Docker and Microsoft partner to drive adoption of distributed applications
blog.docker.com
blog.docker.com
We have 3 apps over 32 servers and 5 environments, and operationally it's like pulling teeth. This has the chance to change everything!
http://research.microsoft.com/en-us/projects/drawbridge/
Or something else entirely.
Regarding Drawbridge, as you pointed out it is an internal research project that we have been innovating on, and that has helped us gain valuable experience with containers. Much of what we announced today was born from the experience that we had with Drawbridge and we are excited to bring container technologies to Windows Server and the Docker ecosystem along with Linux. We think the combination of our own hypervisor for container virtualization and Docker containers for creating a unified deployment and management experience is a compelling scenario for our customers.
Our dev process is quite nice, we have fully automated builds and deployments which are feature branch aware, allowing us to run several features side by side.
On the ops side however, most of the work is done either manually or the scripts are unreliable (again, external company nothing much we can do about it). Setting up new environments can take months. Many times we want to introduce new services but the time it would take to get our infrastructure updated usually outweighs any benefits.
Being able to ship configured containers as smoothly as we can ship code would be nothing short of revolutionary!
But a) original poster is talking about a web application; not a virtual machine or hypervisor and b) VMWare itself still requires an underlying operating system-- one could argue that this case is actually an operating system running another operating system with a thin adapter layer in between (i.e., the VMWare application).
VMware is a company, not a product - the product you're [probably?] referring to is ESX[1], which does not require an underlying OS - it's a Type 1 hypervisor (it runs on bare metal).
How do you think docker containers helps (as Im not sure it would have helped in my previous situation - too many other base issues to deal with first)
We also use database migrations to keep everything updated. It's a simple process but works well.
Code and deployment is all fine, it's maintainaing server configurations across environments (IIS, permissions, ports, ips, firewalls to name a few, a different issue every time and on every env!). Likely this is more to do with having completely seperated operations and development, however bring able to control the apps environment will allow us to deliver more, faster.
thanks for the info!
Great to see this conversation about Octopus. Stay tuned for news about Octopus and Azure.
It can be used to replace deployment tools like capistrano and fabric on Linux. No reason it can't do the same on Windows if you grok it enough to use it.
I haven't used it yet on Windows, but it's good enough on unixes... And it's quite lightweight and easy to setup (it relies on python2 on *nix and powershell on windows)
See: - http://technet.microsoft.com/en-us/library/dn249918.aspx - http://powershell.org/wp/2013/10/02/building-a-desired-state...
From my point of view OS configuration for app level concerns (web servers, services etc), should sit in the development realm, ready for operational teams to distribute and monitor.
Scripting systems does help, but it is not the final solution when you have mixed, varying talent teams. As a side note, every time I have tried scripting Windows systems I get an extra gray hair!
There are certainly people within Microsoft who want to cooperate with the rest of the world, and their behavior has been better in the past several years, but I wouldn't be surprised if it's built into their corporate culture to destroy or coopt anything that might pose a threat to them. And, it may be that the only reason they aren't still destroying and coopting on a wide scale is because they have so much less power today. The web is not owned by Microsoft; in the end, they lost that war, despite trying very hard.
Anyway...Microsoft is not Google (and Google is not flawless in their relation to the rest of the web and Open Source). I think it's wise to proceed with caution whenever interacting with anything Microsoft has touched.
That said, I could be completely wrong, and it's important to trust, but verify.
As a result, I'm pretty comfortable with the terms that ensure we protect our commitment to Open Source and in ensuring there is no special access given by partners, Microsoft or not. The project is the project, governed under the rules set forth by the community. Even members of Docker, Inc. have to advocate and fight for every change they make, as we do not believe in creating different classes of contribution.
Hope that helps, would love to chat more
Time will tell, for now I'll remain optimistic and continue down the path we've started.
Regardless of what others think of Microsoft, I believe my disdain and mistrust for Microsoft is based on a reasonable understanding of historical facts, rather than some subjective vendetta. I've been a nerd for a long time. Microsoft has (until possibly very recently) never been a good citizen in the tech world.
Yes there was a period of a year or two when Netscape entered the abyss and it took a while for a new competitor to IE to spring up. But that happened (Firefox) and shortly after Microsoft resumed development on IE too.
You seem to have selective memory. IE brought A LOT of good to the Web. A LOT of good.
"You seem to have selective memory. IE brought A LOT of good to the Web. A LOT of good."
From where I'm sitting you have a very imaginative memory. IE was better than the competition because it destroyed the competition using the very unethical (and illegal, according to courts in several nations) tactics I've already mentioned.
Honestly, I'm surprised anyone on HN would have so little knowledge of the history of Microsoft and the web that you would interpret their stranglehold on the browser as a positive thing. I simply can't wrap my head around it, it's so absurd to me.
The courts took issue with bundling it with Windows. Not that it was not innovative. There was even a few articles in recent years suggesting the courts went too far over the "bundling" case; sighting comparisons to vendor lock-in all over the industry now not reaching the courts at all. Just one top hit cite: http://readwrite.com/2013/11/12/apple-maps-takes-off-cue-the...
The whole bundling IE case was fundamentally stupid and run by lawyers and judges who had no understand of the technology industry, the web and the wild west stage the web was going through. Back then, it was all about enablement. Getting people just onto the web in the first place was hard. They had to buy a computer, a modem, sign up with an ISP, possibly upgrade their operating system, install a web browser etc. But in those days, installing a web browser meant you had to go buy a Magazine from a shop just to get the CD-ROM. Microsoft viewed this, rightly, as an impediment. So they bundled IE with their OS. A practice that is still common place today by every major operating system that exists including Windows, Linux, OSX, Android, iOS, WinPhone.
Without IE bringing DHTML we would not have the foundations that made modern day JavaScript, SPA, "Ajax" applications today. How is that not fostering innovation? IE 1 to 6 were a technology showcase to show everybody else what the web _could_ be or become. That's one reason why, though they'll never admit it, Mozilla abandoned Mozilla Suite and started work on Firefox.
"So little knowledge of the history of Microsoft and the web"? Er, what? I date back to having to when you had to install a TCP/IP stack manually in order to get onto the web. What you're actually surprised about is that anyone would dare challenge you on what you perhaps believed would be a widely held opinion when it's far from that clear cut.
I do disagree with you, in every regard, but I don't downvote over disagreement (and could not possibly downvote you in this instance). Someone else must also disagree with you.
Microsoft defined the business model of the web browser market as it exists today. I.e. free, widely and easily available, and bundled with the OS. This has not changed. The court cannot prevent free market economics, but they tried to.
Yes Microsoft made Netscape go bust. But Netscape were weak and short sighted. They were like squashing an ant.
Anyway, I think what the justice department went after Microsoft for was a minor part of the wrongs Microsoft committed. Bundling the browser, and effectively prohibiting computer vendors from bundling Netscape, was a nasty trick and it killed Netscape (just as Microsoft killed Lotus, WordPerfect, and numerous others, often through backroom deals to separate those companies from their customer acquisition channels). But, my concerns are less about what happened to Netscape than about what happens to consumers and the market, when consumer choice isn't what decides the outcome.
IE was a proprietary overlay on the web. It was not an HTML browser, it was a Microsoft delivery platform. Having a stranglehold on the browser locked every competing OS out of the web for a decade. I switched to Linux on my desktop in 1995. But, I had to keep a Windows installation around on every desktop I ever owned in order to be able to access bank and government web sites, so I could use IE to access it. That's why IE was destructive, and that's abuse of monopoly power.
And, that's why Microsoft created IE: To embrace and extend the web, to use their existing monopoly in several enterprise markets to subtly take over the new market of the web and destroy the openness that allowed competitors to thrive. And, it succeeded for a long time. For years, we were trapped in this horrible Microsoft-owned world, where the dominant browser was incompatible with every other browser and with the standards and in ways that were intimately tied to the Windows operating system.
So, the courts didn't tackle the most damaging issues, unfortunately, and it took several years for the web to recover from the damage Microsoft caused.
MS was making "illegal market-splitting suggestion"
and
"Microsoft began to use its market power to extract exclusionary deals with many of the largest [PC manufacturers and internet service providers]", threatening Netscape customers such as Compaq that if it tried to replace the Internet Explorer icon with the Netscape Navigator icon "
We can assume they've done many more bad things since their EEE attitude was proven many times (Bill Gates included - vide leaked email in antitrust case http://antitrust.slated.org/www.iowaconsumercase.org/011607/... )
If Microsoft would have made Internet Explorer a boxed piece of software to sit on the shelf beside Netscape Navigator, and if they had priced it similarly, and if people had voted with their wallets in IE's direction (because it was actually a better product for the money), I wouldn't have had a problem with IE taking over the world. As it was, they bundled it, and it was crap in comparison (at least to start with, and many would say up until recently), but Netscape couldn't compete with free.
THAT'S what a lot of us still remember. It was a perfect, easily-visible example of a lot of business moves they have made, and for which they ultimately -- not only weren't punished -- but actually were allowed to prosper because of. THAT'S why people like me are still sore about it.
They won because of BUSINESS SAVVY and LEGAL moves, not TECHNICAL MERIT, and people in the software development world (and everyone else) have paid the price for it for 20 years. (I can't get Sametime in the web version of Lotus Notes to work unless I use Internet Explorer, and in "compatibility mode", to emulate their non-standards-compliant behavior that everyone was forced to code around, as one immediate example.)
All the touchy-feely "openness" they're trying to foist on the world now is going to have to be everything they hope people will interpret it as for the next THIRTY years for me to believe they really want to interoperate with the rest of the world like Linux always has.
Netscape always had a free version* available, so it wasn't really down to price. (*though it was usually the buggier beta version.) Otherwise, its plan was to make money on server side, and despite buying in several server companies, it failed.
While Netscape was technically inferior, it is true that Netscape's own marketing and managerial mistakes contributed to its downfall. For example, you could only download it from Netscape, you couldn't customize or rebadge it, and at one point it decided to withdraw it, so you could only get it as part of a Netscape suite.
All of this was suicidal when Microsoft was shipping a free IEDK and allowing computer mags and ISPs to distribute IE.
Finally, in the anti-trust lawsuit, Microsoft WON the browser case 2-1 on appeal. So bundling wasn't actually illegal, as alleged.
I watched all this closely at the time. It is also well documented in several books, including How The Web Was Won, Competing on Internet Time, and the great High St@kes, No Prisoners.
You can say Netscape imploded, but they experienced a fight-or-flight response forced on them by Microsoft's scummy actions. I never said that bundling was illegal, but that's a purely legalistic distinction. I certainly implied it was unethical.
Before the trial, Microsoft gave about $10K a year in political donations. By the time it ended, they were giving over a million to EACH side. You weren't the only one watching, but not everyone watched the same things.
Remember "Best viewed with Netscape Navigator"? Netscape wanted to control the web by defining and unilaterally introducing web standards, so Microsoft naturally allied itself with the W3C.
It was short lived because Microsoft effectively abandoned browser development following anti-trust action, because it decided it could/would only ship browsers with operating systems. This turned out to be a bad choice because of the long delay between XP (which included IE6) and Vista. The slow take-up of Vista made it even worse. Microsoft didn't want people to use IE6 or XP, but the market decided otherwise.
All this goes back to the consent decree that Microsoft signed with Janet Reno. This specified that Microsoft was allowed to improve the operating system by adding new features. This more or less required it to build IE into the OS.
Microsoft was already under investigation during this period, and for the decade after the case, Microsoft operated under the close supervision of a US Judge. Whether or not you think its actions were scummy, it's unlikely they were illegal. Indeed, during this time, Microsoft was certainly easier for third parties to deal with than Netscape -- see High Stakes for examples.
As for political donations, this was also inevitable. Netscape couldn't compete in either technology or marketing so it played politics instead. Microsoft had operated on the basis that it wouldn't get involved with politics, and then it got screwed for its neglect. It simply decided not to make that mistake again. As anyone would.
Either way, Netscape was doomed in the long run. It was deluded in thinking that people would pay for a client access program. (You're welcome to provide examples of companies that have made pots of money out of this, but usually the client is free and people pay for the service.)
Netscape was even further deluded when it thought that it could charge PC manufacturers to ship its browser. The reality is that companies have to pay PC OEMs for distribution.
The final delusion was that users would pay a price for Netscape Communicator http://en.wikipedia.org/wiki/Netscape_Communicator in order to get their hands on Netscape Navigator.
You really don't need any conspiracy theories to understand why Netscape lost. It had a combination of arrogance and incompetence the like of which I have not seen before or since.
It's almost like you would rather still be paying for web browser software today. Would you?
Speaking of innovation it is true that IE was terrible in adopting standards (this was pain for long, long years and costed tones of money), but, on the other hand, they have invented XMLHttpRequest, which opened doors to modern web applications development.
The real problem was that Microsoft stopped browser development for half a dozen years, and that put it miles behind. The rapid releases from IE8 to IE11 show Microsoft trying to catch up, ie being pretty good at adopting standards.
There are plenty of things to blame, including the US Justice Department, the unexpected delay from the Longhorn disaster, and the backlash against Vista. This left XP dominating the PC market, and XP shipped with IE6.
If that's been holding up the web, it's not because Microsoft willed it. Just the reverse: it's being trying for years to kill it off.
The problem is a lot of Ruby/JS/dynamictypedlanguage hipsters these days who shout so loudly about IE having poor standards support weren't actually around at the time. Everyone loved IE back in those days as it was modern and every release had new exciting features in it (both for users and developers).
The fact that IE got some things "wrong" with CSS and had bugs like transparent PNGs not rendering, er, transparently... were just necessary growing pains. This didn't matter in the wild west days of the web as, well, PNGs weren't even that popular back then and CSS was still something web developers were getting to grips with, slowly.
Compare IE4-6 not to later-generation browsers, but to its direct rivals of the day which was Netscape and Opera and not really much else. Compared to these, IE was king.
At a lower level, this is strengthening a proprietary platform, and might lower adoption of FOSS platforms.
At a higher level, co-opting an open source technology to deliver closed source software may indicate a strategic flaw in our approach to FOSS. On Microsoft's part this is clearly a continuation of 'embrace and extend'.
They'll use docker to provide the functionality and build hype until they can build their own version into the OS by default. Docker on Windows becomes obsolete, and everyone runs apps on Azure.
They are becoming too big to be solo... and I'm thrilled for their team
Much better to focus on what we can control - making docker awesome for as many people as possible.
So much for the portability of modern language runtimes (Ruby, Java, Python, etc), or even being able to cross-compile to other platforms.
Docker solves the wrong end of application packaging by essentially packaging up the entire damn global (and non-portable) OS environment.
While that's the convention, I don't believe that will be the case going forward. I think it's a pretty negative and short sighted response.
I saw a demo very recently of someone creating an extremely barebones container - they were able to trace the exact dependency tree of an application, isolate it, and put it in to a Docker container. All that existed was apps+dependency, no userland. That's the future, imo.
> So much for the portability of modern language runtimes (Ruby, Java, Python, etc), or even being able to cross-compile to other platforms.
Funny you should mention that, because cross compilation happens in Docker all of the time.
Does that address your concern? If not I'd be happy to discuss further.
How do I run that on something that isn't Linux?
The Mac OS X portability story is to run a Linux virtual machine (!).
How is this a sane model when compared to building applications as a self-contained entity?
> Funny you should mention that, because cross compilation happens in Docker all of the time.
I can cross-compile a target for (Mac OS X, Linux, Windows, FreeBSD) and ship it as a self-contained application that runs on any of those systems?
Java did that and everyone went away from it since support became "least of all worlds" for anything non-trivial.
This changes everything!
The MS-bits they're contributing to the Docker project will be contributed in the same way everyone does - under the Apache 2 License, in the open, etc.
> Will it be in Go?
It will be contributed in the main docker repo - github.com/docker/docker
As a result, it will be a community/maintainer decision what language it's written in, but obviously we're heavily biased toward Go.
> Can you comment at this point about what containerization on Windows will look like.
I cannot directly.
> A tree of processes, a lighter, file-based HyperV, or something else?
HyperV integration will exist (and is currently being worked on in the open for Docker) but the technology is fundamental to Windows, like hyper-v. As far as I am aware, not a hyper-v decendant.
> What effect will this have on the filesystem layout?
Each container will have its own.
> Will native ACLs be supported or mapping to Unix permissions?
Can't comment on these details, sorry!
> Has a timeline for initial code in github been announced?
It has not.
They are process based. They do not depend on HyperV and can run on Windows Server on bare metal or inside any hypervisor or cloud.
Just makes this "partnership" a bit more, complicated? Or at least, not what it seems on the surface. If the containers aren't compatible between Linux and Windows, and the tooling will end up .Net in the long run(which I firmly believe), then the only common ground will be some semblance of API compatibility?
That said, it would allow for a more consistent developer experience for application/server development towards how things are deployed with containers on Linux. The biggest advantage OSX really has is that you get a clean UI for when you are simply a user, and a unix environment with the same tooling you use to build applications that will be deployed on Linux.
My true hope for this, was actually to see something beyond boot2linux. And, oh I really wish that HP didn't disable the virtualization support on their lower-end desktops (what I'm currently stuck with at work). VMWare workstation is really the only option for me, and running Ubuntu under that for *nix dev/testing.
Yes!
> How will that work?
The same way it works now. Find an image you want to run, and docker run it.
Why, let's open they are unlimited. :)
I'll be honest and say I don't have an answer. However, we're working on derive one and you'll know as soon as we do.
You can do this today with boot2docker (running a linux VM behind the scenes)
> pull a windows based image and run it on the linux docker platform host?
I suppose the opposite of what I said could be true using something like KVM, but not a use case we are seeing a lot of right now.
The goal is:
Package your Windows app in a docker container, use same tooling you would otherwise use to deploy to a docker engine running on a Windows host
Package your Linux app in a docker container, use same tooling you would otherwise use to deploy to a docker engine running on a Linux host.
Package your Windows app in a docker container, use same tooling you would otherwise use to deploy to a docker engine running on a Linux host and vice versa? If not initially is that an ultimate goal?
There are no current goals to make Linux executables run natively in windows and vice versa.
1. Native Mac OS X support (so Mac Apps can also be in containers) 2. Being able to mix container operating systems with other host operating systems.
If that ever happens then operating systems and their versions would pretty much no longer matter; you could run anything anywhere without compatibility issues. I feel like this is something that has to eventually happen no matter what I'm just always curious what form(s) it will take.
Are there Mac servers out there you want to run apps on? Or just consumer apps you want to run on your mac? Curious about the use case!
> Being able to mix container operating systems with other host operating systems.
VMs are that abstraction today. I think we can do better going forward, but still a lot to do with what we currently have planned.
Maybe they haven't but I feel like Apple has abandoned the Mac OS X Server so no apps at least for me in that case but I think consumer apps would be a pretty cool use case.
> VMs are that abstraction today. I think we can do better going forward, but still a lot to do with what we currently have planned.
Roger; thanks!
This appears to be aimed at SOHO users and not racks and racks of servers in a data centre.
I saw this... https://www.docker.com/community/governance/ but couldn't find any more information...
The first DGAB meeting is on 10/28, where we will propose a different way of working based on all of the feedback.
Stay tuned. We'll have a lot on this in the upcoming weeks.
Having some sort of translation that's not a VM as a part of the distribution mechanism doesn't make much sense, and breaking the portability of Docker comes at a significant cost.
I struggle to see how to make that work cleanly.
"Docker will be able to use either Windows Server or Linux with the same growing Docker ecosystem of users, applications and tools." (emphasis mine)
"bringing together Windows Server and Linux"
"making available some of the best images for Windows Server and Linux."
http://news.microsoft.com/2014/10/15/DockerPR/
Of course, it's hard to talk definitively about something that doesn't exist yet.
UPDATE: It makes more sense after reading this comment: https://news.ycombinator.com/item?id=8460164
Is there a missing comma in each sentence of the goal? It seems like there should be one after "use" or after "engine":
1) [after "use"] Package [on a Windows client] your Windows app in a docker container, use same tooling you would otherwise use [on a linux client], to deploy to a docker engine running on a Windows host Package [on a Windows client] your Linux app in a docker container, use same tooling you would otherwise use [on a linux client], to deploy to a docker engine running on a Linux host.
2) [after "use"] Package your Windows app in a docker container, use same tooling you would otherwise use to deploy to a docker engine [running on Windows or Linux], running [the packaging step] on a Windows host Package your Linux app in a docker container, use same tooling you would otherwise use to deploy to a docker engine [running on Windows or Linux], running [the packaging step] on a Linux host.
I added the implications I understood to highlight the difference the comma placement makes. If there is no comma it's pretty ambiguous / confusing to me on which platform the docker engine is running. I believe from this post of yours that I have misunderstood the announcement, and that windows apps will be able to be made into docker containers that can only run on windows docker engines.
I'm sorry to hear that, we will work to do better.
> and that windows apps will be able to be made into docker containers that can only run on windows docker engines.
This is correct. There's also a thread on how, to the user of docker who just wants to `docker run` something, the distinction doesn't really matter in the end
Containers share a kernel, this would make it impossible. However, with things like boot2docker (25MB Linux distro), this makes it really leight-weight/easy to deploy into a VM and run that way.
Will Docker be shipped as a part of a Windows Component & a paid service from MSFT?
http://www.microsoft.com/en-us/windows/enterprise/products-a...
ETA? :)
This could be huge for DR.
http://news.microsoft.com/2014/10/15/DockerPR/
"Microsoft Corp. and Docker Inc., the company behind the fast-growing Docker open platform for distributed applications, on Wednesday announced a strategic partnership to provide Docker with support for new container technologies that will be delivered in a future release of Windows Server."
I strongly suspected that Windows Server vNext would have some sort of 'container' support after the wild success of Docker.
That's when I knew we had a great potential to partner.
You can set a process's CPU schedule, you can also set it to a "background" priority which reduces both its CPU and IO to the lowest, but you don't really have fine-tune control of how much IO/Network a process has (irrespective of CPU).
This means on Windows in general it is very easy for a single process to run away with all of the machine's IO and there isn't a lot you can do about it except killing the process (I've seen backup clients, anti-virus scanners, etc do this).
It wouldn't be as big of a problem on Linux because first off the IO scheduler seems to be better, but even if it wasn't you can go in and manually configure a process's maximum throughput in a bunch of different ways.
If Windows didn't support this then what is stopping a single Windows-Docker container from stealing all of the system's IO or network capacity?
That gives you virtual LAN bandwidth control, "Storage QoS" (IOPS limits), partitioning, affinity and CPU limits. Not only that you have virtual SAN and network fabric. It's pretty awesome!
If you utilise it by deploying your app to a .wim file and then use wim2vhd and fart it at a Hyper-V host, docker already exists on Windows so I'm not sure what all the fuss is.
Yes we do this.
Hyper-V and similar are great. They aren't as good as containers however. You gain more security with Hyper-V but even assuming full hardware support it is still an expensive thing to be doing.
On a normal desktop machine you can run maybe 2-4 virtual machines (depending on a lot of factors). On that same machine you would want to see 4-10 containers each container doing one "thing" and one thing only.
I suspect that no one really cares. CPUs are pretty cheap. The problem that docker solves is "0 dependency single-step installs, and reliable rollbacks" not "VMs are too slow".
This is true on Linux too, btw- linux containers are still pretty feeble from a security perspective, without draconian seccomp sandboxing.
All the analogous tech to Docker has existed in the AWS+OpenStack ecosystem for a decade; it's just that nobody is willing to pay $50/mo per container. An entire huge company (Heroku) lives off the profit margin that is created by paying for VMs, then selling containers at VM prices.
The infrastructure benefits are completely there in terms of driving up utilization/density of the hardware you're buying. I feel (and know!) there will be a lot of optimization on making this happen.
On the other hand, there are some insecure multi-tenancy concerns for which VMs still offer better isolation. That's ok, because Docker is not about virtualization, it's about a platform for distributed applications.
Thus, finding the right/secure/optimal/? place to run your container should be straightforward and intuitive. That's the direction we're looking to take the industry.
I think what you really mean is that you can't control the overcommit policy on AWS. Otherwise, if VMs were too slow, why would you be running on AWS at all?
> Otherwise, why not just make every container a full EC2 instance?
A lot of people have full EC2 instances running a single container. Or full machines. If you want to upgrade your database software, atomic, full system rollback looks pretty enticing, especially if your database machines are huge, hugely expensive, and you don't have that many spares.
You can treat EC2 instances exactly the same as you treat docker containers—attaching EBS volumes to them in the way you'd attach data volume containers, attaching ENIs like you'd publish ports, etc. You can get exactly your "atomic, full-system rollback" just by having a CF template with a template parameter for the DB AMI to start in an Autoscaling Group, and then pushing a change to that variable. (Effectively, this gives you the same semantics as using Heroku's "config:set" CLI subcommand.)
But people don't do things this way. Why? Because the different pricing models create a different system of incentives around the two ecosystems. EC2 instances are thought of, fundamentally, as machines, rather than as application containers. There are adapters meant to make their usability for such more clear (e.g. Elastic Beanstalk), but in doing so they expose the relative absurdity of the VM pricing model.
As for apps, IIS makes a pretty good container system if you want to use it for that sort of thing.
I guess that at many places you'd just buy datacenter edition, but that's a bit too much for small businesses.
That said, yes, the current plan is that this kernel/service enablement comes with a future version of Windows to be able to run Windows Containers.
Congratulations to MS though, I think this is a good initiative. Not sure why you even need a partnership with docker TBH as they didn't create the underlying OS technologies that make containerization possible on Linux.. But, with the acquisition of Mojang it is apparent MS is placing a lot of emphasis on acquiring mind share.
Secondary, a native Darwin Docker server would be killer.
The mobile implications could be huge if this made it out of Windows Server.
I actually got unofficial approval to push my League of Legends linux image to the hub, which is an absolutely PITA to get installed right. I may do that; I don't play enough to maintain it.
What do you mean by get approval? Isn't LoL F2P? Did you have to get approval from the creator or Docker?
Great idea. I will definitely reach out.
> What do you mean by get approval? Isn't LoL F2P?
Just because it's F2P doesn't mean I have a license to distribute it.
> Did you have to get approval from the creator or Docker?
Docker is an open source project. So, in effect, all contributors (including me!) are creators of Docker :)
You'll have 2 hubs, the "windows hub" and the "linux hub".
This is incorrect.
That said, it's true that Windows won't all of the sudden gain the ability to run ELF.
Running boot2docker on windows is sort of cheating in that it is a virtual machine running docker apps and this effort appears to be a native dockerd running on Windows. Thoughts?
From any machine, you should be able to `docker run` any application, and docker will be smart enough to find (or create!) the best place run it.
The base pieces of the lego are currently in flight in Docker proper to enable this, with great tools built around the Docker API for more advanced topics (like clustering and scheduling)
> Running boot2docker on windows is sort of cheating in that it is a virtual machine running docker apps and this effort appears to be a native dockerd running on Windows.
I wouldn't say it's cheating. If your target is all linux machines, you need a cheap, local, efficient way of developing and testing those applications. Boot2Docker is an option (but not the only one - you could use a cloud service, internal vm infra, etc. etc.)
Love the dialogue - keep it coming.
If you're running docker on windows and you try to run a linux container, it will magically start boot2docker and start your linux container inside that vm if you are running on a windows box.
Now if that is true, my question would be: How do you boot2windows for windows docker images when you can't freely distribute a windows vm?
As I think both being able to support both would be amazing, I don't see how it fits into the current scheme of things.
Maybe! Or you've used something like the new `docker hosts` feature, which could create a new instance for you on any infrastructure provider. Or you could be pointing to a boot2docker node, etc.
> Now if that is true, my question would be: How do you boot2windows for windows docker images when you can't freely distribute a windows vm?
I don't have a good answer for this, as we do not have a boot2windows product or announcement (I assume this is the opposite case, if you have a linux machine and want to run a windows image)
My guess instead is you'd be pointing to a set of infrastructure that can find (or create) Windows Docker Hosts and it would run there.
See https://github.com/bfirsh/docker/tree/host-management
There's also a relevant thread on the docker-dev mailing list
that sounds promising...
The tangential benefits/integrations start to get really interesting.
Your app is a folder and that folder contains everything needed to run the application even up to the point of self-hosting with OWIN - IIS is not needed anymore!
(1) http://www.asp.net/aspnet/overview/owin-and-katana/an-overvi...
Docker will sit on top of that and provide the same client API you're accustomed to, but run Windows based images natively.
Office 365 client only?
I can't speak for Microsoft, but, there's been a ton of thought put in to this, but details subject to change as they start to contribute to the project.
That said, I imagine each container would need those base services (svchost.exe, as an example), its own registry, etc.
> How would you use the build file to install your dependencies without a package manager?
Lots of examples (in child comments), but more will be announced here soon.
[1] https://chocolatey.org/ [2] https://oneget.codeplex.com/
Windows already has a package manager called, creatively, PKGMGR. It is already used to install packages from the command line in core edition (or regular edition).
All great ideas and we are currently thinking though exactly the same ideas as well. Which of these ideas would you like to see happen first, what would be most important to you?
Not sure this is a good idea for Docker though. Doesn't it mean they lose their focus on Linux? Seems to complicate things a lot, and potentially introduce conflicts when prioritising what to do next. Simple is good. Serving a closed source operating system looks unwise, given the nature of their business.
I referred to this above, but big partners working on meaty objectives is actually a positive success criteria that the governance model is built for.
That said, we try very hard to ensure that the flow of the project remains unencumbered.
Docker is a community project; one of the beautiful benefits of that is that Docker can change as long as you can reach critical consensus on any design proposal.
Without the partnership, Docker could be brought to Windows at some undetermined point in the future as soon as those native capabilities existed.
With the partnership, we have a set of dedicated Engineers from Microsoft and the community who are eager to build consensus around getting Docker integrated to the new API/Services that they're creating as soon as possible.
There's no free lunch - if the contributions don't stand the scrutiny of the maintainers, they will continue to work with them until it does.
I hope that helps - happy to discuss further!
I wonder if Steven Sinofsky would have been game for this?
A lot of Windows developers use commercial Windows for development (i.e. Win7) with these features I anticipate more developers using (the more expensive) Windows Server.
If your build output could be a container (VS build process?) that you can ship, or as in a lot of "enterprise" organisations pass on to QA / UAT then this is a big deal and a massive huge step.
Any contribution by Microsoft to the Docker project (currently slated, the Docker Daemon for Windows) will be governed over said license.
What Microsoft chooses to do with Windows/kernel features/how that is licensed is left to their discretion.
How much would this cost and how is the pricing evaluated?
I don't know the price of the next version of Windows Server, but, Microsoft has committed to make Docker available on that.
So, to use Docker on the next version of Windows Server, it will be the cost of the Windows Server license.
I do realize corpo-world is filled with Windows Servers now but it seemed Linux/Docker could change this with containers as a 'standardized server app format' with super easy provisioning process.
Now since Windows will get more or less the same - Linux/Docker and Windows/Docker will compete on tools and raw perf.
Pretty cool evolution, isn't it?
In order words, will there now be Windows images and Linux images, and Windows images will run on Windows hosts and Linux images on Linux hosts?
More on that soon.
Is it something like an integration of Mesos, Fig and Docker?
It sounds like Docker is going to eventually put something out that touches on some of this stuff (although not necessarily self-healing etc.), just wondering how we can take advantage of that or avoid duplicating functionality.
Oracle DB is available. That's got to say something.
Not a bad strategy.
They haven't lost a bit of relevance with the majority of the paying user base. However the tech press like to spin it that way. When I see a Mac or a Linux box in a 2000+ seat corporate network, then perhaps I'll believe it. The only markets they aren't winning are the freshly created volatile ones.
There's a big difference between a software company equipping developers with Linux based machines and a midsize to large non-tech company equipping their users with Linux or even Macs.
Granted, I think Microsoft is losing their grip and non-tech companies (I work for one) are seriously beginning to consider PC alternatives. On the manager and executive level, and power user level, IT departments seem to be accommodating Macs more. For non-power users, web based machines like Chromebooks are becoming a more attractive each day.Spending time in the open source world with younger developers, it's clear the only value they get from Microsoft is for their gaming machines at home. While a number of them respect the development tools, it doesn't matter because Microsoft has done a great job tying it to the Windows titanic.
I recall interviewing with a number of companies in the mid 90's and talking with the programmers who worked with DEC based tools. The ones I talked to had nothing but good things to say about their tools and systems and how they had real implementations of various systems that PC's were trying to implement at the time. I can't help but look back and find myself in the same position on a course towards becoming irrelevant in 10-20 years.
I guess what I was trying to get at is Microsoft is going to have to make a considerable effort to focus on luring the generation they lost if they want to remain relevant. That generation of developers isn't simply going to pony up for slightly better tools. They demand having access to software when it's available so they can download it immediately with one of twenty different package managers.
I'm not predicting Microsoft's doom, but I am saying that Microsoft is facing an incredible number of challenges that will make "do everything" a bad strategy. On too many fronts they're faced with competition that ranges any from inferior to superior, but free. In development tools, they're up against a huge community of open source tools with some of them funded by huge companies whose revenues don't depend on the sale of software. MS Office's share is eroding, not just to direct competition, but to indirect alternatives that posit that complicated word processors and spreadsheets are the wrong answer.
I also don't think Microsoft can do everything because software is going through a Cambrian type of explosion where you're seeing all sorts of manifestations of species and hybrids. Sure a lot of these species will die out, but as we witness new species of databases and operating systems it will be difficult for a large company like Microsoft to predict which ideas it needs to pay attention to and which to ignore. Adapting the complicated markets is an incredible challenge, but there are companies that have done a remarkable job staying on top.
I should have added that I'm a .NET developer who's recently returned working with in an open source shop. The problem I see is not that they don't have good tools or relevant solutions for real world problems. The problem I see is I'm in my late thirties and I find myself to be one of the younger developers when I go out to MS related user groups and events.
Spending time in the open source world with younger developers, it's clear the only value they get from Microsoft is for their gaming machines at home. While a number of them respect the development tools, it doesn't matter because Microsoft has done a great job tying it to the Windows titanic.
I recall interviewing with a number of companies in the mid 90's and talking with the programmers who worked with DEC based tools. The ones I talked to had nothing but good things to say about their tools and systems and how they had real implementations of various systems that PC's were trying to implement at the time. I can't help but look back and find myself in the same position on a course towards becoming irrelevant in 10-20 years.
I guess what I was trying to get at is Microsoft is going to have to make a considerable effort to focus on luring the generation they lost if they want to remain relevant. That generation of developers isn't simply going to pony up for slightly better tools. They demand having access to software when it's available so they can download it immediately with one of twenty different package managers.
I'm not predicting Microsoft's doom, but I am saying that Microsoft is facing an incredible number of challenges that will make "do everything" a bad strategy. On too many fronts they're faced with competition that ranges any from inferior to superior, but free. In development tools, they're up against a huge community of open source tools with some of them funded by huge companies whose revenues don't depend on the sale of software. MS Office's share is eroding, not just to direct competition, but to indirect alternatives that posit that complicated word processors and spreadsheets are the wrong answer.
I also don't think Microsoft can do everything because software is going through a Cambrian type of explosion where you're seeing all sorts of manifestations of species and hybrids. Sure a lot of these species will die out, but as we witness new species of databases and operating systems it will be difficult for a large company like Microsoft to predict which ideas it needs to pay attention to and which to ignore. Adapting the complicated markets is an incredible challenge, but there are companies that have done a remarkable job staying on top.
I had the same response to the predicted demise of Palm and Blackberry. I'm not saying that I know the outcome but that current success is not a predictor of future performance, especially in an industry where disruption is such a focus.
Good luck to Docker.
Our website (docker.com) and the Docker Hub (hub.docker.com), our hosted service, are managed directly by Docker employees and offer a better maintained and security-focused cipher suite than our blog.
Unfortunately, we are aware that the blog's SSL/TLS configuration is not ideal. It is on externally hosted infrastructure and we have been working with our service provider to rectify these issues. We have high expectations of the services we consume and will do what is necessary for the long-term security and confidence of our users.
Or the GUI exe will be also isolated from each container session?
Docker people will spend more time fixing and making bugs due to Microsoft, the documentation will become a mess (it cant possibly be same documentation for that different systems), the Linux-side of docker will be more mediocre compared to what it could have been if all effort went to it.
Remember Internet Explorer 6, Visual Basic, the horror that is Excel and the whole Office suite, Asp.net, Windows Millenium, The attempt to kill Linux by Microsoft through SCO.
This isn't any different than Red Hat contributing the Device Mapper backend to Docker so it can run unmodified on RHEL. Nothing about the UX changed/fragmented from there.
Otherwise existing Dockerfiles will not build on Windows - right?
It's still a compelling option, but I'm not quite certain of the value that it will bring to the table here.