Imagine for a moment that you are an IT Professional charged with the care, feeding, and security of a classic Wide Area Network (WAN). Further, assume that, like any properly-designed WAN, your remote networks (whether MPLS or classic Hub-spoke) egress their internet connections directly, that is to say, internet traffic from remote networks isn’t back-hauled to your datacenter or HQ.
In such a scenario, you will need to have a list of each remote network’s public IP address and other pertinent details in order to manage routing and security at each branch. In my case, I needed up-to-date public IP address information in order to properly segment & report on internet traffic traversing our SSL/TLS proxy inspection service, Zscaler.
So how would you do this? An earlier version of myself, say 15 years ago, would respond this way:
I’d remote desktop to a node in each remote network, open up a browser window, and visit IPChicken.com.Then I’d carefully copy/paste the IP address details into my Excel document, and happy days! – Jeff, 15 years ago
Wrong answer, Jeff from 15 years ago! That’s bad practice, takes way too much time, involves using the cursed mouse, and is fraught with security risk because it involves browser use.
Fortunately, there is a much better, simpler, faster and more secure way to do this. Even better, it involves my favorite tool in the world, Powershell, as well as IPInfo.io, a web service that blows IPChicken.com out of the water.
Best of all, you can do it all without your hands ever leaving your keyboard. Check it out
Let’s use Powershell’s invoke-webrequest cmdlet to see what IPInfo.io returns to us:
JSON, if you’re not familiar with it, is an open standard that has superseded-in practice- XML and other structured document standards. It’s in widespread use across the internet, and it’s really great for us Windows admins that IPINfo.io feeds us a JSON response to our query. Why?
Because we’ve got Powershell to make it look pretty for us! We just need to pipe the results of the invoke-webrequest command into the handy convertfrom-json cmdlet. Voila!
This is great, now I’ve got high-quality IP Information on my workstation. So how do I scale this out to my remote WAN networks? how do I get the public IP address of my Lake Winnepesaukee branch office using Powershell?
Assuming you’ve got a Windows domain and have configured Windows Remote Management in a secure fashion, the way to do this is simple. Let’s use Powershell to tell a WIndows node at each branch to fetch us the public IP address it’s sitting behind, format it in a pretty way, and bring it back to my beautiful blue console. In fact, let’s do all the branches at once by using invoke-command:
Boom! That’s how we do it in 2017! It took less than 20 seconds to invoke our simple invoke-webrequest + convertfrom-json command across five remote hosts. No remote desktop needed….all of it done securely via secure WinRM which I’ve set up my nodes to listen for.
With these results in your console, it’d be trivially easy to dump out each WAN’s public IP information into a CSV, or, even better, create a new Excel spreadsheet using new-comobject and save/send the information from there.
What do you get when you take an IT Systems Engineer with more time on his hands than usual and an unfinished home project list that isn’t getting any shorter?
You get this:
That’s right. I’ve stood-up some IP surveillance infrastructure at my home, not because I’m a creepy Big Brother type with a God-Complex, rather:
Once my 2.5 year old son figured out how to unlock the patio door and bolt outside, well, game over boys and girls….I needed some ‘insight’ and ‘visibility’ into the Child Partition’s whereabouts pronto and chasing him while he giggles is fun for only so long
My home is exposed on three sides to suburban streets, and it’s nice to be able to see what’s going on outside
I have creepy Big Brother tendences and/or God complex
I had rather simple rules for my home surveillance project:
IP cameras: ain’t no CCTV/600 lines of resolution here, I wanted IP so I could tie it into my enterprise home lab
Virtual DVR, not physical: Already have enough pieces of hardware with 16 cores, 128GB of RAM, and about 16TB of storage at home.
No Wifi, Ethernet only: Wifi from the camera itself was a non-starter for me because 1) while it makes getting video from the cameras easier, it limits where I can place them both from a power & signal strength perspective 2) Spectrum & bandwidth is limited & noisy at distance-friendly 2.4GhZ, wide & open at 5ghz, but 5 has half the range of 2.4. For those reasons, I went old-school: Cat5e, the Reliable Choice of Professionals Evereywhere
Active PoE: 802.3af as I already own about four PoE injectors and I’ve already run Cat5e all over the house
Endpoint agnostic:In the IP camera space, it’s tough to find an agnostic camera system that will work on any end-device with as little friction as possible. ONVIF is, I suppose, the closest “standard” to that, and I don’t even know what it entails. But I know what I have: Samsung GS6, iPhone 6, a Windows Tiered Storage box, four Hyper-V hosts, System Center, an XBox One and 100 megabit internet connection.
Directional, no omni-PTZ required: I could have saved money on at least one corner of my house by buying a domed, movable PTZ camera rather than use 2 directionals, but 1) this needed to work on any end-point and PTZ controls often don’t
And so, over the course of a few months, I picked up four of these babies:
I liked these cameras from the start. They’re housed in a nice, heavyweight steel enclosure, have a hood to shade the lens and just feel solid and sturdy. Trendnet markets them as outdoor cameras, and I found no reason to dispute that.
My one complaint about these cameras is the rather finicky mount. The camera can rotate and pivot within the mount’s attachment system, but you need to be careful here as an ethernet cable (inside of a shroud) runs through the mount. Twist & rotate your camera too much, and you may tear your cable apart.
And while the mount itself is steel and needs only three screws to attach, the interior mechanism that allows you to move the camera once mounted is cheaper. It’s hard to describe and I didn’t take any pictures as I was cursing up a storm when I realized I almost snapped the cable, so just know this: be cognizant that you should be gentle with this thing as you mount it and then as you adjust it. You only have to do that once, so take your time.
Imaging and Performance:
Trendnet says the camera’s sensor & processing is capable of pushing out 1080p at 30 frames per second, but once you get into one of these systems, you’ll notice it can also do 2560×1440, or QHD resolutions. Most of the time, images and video off the camera are buttery smooth, and it’s great.
I’m not sharp enough on video and sensors to comment on color quality, whether F 1.2 on a camera like this means the same as it would on a still DSLR, or understand IR Lux, so let me just say this: These cameras produce really sharp, detailed and wide-enough (70 degrees) images for me, day or night. Color seems right too; my lawn is various hues of brown & green thanks to the heat and California drought, and my son’s colorful playthings that are scattered all over do in indeed remind me of a clown’s vomit. And at night, I can see far enough thanks to ambient light. Trendnet claims 100 foot IR-assisted viewing at night. I see no reason to dispute that.
Let the camera geeks geek out on teh camera; this is an enterprise tech blog, and I’ve already talked abou the hardware, so let’s dig into the software-defined & networking bits that make this expensive project worthwhile.
Power & Networking
These cameras couldn’t be easier to connect and configure, once you’ve got the power & cabling sorted out. The camera features a 10/100 ethernet port; on all four of my cameras, that connects to four of Trendnet’s own PoE injectors. All PoE injectors are inside my home; I’d rather extend ethernet with power than put a fragile PoE device outside. The longest cable run is approximately 75′, well within the spec. Not much more to say here other than Trendnet claims the cameras will use 5 watts maximum, and that’s probably at night when the IR sensors are on.
From each injector, a data cable connects to a switch. In my lab, I’ve got two enterprise-level switches.
One camera, the garage/driveway camera, is plugged into trunked, native vlan 410 port on my 2960s in the garage,
The other switch is a small CIsco SG-300 10p. The three other cameras connect to it. The SG-300 serves the role of access-layer switch and has a 3x1GbE port-channel back to the 2960s. This switch wasn’t getting used enough in my living room, so I moved it to my home office, where all ports are now used. Here’s my home lab environment, updated with cameras:
Like any other IP cam, the Trendnet will obtain an IP off your DHCP server. Trendnet includes software with the camera that will help you find/provision the camera on your network, but I just saved a few minutes and looked in my DHCP table. As expected, the cameras all received a routable IP, DNS, NTP and other values from my DHCP.
Once I had the IP, it was off to the races:
Set DHCP reservation
Verify an A record was created DNS so I could refer to the cameras by names rather than IP
Trendnet is nice enough to include a fairly robust and rebadged version of Luxriot camera software, which has two primary components: Trendnet View Pro (Fat Client & Server app) and VMS Broacast server, an http server. Trendnet View Pro is a server-like application that you can install on your PC to view, control, and edit all your cameras. I say server-like because this is the free-version of the software, and it has the following limits:
Cannot run as a Windows Service
An account must be logged in to ‘keep it running’
You can install View Pro on as many PCs as you like, but only one is licensed to receive streaming video at a time
Upgrading the free software to a version that supports more simultaneously viewers is steep: $315 to be exact.
Naturally, I went looking for an alternative, but after dicking around with Zoneminder & VLC for awhile (both of which work but aren’t viewable on the XBox), I settled on VMS Broadcast server, the http component of the free software.
Just like View Pro, VMS Broadcast won’t run as a service, but, well, sysinternals!
So after deliberating a bit, I said screw it, and stood-up a Windows 8.1 Pro VM on a node in the garage. The VM is Domain-joined, which the Trendnet software ignored or didn’t flag, and I’ve provisioned 2 cores & 2GB of RAM to serve, compress, and redistribute the streams using the Trendnet fat client server piece as well as the VMS web server.
On that same Windows 8.1 VM, I’ve enabled DLNA-sharing on VLAN 410, which is my trusted wireless & wired internal network. The thinking here was that I could redistribute via DLNA the four camera feeds into something the XBox One would be able to show on our family’s single 48″ LCD TV in the living room via the Media App. So far, no luck getting that to work, though IE on the XBox One will view and play all four feeds from the Trendnet web server, which for the purposes of this project, was good enough for me.
Additionally, I have a junker Lenovo laptop (Ideapad, 11″) that I’ve essentially built into a Kiosk PC for the kitchen/dining area, the busiest part of the house. This PC automatically logs in, opens the fat client and loads the file to view the four live feeds. And it does this all over wifi, giving instant home intel to my wife, mother-in-law, and myself as we go about our day.
Finally, both the iOS & Android devices in my house can successfully view the camera streams, not from the server, but directly (and annoyingly) from the cameras themselves.
The Impact of RTSP 1080p/30fps x 4 on Home Lab
I knew going into this that streaming live video from four quality cameras 24×7 would require some serious horsepower from my homelab, but I didn’t realize how much.
From the compute side of things, it was indeed alot. The Windows 8.1 VM is currently on Node2, a Xeon E3-1241v3 with 32GB of RAM.
Typically Node2’s physical CPU hovers around 8% utilization as it hosts about six VMs in total.
With the 8.1 VM serving up the streams as well as compressing them with a variable bit rate, the tax for this DIY Home surveillance project was steep: Node2’s CPU now averages 16% utilized, and I’ve seen it hit 30%. The VM itself is above 90% utilization.
More utilization = more worries about thermal as Node2 sits in the garage. In southern California. In the summertime.
Node2’s average CPU temperature varies between 22c and 36c on any given warm day in the garage (ambient air is 21c – 36c). But with the 8.1 VM, Node2 has hit as high as 48c. Good thing I used some primo thermal paste!
From the network side, results have been interesting. First, my Meraki is a champ. The humble MR-18 802.11n access point doesn’t break a sweat streaming the broadcast feed from the VM to the Lenovo Kiosk laptop in the kitchen. Indeed, it sustains north of 21mb/s as this graph shows, without interrupting my mother in law’s consumption of TV broadcasts over wifi (separate SSID & VLAN, from the SiliconDust TV tuner), nor my wife’s Facebooking & Instagramming needs, nor my own tests with the Trendnet application which interfaces with the cameras directly.
Meraki’s analysis says that this makes the 2.4ghz spectrum in my area over 50% utilized, which probably frustrates my neighbors. Someday perhaps I’ll upgrade the laptop to a 5ghz radio.
vSwitch, the name of my Converged SCVMM switch, is showing anywhere from 2megabits to 20 megabits of Tx/Rx for the server VM. Pretty impressive performance for a software switch!
Storage-wise, I love that the Trendnets can mount an SMB share, and I’ve been saving snapshots of movement to one of the SMB shares on my WindowsSAN box.
I am also using Trendnet’s email alerting feature to take snapshots and email them to me whenever there’s motion in a given area. Which is happening a lot now as my 2 year old walks up to the cameras, smiles and says “Say cheeeese!”
So I was tooling around one day in the lab, reading Ivan Ristic’s book on SSL/TLS, when I came across his advice on securing Windows-based Infrastructures from offering up the use of out-of-date/obsolete or otherwise insecure cipher suites to hosts on the other end of an https connection.
I read Ristic’s chapter a couple of times, reviewed TechNet, and selected a set of cipher suites in Group Policy in the order I wanted them used, based largely on Ristic’s text, but with a few others I knew I’d need after the policy went live. Then I pushed out the new policy, named “Strong Crypto,” to all physical, virtual and laptops in my home lab.
A few gpupdates later, I was pleased to see that nothing was broken. Schannel wasn’t showing any errors, User & Computer accounts were authenticating and getting kerb tickets, and pleasantly, my Outlook fat client didn’t even hiccup; it happily was using TLS 1.2 cipher suites to talk with my Office 365 Exchange instance.
And then, two days later, I noticed it. OneDrive for Business was busted, had gone Pear Shaped, and was now totally t***-up as my English friends would say.
A couple hundred gigabytes of files no longer syncing to my Sharepoint Online site, as evidenced by these Microsoft Icons of Distress:
So, what’d I break?
I’ll get to that in a moment, but first: why would you bother with something as obscure as cipher suites and their order? I mean beyond the fact that toggling the cipher suite sounds cool?
Why Cipher Suites are Important
Cipher suites are a critical part of your AD infrastructure. They’re critical as they represent a sort of baseline set of standards that client & server negotiate over during the complicated and very important tête à tête that is the TLS/SSL handshake between client/server.
You can and should read more about TLS handshakes in this RFC, but the bottom line is this: client & server are supposed to negotiate with each other, find the most secure and common set of cipher package, and use it during the secured session.
If client & server can’t find at least one common cipher suite, you have a busted TLS connection. And that’s no bueno,unless it was your intent.
In Microsoft-land, the default set of cipher suites is pretty good. Who am I kidding, it’s an acronym rich playground of security paradigms, as evidenced by the Group Policy editor:
Don’t be intimidated by all the crypto terms on this screen. What you see is the list of cipher suites -and the order in which they are presented to a host- by default.
The way to read one of these cipher suites is by breaking it down into its constituent parts:
So, the Cipher suite above uses TLS as its protocol (vs SSL), can exchange keys via the Elliptic Curve Diffie Helman ephemeral mechanism, accepts an RSA x.509 certificate, and is willing to encrypt the session via the AES 256 bit block cipher. The last bit, we’ll get to in a moment.
Be cautious when modifying
Since I was doing this in my lab, I had no concern about legacy applications, but in a production environment, you’ll want to tread lightly and deliberately here. Consider:
If your’e in a typical Microsoft IT shop, you probably have a few legacy applications hanging around that may rely on old cipher suites, or vice-versa, the application server can’t use the newer cipher suites that come built into your desktops & laptops.
Take Windows Server 2003, for example. The base OS doesn’t support Elliptic Curve Diffie Helman for Key Exchange, so right off the bat, if you’ve got 2003 Hosts serving up https Sharepoint or Exchange in-house, your clients & servers will never utilize TLS_ECDHE as that suite is not common to both of them. The contrary is also true; your Windows 8.1 laptop isn’t going to support the oldest suites that your 2003 server does; TLS_RSA_WITH_DES_CBC_SHA is never going to be the cipher suite watering hole your clients/servers meet around ((thank Goodness!!)) unless you go out of your way to make it happen.
The lesson here is that old cipher suites never die, dependency on them just fades away as your modernize/replace your legacy in-house applications with modern, streamlined, and properly TLS-secured ones. So be cautious, lest you break a legacy application.
You might be thinking I’m full of great advice, yet I still managed to wreck my OneDrive for Business sync app. And you’d be right!
So what happened?
Essentially, I broke my little OneDrive for Business sync app because I didn’t include SHA1 as possible hash algorithm in any of the cipher suites I selected.
And SHA1 is used by Microsoft IT ((as a side note, it’s really awesome to see Microsoft IT’s PKI, built out as it should be. Here’s a PKI serving not just Microsoft internal employees -all 100k of them- but millions of customers. If Microsoft IT can build a PKI to that scale, surely you and I can build one for the users dependent on us!)) in at least two places: as the Signature Hash algorithm on the root certificate of my Sharepoint site, and as the hashing mechanism for the Thumbprint on *.sharepoint.com certificate.
Had I visited my Sharepoint site in IE, I would likely have seen an error message in my browser; but I use Opera normally, and Opera -like Chrome & Firefox- have cipher suites apart from Windows’ so I never saw an error.
Adding the strongest cipher suite that included SHA1 fixed the error right away. ((Interesting aside: Google, and many security researchers, consider SHA1 to be end-of-life as it is now, or will be very soon now, computationally feasible to crack it, if that’s the right word. Google wants to sunset SHA1 in its browser this year; Ivan Ristic’s site will give https sites that use SHA1 a D- rating by the end of 2015. Microsoft IT, meanwhile, still uses it in production, but plans to deprecate it at the end of 2016. What gives? You could say there’s a pissing match between these leviathans of technology, or that one is trying to screw the other. But in essence, all parties agree SHA1 should fade away, they just differ on how aggressive deprecation efforts should be.))
Just how hard is it in 2015 to order & deploy a cheap commodity internet circuit to connect a remote office/branch office (ROBO) to the rest of your corporate WAN via the internet? ((Commodity = business class internet, something less reliable but orders of magnitude less expensive than a traditional private line, T1, or managed MPLS circuit. Commodity also means fat, dumb internet pipe, a product that cable internet companies consider an existential threat))
Pretty damned hard.
Why so difficult Jeff?!? you’re thinking. I stand-up tunnels and tear them down all day long, I route/switch in my sleep and verily I say unto you that my packets always find their way home, tags intact, whether on the WAN, between switch closets in the campus, or between nodes in the datacenter!
Verily they do indeed, and I salute you, you herder of stray packets!
It’s not that the technology connecting core to branch is hard or difficult, no, what I’m bitching about today is connecting the branch site to the internet in the first place.
It’s layer 1, stupid.
Truly, ordering internet service for a small or even medium-sized branch office is one of the most painful exercises in modern IT.
Here, let me show you:
You Bing/Google various iterations of “Lake Winnepesaukah ISPs,” , “Punxatawney Packet Delivery,” , “Broadband Service in Topeka,” “Ethernet over Copper + Albuquerque,” “Business Cable Internet – Pompano Beach, FL” and such. Dismissing the spam URL results on Page 1-12, you eventually arrive at Comcast, Time Warner, or Charternee Spectrum Business, or whatever little coax fiefdom has carved out a franchise at the edge of your business. You visit their website, click “Business” and fight your way through pop-ups and interstitials to a page that says it can verify service at your branch office’s address.
Right, you think,I’ll just Tab-tab my way through this form, input my branch office address here, punch that green submit button there, and get these nasty Layer 1 bits out of the way. But this isn’t the old days of 2009 when you could order a circuit online or at least verify service…oh no, no sir, this is the future…this is 2015. In 2015, you see, the Cable providers demand audience with you, so that they can add value.
Pay the Last Mile Toll: So you surrender your digits and wait for a phone call. When it rings 36-72 hours later, you’re determined to keep it short. What you want is a simple yes/no on service at your ROBO, or an install date, but what you get is a salesperson who can’t spell TCP/IP and wants to sell you substandard VoIP & TV. “Will you be uploading or downloading with this internet connection?” is just one of the questions you’ll suffer through to mollify the last mile gatekeepers standing between you and #PacketGlory on the WAN.
At long last, install day arrives: You’ve drop-shipped the edge router/overlay device, you’ve coordinated with the L-con, and the CableCo tech is on site at your ROBO to install your circuit. Hallalelujah, you think, as you wait for the tunnel to come up. But it never does, because between your awesome zero-touch edge device & your datacenter lies some crazy bespoke 2Wire gateway device that NATs or offers up a free wifi connection to the public on your dime. Another phone call, another fight to get those things turned off.
Nuts to all that, I say.
This is America jack, and the great thing about America is choice. Even when you don’t have choice (and you don’t in the case of cable franchises & municipalities), all you may need is line of sight to one of these things:
That’s right. Fixed wireless, baby. I’m hot on fixed wireless in 2015. It’s everything CableCo isn’t. It’s:
Friction free: In place of the coax fiefdoms and gatekeepers, the 1-800 numbers, and the aggressive salespeople, there’s just Joe, a real engineer at a local fixed wireless ISP. Joe’s great because Joe’s local, and Joe takes your order, gives you his mobile, installs the antenna at your branch, and hands you a blue wire with three static IPs.
Super-fast to deploy. You want internet at your ROBO? Well guess what? It’s already there, you just need the equipment to catch it.
More reliable than it used to be: Now of course this all depends on the application you’re trying to deliver to your ROBO, but I’ll say this: Fixed Wireless has improved. You don’t need to fear (as much) a freak snowstorm, a confused flock of Canada Geese, or rain. For a small ROBO, a fixed wireless connection might be enough to serve as the primary WAN link. For larger ROBOs, I think the technology is mature enough to serve as a secondary WAN link, or even your primary Internet circuit. ((Routing business traffic over the expensive wired link and internet over the cheap fixed wireless link is a recipe I’d recommend all day long and twice on Sundays ))
As Secure as Anything Else These Days: How difficult would it be to perform a man in the middle attack via interception of a fixed wireless connection? I’m not sure, to be honest, but if you aren’t encrypting your data before it leaves your datacenter, you have a whole lot more to worry about than a blackhat with a laptop, a stick, and a microwave antenna.
Cost competitive: I’ve deployed a couple of fixed wireless connections and I find the cost to be very competitive with traditional cable company offerings. Typically you’ll pay about $200 for the antenna install, but unlike the fee Comcast would charge you to install their modem, I think this is justified as it involves real labor and a certain amount of risk.
Regional/Hyper-local but still innovative: For whatever reason, fixed wireless ISPs have proven resistant to the same market forces that killed off your local dial-up/DSL ISP. Yet this isn’t a stagnant industry; quite the opposite in fact, with players like Ubiquiti Networks releasing new products.
I’ve been working on the WAN a lot lately and I’ve deployed two fixed wireless circuits at ROBOs. If you’ve got similar ROBO WAN pains, you should have a look at fixed wireless, you might be surprised!
I’ve not always had a bromance with Microsoft’s Office suite. I cut my word processing teeth on WordPerfect 5.1, did most of my undergrad papers in BeOS’ one productivity suite ((GoBe Productive, still the best Office suite name)) , and touch-typed my way to graduating cum laude in grad school with countless Turabian-style Google Docs papers.
That was for corporate suits, man. Rich corporate suits.
But all that’s ancient history. Or maybe I’ve become a suit. Either way, I’m loving Office today.
In 2015, Office has transformed into the ultimate agnostic git ‘r done productivity package. It’s free to use in many cases, but if you want to ‘own’ it, you can subscribe to it, just like HBO ((For the IT Pro, this is a huge advantage, as a cheap E-class sub gives you access to your own Exchange instance, your own Sharepoint server, and your own Office tenant. It’s awesome!)) . It’s also available on just about any device or computing system you can think of, works just as well inside a browser as Google Docs does, and has an enormous install base.
Office has become so impressive and so ubiquitous that it’s truly a platform unto itself, consumed a la carte or as part of a well-balanced Microsoft meal. I’m bullish on Windows but if Office’s former partner ever sunsets, I’m convinced my kid and his kid will still grow up in an Office world.
All of that makes Office really important for IT, so important that you as an IT Guy should consider standing-up some easy instrumentation around it.
Enter Office Telemetry, a super-simple package that flows your Office data to a SQL collector, mashes it up, and gives you important insight into how your users are using Office. It also surfaces the problems in Office -or Office documents- before your users do, and it’s free.
Oh, did I mention it’s called Office Telemetry? This thing makes you feel like an astronaut when you’re using it!
Here’s how you deploy it. Total time: about an hour.
Spin-up a 2008 R2 or 2012 VM, or find a modestly-equipped physical box that at least has Windows Management Framework 3.0/Powershell 3.0 on it. If it has a SQL 2012 instance on it that you can use, even better. If not, don’t stress and proceed to the next step.
Set-aside a folder on a separate volume (ideally) for the telemetry data. If you’re going to flow data from hundreds of Office users, plan for a minimum of 5-25 megabytes per user, at a minimum.
If your users are on the WAN, plan accordingly. Telemetry data is pretty lightweight (50k chunks for older Office clients, 64k chunks for Office 2013)
Install Office ProPlus 2013 or 365 on the VM. You do not need to use an Office 365 license for it to run.
Because it’s a script, you’ll need to temporarily change your server’s execution policy, self-sign it, or configure Group Policy as appropriate to run it. TechNet has instructions.
Run the script; it will download SQL 2012 express and install it for you if you don’t have SQL. It will also set proper SMB read/modify permissions on that folder you set up earlier.
As if that wasn’t enough, the script will give you a single registry keyfile you can use to deploy to your user’s machines.
But I prefer the Group Policy/SCCM route. Remember the ADMX files you deployed? Flip the switches as appropriate under User Configuration>Administrative Templates>Microsoft Office 2013> Telemetry Dashboard.
Sit back, and watch the data flow in, and pat yourself on the back because you’re being a proactive IT Pro!
As I’ve deployed this solution, I’ve found broken documents, expensive add-ons that delay Office, and multiple other issues that were easy to resolve but difficult to surface. It’s totally worth your time to install it.
It’s been awhile since I posted about my home lab, Daisettalabs.net, but rest assured, though I’ve been largely radio silent on it, I’ve been busy.
If 2013 saw the birth of Daisetta Labs.net, 2014 was akin to the terrible twos, with some joy & victories mixed together with teething pains and bruising.
So what’s 2015 shaping up to be?
Well, if I had to characterize it, I’d say it’s #LabGlory, through and through. Honestly. Why?
I’ve assembled a home lab that’s capable of simulating just about anything I run into in the ‘wild’ as a professional. And that’s always been the goal with my lab: practicing technology at home so that I can excel at work.
Let’s have a look at the state of the lab, shall we?
Hardware & Software
Daisetta Labs.net 2015 is comprised of the following:
Five (5) physical servers
136 GB RAM
Sixteen (16) non-HT Cores
One (1) wireless access point
One (1) zone-based Firewall
Two (2) multilayer gigabit switches
One (1) Cable modem in bridge mode
Two (2) Public IPs (DHCP)
One (1) Silicon Dust HD
Ten (10) VLANs
Thirteen (13) VMs
Five (5) Port-Channels
One (1) Windows Media Center PC
That’s quite a bit of kit, as a former British colleague used to say. What’s it all do? Let’s dive in:
The bulk of my lab gear is in my garage on a wooden workbench.
Nodes 2-4, the core switch, my Zywall edge device, modem, TV tuner, Silicon Dust device and Ooma phone all reside in a secured 12U, two post rack I picked up on ebay about two years ago for $40. One other server, core.daisettalabs.net, sits inside a mid-tower case stuffed with nine 2TB Hitachi HDDs and five 256GB SSDs below the rack.
Placing my lab in the garage has a few benefits, chief among them: I don’t hear (as many) complaints from the family cluster about noise. Also, because it’s largely in the garage, it’s isolated & out of reach of the Child Partition’s curious fingers, which, as every parent knows, are attracted to buttons of all types.
Power & Thermal
Of course you can’t build a lab at home without reliable power, so I’ve got one rack-mounted APC UPS, and one consumer-grade Cyberpower UPS for core.daisettalabs.net and all the internet gear.
On average, the lab gear in the garage consumes about 346 watts, or about 3 amps. That’s significant, no doubt, costing me about $38/month to power, or about 2/3rds the cost of a subscription to IT Pro TV or Pluralsight. 🙂
Thermals are a big challenge. My house was built in 1967, has decent insulation and holds temperature fairly well in the habitable parts of the space. But none of that is true about the garage, where my USB lab thermometer has recorded temps as low as 3C last winter and as high as 39c in Summer 2014. That’s air-temperature at the top of the rack, mind you, not at the CPU.
One of my goals for this year is to automate the shutdown/powerup of all node servers in the Garage based on the temperature reading of the USB thermometer. The $25 thermometer is something I picked up on Amazon awhile ago; it outputs to .csv but I haven’t figured out how to automate its software interface with powershell….yet.
Anyway, here’s my stack, all stickered up and ready for review:
Beyond the garage, the Daisetta Lab extends to my home’s main hallway, the living room, and of course, my home office.
Here’s the layout:
On the compute side of things, it’s almost all Haswell with the exception of core and node3:
Node4, Haswell, i5-4670, 4, 32GB, Cluster node/storage, 2012r2 core, Asus
I love Haswell for its speed, thermal properties and affordability, but damn! That’s a lot of boxes, isn’t it? Unfortunately, you just can’t get very VM dense when 32GB is the max amount of RAM Haswell E3/i7 chipsets support. I love dynamic RAM on a VM as much as the next guy, but even with Windows core, it’s been hard to squeeze more than 8-10 VMs on a single host. With Hyper-V Containers coming, who knows, maybe that will change?
While I included it in the diagram, TVPC3 is not really a lab machine. It’s a cheap Ivy Bridge Pentium with 8GB of RAM and 3TB of local storage. It’s sole function in life is to decrypt the HD stream it receives from the Silicon Dust tuner and display HGTV for my mother-in-law with as little friction as possible. Running Windows 8.1 with Media Center, it’s the only PC in the house without battery backup.
About 18 months ago, I poured gallons of sweat equity into cabling my house. I ran at least a dozen CAT-5e cables from the garage to my home office, bedrooms, living room and to some external parts of the house for video surveillance.
I don’t regret it in the least; nothing like having a reliable, physical backbone to connect up your home network/lab environment!
At the core of the physical network lies my venerable Cisco 2960S-48TS-L switch. Switch1 may be a humble access-layer switch, but in my lab, the 2960S bundles 17 ports into five port channels, serves as my DG, routes with some rudimentary Layer 3 functions ((Up to 16 static routes, no dynamic route features are available)) and segments 9 VLANs and one port-security VLAN, a feature that’s akin to PVLAN.
Switch2 is a 10 port Cisco Small Business SG-300 running at Layer 3 and connected to Switch1 via a 2-port port-channel. I use a few ports on switch2 for the TV and an IP cam.
On the edge is redzed.daisettalabs.net, the Zyxel USG-50, which I wrote about last month.
Connecting this kit up to the internet is my Motorola Surfboard router/modem/switch/AP, which I run in bridge mode. The great thing about this device and my cable service is that for some reason, up to two LAN ports can be active at any given time. This means that CableCo gives me two public, DHCP addresses, simultaneously. One of these goes into a WAN port on the Zyxel, and the other goes into a downed switchport
Lastly, there’s my Meraki MR-16, an access point a friend and Ubiquity networks fan gave me. Though it’s a bit underpowered for my tastes, I love this device. The MR-16 is trunked to switch1 and connects via an 802.3af power injector. I announce two SSIDs off the Meraki, both secured with WPA2 Personal ((WPA2 Enterprise is on the agenda this year)). Depending on which SSID you connect to, you’ll end up on the Device or VM VLANs.
The virtual network was built entirely in System Center VMM 2012 R2. Nothing too fancy here, with multiple Gigabit adapters per physical host, one converged logical vSwitch and a separate NIC on each host fronting for the DMZ network:
Thanks to VMM, building this out is largely a breeze, once you’ve settled on an architecture. I like to run the cmdlets to build the virtual & logical networks myself, but there’s also a great script available that will build a converged network for you.
A physical host typically looks like this (I say typically because I don’t have an equal number of adapters in all hosts):
We’re already several levels deep in my personal abstraction cave, why stop here? Here’s the layout of VM Networks, which are distinguished from but related to logical networks in VMM:
I get a lot of questions on this blog about jumbo frames and Hyper-V switching, and I just want to reiterate that it’s not that hard to do, and look, here’s proof:
And last, and certainly most-interestingly, we arrive at Daisetta Lab’s storage resources.
Well, I had so much fun -and importantly so few failures/pains- with Microsoft’s Tiered Storage Spaces that I’ve decided to deploy not one, or even two, but three Tiered Storage Spaces. Here’s the layout:
[table]Server, #HDD, #SSD, StoragePool Capacity, StoragePool Free, #vDisks, Function
Core, 9, 6, 16.7TB, 12.7TB, 6 So far, SMB3/iSCSI target for entire lab
Node1,2, 2, 2.05TB, 1.15TB,2, SMB3 target for Hyper-V replication
Node4,3,1, 2.86TB, 1.97TB,2, SMB3 target for Hyper-V replication
I have to say, I continue to be very impressed with Tiered Storage Spaces. It’s super-flexible, the cmdlets are well-documented, and Microsoft is iterating on it rapidly. More on the performance of Tiered Storage Spaces in a subsequent post.
Like a lot of IT Pros, I’ve been studying up on security topics lately, both as a reaction to the increasing amount of breach news (Who got breached this week, Alex?) and because I felt weak in this area.
So, I went shopping for some books. My goals were simply to get a baseline understanding of crypto systems and best-practice guidance on setting up Microsoft Public Key Infrastructures, which I’ve done in the past but without much confidence in the end result.
This 3.2lb, 800 page book has a 4.9 out of 5 star rating on Amazon, with reviewers calling it the best Microsoft PKI guide out there.
Great! I thought, as I prepared to shell out about $80 and One Click my way to PKI knowledge.
That’s when I noticed that the book is out of print. There are digital versions available from O’Reilly, but it appears most don’t know that.
For the physical book itself, the least expensive used one on Amazon is $749.99. You read that right. $750!
If you want a new copy, there’s one available on Amazon, and it’s $1000.
I immediately jumped over to Camelcamelcamel.com to check the history of this book, thinking there must have been a run on Mr. Komar’s tome as Target, Home Depot, JP Morgan, and Sony Pictures fell.
The price of this book has spiked recently, but Peak PKI was a full three years ago.
I looked up security breaches/events of early 2012. Now correlation != causation, but it’s interesting nonetheless. Hopefully this means there’s a lot of solid Microsoft PKI systems being built out there!
Rather than shell out $750 for the physical book, I decided to get Ivan Ristic’s fantastic Bulletproof SSL/TLS, which I highly recommend. It’s got a chapter on securing Windows infrastructure, but is mostly focused on crypto theory & practical OpenSSL. I’ll buy Komar’s as a digital version next or wait for his forthcoming 2012 R2 revision.