Slashdot Mirror
Microsoft Releases Windows Server 2012
Barence writes "Microsoft has released Windows Server 2012, letting businesses test it for 90 days on the Azure cloud platform for free. There are two versions of the main edition of Windows Server 2012: one with virtualization support and one without. The former, the Data Center version, costs $4,809, while the Standard edition will cost $882. There's also an Essentials version, which replaces Small Business Server, for $501 per server, and Windows Server 2012 Foundation, which will only be available pre-installed on hardware."
Ars has a detailed look at the new edition.
$4k to enable visualization support (that the code already is there for?)
Yet MS wonders why they have such a comparatively tiny market share of the server market...
For large sets, this will be our guide even unto death, for the LORD will work for each type of data it is applied to...
... it will need Metro-style management tools!
.sig: No such file or directory
Functionally, Standard and Datacenter are the same. Even things like clustering, which used to be the sole preserve of the higher-end Windows Server SKUs, are found in Standard. The only difference is the number of Windows Server virtual machines supported per license.
So again: The only difference between the Standard and Datacenter is the licensing. Same software, two licenses.
That is up to you.
There is no increased CPU count. Both Standard and Datacenter support 2 CPUs per license.
With Datacenter you get unlimited (Windows) VMs, so if you run more than 10 Windows VMs on a (2 CPU) box, it is cheaper.
For less dense virtualization, use Standard licenses, as each give right to two VMs.
3 days of grubbing around in the registry and it still doesn't work.
On the linux servers, The same task was done with 3 iptables lines.
including the "service iptables save" .
I was underwhelmed.
Run the 99% of commercial apps that are coded agaist the win32 api in a supported manner? Have vm management tools that don't suck horribly? I could go on but I'd just be further feeding the troll.
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
1) You are unable to grow a neckbeard.
2) You've had sex without having to pay for it.
So bad and yet still miles better than any Linux based operating system or OS X.
That'll be why the world runs on Windows servers and no-one would think of putting any critical service on Linux.
there will be a disaster in 2012
"Lame" - Galaxar
We run a heterogeneous shop split about 50/50 between Linux (Debian) and Windows (2003/2008). Windows excels at certain things, Active DIrectory, and running .net apps delivered to us by various contractors. Our Linux systems run mission critical services as well as file-servers, and virtualization via VMWare's ESXi products (horribly overpriced but it's the situation that I inherited). I poke fun at the Windows guys fairly often and I get joked at in return, but the reality is that we all realize that it's about the right tool for the job. I don't have a single metal Windows install at home and I don't feel at all left out of the commercial loop, but like everything in life your own mileage will vary.
"Never let your sense of morals prevent you from doing what is right" - Salvor Hardin
(posting anonymous for obvious reasons)
It only took you three days. We were dealing with a screwy Microsoft Lync mobility issues whereby the iOS client just wouldn't work (but every other client under the sun worked). The only odd-ball thing about our setup was one of the four servers (at least four are required for any Lync deployment) was a Linux box acting as a reverse proxy. We opened up a ticket with Microsoft on April 30, 2012. The time spent with them since is a waste of time:
* We repeatedly requested the actual HTTP request/response data from the iphone's perspective, annotated with notes on how it differs from what the iphone expected. Every time we requested it, they provided us with the client's general iphone debug log (which was useless to us), even though we explained that it doesn't fulfill our request.
* We asked for details on what is expected of the Lync reverse proxy. They provided us with instructions on how to set up TMG. We replied that the provided information did not fulfill the request. Their response was a shrug and another link to the same instructions.
* We asked if there was anything specific to the iOS client that required ISA or TMG. They demurred on it, refused to research it, refused to acknowledge the bug for *four* months. I'm not exaggerating. It was August 31 when we inferred from the continued back and forth that the only way Microsoft can hope to grasp the problem is to make the reverse proxy an ISA server.
From this, I learned that Microsoft support really isn't much better than doing it yourself. They have no inside tricks, they have no way of getting a guru to weigh in on anything, and they hope that by sending you the same wrong information over and over they won't have to acknowledge faults in the product.
For my part, calling Microsoft support isn't an option any longer. It is a waste of time and money that could be better spent solving the problem myself.
It's probably time to seriously consider moving from 2003 to 2007.
Some mornings it's hardly worth chewing through the restraints to get out of bed.
So basically, Windows is the right tool for things that only run on Windows ... otherwise, use Linux.
Sure, but right now you can get server core and hyper-v standalone and run many virtual servers, but in the future if you want to stay current, at some point you're going to have to pay through the nose.
That has got to be one of the better examples of properly applied sarcasm I've see here in a while.
Good play...
It's POSIX-complaint? :|
That'll be why the world runs on Windows servers and no-one would think of putting any critical service on Linux.
The Oracle world (big business, government) is definitely running on Linux instead of Windows. With the decline of Unix running on "big iron", with the exception of IBM's RS/6000 and AIX being the last holdout, everyone is moving their enterprise, mission critical apps to Linux. Especially with Oracle themselves releasing a tweaked version of RHEL, Linux is an "officially supported" platform that even satisfies the corporate PHBs and bean counters.
I make a pretty good living porting Oracle enterprise databases and apps to Linux. Just a couple weeks ago, we ported a Windows-based Oracle WebLogic middleware server from Windows to OEL Linux running on the very same piece of hardware, and got a tenfold boost in performance. With results like that, business loves Linux now.
Granted, only server-side things on Linux are welcome in the business world. The desktop will sadly *never* be adopted in any significant numbers in any enterprise. All because Windows and Active Directory rule that market segment.
So basically, Windows is the right tool for things that only run on Windows ... otherwise, use Linux.
That sounds about right to me.
"Never let your sense of morals prevent you from doing what is right" - Salvor Hardin
Active Directory is worth the price of Windows Server alone, and I say that as a Linux sysadmin who's implemented an OpenLDAP infrastructure (everything from AuthZ/AuthN to Puppet ENC backend to a single point of truth for Nagios). AD is miles away from anything any Open Source or Apple product has ever implemented.
For server functionality pure bullshit. I have a decade's experience running Windows and *nix servers, often in the same networks and while Windows has AD and GPOs to its benefit, in other respects it is horribly backwards and painful to use. Just backing up the system config in Windows is appallingly difficult compared to *nix.
The world's burning. Moped Jesus spotted on I50. Details at 11.
You are correct, and the two replies to you are lies. Datacenter gives you UNLIMITED guest OS CALs.
This site is pathetic. The amount of linux shilling that goes on here is sad.
Haha.
On a serious note, though, you actually can run POSIX apps on Server 2012. NT has, since its inception, included support for POSIX APIs and filesystem behavior. These days it's called SUA (Subsystem for UNIX Applications) and a smallish but fully functional operating environment for it, called Interix, is available for free. The installer will also let you enable various tweaks such as SetUID/SetGID behavior and filesystem case sensitivity, things you can't get with Cygwin or the like. It's implemented as an NT subsystem, same as Win32, so the speed is basically native as well. Interix comes with a working build toolchain, plus you can get a package manager for a repository of precompiled software and updates from http://suacommunity.com./
I'm not sure I'd advocate adopting it at this point if you haven't already - MS has been making moves toward discontinuing support for some years now, and it appears to no longer be in any of the client editions but Enterprise - but it exists, and it works. MS themselves used it to host Hotmail on Apache before they ported it to run on IIS. I use it (on client) both for various utilities that I prefer the POSIX versions of (git and ssh and such, plus sometimes there is no Win32 version) and for bash (my primary shell).
There's no place I could be, since I've found Serenity...
So what exactly does it do that similarly equipped Linux machines/vps' can't do that justify the cost?
* New resilient file system ReFS (think BtrFS when completed)
* Storage Spaces (think ZFS storage pools)
* SMB 3.0 - higher performance network transfer, transparent failover, SMB scaleout (multiple servers serve same shares and aggregates bandwidth), SMB Direct (efficient remote direct memory access), SMB Multichannel, Volume Shadow Service (VSS) for SMB file shares, SMB encryption, SMB Directory Leasing (negotiates and updates local caches of metadata over slow networks)
* Dynamic access control (claims and policy based access control). Think SELinux, grsecurity. Access control based on what application the user is running (sandboxing), from what type of device the user is accessing the resource, on other user attributes than security groups (e.g. who is the manager, what department does the user belong to etc), access control based on attributes of the file (e.g. classification, select words of a Word document)
* RemoteFX improvements, e.g. virtualized GPUs (can use local or remote shared GPUs during RDP sessions), remote low-latency multitouch.
* Direct Access over IPv4. Think hassle-free VPN.
* Hyper-V 3: ethernet cable live migration (neat trick) lets you migrate VMs off one server onto another server over the network without the servers sharing anything. Many Hyper-V manageability improvements. Crazy scalability, e.g. a 63-node Hyper-V cluster runs 4000 concurrent VMs simultaneously. Hyper-V replica.
* Server manager: Yes, a Metro (oops - "Modern") style management app for multiple servers. Integrates with response files and powershell workflow scripts to manage multiple computers (servers/workstations) at once - e.g. install new software, perform configure actions.
* PowerShell 3 with new features such as resilient remote connections (you can detach from a remote session and pick it up later/from another device), workflow scripts which can perform actions with suspend/restart/repeat semantics. No, not just "suspend process" - but actually persisting the state of a script to be continued later, e.g. after a computer restart (or from another machine).
* Thousands of new PowerShell cmdlets (many/most automatically derived from WMI providers) to control virtually anything on local or remote computers.
* Block sized data de-duplication
These are features I could find by googling. I'm sure there are more. Obviously not all of them will appeal to Linux enthusiasts. But still...
Reading slashdot one-liner: (irm http://rss.slashdot.org/Slashdot/slashdot).rdf.item | fl title,desc*
Yet MS wonders why they have such a comparatively tiny market share of the server market...
According to this arstechnica article (2011), Microsoft had a 25% webserver market share (IIS) as of 2010, and 15% as of 2011. For standard servers, they accounted for 71% of all quarterly server shipments (original source, IDC). According to a survey in 2010 (the only one I could find on smtp market share, and was linked in Wikipedia), Exchange is the third most popular SMTP server (17%-- behind exim @ 34% and postfix @ 21%, and just ahead of sendmail).
You can call that many things, but "comparitively tiny" it isnt. Microsoft server is remarkably popular in SMB situations, and even in larger companies, and trying to write it off as irrelevant or whatever your angle was is silly.
Also silly is the comment about "code already there"-- EVERYONE does this, from RedHat to VMWare to Adobe any other company that sells multiple tiers of its software product.
Wait, every single other client works, IOS doesnt, and your analysis is "Must be Microsoft's fault"? And you asked MS support for IOS details, and then wondered why they gave you the cold shoulder?
Seems to me youre better off bringing apple support in on this, or focusing on the "what is IOS doing wacky" rather than "what is IIS doing wacky".
If all they need is a file server and theyre happy with a workgroup, theres no reason to do with Windows Server at all-- there are many NASes out there that will fit the bill, or you could build your own and stick some distro on it (not like theres a shortage of SOHO fileserver distros out there).
I find backing up the registry in a fashion that allows me to easily restore configurations a real pain. NTBackup and it's descendants are hardly backup wonders. Configuration via text file is infinitely easier to deal with than binary hives.
I don't even bother restoring failed domain controllers any more. I have other DCs replicating AD data so I just build a new server, promote it to a DC and let replication do the heavy lifting. Hrlluva lot easier than what passes for bare metal recovery in the Windows world.
The world's burning. Moped Jesus spotted on I50. Details at 11.
The thing with *nix, or at least any version I've worked with, the functionality is already there. Configurations are almost always in human-readable text files, and I have a toolset that has been around in one form or another for decades to work with those files. I can easily make backups of daemon configurations, and indeed have been able to restore a server with the contents of /etc and the data files.
The world's burning. Moped Jesus spotted on I50. Details at 11.
For server functionality pure bullshit. I have a decade's experience running Windows and *nix servers, often in the same networks and while Windows has AD and GPOs to its benefit, in other respects it is horribly backwards and painful to use. Just backing up the system config in Windows is appallingly difficult compared to *nix.
So, how does Linux handle online backups of running server workloads? Does Linux have a way to signal to running services (like RDBMSs, hypervisors, file servers) that a backup is about to happen, negotiate which files are to be included in the backup and then in a fragment of a second work with the running service to synchronize disk content so that the backup will be consistent?
A running database server will almost invariably hold some state in memory. If the power is lost it will be able to rebuild from the disk state, but that can be a time consuming task. If the backup system is simplistic it will just back up the disk state of any file. Upon restoring it will appear as if the power was lost and the roll-forward log will have to be played.
A more advanced backup system will integrate with the services to ensure that for a very brief time (just enough to take a snapshot) the disk state is consistent and thus will not require a rebuild/roll forward if it is ever restored.
Windows comes with Volume Shadow Copy Service (VSS) and a file system which supports block level snapshots. VSS works with VSS aware applications (VSS writers) such as Microsoft SQL Server, Oracle Database Server, Exchange Server, Active Directory, NTFS and Hyper-V server. When a service is a VSS writer it participates in VSS coordination/synchronization to create consistent disk state.
It even works through Hyper-V: When you back up the Hyper-V host, Hyper-V itself is a VSS writer which recursively invokes the VSS running inside guest OSes (if Windows) to ensure that any service inside the Hyper-V guest OS is also disk consistent exactly when a snapshot of the virtual hard disk image is created.
To my knowledge, Linux doesn't have anything like VSS. Which means that each application/service must be handled separately. Typically you will stop the service during the backup. Some services such as PostgreSQL can recover from a non-consistent disk image; others can not. Individual applications may have commands/services which allow admins to "dump" state to a file to be backed up separately. All in all reliably backing up a running Linux server is more complicated compared to backing up a running Windows server with VSS aware services.
Reading slashdot one-liner: (irm http://rss.slashdot.org/Slashdot/slashdot).rdf.item | fl title,desc*
That's utter nonsense. Windows Server Backup is about a billion times better then NTBackup. Pure image based backup, allowing multiple versions of files to be stored, Exchange aware, SQL aware and allowing individual files to be restored, easily. I would use WSBU over NTBackup any day of the week (and do). It works every time - and offers damn near instant bare metal recovery of corrupted servers. NTBackup, on the other hand, required you to rebuild from scratch and then manually restore files, apps, etc, painfully.
Just because you never learned how to use a tool doesn't make it bad. It is trivial to configure WSBU to backup individual components, such as system state, volumes or yes, even individual folders. Again - *you* not knowing how to do something doesn't make it impossible.
And for the obligatory Slashdot 2012: no, I am not paid or affiliated in anyway with Microsoft. Sometimes people like the changes they make because they actually tried them and found them better.
Personally with all of the Enterprise level support I've dealt with (e.g. IBM, EMC, HP, Dell, Oracle, CA, etc.) Microsoft is among the best.
(I'm talking Enterprise support, as in paying 7 figures/yr for licensing and support. Not calling an 800 number to India for someone to tell you to reboot your computer as you would get from a el-cheapo Wal-Mart laptop.)
Don't get me started on Oracle. Most of the time the problem I'm calling about is less painful than dealing with Oracle support.
Any insufficiently advanced magic is indistinguishable from technology.
Hyper-V Server is still free.
Yep, and regedit is as intuitive as vi.
That ars technical article makes the same mistake so many others do. It confuses hostnames with servers. It assumes a 1:1 ratio of servers to host names, and that is nowhere near the case. It also confuses "apache" and "iis" with windows and non-windows. There are lots of apache servers running on Windows out there (mostly because they have apps that require a java application server like tomcat and apache is typically used on the front end of tomcat, although IIS can be used as well).
The fact is, Windows web servers tend to have fewer domain names per server than Apache because Windows is used more commonly in enterprise environments while Apache is used more commonly in web hosting environments.
What that all boils down to is that any technical "journalist" who quotes Netcraft's host survey as evidence of server installation numbers is a moron.
If you need web hosting, you could do worse than here
Sorry, but at least two of your points are factually incorrect.
* ReFS is lacking a few notable features, including file compression / encryption, sparse files, hard links, extended attributes, disk quotas, and others[1]. You could say that the only notable improvements over NTFS that it has would be much improved resiliency and higher capacity limits. You can't compare this to BrtFS. At all. The two aren't even in the same ballpark. ReFS is there to store millions of large files and managed bad blocks in a smart way without taking the volume offline. It supports little else.
* Dynamic access control can't even be compared to SELinux. SELinux can restrict a program to running from a certain location, it can restrict which ports in the TCP/IP stack it can/can't open, it can restrict which hosts a specific process can talk to, and yes, it can alter the fundamental view of the file system hierarchy based upon access levels granted. Dynamic access control is really just more complexity in the form of an ACL on top of the already present windows file system ACLs, and it impacts nothing outside of files[2]. Now, you can use claims (which dynamic access control is built upon, at least partially) to control other aspects of your environment, but that isn't "dynamic access control" as far as MS is concerned. Further, it really is another layer of complexity -- if your claims server (which is a web server(!)) goes down, you're losing access to stuff (but if you're a decent sized MS shop, this will likely not be an issue, as you're already maintaining decent uptime on your DCs). Then the file system level ACL comes into play again. It's going to be crazy stupid hard to diagnose a claims access issue in a large production environment, no matter what MS has done towards fixing these issues. Somewhat amusingly, dynamic access control isn't supported on ReFS at all [2].
Now normally I'd just trust you that you googled around to find this stuff, but you've got some powershell in your signature, which leads me to believe that you've done a bit more checking than the "stereotypical slashdot linux sysadmin" and this only goes towards scaring me a bit.
[1] http://blogs.msdn.com/b/b8/archive/2012/01/16/building-the-next-generation-file-system-for-windows-refs.aspx
[2] http://technet.microsoft.com/en-us/library/hh831717.aspx
So in other words, by your own description, things that you can already get in linux.
BtrFS has not been completed yet. ReFS is shipping. ReFS will not have all the features of the completed BtrFS, but for now ReFS offers features not available in any shipping Linux.
I don't think ZFS is production quality on Linux yet either. Storage Spaces under Windows is nor shipping.
Dynamic Access Control actually ups the ante for SELinux, grsecurity apparmor etc. While it still protects access to resources it does so based on potentially very fine grained policies which can express rules based on a very wide range of properties. And it brings claims based security all the way into the primary access control of an OS. Linux does not sport claims based security.
Ok, things linux doesn't have yet, but are on the way.
Sure. I am not aware of any effort to bring something like VSS to Linux, though. Windows now extends VSS to remote file systems (shares), which means that clients can ensure consistency even if an application/service stores files remotely (e.g. a SQL Server keeping it's data files on a remote server).
I really don't understand what this is. An automagic VPN? Doesn't sound all that special. NetworkManager has been able to do system-wide VPN connections for a while now.
Yes, an automagic always-on, bi-directional VPN on steroids. No calling, no VPN client installations. Just take the laptop outside the perimeter and it is still connected, still secured, still managed.
So the equivalent of what you can already do on linux with a combination of SSH, Puppet/Chef, and Screen. Admittedly an improvement for Windows, but this has always been a strength with linux.
All in all a meh, in my opinion. If you really have a need for the high-end features, perhaps Microsoft is offering at a competitive price. But otherwise doesn't seem to offer much that you can't already get with a linux, bsd, or solaris distribution.
Uhm, not quite. But unless you experience the new Server Manager you are not going to understand. It has this "declarative" feeling - comparable to controlling your network with declarative network policies as opposed to relying on scripts running on each node to set thing up.
Reading slashdot one-liner: (irm http://rss.slashdot.org/Slashdot/slashdot).rdf.item | fl title,desc*
Last time I looked at it was when Server 2008 was released. This isnt an issue of "didnt take the time to learn it"-- at that time, the official stance as given on the official Exchange team blog was "it was crippled it so that noone would make the mistake of using it for business". To reiterate-- this was the OFFICIAL exchange blog, ie microsoft employees.
Its entirely possible that in the time since they have corrected the issues I mentioned, or brought it back as something new-- but they definately DID cripple the built in backup on the release of 2008. Im not sure how possible it would be to find that article as it was a blog entry and it was 5 or so years ago, but Ill give it a shot and post it here if I do manage to find it.
I found it:
Windows server backup not exchange aware
We have decided to develop and release a VSS-based plug-in for Windows Server Backup that will enable you to properly backup and restore Exchange 2007 with a built-in Windows 2008 backup application.
While you will be able to backup and restore Exchange 2007 on Windows 2008, you should not expect feature parity with the Windows 2003 NTBackup experience.
The removal of NTBackup / its (known) inferior successor:
(Reasons listed there roughly boil down to, 1) most people get third party software; 2) ntbackup was never meant to be an enterprise solution; 3) we think optical media is the future and that tape sucks)
There are lots and lots of other posts on this. More to the point, the features you mention are brand new as of R2-- they were not there in the original release:
Windows Server Backup in Windows Server 2008 R2 includes the following improvements:
More flexibility in what you can back up. Windows Server Backup enables you to back up selected files instead of full volumes. You can also exclude files based on file type and path.
That is, you simply couldnt do this prior to R2, which, along with no tape and no exchange support, made it utterly fall off of my (and many others') radars as utterly irrelevant. Basically all of the cool features you mention simply werent there in the initial release-- it was a straight dumb "image the whole box or nothing at all" program, except it wouldnt even work if you had stuff like Exchange or HyperV and no VSS plugin.
Not only that, but even if I had noticed that release-- which TBQH i did not-- NTBackup was already such a disaster that I would be hesitant even now to return to something like WSB.
It sounds like your experience is mostly with Win Server R2 and above, which is fine; if thats true, just keep in mind that there are a lot of us with horror stories of NTBackup, and that WinServer2008 was not always as polished as it is now.
You are correct, and the two replies to you are lies. Datacenter gives you UNLIMITED guest OS CALs.
This site is pathetic. The amount of linux shilling that goes on here is sad.
Funny that somebody posting anonymously accuse others of lying without presenting any references to support the claim.
Well, here is the fact: http://www.microsoft.com/licensing/about-licensing/client-access-license.aspx#tab=2 "Windows Server Per-Processer licensing also requires a CAL"
The truth is never shilling (or penny, or pound...)