About subnetwork

Jonathan Davis loves network engineering. He's been in IT for over 17 years. In that time, he has worked in almost every aspect of IT. Nothing holds his attention and imagination like building wireless networks for unique environments.

Mist Systems Unveils an Environmental Sensor that is also a Wi-Fi 6 AP

At Mobility Field Day 4, we heard from a few companies which are working hard to extend the capabilities of their AP’s well past only serving traditional 802.11 clients.
Mist Systems, a Juniper Company, was one such presenter, and they might have a fantastic new platform with their latest hardware, the AP43.

Mist Wi-Fi 6 AP Specs

The idea is simple. Most campuses have AP’s covering their entire environment. In many large environments, they share that ceiling space with other types of sensors or networks. These overlay networks may include building and security sensors, Zigbee control of lights or door locks, and test sensor networks.

In many ways, Mist has been a bit ahead of this curve. Their AP’s already included an IoT port, which enabled triggering devices like door locks or sensing through a variety of GPIO sensors.

Their new AP43 is a dual 5Ghz capable 802.11ax access point. It includes 802.3bz NBASE-T port to ensure the network port never becomes a bottleneck. That port also includes 802.3bt power capabilities so that it can pass power out of its secondary port, enabling it to daisy chain any 802.3af network device. The obvious candidate here is the BT11, Mist’s BLE sensor.

Further, each AP43 includes built-in sensors to provide temperature, humidity, barometric pressure, and angle/orientation. The inclusion of these sensors come with some unique engineering challenges. If Mist is successful in getting them to work appropriately, it could be a game-changer.

The biggest challenge when considering environment monitoring on an access point is heat. Anyone who has ever touched an AP that has been on for a while knows it can be hot. Thanks to the first law of thermodynamics, we know that all energy consumed by the AP that doesn’t get radiated as RF is instead transformed to heat. But that heat output isn’t consistent. It will vary based on the transmitter duty cycle or CPU load of the AP.

Additionally, that heat creates a micro-climate around the AP, which will lower the humidity percentage since warm air holds more water than cool air. Warm air is also less dense, which may affect the barometric pressure sensor.

The humidity/heat problem is further exacerbated by the fact that all water in the air is absorbing a small amount of the radiated RF power.

Finally, the ceiling can be many degrees warmer than in the same room at desk level.
These are challenges that I am sure Mist has taken into account, and the fact that they can work through them is impressive. Having environmental reporting built into the AP could make for a fantastic use case for building managers.

Moving down the list, the barometric pressure and orientation/angle sensor have some compelling use cases. By comparing atmospheric pressure among AP neighbors, Mist should be able to tell which AP’s are on the same floor in multi-floor buildings. This information could significantly impact 802.11k neighbor reports. By excluding AP’s which may be heard by the AP, but are obviously on a different floor, the chances of a client choosing a better roam candidate increases.

By comparing atmospheric pressure among AP neighbors, Mist should be able to tell which AP’s are on the same floor in multi-floor buildings. This information could significantly impact 802.11k neighbor reports.

Finally, the angle sensor can help identify AP’s mounted on a wall versus a ceiling. With that information and Mist’s ML backend, it should be able to better locate clients in RTLS environments.

These new sensors extend the AP capabilities well past the traditional use cases. Can Mist pull off the environmental monitoring? Can they adjust their neighbor report automatically based on elevation? I’m excited to play with these features in the future and get to the bottom of these answers and more.

Either way, it is clear that Mist has built the AP43 as a platform they can innovate with and I’m excited to see where they take it.

Take a look and tell me what you think:

Mist Systems Mist AI for AX – Wi-Fi 6 from Gestalt IT on Vimeo.

SD-WAN Reimagined – 128 Technology

128 Technology LogoMaybe it’s just me. I’ve always felt like SD-WAN was kludgy. Every time I listen to an explanation of how it works, I think of the picture of a mechanic putting duct tape on the wing of an aircraft while passengers sit inside awaiting departure. I imagine sitting in the window seat, watching it take place and asking the questions: “Is that really the best way to fix this problem?” “Are we trusting duct tape to hold the wing together?” and even “Shouldn’t the wing hold itself together?”

Despite having those questions, I hopped onto an aircraft and flew off to a Tech Field Day Exclusive with 128 Technology in July. After arriving, I didn’t think about duct tape once.

128 Technology is a five-year-old company focused on creating the best SD-WAN solution. As a new company, building a new product to answer a specific set of challenges, 128 Technology had an empty toolbox. That also meant they had no baggage to bring with them. It was a fresh start. They could make their solution be anything they wanted it to be.

Sue Graham Johnston introduces 128 Technologies to the Tech Field Day DelegatesAccording to Sue Graham Johnston, “…we decided to reorient networking to focus on the session, we can get rid of about 30 years’ worth of technology workarounds and overlays…” In case you are wondering, yes, that is duct tape she’s talking about.

That one statement piqued my interest, set the stage, and explained much of how their model works. It is simple enough to brilliant.

128 Technology uses a 5-tuple to identify each session: source and destination IPs, source and destination ports, and the protocol. When the session is built between the ingress and egress router, the first packet is encapsulated with 150-200 bytes of metadata to establish the session. After the session is established, no further encapsulation is needed as the ingress and egress routers have all of the data that is necessary. When each packet hits the ingress router, the source and destination addresses are changed until it hits the egress router. (Does this sound a bit like NAT? Because its NAT for SD-WAN.)

That’s their magic: No encapsulation, lower overhead, no need to fragment larger frames to provide space for additional headers, and significant bandwidth savings.

Now that you understand the basics of how 128 Technology builds sessions, it’s also essential to see how they integrate security. After all, this is intended to be an SD-WAN solution where data will traverse the internet.

Here again, there are a few basics to understand. All metadata to establish sessions is encrypted. Unencrypted traffic between Ingress and Egress routers is encrypted with AES128 or AES256. SSL or other encrypted traffic doesn’t need to be re-encrypted, so 128 Technology doesn’t. That reduces latency, complexity, and overhead. The last important piece of the puzzle is that the 128 Technology network operates as a zero-trust security environment. All data sessions must have a service policy created to allow traffic to flow. No service-policy means no traffic.

The last consideration is how to manage the SD-WAN environment. One router in the network is assigned the role of Conductor. All routers and the Conductor run a single code base, ensuring consistency in bug fixes and behavior. The Conductor is not required for configuration or operation but provides a central point configuration of all devices.

When I consider the takeaways from Networking Field Day Exclusive with 128 Technology, one thing jumps out far above the rest. Their focus on simplicity and the most critical part of data networks: the data session. I feel the solution is well thought out, and based on the customers that are using it in production; it seems the execution delivers on their promises.

The only remaining question I have do not relate to their technology at all.

  1. When will 128 Technology be acquired?
  2. Who will acquire them?
  3. Will it be to include them into an existing full-stack solution or acquired to be used by a service provider in their internal networks?

I hope that this will be a product that we can all benefit from as direct customers.

Take time to watch the videos and see if you agree.

128 Technology Networking Platform Overview from Gestalt IT on Vimeo.

MFD3 – Link-Live Updates

This is the third blog from the Company-Previously-Known-As-Netscout’s session at Mobility Field Day 3. You can read about the AirCheck G2 v3.0 update and also the LinkRunner G2 v2.0 Update.

To catch you up, I came into MFD3 less enthusiastic than most regarding Netscout and their lineup of handheld network tools. With that said, I took notice in 2017 at MFD2 that the company was paying attention to feedback and looking for suggestions on how to improve their product offerings.

One of those improvements for MFD3 was a further expansion of the capabilities of Link-Live.

Link-Live has matured into a tool for consolidating all of your test results AND managing the tools at your disposal.

Many of these updates were covered in the LinkRunner and AirCheck updates, but bear repeating:

  • AirCheck software updates
  • AirCheck G2 Profile sharing
  • Packet capture sharing
  • Simplified App search for the LinkRunner G2
  • Files Folder – There is a lot more available that can be uploaded and saved to a project folder
  • Full AutoTest results

The most significant aspect of the Link-Live updates is a clear direction to make the LinkRunner and AirCheck entirely manageable without a Windows PC. This is a substantial shift from the past, and I am very excited to see it taking place because I stay away from Windows as much as possible.

So, the ultimate question, does the updates to the AirCheck G2 and LinkRunner G2, along with the new features of Link-Live make me change my opinion? Do I now see the ROI? Would I spend my budget, either personal or business on either tool?

The answer is “yes” to all of the above. With the divestiture of the handheld tools from Netscout into its own company, I expect the future to be bright. I think we will continue to see updates, new use cases, and great support. The handheld network tools team has won me over, and I’m happy to change my previous opinion. I will acquire both tools over the coming months for my personal toolkit, as I know my employer doesn’t have the budget. I don’t think there is more to say.

MFD3 – AirCheck G2 V3.0 Announcement

aircheckg2In case you missed it, MFD3 was an opportunity to reevaluate my opinion on the Aircheck G2 and LinkRunner G2. After my experiences at MFD2, I was no longer openly hostile towards the tools and saw that there was a legitimate desire to be open, fill the needs of users, and provide regular updates with new features.

As someone who identifies explicitly as a Wireless Network Engineer, the AirCheck G2 has been on my radar for a while, so I was interested, maybe even excited, about the opportunity to see the latest updates.

The AirCheck G2 v3.0 update adds:

  • Over-the-Network firmware updates – Sadly the V3.0 software update will have to be loaded from a PC, but from that point forward, a user with an active support contract can update the device directly.
  • Over-the-Network profile sharing – If your organization has more than one AirCheck G2, you can now ensure that everyone is testing using the same profile, all over-the-air through Link-Live.
  • Improved Link-Live interaction – manage device profiles, get test results, packet captures, etc.
  • Improved Locator Tool accuracy the Locator Tool now uses all three radio chains to enhance signal strength and accuracy
  • Enhanced AP name support – can now read AP names from Aerohive, Aruba, Cisco, Extreme Networks, and Huawei
  • Improved iPerf test performance – can now test using iPerf2, up to 300Mbps
  • Improved packet capture workflow – now users can be particular regarding the type of traffic they want to capture
  • Voice VLAN on ethernet test – if there is a voice VLAN assigned to the ethernet port, it will be displayed
  • Import certificates with a thumb drive – This simplifies importing certificates and is especially crucial for wireless engineers who might work at various customer sites.
  • Static IP’s can be assigned to the ethernet port
  • Other updates, which you should watch the video to see:

 

So, have I changed my mind? Am I ready to own a LinkSprinter G2 or AirCheck G2? Well, I think we should discuss Link-Live. That definitely factors into my decision.

MFD3 – LinkRunner G2 v2.0 Update

I have an admission to make. Before Mobility Field Day 2 of 2017, I was openly hostile towards the biggest player in the handheld network tools market. Through a series of lousy blue and gold experiences, I decided I no longer had room for those tools in my budget. Even after receiving a blue and gold LinkSprinter at a WLPC, I was apathetic at best.

But, I like reexamining my strongly held opinions. I believe that admitting I am wrong is much better than holding firmly to an incorrect conclusion.

linkrunnerSo, in 2017 when Netscout presented at MFD2, I got my opportunity to reconsider. They were working to expand the capabilities of the toolset, and they were open to feedback and requests for new features. I even considered purchasing an AirCheck G2, but ultimately found that I hadn’t budgeted for it. (Shocking!)

So, let’s fast forward to 2018 and MFD3. Over the two hour window Julio Petrovitch, from the handheld network tools group previously owned by Netscout, covered many topics, but the topics of most interest were the AirCheck G2 v3.0 and the LinkRunner G2 v2.0 software updates. So again, I got to reevaluate my opinion.

The very first revelation to me was this team now truly believes in updates! The LinkRunner was released last October, so approximately a year later they are adding features with v2.0. The announcement included significant new improvements and features, not just small dot revision updates and bug fixes.

The LinkRunner G2 v2.0 update adds:

  • 802.3bt support – provides both loaded and unloaded voltage and wattage reporting of class 5-8 PoE PSE equipment
  • Injector support – measures from 12-60 volts
  • More VLAN information – the LinkRunner G2 can provide lots of information regarding the VLAN’s that are accessible from a switchport; useful if you have ports configured with a voice and data VLAN.
  • Enhanced DHCP Test – Now supports providing information from DHCP Options 43, 60, and 150.
  • Auto Test Improvements – allows a user to refine how they would like a test to run.
  • VLAN Monitor Tool – plug the LinkRunner G2 into a trunk port and monitor all of the VLAN’s that are available and the amount of traffic broadcast on each
  • Packet Captures – or as Julio Petrovitch correctly called it frame captures. Plug the device into a mirror or span port and capture traffic directly to the LinkRunner G2.

One more note; the LinkRunner G2 can charge from PoE! That isn’t a new feature, but it was one that I missed. I am mentioning it here for those others who might also be unaware.

So, the real question, have I changed my mind about Netscout? Maybe, but first, I think we should discuss the AirCheck G2.

Watch the whole presentation and then tell me what you’re most excited about in the comments.

 

MFD3 – Huge updates for AirCheck G2 and LinkRunner G2; then Netscout announces their sale

Mobility Field Day 3 was great! If you missed it, I will be releasing a few blogs over the coming weeks from my experience at the event. In the meantime, you can watch all of the videos here:

https://techfieldday.com/event/mfd3/

One of the most interesting developments this morning was the announcement from Netscout that it was divesting its handheld network test division to StoneCalibre.

The press release can be found here:
https://www.netscout.com/news/press-release/netscout-divests-handheld-network-test-business

While this announcement creates quite a few questions around the future, I firmly believe that the great group of people who have brought us the recently announced LinkRunner G2 v2.0 and AirCheck G2 v3.0 software updates are going to keep killing it. I’m excited to see what they bring to us in the future and hope to see them presenting once again at Mobility Field Day 4.

VIAVI Observer Apex- Finding the needle faster

I participated in the Tech Field Day Extra events at Cisco Live. One of the presenters, VIAVI has been floating near the edge of my awareness for a while, so it was great to see their presentation and get a better understanding of the VIAVI Observer Platform.

Anytime I see a presentation from a monitoring solution there are three questions that I ask:

“How useful would this be for tier one technicians?”

I usually consider that question from both the perspective of a NOC and also a helpdesk technician. If a monitoring tool isn’t practical for those roles, I am the one who gets stuck using it all of the time, and therefore, it has no place in my environment.

“How useful would this tool be for me?”

If the tool can’t offer enough information to be useful for a senior engineer, I don’t want to pay for it. It also increases the complexity of passing trouble tickets up the chain as each person has to start back at zero in their own tool.

“Does this make it easier to find the problem, or just add another step?”

Monitoring tools which only show up/down status and system logs have very little use for me. I can easily find those by other means, or on the device itself, faster than I can fire up a browser, click on a bookmark, log in, navigate through a device tree, etc.

VIAVI has provided the right answers to all three questions.

product-obsever-apex-welcome

The starting page for Observer is simple. It doesn’t take forever to load as it attempts to pull data from many different sources to provide a general health overview that rarely has anything to do with the reason you opened the application. Instead, Observer’s search box is ready for any relevant text the technician may know about the problem. If you have an IP, MAC address, VLAN, or hostname, those are all great places to start. You can also choose to push into a more generalized monitoring view like Application Performance, Network Performance, etc.

The search box is the beauty of the application for me. VIAVI indexes all of the monitoring sources for things like MAC addresses, IP addresses, interfaces, usernames, and other metadata and then correlates that information together. A technician doesn’t need to look up an IP address in the ARP table, get the MAC address, look up the MAC address in the MAC address table to get the port, then check the port for errors. A search on the IP address will provide all of that information, quickly! Since VIAVI also knows the assigned VLAN, it quickly displays “Here’s a bad actor on the same VLAN that is flooding the VLAN with bad frames.” The technicians can find problems without looking directly for them. That’s a huge win. This is not looking for a needle in a haystack. This is turning on an extremely powerful magnet and letting the needle come to you.

Another great feature is that Observer creates a baseline from the information that it acquires. With that baseline that understands system X typically runs at 75 percent utilization, but is now running at 90 percent, more problems quickly float to the surface. Additionally, the baseline filters out the normal abnormal. Is it “normal” for that system to run at 75 percent utilization all of the time? Maybe so. If it is, a technician doesn’t need a warning about it. It might be operating as designed.

If a technician can’t find a solution through the dashboard, the next engineer who picks up the problem will want to dig deeper. Thanks to the stored packet traces which provided all of the metadata the technician used, the engineer can take a look at the actual packets. Aside from the standard fields like source and destination, IP’s and ports, Observer also includes a patent-pending User Experience Score which is a 1-10 scale to aid in finding problems faster within the trace files.

Taking the click-through troubleshooting one step further, Observer creates Application Dependency Maps which aid an engineer to understand all of the dependent systems quickly and which are affecting performance.

When considering my initial three questions I proposed, I feel VIAVI’s Observer is providing pretty compelling answers for each. I look forward to learning more.

In many ways, Tech Field Day offers a similar solution to VIAVI Observer. TFD allows me to filter through the marketing hype, and get to the bottom of a product or solution and whether it will be useful to me. Don’t forget to check out the many other videos and content created by Tech Field Day at Cisco Live.