What went wrong, right, and how we could have fixed it

Today was a very hectic day. My group was given the task of troubleshooting PCs and getting the old computers up and running again. In the beginning, a lot of things went wrong. The PC we got didn't have a hard drive or RAM or power supply and because of that we couldn't get that one running. Then we got a new PC with everything in it already and that made our lives so much easier because we could not find the parts for the old PC. Then we could not find a monitor, so we had to trade in a monitor for another one. In order to do so, my partner and I had to carry monitors back and forth from computer labs. When we got the new monitor, it was missing an audio plug so we had to bring it back and get another one. That was pretty much all that went wrong today. Then we just plugged the monitor and PC in and got them working. Then the final thing went wrong, there was no Internet connection because we did not have Ethernet cables to connect to the Internet. However, we were not able to fix this because we didn't have time or else we would have fixed it by finding a way to get wireless internet.

UDP and the Transport Layer

UDP is a type of protocol that stands for User Datagram Protocol. It is an integral part of the Transport Layer in the TCP/IP Model. The UDP protocol allows computer applications to send messages on an IP network. UDP is an unreliable datagram protocol and it is also connectionless. The word "unreliable" simply means that there is no possible way to confirm that the message has reached the host destination or not. Furthermore, there is no way to verify that the message reached the destination correctly or not. According to docstore.mik.ua, UDP is the most efficient protocol for the Transport Layer. The site gives the example that if the data being sent over the TCP/IP Network is small, then the overhead of making connections and assuring reliable delivery is greater than re-sending the data packet itself. As we learned in class today, Transport Layer protocols require minimum overhead and the UDP protocol gives just that and that is why it is helpful to the Transport layer.

Source: http://docstore.mik.ua/orelly/networking/tcpip/ch01_06.htm

Transport Layer Problems

Sony

Sony's main problem was security. A group of hackers hacked into Sony's Playstation Network and was able to steal the information of thousands of users including credit card information, email addresses and home addresses. The reason this monumental hacking occurred was a fault in Sony's security of the Playstation Network. There was a major lapse in security and this allowed hackers in Anonymous to infiltrate the network and steal this information. Sony responded quickly as soon as they were made aware of the hack and shut down nearly 77 million PSN users' accounts. As a precautionary measure, Sony fixed the problem and upgraded their security and then turned their heads towards the customers. Sony offered PSN users an apology and a one month of Playstation Plus+ for free, access to Qriocity for a month, two free games from a selection of five games, as well as a identity-theft program. The total cost of this event for Sony was nearly $170 million.

Sources: http://www.esecurityplanet.com/headlines/article.php/3934276/Sony-Says- PlayStation-Security-Breach-Will-Cost-170-Million.htm

http://www.quora.com/PlayStation-Network-Hacked-Outage-April-May-and-June- 2011/What-caused-Sonys-PSN-outage-in-April-2011

Blackberry

Blackberry suffered an outage in service when it's core switch at Slough data center crashed. RIM, the maker of Blackberry, is still clueless as to what caused the biggest network crash in its history. When RIM was asked about the lapse, it gave no straight answer. However, according to Pcworld.com, the real reason this lapse occurred is because RIM does not have the capacity to re-route traffic to countries that are supported by the Slough switch. Blackberry is slowly resolving the outage and the network is limping back to normal after Blackberry replaced the switch. However, Mike Lazardis apologized to the public via a YouTube video.

Sources: http://www.telegraph.co.uk/technology/blackberry/8824912/BlackBerry-services- restored-worldwide.html

http://www.pcworld.com/article/241917/rim_clueless_about_cause_of_blackberry_outage.html

What is DHCP and how, exactly, does it work?

DHCP stands for Dynamic Host Configuration Protocol and it is a server service. Its main purpose is to lease IP addresses to network clients. Although it's main purpose is to lease IP addresses to network clients, it also provides related IP information such as subnet masks, default getaways, and the IP address to DNS servers. DHCP works by leasing an unique IP address to a network client for a certain period of time. DHCP works through four main steps:

1. When a network client's computer boots up, it sends out a packet called DHCPDISCOVER and this packet contains the computer name and MAC address and basically requests for a IP address from DHCP servers.
2. DHCP servers then respond to the packet with a DHCPOFFER.
3. Then the client computer responds with a DHCPREQUEST packet and this basically means that the client has accepted the offer and would like to lease the IP address.
4. This is the final step of the DHCP process. The DHCP server responds with a DHCPACK message that contains all the IP information and acknowledges the acceptance of the offer.

The client computer is part of the TCP/IP network and can communicate on it now since every computer on this type of network model needs an unique IP address.

-Information paraphrased from http://www.dummies.com/how-to/content/installing-and-configuring-dhcp.html

Peer to Peer (P2P) & Client-Server Models

Advantages of Peer to Peer

-The advantages of peer to peer (P2P) are that information gets from the source to the destination faster. In addition, these networks are easier to setup than client-server models. There is also no need for a dedicated server since all the pieces of equipment attached to the network act as servers and clients. Also, in P2P networks, when the demand on a network increases, so does its capacity. There is also no need for network administrators on P2P networks, the network is cheap to setup and fast to run and finally, every computer in the network can make backups of its data to the other computers for security.

Disadvantages of Peer to Peer

-There are not many disadvantages to using a Peer to Peer network but the biggest one would be that it does not a support a wide range of computers. These networks are usually used for a small amount of computers usually around 10-15. Another problem is that these networks are most commonly used in applications that don't have much security such as LimeWire and FrostWire, therefore, these networks are not advisable for businesses and other large corporations. As you can see, another problem with Peer to Peer networks is the lack of security and their vulnerability. Furthermore, these networks are not centralized and therefore make network administration is virtually impossible even though it is not needed for such a small network. Finally, no computer in the network can be relied upon to keep information safe for an extended period of time.

Advantages of Client-Server

-Client-Server networks are centralized and this is one of their biggest advantages. This means that all resources and data security is controlled by a server. Another advantage is its scalability. Nearly all elements in a client-server network can be replaced with ease as the need for it increases. These networks are also very flexible and can have new pieces of technology integrated into them easily. Client-Server networks also have great accessibility and can be accessed from multiple platforms.

Disadvantages of Client-Server

-Client-Server models have a long list of advantages and a fairly short one of disadvantages. For starters, these networks can fail more often than Peer to Peer networks because they have to depend on one central server to operate the network. Client-Server networks differ from Peer to Peer networks because they can foster more people on their network. They do this by relying on one server to operate things smoothly and because so many people can be on the network can be on at once, the server can easily be overloaded and although this problem can be fixed easily, it takes time to do so. The final disadvantage to client-server networks is that they are quite expensive and difficult to set up.

Which is more centralized?

- A Client-Server Model is more centralized because it depends on one central server to run the network.