I am on 13.10 with xmonad but will try to move to 14.04 LTS with Unity.
This will happen as soon as I finish my Ansible playbook to get all my current Ubuntu tweaks and dev env setup implemented so I have a repeatable setup and can safely do a fresh install of my desktop as needed with zero manual work.
Dotfiles alone didn't cut it and VPS provisioning frameworks like Ansible are great examples how desktops could be built as well :-)
I've wanted to be able to automate setting up a new Linux desktop on a new laptop for a long time. It's a drag setting up a new box, it takes nearly a month to get everything tweaked right after I switch to a new distro (which is far more often that I should be responsibly doing it).
I've been thinking about setting up an Ansible script, I've never used it before though. How would this work on a new machine? Is it something you run post-OS install? I want to try it out with Arch Linux.
Yes, a playbook would run post-OS install. It's either executed locally or anything you can SSH to. If being careful enough one can support multiple OS-s with the same playbook (at least different Linux flavors) but it's definitely quite some maintenance like different package names and conf file locations etc.
A playbook contains tasks like installing, removing packages, editing config files etc. The beauty of it is that if done correctly you can run the same playbook over and over again against the same computer and each task can be smart enough to know if it needs to be run or not. This makes updating your computer with the same or updated playbook very fast.
With Ubuntu it's easy to get to the post-OS install, it's more involved with Arch Linux. I would first create an Arch box with Vagrant using Virtualbox. And then create a playbook that works with that virtual Arch box. If the box and ansible playbook is good enough then
https://wiki.archlinux.org/index.php/Moving_an_existing_inst...
At least in theory, I haven't moved Arch between VM and a real host myself... Of course you can also just do a manual Arch install and then run the playbook but that would miss half the fun of doing less by doing more.
I'm not sure Ansible would be very useful for the use-case of "getting your desktop right". Generally there's three things that is needed to get a (personal) dekstop set up:
1) Get the right packages (eg: dpkg --get-selections > my-pacakges.list; dpkg --set-selections < my-packages.list). This part ansible can do fine -- but not really any better than your distributions package tool (for this particular use-case)
2) Set up your account (This can be done with kicstart, FAI or preseed.cfg) If you only have one user, this might not be worth automating. This is also a task ansible can handle fine.
3) Set up your preferences/restore your home folder etc.
This is the area that tends to take the most time and care.
Personally I've pretty much stabilized on a very spartan xmonad config (driven from ~/.xsession), and have a mercurial repo with various dot-files that I just clone and symlink to (and I have a bash-script that takes care of setting up the symlinks) -- I also manually set up a python virtualenv (which bin I add to my path), an ~/opt hierarchy managed by xstow (I usually recreate this, as the whole point is to follow upstream, unpackaged software) and a ~/bin folder that's also in a mercurial repo (a few shell scripts for toggling vga/internal displays for my laptop and some similar small utilities).
Now, "manually" symlinking your dot-files probably isn't the best idea -- the main reason I do it this way, is that I've had to manage my profile across different distributions (and for a time also on Solaris) -- and then just keeping home in version control can be a little to simplistic (now with mercurial/git I suppose using named branches for the various machines might be a viable option though -- not sure if I want a check-out command in my .bashrc/.xsession though -- too much can go wrong... eg if git/mercurial for some reason isn't available won't run without errors...).
I'm of the "latest is greatest" school. I switch my desktop on release day. Week-month later (depending on issues) I deploy to testing machines. Month-three later (depending on issues, my work load, need for any upgrades) deploy on subset of production. If all is well, then to rest of production shortly thereafter.
OTOH, Long Term Support. You shouldn't have to upgrade for awhile if its a pain for you.
For all personal/hobby projects, I can see myself switching within few weeks. For the "serious" bunch, I may be waiting until the middle of 2016, then I'll start migrating.
since noone actually responded to your question, and assuming you are talking about on servers, i began testing our configs against trusty some time ago. I'm not going to switch right when it's released, but I am eager to move to it. You're going to get lots of things that you may have had to work for in 12.04, like a newer kernel and newer versions of system tools and libraries and your key server software, from OpenSSL to nginx/apache2, python, ruby, libc, etc.. The LTS cycle works, and I encourage you to begin testing as soon as you can, and in the future to test while it's in testing simply because you can - it's not expensive to throw up an instance here or there.
Ubuntu's user interface upgrades as of 11.10 (the "unity" interface) sucked so bad that I refuse to use Ubuntu beyond verision 10.04 on desktops/laptops.
11.10 was when I switched over to Mint and never looked back, and it seems that doing so was a wise move, given the Amazon adware/spamware/spyware that Canonical saw fit to include in more recent versions.
Even though Mint includes proprietary binaries (like Flash and Audio/Video codecs), which may or may not contain opaque questionable material, at least the third party non-open-source software is something that (arguably) improves the distribution and actually serves a purpose for me, as the end user.
Mint has changed over time too, though, and now I'm thinking about moving to a personally customized Debian image, and a hobbyist project. Hopefully it won't prove to be too demanding to pull off.
Your feelings about Unity, and your preference for Mint because things regularly break and it gives you learning opportunities, may be things you really want to get said. But they are not really relevant to the question you are replying to.
I too was pushed to mint after 11.10 but I'm sick of Mint now too. Not only do upgrades break it frequently, but I find small 'glitches' that end up taking up too much time tracking down. Stuff like Wifi dropping, and icons in my taskbar dissapearing, random browser crashes etc.
What I did was decide to sit down and do an Arch install. Yes it takes time, and yes you have to know what you're doing. But I invested the time up front, and now it runs very reliably, and faster on the same machine.
I say if you want to GSD the best thing to do is set up something like Debian, Arch or Gentoo and invest the time setting it up so you can use it without problems later. I don't know about you but I have better things to do than screw with an OS all time, I have real work to do. These "harder" distros are great for that.
I am using Mint 16 now; first time using it. Previously I was using Ubuntu. I had to switch to Mint because we got W540's at work which has a lot of new hardware that is not currently supported (well). I wasn't able to get Ubuntu working on it properly, but I have been impressed with Mint so far. The only issue right now is that it doesn't see my nvidia card and so I have to use the integrated chip instead of the discrete one. I'm hoping that Mint 17 fixes the issue.
How hard is it to get a custom system up and running from Arch? I haven't done anything like that in a few years although I've had a lot of experience with setting up custom FreeBSD systems. Is it more or less like that?
What I'm reaching for is something that "just works" and that I can work on reliably instead of having to fix obscure problems all the time. I figure once I set something up that works, I can simply create an image of it to use later.
Arch isn't that hard to set up, it's still easier than Gentoo. You just have to set it up from an explicit point of view. You must know every detail of what you want, and each item up, rather than a "10 clicks and I have an OS now" type of setup.
It helps to have a good knowledge of Linux to do it, because you know where things should go and where to look if there is a problem, but it doesnt' require you to become a kernel hacker just to get it to a prompt.
Honestly though, the only thing that attracts me to Linux in the first place is digging into the source and pulling apart the pieces and putting them back together again, so that I can understand the tools I use more intimately.
To me, distribution projects like Debian, Arch & Gentoo serve as a useful source for complete repositories of working, interoperable packages, moreso than they serve as a convenient provider of a working operating system.
I tend to chalk up buggy, quirky work-arounds and long waits for bug fixes in Linux as the price I'm willing to pay to be able to see the source, and receive the software for free. Annoying glitches are something I wouldn't tolerate from Apple or Microsoft, if I'm going to shell out for the high price tags placed on their operating systems.
I'm with you 100%. Often times I do use Linux as a dumb platform to do work like I would Windows or OSX, but occasionally I love tearing into it and messing with it, which is priceless in my book.
In your experience, is Debian less prone to breakage when upgrading from stable to stable than Ubuntu with its release upgrades? I've run Debian but I have never been patient enough to stay with a stable release and find out.
To be fair, Unity in 11.10 was quite broken and slow and resource hungry, but they fixed that in 12.04 and 12.10. I understand if people don't like Unity, but it's been getting faster and better than in the earlier versions. Personally I use cinnamon.
IIRC it's Canonical's official stand to suggest users to wait for the first point release. I am sure I read this some days ago, from following a link posted here.
Indeed, upgrades aren't prompted for LTS users in Update Manager (Software Updater) until the .1 release hits. Intrepid souls are of course free to upgrade before that.
I decided to go the LTS route as well. I actually don't know when they will actually release the new version, but I assume it will be before the end of April (do they always hit their target release month?).
I will be upgrading fairly early; probably at the end of April. I have a big deadline in late April and upgrading Ubuntu will be my reward/punishment. I suppose I will listen for any disgruntled users before I make the plunge, however, and reassess if things look too problematic.
Note: this is for a work machine, but a PC, not a server or something like that.
You have to follow the bug reports if you rely on stuff, e.g. this bug in 12.04 (which was not in 10.04) https://bugs.launchpad.net/ubuntu/+source/grub2/+bug/669481 has no released fix two years later - if you need your servers to consistently reboot after a power down without human intervention, you need to apply the fix manually (just inserting a line in a config file, but still ...). I know it says "fix released" - but after having to drive to reboot servers, I can assure you that is not the case.
Personally, I'd wait at least 3 months, and follow bug reports before switching.
Ideally I would want to upgrade within a month or so (after major kinks have been worked out) but there is stuff that may not work with the new kernel (I extensively use NoMachine), so I want to make sure if I upgrade I can still use it. I'll first install it in VirtualBox and give it a try. If it works, I'll upgrade. There are disadvantages to sticking with 12.04 for too long. You get stability but you start missing out on the new features of the kernel and other daily use software like LibreOffice. So, I really want to upgrade as soon as possible.
The official Canonical/Ubuntu line on this is to wait for the first point release, 14.04.1... however long that takes. I think for 12.04 it took 2-3 months.
Our organization is currently running on Ubuntu 10.04, which is reaching end-of-support next year, so I've begun a push to move to 14.04. All of our server installations are entirely automated so a lot of it will just be porting over the preseed variables to the installer, answering any new questions, and possibly moving to newer versions of some of our dependent software (a lot of which is equally out of date).
Why wouldn't you build a newer package for 13.04, or try using the trusty package on it? I can't imagine waiting for a new OS release every time I needed a newer package.
It is not really often I need a newer package than what is available, I can only think of Beanstalk because if this issue.
I am not really that used to build my own packages, but it is something I want to learn. And I didn't know that I could download from another version, I will look into it, thanks ! :)
apt-get build-dep beanstalkd
apt-get source beanstalkd # or in your case, download the newer source-package
cd ./beanstalkd-1.x.x
dpkg-buildpackage -rfakeroot -us -uc
This should do the trick. There are a lot of variations regarding the last line. Just google for Debian packaging or Ubuntu packaging tutorials. It really works well for minor changes. If you need to deploy it to several machines it's also possible to add a custom flag to the package version...
And it's an unbelievable pain in the ass when they change packaging styles, and the new src packages aren't compatible with the older build tools. IIRC, that happened between 10.04 and 12.04.
It did, but it's only relevant for packages which actually updated their build scripts. A huge proportion of packages from 12.04 will still build fine on 10.04 (which I've been doing in order to backport certain packages to install).
Takes a year for me on average to roll all production to a new LTS. This includes giving package authors to release good package for the new LTS and testing.
Moving to new LTS right away is often a suicide mission.
I'm thinking of trying to switch to it completely. I'll probably still need Windows for some stuff, but I'll try to use Ubuntu as my main OS once this LTS comes out.
Currently on 13.10 at home, 12.10 at work desktop, 12.04 on servers. I feel no need yet to upgrade my desktop and my servers. All my previous upgrades had been motivated by better config tools and installers or new software. All these issues have dealt with by earlier releases. I think the distro is no longer the conduit for delivering new software and 14.04 doesn't bring any compelling upgrades outside of Unity tweaks, which I don't use. I'm going to wait this one out.
