That said, there are 3 networks designed per the RFC to never be routed across the public internet.

10.0.0.0/8 172.16.0.0/12 192.168.0.0/16

You design your local networks to use addresses within one of these ranges.

I'm sure there is a video for this.

@mary said in Binary Math - CompTIA Network+ N10-007 Prof Messer:

While the math is pretty straightforward, the why would I use this wasn't covered. How often do you use this and for what?

Jared is correct. Basically once you understand how binary makes things work, and knowing how it works in a mask, you never use it. You won't ever need to work in binary, but knowing that it is underneath things helps make something that is super complex, trivially obvious. Basically, if you understand the binary, subnet masking is SO simple. If you don't, it seems like black magic.

@mary said in Software-Defined Networking - CompTIA Network+ N10-007 Prof Messer:

Is there specific software that is used for creating a virtual environment?

SDN refers to a great number of potential solutions. In most cases, it is software built into the hardware. So Cisco, Juniper, Brocade and other networking makers offer software solutions, often that have no particular name. They are just the SDN options of that gear.

Pure software does exist and a common example of it is ZeroTier which is designed as a complete virtual networking layer on top of your physical network and has no interaction with the underlying hardware so it completely agnostic.

@mary said in Circuit Switching and Packet Switching - CompTIA Network+ N10-007 Prof Messer:

Is there anything not considered packet switching?

Yes, the most common example is the voice network (the PSTN / Public Switched Telephone Network). It establishes a "circuit" for every call, and it is permanent until that call ends. It switches on a "call by call" basis. Every packet associated with that call takes the same path.

It is almost always considered archaic to use circuit switching, it is rarely useful or efficient. It was popular in the "pre-data" era of networking when it was audio and video, generally analogue, going over the path rather than individual packets. Now that we have packetized data and the power to evaluate the route every packet, there is little to no purpose for circuit switching.

@mary said in Access Control Lists - CompTIA Network+ N10-007 Prof Messer:

Is there a way to deny all incoming traffic to your firewall?

Sure. But if you think about it, you'd absolutely never want this. This would be the same as disconnecting the network. If you wanted to do this, you would simply unplug the network instead (as that is more reliable.) The only reason to truly "deny all" is to go offline. And if you want to be offline you logically want it to be really reliable and since you'd want no traffic to make it no matter what in that situation, pulling the plug is the better choice 99.99% of the time.

@mary said in Network Address Translation - CompTIA Network+ N10-007 Prof Messer:

Is there any kind of slowdown when using just one port if you are getting a lot of traffic?

No not really. The most commonly used ports are 80 and 443. They process quite a bit of traffic on your average workstation.

In fact, most servers are designed to work with a single port or just a handful of ports open. For custom applications using a specific port makes it easier to troubleshoot issues and restricts non application traffic. Many apps are defaulting to 443 these days. Although, keep in mind SSL /TLS can operate on other ports.

Yes, if you are setting physical ports to a VLAN, then they are acting like a physically different switch on those ports. So attaching another physical switch to one of those ports would make it a switch on that VLAN.

In big companies how use VoIP for production for example, this is a most, QoS needs to be very well deployed, to improve users/clients experience using the phone.

@mary said in Configuring IPv6 - CompTIA Network+ N10-007 Prof Messer:

When using ipconfig on a network I've seen the IPV4 and IPV6 addresses shown at the same workstation. Sometimes it shows IPV4 (preferred). Why is that?

I don't have a specific reasoning - but I'm guessing it's because the computer received a DHCP reply on IPv4 - so the computer assumes that IPv4 is the primary way this network wants to communicate.

@mary said in IPv4 and IPv6 Addressing - CompTIA Network+ N10-007 Prof Messer:

If you are using :: to abbreviate sets of 2 or more zeros, how can you tell how many there are? In the example he condensed 3 sets down to ::?

IPv6 always has 8 sets of 4 numbers for addressing. So it should be easy to figure out with a little math. IE: FABD:4976:5AC3:: would mean that 5 more sets of 4 zeros is tacked on FABD:4976:5AC3:0000:0000:0000:0000:0000

@mary It depends of manufacturer or vendor. Now in this times everyone is using "Standard Protocols"

@melvinsilva said in IGP and EGP - CompTIA Network+ N10-007:

@scottalanmiller I Will Add; "Only by WAN Network Administrators".

No, internal too. In fact, most, by far, are internal. You only have a few big WAN connections with most companies. But you might have hundreds or thousands of internal routes that have to be managed.

Both required a good initial configuration, but when "issues" occurs like link flaps or ISP outage (when MPLS fails), Dynamic may affect Router performance (CPU, Memory, etc).

Using Dynamic Routing; if you have no backup link or you dont have a proper failover configuration with correct threshold, the network updates will cause router performance degradation. It constantly will try to reach destination via default gateway.

Using Static Routing; If you have not a recovery plan or a back door to enter the remote router, you will have zero access to that device until link or issue is restored. Packets will be forwarded to a dead route.

For both, Initial design and configuration is the Key, when issues happens troubleshooting is a nightmare when things are not well done.

@mary said in Switch Interface Properties - CompTIA Network+ N10-007 Prof Messer:

Are port mirrors common for companies that monitor what employees are doing on the network or is something else used?

No, that would be insanely impractical. A port mirror doesn't give you a copy of what someone is actually doing, it gives you a copy of the network traffic. Which is a butt load of disconnected data. It's not like you would know what the end user was doing, only what was being transferred on the wire.

If you think about what you'd see on the wire... a single file transfer or LAN based action might generate a huge amount of traffic. Or going to a website might create a bunch of unintentional traffic from ads that aren't something that the end user cared about. Or a website or app open in the background might generate gops of traffic for something that isn't being used.

It would be able to tell you if the person is on YouTube or Spotify, but would not tell you what they are doing that for or if they are actively "using" that thing. Basically you'd be flooded with information that would take ages to sort through, and the resulting information would tell you essentially nothing about what the end user was doing.

@melvinsilva said in Spanning Tree Protocol - CompTIA Network+ N10-007 Prof Messer:

I had experience when "users" manipulate patch panel or switch and plug both end of the cable (patch cord) to the same switch, creating a physical loop, but STP blocks this and prevent network issues.

Yeah, having switches that have STP turned on by default can be such a life saver in those situations. One place I worked liked to put in unmanaged switches all over the place. Caused me multiple long nights just tracking down which switch was the issue and then finding the loop cable. I knew exactly what was going on, finding it without good documentation and people saying "Oh, there's another network thingy over here." was just so much.... fun.

Is there a limit to how many trunks you can use?

@scottalanmiller said in Protocol Data Units - CompTIA Network+ N10-007 Prof Messer:

@WrCombs said in Protocol Data Units - CompTIA Network+ N10-007 Prof Messer:

Why would you not want to fragment a frame?

Performance. Fragmentation takes time and effort. If you can avoid it, there is less work to be done and, all other things being equal, that means cheaper equipment going faster.

Ah, That makes sense.

It's more purpose based. Unicast is for normal communications. Broadcast is normally reserved for special cases like DHCP where there is no way to do unicast.

Can anyone give me an example of a broadcast domain? This seems a bit basic. It made me think of things like push notifications, so I'm not sure if I'm getting the concept.

Worth noting that STP (Spanning Tree) is common in good switches, but it's also common to have switches that don't have it.