Studying CompTIA Network+ - Need Advice

Soulzz

Soulzz

Founder
Retired Founder
Joined
Jan 2, 2008
Messages
19,100
Age
36
So, I'm having a little problem understanding subnetting, and base 2 to decimal conversion.

I understand the purpose of subnet masks, and how to read the first two octets of an IP, and the last two. I understand Class A, Class B, and Class C address ranges.

The issue is understanding specific details on how to find the subnet mask of an example like 10.50.40.10, or 255.255.0.0/16 (I'm not sure if that's the correct CIDR value). Basically, the theory of it. To be honest, in most environments, creating the network mask and infrastructure is a small part of supporting the client. You really don't do it that often.

All I need is an ebook recommendation going in-depth on mastering subnet masks, creating an accurate subnet mask, and converting decimal to binary, etc.

Thanks! @Irishian88J @RainMotorsports @Mystic anyone else work as a help desk professional or network engineer?
 
Since you're probably taking a test you have to learn this, and supernetting which is a real pain in the ass, and I never really got. I can tell you in the real world, unless you're working on massively large projects, you will probably never us past a /24 - /32

I have done some white listing past /24 when white listing large networks for other companies.
 
i am not a network engineer but i do deal with alot of that area. our WTG parks have dual Firewalls and a single router plus 3-5 stacked switches all routed via MPLS to our corp office (we have to comply with NERC regulations).

i have used https://networkproguide.com/cidr-subnet-mask-ipv4-cheat-sheet/ as a cheat sheet to help me understand what is actually going on as our Real Time Automation Controllers only accept CIDR's no specific set mask's.

on each of our parks we separate different VLANs for different subnets. we use /16, /24 and if i need to set a static route a /32.

i actually just had a problem that was cool to solve, i needed to dual home one of my RTAC's one had an IP of 172.20.21.201(not really the IP) and the other one 172.16.14.155. so with this problem the RTAC's interface was pretty shitty and would only let me operate on the 14.155 network even though it all looked the same. the routing tables forced everything on that network. so what i needed to do is set my subnet to very specific range as i was attempting to hit devices at 172.16.16.xxx and 172.16.14.xxx. i had to set static routes so that the device would work on all of the tables. it was interesting and a crash course'ish into networking which i still dont understand.

but what im saying is make yourself a problem and solve it using your resources.
 
We're divorcing ourselves from MPLS. I just did a maintenance Tuesday night where we reconfigured all phone and VTC traffic for a branch office to route out of a secondary network via SDWAN, with some traffic shaping and QOS policies in place. It's going to save us a bundle on overpriced MPLS circuits. SDWAN is here, it's now, it's WOW! (yes I stole that lol)
 
Statistic68 said:
We're divorcing ourselves from MPLS. I just did a maintenance Tuesday night where we reconfigured all phone and VTC traffic for a branch office to route out of a secondary network via SDWAN, with some traffic shaping and QOS policies in place. It's going to save us a bundle on overpriced MPLS circuits. SDWAN is here, it's now, it's WOW! (yes I stole that lol)
Click to expand...
SD-WAN is nice, but it still has issues in my book. You are relying heavily on the cloud to complete a connection, instead of knowing exactly where the endpoints are. in an MPLS.
 
I would suggest using this

:
casio_fx_85wa_calculator-s.jpg
 
Soulzz said:
So, I'm having a little problem understanding subnetting, and base 2 to decimal conversion.

I understand the purpose of subnet masks, and how to read the first two octets of an IP, and the last two. I understand Class A, Class B, and Class C address ranges.

The issue is understanding specific details on how to find the subnet mask of an example like 10.50.40.10, or 255.255.0.0/16 (I'm not sure if that's the correct CIDR value). Basically, the theory of it. To be honest, in most environments, creating the network mask and infrastructure is a small part of supporting the client. You really don't do it that often.

All I need is an ebook recommendation going in-depth on mastering subnet masks, creating an accurate subnet mask, and converting decimal to binary, etc.

Thanks! @Irishian88J @RainMotorsports @Mystic anyone else work as a help desk professional or network engineer?
Click to expand...

LOL. So I was reading someone bitching awhile back about having to do subnet calculations on paper. I told him look when the network is down you can't just google it. You have to learn this shit for the day you end up in that edge case scenario.

I am not your guy for sure. I do basic networking and thats it. When our companies internet went to shit they tried to tell us it was our equipment. I told my boss the truth is I am not qualified to diagnose this shit but no question their modem is defective. About a week later we got a letter in the mail asking us to call them to replace our modem that has a known defect. That's about as far as I have gone into serious networking lol

The Network+ is the one I am scared to take but most interested in. The Security+ practice tests are a piece of cake but I still feel under qualified for.
 
Soulzz said:
SD-WAN is nice, but it still has issues in my book. You are relying heavily on the cloud to complete a connection, instead of knowing exactly where the endpoints are. in an MPLS.
Click to expand...

I'm not sure what you're referring to, but SDWAN is local firewall/or SDWAN devices ran service that can load balance and traffic shape multiple disparate internet connections. For example, I could have a T1, a 50/50 synchronous dedicated fiber, and a 120/20 asynchronous cable modem, and set various policies in my firewall to do things like load balancing between them all, or say putting all streaming traffic on one, backup traffic on another, and SIP phone traffic on yet another, but have them all back each other up in case there are either outages or connectivity issues like packetloss. It almost eliminates the need to route though someone's MPLS cloud, which basically gives you QOS on the internet. SDWAN can handle as many circuits as your SDWAN device has ports. One of the techs I know actually uses it at home leveraging some cellular hot spot devices to back up his primary consumer broadband connection.
I would say SDWAN is the future, but as I understand it it's been around for about 10 years. Everyone is moving to it, even Windstream, who are the one's who provide MPLS to us! lol.... they're trying to sell us SDWAN now.
 
Statistic68 said:
I'm not sure what you're referring to, but SDWAN is local firewall/or SDWAN devices ran service that can load balance and traffic shape multiple disparate internet connections. For example, I could have a T1, a 50/50 synchronous dedicated fiber, and a 120/20 asynchronous cable modem, and set various policies in my firewall to do things like load balancing between them all, or say putting all streaming traffic on one, backup traffic on another, and SIP phone traffic on yet another, but have them all back each other up in case there are either outages or connectivity issues like packetloss. It almost eliminates the need to route though someone's MPLS cloud, which basically gives you QOS on the internet. SDWAN can handle as many circuits as your SDWAN device has ports. One of the techs I know actually uses it at home leveraging some cellular hot spot devices to back up his primary consumer broadband connection.
I would say SDWAN is the future, but as I understand it it's been around for about 10 years. Everyone is moving to it, even Windstream, who are the one's who provide MPLS to us! lol.... they're trying to sell us SDWAN now.
Click to expand...
That's my point. SD-WAN goes through an ISP, then you rely on a host of cloud networks. At least with a private MPLS, you know exactly what networks are connected.

I suppose it's just like AWS as far as scalability, but SD-WAN in my research is no where near the stability of AWS. If you can show me a rock solid provider, and Windstream.. meh.

I know how SD-WAN is beneficial. But just like the cloud, we can't put all of our eggs in one basket. There is a purpose for on-prem equipment.
 
Soulzz said:
That's my point. SD-WAN goes through an ISP, then you rely on a host of cloud networks. At least with a private MPLS, you know exactly what networks are connected.

I suppose it's just like AWS as far as scalability, but SD-WAN in my research is no where near the stability of AWS. If you can show me a rock solid provider, and Windstream.. meh.

I know how SD-WAN is beneficial. But just like the cloud, we can't put all of our eggs in one basket. There is a purpose for on-prem equipment.
Click to expand...

I'm not sure what you mean by cloud networks. MPLS is a cloud network, SDWAN utilizes disparate internet providers (no cloud). With MPLS, you have a single point of failure, the MPLS network provider. With SDWAN, you have no point to point QOS, but you can mimic the same things via traffic shaping. IE. prioritizing applications before they hit the internet, (on both sides of the connection) and monitoring the health of circuits so that if there is a problem like packetloss, you can utilize a different carrier/backbone. Everyone is moving to SDWAN because it's cheaper. The first office we put it at was paying $1300/mo for an MPLS circuit. Now they have 2 circuits on two different backbones, and pay about $600/mo, and SDWAN gives them more control over how those circuits are used.

Also, MPLS is still dependent on your last mile provider. There is no true dedicated MPLS circuit. In the above example, when we had MPLS, ATT was the last mile provider. You have to traverse their circuit to get a hand off to the MPLS provider, then you're in the MPLS network, then they dump you on another last mile provider, in our case Comcast. In the above example we now use both ATT and Cox for the branch office, with SDWAN.

There is still a case for MPLS if you don't mind the cost. Larger companies will use both, SDWAN and MPLS, using MPLS for SIP traffic (phone system). This is what Windstream was trying to sell us. For most companies though, SDWAN is a huge breakthrough in cost savings.
 
What I mean by MPLS is that it is generally a private network. My private I don't mean private cloud I mean it's a low utilization network.

I think you and I are approaching this from a different perspective.

I remember when cloud computing just came out in everyone wanted to go to it. After about 6 years of rhetoric from cloud providers and marketing people everyone realized that you can't put all your eggs in one basket.

The same is true for SD-WAN.

My point is you need to connect to your ISP then SD-WAN. How is that any more reliable then going through a less utilized MPLS?

I'm pretty sure there is no AWS equivalent in terms of magnitude and scale of SD-WAN.
 
That's not what SDWAN is. It's not something you connect to and it's not a cloud of any type. It's simply intelligence built in to a firewall, or a dedicated SDWAN device, like Silverpeak, that has the ability to handle multiple internet connections at your site. Think of it as a switch, or a router, with intelligence. The intelligence means that you don't have to rely on an MPLS cloud. MPLS "is" a cloud service. You can connect straight to another office, you 'could' connect to cloud services, you can even use SDWAN to control how you connect to an MPLS service. There are allot of tricks sdwan can do.

https://www.netbraintech.com/ftp/EE61/OnlineHelp/mpls_cloud.htm

SDWAN is kind of the opposite of putting all of your eggs in one basket. Because you can have multiple, often cheap, circuits at any given office, and basically treat them as a kind of bonded circuit, you're divorcing yourself from one basket. MPLS on the other hand is putting all of your eggs in one basket. If your MPLS cloud provider has an issue, your traffic is dependent on their network, their backbone, their services.

Don't get me wrong, as I said there is still a place for MPLS if you can handle the cost, but i would put them together in that case.
 
Any way, this is getting off topic.

I'm fine with having discussions, but this was created for Network+ study guide recommendations, not a debate on the merits of SD-WAN.

Thanks.

I think you replied just before I wrote this.

Please stay on topic. I created this thread for Network+ advice, thanks :).
 
Anyone else? I'm checking with my boss to see about getting reimbursed for the book. If no one says anything, I'll stick with the network+ by Mike Myers.
 
You must log in or register to reply here.
Back
Top