This topic is to discuss the following lesson:
1 Like
Hello,
Interesting lesson !
I would like to build my network home lab and I I want to implement a landing server for remote access RDP/SSH to the whole devices i.e routers, switches, firewall, etc.
How can I do that to get access to this lab even outside my home ?
With regards!
Serge
1 Like
Hello Konan
The simplest way to do this would probably be to create a remote access VPN. YOu can do this if you have either an ASA or a Cisco IOS router. More info on how to do this can be found below:
If you do this, then you can remotely connect to your internal subnet and then you will have access to all of your internal devices, either via CLI, or via any other protocol (FTP, TFTP, HTTP, Telnet, SSH, etc…)
I hope this has been helpful!
Laz
1 Like
Thanks for your help.
Regards
Serge
1 Like
Hi experts, sorry to ask a simple and routine question but it will be helpfull for me to invest in correct devices. I’m Network Enginer by profession. I’m preparing for CCNP not only for certificate but also for my day to day work. I would like to buy used routers and switches for hands on experience. Please guide me which series of routers and swwitches would be good to Practice. Awaiting for your valuable inputs. Thanks in Advance.
Hello Vasanth
It’s good that you see your study for CCNP not just as a way to get a certification but as a practical learning experience for improving your real on-the-job skillset. We’re glad to be of help with the lessons, as well as via the forum responding to questions.
For almost all of the certification requirements for CCNP, using emulators/simulators should cover you for your labbing and study purposes. This includes both free as well as paid options. You can learn more about what’s available at this NetworkLessons note on practice labbing best practices.
Typically emulators are the cheapest solution. Purchasing actual devices is obviously more expensive, and strictly speaking, you don’t need to do so. However, some prefer the “feel” of real devices over a simulated environment.
For equipment that focuses on switching, I suggest you take a look at this lesson:
These are older models, but they should suffice for most of what you need. And the older they are (and if they are EOL too) the cheaper you can find them as used or refurbished equipment.
For routers, take a look at the Cisco 1841, 2801, or 2811. All of these support version 15.X IOS and should do nicely. They too are quite old, so you should find them cheap.
Now if you want to get your hands on an ASA as well, your best bet would be the ASA 5506-X which supports the newer FirePower features.
For wireless, consider the 2504 controller, and at least two 3502i APs, and a single stand-alone AP like the1142.
Now you won’t need the ASA and wireless gear for CCNP, but I just added those in for completion.
I hope this has been helpful!
Laz
1 Like
Hello Laz,
Thank you very much for your suggestion. I’m really excited to have a real hardware and also emulator for practice, thanks again.
1 Like
Hi,
Great article!
I’d like to ask about what René said:
“If you buy real switches, I would get at least three. This allows you to test protocols like spanning-tree without any issues.”
I understand that one switch isn’t enough, because you couldn’t experiment with scenarios involving BID configurations. But why aren’t 2 switches enough? You could connect multiple ports on your switches to create a loop.
One benefit I could think of of having 3 switches is that you can experiment with Multi-chassis EtherChannel technologies (StackWise, VSS, vPC), and another benefit might be that you can play around with 3-tier topologies as well if you have 3 switches instead of just 2.
Have a nice weekend.
Attila
Hello Attila
Yes, you are correct. You could explore most features of STP using two switches in this manner. However, you would not get the full experience. Having three switches allows you to explore more complex and more realistic scenarios, including gaining a better understanding of root bridge elections, multiple port roles, as well as more complex convergence scenarios.
Yes, these, too, are all valid reasons to get three switches. Although things like StackWise, VSS, and vPC require specific models that are typically more expensive and would probably go beyond what one would be willing to pay for a lab. Take a look at this NetworkLessons note on the topic of switch high availability options for more details.
I hope this has been helpful!
Laz
1 Like
Hello,
I’d like to ask a question about this sentence:
“The more devices you have, the more cables you have. I have like 6-8 UTP cables to a single server. Having to follow the UTP cable is a pain. If you have a cable wrap label on both ends, you can quickly identify the cable. I use a random 4-digit hexadecimal value.”
Why isn’t it better to label the cables according to which ports on which devices they connect? For example, if I name one of my switches SW1, and another SW2, then I’d label the cable that connects their ports accordingly. Meaning, if SW1’s G0/1 connects to SW2’s G0/2, then I’d label the cable with a label that says “SW1 G/1 to SW2 G0/2”. That way, even if I accidentally unplug the cable, I can see which two ports it must be connected to. (Having all of this documented in an Excel file would be a useful bonus as well.)
Why is the “random 4-digit hexadecimal value” method better than this one? Can someone please help?
I’d like to ask another question as well. I can see that some of your cables are criss-crossing each other. I assume that’s because the connections between your devices developed organically: as you worked on your labs, you thought about new ways to connect the ports according to the specific topologies that you wanted to create. It would be pretty difficult to predict precisely how you’d want to experiment as your studies unfold (a real-world scenario where you design everything in advance is, I assume, a different thing), but wouldn’t it be beneficial to connect some ports in advance so that the cables would mostly line up nicely? Meaning you could predict that in the future, you’ll connect maybe 60% of your ports, so you’d connect them in advance, even though the interfaces are shut at that point in time when you’re connecting them.
So the reason why you’d connect them at that point in time would be in anticipation of a possible future use of those ports. You’d then have cables that all line up nicely (e.g. SW1’s G0/1 to SW2’s G0/1, SW1’s G0/2 to SW2’s G0/2, etc.)
Kind of like this:
Pic if the link dies later (I’ve kept the link as well because it has a collection of other nice pictures, too):
Thanks.
Attila
Hello Attila
The important thing here is to label the cables. You can use whatever system you like to label them. Rene has chosen to use random hex numbers. Maybe because they look consistent or you can represent up to 65000+ numbers with four hex digits, so you never run out! The reasoning behind the specific format doesn’t really matter. Just use what works best for you.
I would avoid labels that give specific port numbers because if for whatever reason you change the port you use, you will have to go back and re-label the cable. It’s better to use a system that is more flexible. You will find that over time, a nice clean initial installation can quickly become a spaghetti mess simply because things change.
In datacenters and telecom closets that I have worked with, I find it very difficult to maintain a setup that is as clean and beautiful as the image you shared. That’s the ideal, however, when you have multiple people accessing a telecom closet, making changes at regular intervals, it’s difficult to maintain. It also depends upon the environment you are working in. Some enterprises will have very few changes, so it’s easier to maintain this level of tidyness, while others change all the time. If you can anticipate how cabling will be in the near future, that’s great, but if not, then you will have to either spend a lot of time rewiring occasionally, or be content with a little bit of cable crisscrossing… 
I hope this has been helpful!
Laz
1 Like
Hello Laz,
Thank you.
So then the idea behind picking a random number is that when you plug a cable into different ports, you update your digital documentation of that cable? For example, if you’re using Excel, you have a cell with those random numbers, and next to that cell you document which two ports they plug into. And when you plug any cable into different ports, you just rewrite the info in the cell that’s next to the random number’s?
Like so:
If you filter the columns, you can easily check a specific device only, or all connections between any devices.
Thanks.
Attila
Hello Attila
Again, I’m not sure of the actual logic that Rene is using for the hex numbers, however, the idea I have in my head is that the labeling follows the cable rather than the switch port. Throughout the lifetime of the cable, it may be plugged into various ports, moved, and replugged in, but the hex number follows the cable. This eliminates any relabeling of cables. At the same time, you can update your spreadsheet accordingly, as you have suggested in your post.
That way you may have a network with thousands of switch ports in multiple telecom closets or even in multiple sites, but your cable labels will be unique throughout your whole enterprise. That is beneficial and does make things easier I believe. Does that make sense?
I hope this has been helpful!
Laz
1 Like
Hi, great post here it was very informative. I had a few questions.
What exactly creates the internet our isp gives us when we use the coaxial cable to our modem? Is it an electrical current? I understand we get access to the ISPs network so we can access online websevers and stuff but what even generates this connection for us in the first place? How does the ISP get an internet connection? Is there like a power plant or something that is needed to create internet? What underlying mechanism or device does this? It cant just come out of thin air because we twist some cables the right way or is that how it works when we send electricsl currents through it?
I was also wondering how this works in an enterprise or a big business. Do they just have one modem that would connect to a bunch of routers in a network closet or would they have multiple modems that connect to a bunch of different coax connections?
Are there ISP plans that exist for these big companies that need to support all these devices and nodes?
I am just trying to understand the bigger picture.
Thank you guys!
Note:
I have done more research and I understand the infrastructure we used and have created needs power and electrical currents to work and make data travel. I still and wondering about the last two questions I had though.
Hello Cameron
The coaxial cable that comes into your premises and terminates on a Cable Modem is simply the physical infrastructure that interconnects your internal network with the network of the ISP. The coaxial cable carries data in the form of electrical signals. Your modem then converts these signals into a format that your devices can understand. In other cases, you may have other technologies and media that connect your premises to the ISP. These include fiber optics as well as telephone cables (for xDSL) or even a wireless link. This connection to the premises is often called “the last mile” or the “local loop”. It’s the method via which your local network is physically connected to the ISP.
This depends upon the size of the enterprise network, as well as the physical locations at which it exists. Typically, each physical location should have a link to an ISP. For large networks, this is usually fiber optic connections delivering multi-gigabit speeds. Such fiber optics will terminate on customer premises equipment (CPE) which usually belongs to the telco, that in turn, connects to the edge network devices of the enterprise network. Edge devices are typically firewalls or routers that route traffic further into the private enterprise network.
Typically, an enterprise will have two or more such links to ISPs in order to ensure redundancy in the event that one link fails.
Each ISP has its own network with points of presence (POPs) that physically exist throughout the geographical area where the ISP provides service. That ISP network is considered a Tier 3 ISP network. It, in turn, interconnects with other Tier 3 ISP networks as well as with Tier 2 ISP networks, as part of the hierarchical structure of the Internet. For more information about this structure, take a look at this NetworkLessons note on the topic.
I hope this has been helpful!
Laz
Hi Laz,
Thank you for taking your time to reply to me. Your help is greatly appreciated. I understand it a lot more now.
1 Like