefully by the end of the month I'll have one worthy of being Youtubed and distributed as a 0.5.0 release.

Re: The State of MineOS Alex 9/18/12 8:43 AM
Hurray! I knew you could do it!
Re: The State of MineOS Dragoon209 9/18/12 8:45 AM
Awesome!

I have started using the console for most things, but as long as its functional, I don't think it matters as much if it looks nice. As it is, it's organized, and self-explainitory, and you should be proud. I'm excited to start using it!
Re: The State of MineOS Nick Lewis 9/18/12 2:57 PM
Will, I'm thrilled to hear how well it's coming along. I'll do my best to help with testing once you're ready. Thanks for such an awesome tool!
Re: The State of MineOS Martin Ramshøj 9/19/12 11:37 PM
Hi.
Great to see you are making progress on this. Looks good. I'm looking forward to do a clean install on my server with this to get it up and running again.

Thanks for all the hard work you do :)
- Martin 
Re: The State of MineOS Will 9/21/12 9:36 PM
After numerous revisions and refactorings and more,the web ui is in terrific shape. It may still be a while before it matches the functionality of the current web ui, but soon and when it's complete I foresee its functionality and usability far exceeding the current one.

At this point, it's fully styled and you can create, start, stop, restore, backup, archive, prune...still remaining is importing and modifying existing configs. Memory is now reported in the Ui, all known server mods are supported and you can download and update and add/remove jars from the ui.

Foreign and templates are things of the past, too and there's a sweet new system for extending custom functionality. You can also make macros which can simultaneously do console commands and in game commands, and they can be scheduled as well.

Lastly, the latest version will always appear on the web Ui so you know when an update is needed, and you can toggle between stable and beta script versions at will. All in all, it's pretty sweet.  

Re: The State of MineOS pareidol...@gmail.com 9/22/12 5:44 PM
I've been using MineOS for over a year, and I wanted to say that it has been a very nice experience so far!
Also, I am interested in helping to test new versions functionality, especially tekkit support.

Currently I am running 0.4.11e, but I do not have 'beta channel' scripts available.  (Those are in 0.4.12, right? I haven't yet figured out where that is being distributed..)

Is there a version of MineOS with tekkit support built in that I could be helping to test, or should I wait for the 0.5.0 release?
Re: The State of MineOS Will 9/22/12 7:28 PM
You should wait, or look into those "modified scripts" floating around in these forums.  From what I've read, there was success with tekkit for the most recent versions.  0.5.0 will be extra nice, but don't feel tethered until I release it (even though it will hopefully still be soon!).  That said, 0.4.12 isn't around anymore I don't think (it might be online, but I don't work on that series of scripts anymore, so I can't bet that I'll be doing much work on em anymore ever).

Seriously though, 0.5.0 will be great! 
Re: The State of MineOS marsla...@gmail.com 9/26/12 4:52 AM
Nice to see that you are making process!


On Tuesday, September 18, 2012 1:19:29 AM UTC+2, Will wrote:
Re: The State of MineOS Will 10/19/12 7:48 AM
The panel is done!
Re: The State of MineOS Zone 10/19/12 10:49 AM
Woot!
Re: The State of MineOS marsla...@gmail.com 10/21/12 4:29 AM
Awesome! when do you plan on releasing it to the public?
Re: The State of MineOS Will 10/21/12 9:59 AM
It is released to the public but I'm wary to force it upon people whose servers are already set up--check the thread 0.5.0 scripts--at the top of the list. It has the 3 instructions required to upgrade.  Because a "profile" has to be set up under the new system, it might be an inconvenience I anybody not starting fresh.

Soon I'll be writing something that should ease that process. Thanks for the headsup

Re: The State of MineOS marsla...@gmail.com 10/22/12 12:25 PM
okay awesome

and the iso is updated also?
Re: The State of MineOS az...@jcaks.net 10/24/12 1:04 AM
I'd love to see you put your files into a github repo, and then allow others to commit code to the project, to get things rolling.

I say keep with Crux, i only just started playing with it but it is quite awesome.
Re: The State of MineOS Will 10/24/12 8:52 AM
On Wednesday, October 24, 2012 1:04:14 AM UTC-7, az...@jcaks.net wrote:
I'd love to see you put your files into a github repo, and then allow others to commit code to the project, to get things rolling.

I say keep with Crux, i only just started playing with it but it is quite awesome.

I've fielded this request a number of times before, but I've declined to actually follow through in opening a github because of the low follow-through rate of the would-be committers.  For the most part, I barely have received any code contribution or improvement from users even without the barrier-to-entry that github provides.  

I most certainly do recognize the value of versioning and "pull" availability for casual updaters, but there does not appear to be many people who deploy mineos who are interested in the area of developing it. That said, I'd probably consider it even if I had one person send me a modified .py file and only he wanted it, but without even that single contributor, github just adds complexity to my update process.

On another note, its up in the air how the next few months will go regarding crux.  On one hand, python & js/webui development is the most exciting thing in the world to me right now (and a few months ago, js made me want to slit my throat), and going through the kernel rebuild and bash scripting process is much less desirable now.  Actually, I'm having trouble getting the new 3.6.3 kernel onto the distribute-able iso (because my build box VM was lost) and its getting quite rough. That said, I love crux as well and I just hope that these scripts and web-ui can even thrive when not married to the OS.
Re: The State of MineOS Seth 10/27/12 2:55 PM
I'm relatively new to MineOS but experienced with Linus. I like the current build.

Unfortunately I wrote my add-ons in PHP and bash. They allow you to manage bans in the web-ui (using the MineOS "api") as well as to automatically install IP bans in Minecraft into iptables for greater protection (*all* your server instances are protected from all attacks). If you pardon an IP ban it takes it out of iptables.
Re: The State of MineOS Will 12/11/12 5:58 AM
MineOS seems to doing quite nicely with the latest web-ui, and I'm very pleased with my work in both the javascript and the python backend.  While MineOS will continue to get updated when required--such as when bugs/erratic behavior is found--most of my dev efforts are currently going toward the interest-of-the-month, which is putting MineOS scripts on other platforms.

Already, Ubuntu Server 12LTS can be fully converted to a MineOS-enabled server through the comprehensive instructions on the wiki: http://minecraft.codeemo.com/mineoswiki/index.php/Ubuntu

In addition, MineOS is now available on Turnkey (see this post: https://groups.google.com/forum/?fromgroups=#!topic/mineos/ZUTuE9Gkxvs).  While Turnkey is currently the least recommended option (32-bit and highest mem footprint), it has the advantage of actually being probably the most secure distro offered, as well as having the largest package repositories.  Soon (tm) when Turnkey gets released in x64, Turnkey is likely to be the only actively-developed operating system for MineOS, as the customization process is second-to-none.  To illustrate its ease, when Turnkey x64 releases, I can release Turnkey MineOS x64 in less than 3 minutes using the architecture-independent package TKL facilitates making.
Re: The State of MineOS SBHouse 12/15/12 3:58 PM
Hi Will!  Looking forward to the future "3 minute" package for MineOS on TKL 64 someday!  I've used both TKL and Ubuntu Server.  I've used TKL back when it was Ubuntu based and recently now that it's Debian based.  And while both can be secure, I agree that the TKL way is to secure things down first.  And if all you do is just use the TLK LAMP appliance, then you have EVERYTHING you need to do MineOS and Bukkit plugins that need a MySQL server.  Plus you get nice web administration and you can add anything that has  Debian package.  Very nice!!

- SBHouse
Re: The State of MineOS Will 12/18/12 2:04 PM


On Saturday, December 15, 2012 4:58:07 PM UTC-7, SBHouse wrote:
Hi Will!  Looking forward to the future "3 minute" package for MineOS on TKL 64 someday!  I've used both TKL and Ubuntu Server.  I've used TKL back when it was Ubuntu based and recently now that it's Debian based.  And while both can be secure, I agree that the TKL way is to secure things down first.  And if all you do is just use the TLK LAMP appliance, then you have EVERYTHING you need to do MineOS and Bukkit plugins that need a MySQL server.  Plus you get nice web administration and you can add anything that has  Debian package.  Very nice!!

- SBHouse

In the spirit of MineOS, I actually tried to mimic the packages to CRUX as much as possible.  Therefore, no LAMP, but instead LHMP (not a real thing lol).  Because as always, low-footprint, easy-config Hiawatha returns! 

And with TKLs amazing "patch" system, it is as hardened and well-made an appliance as TKL itself could release, because all the package config and lockdown is all on their end too-- its such a nice system!
Re: The State of MineOS Clayton Fields 12/27/12 9:07 PM
Personally, I'd like to see the whole mineOS setup as a Arch Linux package in the AUR.  A single step install could be done with pacman/packer.  https://wiki.archlinux.org/index.php/Creating_Packages, just for reference if you're not an an Archer... 

On Thursday, February 23, 2012 4:07:33 PM UTC-6, Will wrote:
MineOS CRUX is--I feel--nearly as final as it will get.  While there are countless ideas I have come up with that have gone undeveloped and would still be of great use, it's at the point where I believe the overall usability/educational experience has balanced out exactly where I want it.
  • Ideas such as the auto-refreshing AJAX-driven web-ui I am technically competent in pulling off, but aesthetically incompetent for.
  • The log parser/MySQL plus visual graph would be nice,  but overall is more fun than function. 
  • My share-your-world-on-the-cloud / s3 backup idea I still think would be amazing, but  would probably be unpopular
  • My MySQL filesystem (database held chunks, rather than file-based) would be amazing, but may be impractical to spend time developing if API modding changes code drastically in the next few months.
  • A couple of other great ideas, etc.
That said, while I'm at no shortage of ideas and ways to dump hundreds man-hours programming, the returns might not be as great as I'd expect and 'if it isnt broken, dont break it'.

So where does this leave me now?

Right now, with no new programming challenges (or at least, none being developed...I spend a lot of time theorizing, but not prototyping), I'd probably be happy with simply providing support for issues. However, the direction of the questions lead me to believe that MineOS it self does not need much more updating, but instead there could be use to a more general-education of Linux. I have a few ideas that I'd like to feel out to see if it interests any of you:
  1. Creating a collaborative fork
    Right now, MineOS CRUX is pretty much exclusively handled by me. Some code contributions come in here and there and dozens of feature requests, but I stand as the single decider and implementer of new code and functionality.  This sort of centralized management means updates come at my timeline, and in this case, last ISO update in October.

    Would any be interested in creating a non-crux based MineOS?  That is, I'm asking if anybody would be interested in learning the steps of transforming an OS into a MineOS while also being themselves decision makers.  Things like mineos.py are transferable, but getting a distributable ISO, automating installation and deployment tasks, as well as documenting are things that are both immensely educational (you bet it was for me!) as well as intensely rewarding. Realistically, one could consider making a MineOS to the likes of:
    * MineOS FreeBSD (oh, unix!)
    * Turnkey MineOS (http://www.turnkeylinux.org/)
    * MineOS EC2
    * Virtualbox Appliances
  2. Creating an unrelated Desktop OS
    While having a graphical GUI on a server has been largely requested, it has never been interesting to me to make a MineOS + GUI version.  On the other hand, if we wanted to create a Desktop OS, learn the process, maintain and distribute, this would be a huge learning opportunity for us all.  That said, I see no reason why existing materials couldn't be made to work with MineOS script, but the primary focus here is MineOS-independent: creating a usable, pretty, and customized desktop distro.
Keep in mind, the primary focus of these options is the learning. Neither of these will be options that give some perfect, immediate product. It will, however, give you a chance to learn the process that I have gone through--actively, and with my guidance and accumulated knowledge--to create something you may find useful. For me, it will give me the chance to finally coordinate a team-oriented project rather than a solo-work, employing things such as git, buildboxes, and--most significantly--it is an opportunity for me to see how a project grows organically.  I do not expect people to make unrealistic 'commitments' to the project, dedicating time you simply don't have, but if enough people are interested, hopefully the distribution will be thinned quite nicely and progress can be made even if often untouched. (think of CRUX the distro... they release a minor version (.1) every 3-4 months, and a major version (1.0) every 2-8 years).

So what do you all think?

Will


Re: The State of MineOS Will 1/31/13 12:58 PM

On Thursday, December 27, 2012 10:07:45 PM UTC-7, Clayton Fields wrote:
Personally, I'd like to see the whole mineOS setup as a Arch Linux package in the AUR.  A single step install could be done with pacman/packer.  https://wiki.archlinux.org/index.php/Creating_Packages, just for reference if you're not an an Archer... 

 
MineOS doesn't really fit as a package...mostly because MineOS aims toward being more of an operating system (well, strictly speaking, a distribution...and not even close to an OS), but you see the point.  Put another way, MineOS is convenient because it does more than what a package does/strives to do: which is install one thing, its dependencies, and configure it.  Creating such a package as a "turnkey" solution for arch wouldn't work so well because it makes too many assumptions.

1) mineos scripts operate as a non-privileged user.  packages don't typically create usable-users; typically, the most a package will do is create non-interactive users with no login capability
2) mineos depends on several other packages to work in a way not out-of-the-box.  For example, it uses hiawatha on port 80.  So a few things can happen here
  Arch can see 'hiawatha is installed--skip this dependency and just apply the mineos' hiawatha.conf.... problem: if its already installed, it shouldn't blindly overwrite configs.
  Arch can see 'hiawatha is not installed--install it and apply the mineos' hiawatha.conf'--problem: it has no knowledge that it'll try to use port 80/443, which could already be running a lighttpd, nginx, apache server, etc.
  Arch can see 'hiawatha is installed--preserving existing configuration'--problem: this mineos package is no longer a one-and-done.
3) mineos uses crontabs.  I don't think its particularly responsible for a package to add crontabs (bad style)

So we can see that a package should typically deal with just itself--and any dependencies typically shouldn't need post-configuring as well.  That makes an arch package a less useful (or at least, less predictable) package than arch deserves.  It'd be much easier to do MineOS-on-Arch akin to MineOS-on-Ubuntu...really easy, actually...but it all boils down to: do people want to?
Re: The State of MineOS Peter Murdock 7/3/13 2:33 AM
The webui seems to be really buggy with the new 1.6.1 craftbukkit. If I start the server, nothing really loads in the webui and server status reads DENY even though I'm in the server playing with no problem (hardware isn't the least stressed either).

Is this something that can be fixed?
Re: The State of MineOS Will 7/6/13 11:29 AM
On Wednesday, July 3, 2013 2:33:18 AM UTC-7, theratio...@gmail.com wrote:
The webui seems to be really buggy with the new 1.6.1 craftbukkit. If I start the server, nothing really loads in the webui and server status reads DENY even though I'm in the server playing with no problem (hardware isn't the least stressed either).

Is this something that can be fixed?

I just fixed this today with the BETA CHANNEL.  Please give it a try and report if it is working properly for you (and if not, what jar/url you're using). 
Re: The State of MineOS Peter Murdock 7/9/13 5:27 AM

I updated to 1.6.2 before I saw there were updates. Anyhow, it still persists with newest craftbukkit. It doesn't really render anything useless, it's just buggy.

URL is defualt I think, /admin. JRE is (build 1.7.0-b147).
Re: The State of MineOS Hexagon 7/11/13 11:51 AM
Hey Will,

I'd love to create a non-crux based MineOS! Since I have quite a bit of free time (with some exceptions) I could really work on it a lot. Feel free to contact me if you need help or anything else.

Hexagon
Re: The State of MineOS Will 7/11/13 12:35 PM
On Thursday, July 11, 2013 11:51:23 AM UTC-7, Hexagon wrote:
Hey Will,

I'd love to create a non-crux based MineOS! Since I have quite a bit of free time (with some exceptions) I could really work on it a lot. Feel free to contact me if you need help or anything else.

Hexagon

Heya,

What did you have in mind?  Thus far, there have been a few non-crux based MineOSes (turnkey, microcore), and then some atop-other-distros (ubuntu, freebsd)--so at this point, its pretty easy to get it on any distro you please, but if you have free time and a drive to build something, I'm sure I can be of service!
Re: The State of MineOS Alex Yorke 7/11/13 12:40 PM
Hey,

Thank you for your very quick reply!

Since you said there are a few already, would it be good to invest my time into creating another non-crux based mineos or is there something else I could help with?

Hexagon
Re: The State of MineOS Will 7/11/13 1:43 PM
On Thursday, July 11, 2013 12:40:29 PM UTC-7, Hexagon wrote:
Hey,

Thank you for your very quick reply!

Since you said there are a few already, would it be good to invest my time into creating another non-crux based mineos or is there something else I could help with?

Hexagon

Keep in mind, there is no pressing need for any new distros, as the ones available are already fairly varied, nor am I in need of more manpower to reach out to more....this was more of  an offering to others in the community who wish to have a chance to build something themselves.

In other words, I'm offering education and advice to users so they can go through the process I did: learning Linux, anticipating user issues, providing support, configuring and distributing an ISO.

You can invest your time in any area you please, but I'll let you in on a little secret!  The time you spend on a project often only feels rewarding if it somehow improves your current situation!  Thus, if you don't think you could use MineOS on another distro, you probably wouldn't want to tackle that kind of issue.  If you think MineOS could use a better web-ui, then you'd start at HTML/css/js (or even python!).  

I appreciate your volunteering of effort, but lets direct it at something that would personally drive you (with me to help along the way)!

Will
Re: The State of MineOS Alex Yorke 7/11/13 3:39 PM
That sounds great!  I kinda felt like MineOS could use a different web-ui (before your comment), so I'll give that a shot.
--
You received this message because you are subscribed to a topic in the Google Groups "mineos" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/mineos/EeMDWVzS380/unsubscribe.
To unsubscribe from this group and all its topics, send an email to mineos+un...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 
unk...@googlegroups.com 7/12/13 3:15 PM <This message has been deleted.>
unk...@googlegroups.com 7/14/13 1:29 PM <This message has been deleted.>
unk...@googlegroups.com 7/14/13 1:53 PM <This message has been deleted.>
The State of MineOS gradymcd 7/30/13 5:57 PM
A desktop OS seems like a very good idea. We could have a very lightweight OS that is optimized for mineos. I would live the help work on this. There is a very good to making a Linux distro at http://www.linuxfromscratch.org/lfs/.
Re: The State of MineOS Will 10/9/13 10:56 AM
I'm officially phasing out 0.5.0 in its entirety.

0.6.0 hasn't been met with critical acclaim and there has been pretty unsubstantial usage so far.  It could be that the UI is a complete bust or it could be that asking a user to do installation steps was a bust.

At any rate, 0.6.0 is now being distributed as an ISO, retaining the familiar 'mc' user and thus removing the requirement for 'root' to be used by the web-ui from fresh install.  So, unlike installing it from scratch (which does still require root), 0.6.0 is trying to be a return to the do-everything-for-the-user ISO mentality, which I hope will improve adoption.

Unfortunately, if 0.6.0 is a bust (and people were hesitant about 0.5.0, you know!) that could very well mark the end of development of MineOS, as 0.5.0 is not something I wish to continue work on (and 0.6.0 very much is).  Now that all the download urls as of today have been updated to 0.6.0 ISOs we'll see if it can rekindle heavy dev as it was over the past few months.
Re: The State of MineOS FUP 10/10/13 6:54 AM
Will, "asking a user to do installation steps" is a bust.". Users of today are spoilt with playstore and appstore, and windows install and update...

I find 0.6.0. excellent as is. The needed functionality is all there. Version 0.5.0. paved the way for that to happen.
Could think of things that would be nice to have like more user friendly language maintanance support or some sort of autorisation management of users/groups/servers. Like I said nice to have.

Thanks for creating and share MineOS for free. Also for taking time to give Linux-install advice, tips and tricks!


Op woensdag 9 oktober 2013 19:56:08 UTC+2 schreef Will:
Re: The State of MineOS IRQ 10/11/13 6:57 AM
User/group management can be done throug webmin. In my Turnkey ISO with 0.5.0 it was installed by default. Don't know about 0.6.0 ISO.

четверг, 10 октября 2013 г., 17:54:17 UTC+4 пользователь FUP написал:
Re: The State of MineOS FUP 10/11/13 10:20 AM
It's the only basic setting that can't be done with MineOS itself.

I know it can be done with Webmin 'System'>'Users and Groups'. Or, as Will instructed me to do, from the command line use 'adduser', 'addgroup' and 'usermod -a -G'.


Op vrijdag 11 oktober 2013 15:57:12 UTC+2 schreef IRQ:
User/group management can be done throug webmin. In my Turnkey ISO with 0.5.0 it was installed by default. Don't know about 0.6.0 ISO.

...
unk...@googlegroups.com 10/16/13 6:31 PM <This message has been deleted.>
Re: The State of MineOS rem...@gmail.com 10/30/13 10:55 PM
Will,

Thank you so much for this. I've been using MyMcadmin on the same pc that I play games on, and it's been frustrating having the computer on in the room 24/7 to keep a server up when it's heating my room like a space heater all the time. Now, I have a dedicated Mineos Linux box in the other room and this is way cooler than mcadmin. I've had issues with Feedthebeast (FTB) mods crashing, but I read online that many people are having similar issues today because of a special mob spawn in IndustrialCraft. I'll try running the server again tomorrow (the special mob spawn is for Halloween and will go away tomorrow).

Some feedback regarding the MOS Web User Interface:

The new way the tabs are set up is a little bit buggy. I watched the video you posted about how to set up a FTB server in MineOS and the interface you had there, while less flashy and much more simple, was better in my opinion. My biggest beef is that you have to refresh the page when managing multiple servers at once. For example: I would set up a FTB Unleashed and a FTB Ultimate (different flavors of FTB) server. To modify the settings in Unleashed, I have to go to dashboard, select Unleashed, modify the settings. In order to correctly switch to Ultimate, I can't just go back to the dashboard and select Ultimate because while it says Ultimate, it actually has the settings for Unleashed still cached and if modified will edit Unleashed settings. If I refresh the page and go back into the dashboard, it correctly modifies the proper server. I would really love a dropdown menu to select the server settings to modify like the way you had it in your video from early this year. 

Anyway thanks a ton for this awesome OS and looking forward to future releases.

Jeremy
Re: The State of MineOS adm...@gmail.com 10/31/13 10:15 AM
For me, I just checked in - my 5.0 install was working and not giving me fits.  I think slow adoption may just be the fact that people got it working, it kept working and they didn't mess with it...  I am not at the point I want to rebuild my server and 6.0 it will be!  

Thank you Will for a great platform to play Minecraft with my kids and introduce them to the command line and terminal interfaces and managing servers etc... 
Re: The State of MineOS steven...@gmail.com 11/18/13 8:33 PM
I was in that boat.  Just fired up 6.0 this evening, and it is pretty neat.  Still have some digging around to do, but so far, so great.  Thanks!!


On Thursday, October 31, 2013 1:15:52 PM UTC-4, adm...@gmail.com wrote:
For me, I just checked in - my 5.0 install was working and not giving me fits.  I think slow adoption may just be the fact that people got it working, it kept working and they didn't mess with it...  I am not at the point I want to rebuild my server and 6.0 it will be!  

Thank you Will for a great platform to play Minecraft with my kids and introduce them to the command line and terminal interfaces and managing servers etc... 
Re: The State of MineOS joshsc...@gmail.com 11/26/13 4:53 AM
Hello, I am new to this MineOS thing, and I did not get to install it yet, but I do have some Linux knowledge, and I may as well share my two cents:

I do like your idea of minor updates with some major ones, but that can result in stagnation, and some people would for competing projects etc.
A good fork would be off TKL, they even offer a dev environment to use so that you can create the distro. A desktop would also be interesting, you could ask mojang if you could bundle the minecraft client on it.
For the desktop, I would base it off Gentoo, as it is very nice and expandable.
Re: The State of MineOS rikber...@gmail.com 6/17/14 8:32 AM
i think that a nice "fix" for the whole "WE WANT AN GUI ON THE SERVER ITSELF NAO" could be just having it run the webui/ installing an existent linux gui on it (openbox, kde, xfce, unity (i would recommned openbox because of it being ligtweight))

Op donderdag 23 februari 2012 23:07:33 UTC+1 schreef Will:
MineOS CRUX is--I feel--nearly as final as it will get.  While there are countless ideas I have come up with that have gone undeveloped and would still be of great use, it's at the point where I believe the overall usability/educational experience has balanced out exactly where I want it.
  • Ideas such as the auto-refreshing AJAX-driven web-ui I am technically competent in pulling off, but aesthetically incompetent for.
  • The log parser/MySQL plus visual graph would be nice,  but overall is more fun than function. 
  • My share-your-world-on-the-cloud / s3 backup idea I still think would be amazing, but  would probably be unpopular
  • My MySQL filesystem (database held chunks, rather than file-based) would be amazing, but may be impractical to spend time developing if API modding changes code drastically in the next few months.
  • A couple of other great ideas, etc.
That said, while I'm at no shortage of ideas and ways to dump hundreds man-hours programming, the returns might not be as great as I'd expect and 'if it isnt broken, dont break it'.

So where does this leave me now?

Right now, with no new programming challenges (or at least, none being developed...I spend a lot of time theorizing, but not prototyping), I'd probably be happy with simply providing support for issues. However, the direction of the questions lead me to believe that MineOS it self does not need much more updating, but instead there could be use to a more general-education of Linux. I have a few ideas that I'd like to feel out to see if it interests any of you:
  1. Creating a collaborative fork
    Right now, MineOS CRUX is pretty much exclusively handled by me. Some code contributions come in here and there and dozens of feature requests, but I stand as the single decider and implementer of new code and functionality.  This sort of centralized management means updates come at my timeline, and in this case, last ISO update in October.

    Would any be interested in creating a non-crux based MineOS?  That is, I'm asking if anybody would be interested in learning the steps of transforming an OS into a MineOS while also being themselves decision makers.  Things like mineos.py are transferable, but getting a distributable ISO, automating installation and deployment tasks, as well as documenting are things that are both immensely educational (you bet it was for me!) as well as intensely rewarding. Realistically, one could consider making a MineOS to the likes of:
    * MineOS FreeBSD (oh, unix!)
    * Turnkey MineOS (http://www.turnkeylinux.org/)
    * MineOS EC2
    * Virtualbox Appliances
  2. Creating an unrelated Desktop OS
    While having a graphical GUI on a server has been largely requested, it has never been interesting to me to make a MineOS + GUI version.  On the other hand, if we wanted to create a Desktop OS, learn the process, maintain and distribute, this would be a huge learning opportunity for us all.  That said, I see no reason why existing materials couldn't be made to work with MineOS script, but the primary focus here is MineOS-independent: creating a usable, pretty, and customized desktop distro.
Keep in mind, the primary focus of these options is the learning. Neither of these will be options that give some perfect, immediate product. It will, however, give you a chance to learn the process that I have gone through--actively, and with my guidance and accumulated knowledge--to create something you may find useful. For me, it will give me the chance to finally coordinate a team-oriented project rather than a solo-work, employing things such as git, buildboxes, and--most significantly--it is an opportunity for me to see how a project grows organically.  I do not expect people to make unrealistic 'commitments' to the project, dedicating time you simply don't have, but if enough people are interested, hopefully the distribution will be thinned quite nicely and progress can be made even if often untouched. (think of CRUX the distro... they release a minor version (.1) every 3-4 months, and a major version (1.0) every 2-8 years).

So what do you all think?

Will


More topics »