Saturday, June 2, 2012

Don’t Sell – Help!


Don’t Sell – Help!

There are another three components that are critical to website success—besides avoiding the pitfall of focusing on product before knowing who your target market is and where they hang out.
The first is help, don’t sell. By helping you will sell. The second is build credibility and rapport through educating.
You’ll go to a sloppily put-together webpage and get these annoying boxes that come up asking if you want something you weren’t thinking about before but are certainly too annoyed by now to entertain considering.
When you’re marketing online, remember that people don’t shop online. People research online.
Most people go online for research on how to buy a new car, or increase their business, or to learn about raw foods, or whatever it may be. So your website has to be designed around helping, not selling. Informing, revealing, educating.
If you sell, you will fail. By helping you build credibility and rapport. People will believe in you, and because of that, you can show them the value in your product or service. And they will buy from you. It’s absolutely critical to what you do online.
Everything on your website—every button, every graphic, every word you put on the page has got to be built around helping, not selling. Make it most appealing to researchers, not to shoppers. Save the two-page sales letters and the paper. Whatever it is you’re the expert in, whatever you do, they want more information. They’ll call you, email you, or order right off the website.
The mindset of the online researcher-pre-shopper is, What’s in it for me? That’s the first question you want to answer on your website. If you can’t answer that question at the top of your webpage, change it or you’re going to fail. Why should anybody stay there? They’ve got hundreds of websites to go look.
Don’t have a mission statement at the top of your page talking about who you are, what you do, how great and fantastic you are, and how you have a Ph.D. Nobody cares about that.
They care about what you’re going to do for them. That doesn’t mean not making available credentials and testimonials, but it all goes back to writing headlines—something that will make them stay. Suck them in. Make them feel like they’re going to be learning something they didn’t know before.
Nearly the same marketing rules still apply. Walk a mile in your prospects’ shoes. Empathize with them in plain language. Pharmaceutical ads are written and spoken so that children understand them. Use simple and/or precise language (depending on the uniqueness of your market) as your keywords—the words prospects would be typing into Google or Bing to find you. If you’re selling cow juice but everyone calls it milk, you better call it milk too because people don’t search for cow juice, they search for “milk.” If you get stuck, there are writers at reasonable prices who can do it for you.
It’s not how well you know your product that is going to determine the level of your sales and your success; it’s how well you know your customer, so that you’re ready for them when they visit your web page.
Next month we’ll cover that last key to success in building an online business—having the best possible sales process you can have—one that costs you nothing and boosts your profits!

Thursday, May 3, 2012

How to redirect traffic in OSPF

How to redirect traffic in OSPF

VERSION 1 


Introduction


This document provides a sample configuration on how to redirect traffic in OSPF

Requirements


It is assumed that a person reading this document is familiar with general Ospf operation.

Components used


For this document 3700 Series router with IOS 12.4(2)T loaded in it.

Network Diagram


ospf.jpg

Configuration


The configuration files are attached with this document.

For this example we will take network 10.1.23.0/24 and manipulate routing in a way that R2 is the router which is always preffered over R3 by R1.

Before we start making changes in routing table of R1:

Lets have a look at the rouring table of R1 and see how it reaches network 10.1.23.0/24:

R1#show ip route ospf
2.0.0.0/32 is subnetted, 1 subnets
O 2.2.2.2 [110/65] via 10.1.12.2, 00:24:06, Serial1/0.12
3.0.0.0/32 is subnetted, 1 subnets
O 3.3.3.3 [110/65] via 10.1.13.3, 00:24:06, Serial1/0.13
10.0.0.0/24 is subnetted, 3 subnets
O 10.1.23.0 [110/74] via 10.1.13.3, 00:24:06, Serial1/0.13
[110/74] via 10.1.12.2, 00:24:06, Serial1/0.12

Now to make sure that Router R2 is always preffered over R3, we have number of options available like, we can change bandwidth or cost of links. However, the option we are going to use involves only one command and is easy to use. The command we will be using is max-metric router-lsa

We will configure this command on Router R3, here is the syntax:

R3#config t
Enter configuration commands, one per line. End with CNTL/Z.
R3(config)#router ospf 1
R3(config-router)#max-metric router-lsa

This will configure R3 to advertise a maximum metric so that the other routers do not prefer the router as an immediate next-hop in their Shortest path first calculations.

Verify


Lets see the routing table of Router R1:

R1#show ip route ospf
2.0.0.0/32 is subnetted, 1 subnets
O 2.2.2.2 [110/65] via 10.1.12.2, 00:03:12, Serial1/0.12
3.0.0.0/32 is subnetted, 1 subnets
O 3.3.3.3 [110/65] via 10.1.13.3, 00:03:12, Serial1/0.13
10.0.0.0/24 is subnetted, 3 subnets
O 10.1.23.0 [110/74] via 10.1.12.2, 00:03:12, Serial1/0.12

We can see that R1 only uses R2 as next-hop for networl 10.1.23.0/24 now.

Thursday, April 26, 2012

了解交換器如何傳遞封包-1

了解交換器如何傳遞封包

胡凱捷
交換器是網路中一個相當重要的設備,肩負著將網路封包送到正確目的地的任務。本文將講述交換器的主要功能、各種傳送Frame的運作模式以及如何學習來源端MAC位址等等,並且說明交換器在虛擬區域網路內的運作原理。
網路交換器(Switch)設備在網路中負責傳遞網路封包。在傳遞網路封包之前,交換器必須先知道目的地在哪裡、網路封包應該要從哪一個埠傳送出去。為了達到這樣的功能,交換器必須學習並且記錄哪個埠可以到達哪些網路位置,而這個學習過程牽扯到傳遞網路封包的方式。 

接著,還會介紹更複雜的環境:虛擬區域網路(VLAN)。交換器若處於多個VLAN的情況時,如何傳遞資料呢?畢竟有VLAN的環境和沒有VLAN的環境還是不一樣,而VLAN剛好又是交換器一個很重要的功能。因此,本文會詳細說明交換器網路設備在不同的網路環境中如何傳遞網路封包。 

交換器的主要功能 

用於乙太網路中的交換器能夠增加網路的可用頻寬,同時也能根據發送過來的網路封包,分析其中的來源端與目的端的MAC位址,製作出MAC位址表,並藉由這樣的MAC位址表正確地轉發網路封包。 

交換器雖然都是運作在網路七層架構當中的第二層,但是由於交換器擁有較高速的內部結構以及擁有比較多的埠介面,所以與傳統的閘道器(Bridge)比較起來,交換器能夠提供更多的網路流量。 

關於MAC位址的學習,交換器會從它們的埠上監聽所傳進來的Frame(經過交換器的網路封包稱之為Frame),檢測這些資料的來源端MAC位址,並且把MAC位址與埠號的對應關係記錄下來,儲存在本地端的MAC資料庫內,而這份MAC資料庫通常稱之為MAC Address Table或是Content-addressable Memory (CAM) Table。 

當交換器再次收到Frame時,會先到MAC資料庫中查看從哪個埠出去可以到達這個目的地MAC位址所指定的機器,如果目的端的MAC位址能夠在MAC資料庫內被找到,則這個Frame就只會從學習到的埠號轉發出去。 

倘若在MAC資料庫中找不到這樣的對應關係,這個Frame就會從所有其他的埠轉發出去(來源埠除外)。這裡所指的Frame,有人用中文稱它為「幀」,但是筆者建議還是使用原有英文來指稱會比較好。 

交換器網路設備傳送Frame的模式 

交換器在傳送Frame時,有「Store and Forward」、「Cut Through」、「Fragment-Free」三種可用的運作模式,以下分別加以介紹。 

Store and Forward模式 

在Store and Forward模式下,交換器會先把Frame完整地接收下來,然後才進行轉發的動作。因為在這個模式下,整個Frame都會被讀取,所以來源端及目的端的MAC位址都能夠被讀取。 

另外,CRC(Cyclic Redundancy Check)的錯誤檢查動作也會被執行,並採用相關的篩選動作,之後才會進行轉發。如果CRC的檢查失敗,這個Frame就會被遺棄。這種轉送模式雖然能夠確保Frame中資料的正確性,但是也比較費時,其所需要的延遲時間與Frame的資料長度有關。 

Cut-Though模式 

在Cut-Through轉送模式中,就不會像Store and Forward模式這樣的麻煩,一旦接收到Header之後,看到目的端的MAC位址,就會馬上執行Frame的轉送動作。 

因此,Cut-Though模式比Store and Forward模式快上許多,而且Cut-Though模式所需要的時間,並不會因為Frame的長度變長就跟著變多,因為閘道器及交換器只想取得目的端的MAC位址,而目的端的MAC位址都一定會存放在Frame資料內最前面的Header之內。 

不過,雖然有些交換器在這種模式下只想讀取MAC位址,但還是有某些交換器會讀取CRC值並記錄下錯誤數目,不過一旦有錯誤發生,在這種模式下,是不會將這個Frame遺棄,而是針對設定值自動或手動切換成Store and Forward模式。 

這種可以允許切換的Cut-Though又稱為Adaptive Cut-Though模式,此種模式結合了Cut-Though模式與Store and Forward模式的優點。 

Fragment-Free模式 

在Fragment-Free模式中,交換器只會讀取Frame的前64個bytes(這也是乙太網路中Frame最小的資料單位),而最主要原因是,在乙太網路中碰撞幾乎都是發生在前64個bytes之中,因此一旦發生碰撞,就會產生小於64個bytes的Frame資料。所以,交換器經由讀取前64個bytes資料,就可以初步篩選經由碰撞所產生的Frame資料。 

Fragment-Free模式的速度比Cut-Though更快,而且一旦發現有錯誤的Frame,將會把這樣的Frame丟棄,這是與Cut-Though模式不同的地方。這個模式也稱為Modified Cut-Though模式。 

了解交換器如何傳遞封包-2

了解交換器如何傳遞封包

胡凱捷
交換器是網路中一個相當重要的設備,肩負著將網路封包送到正確目的地的任務。本文將講述交換器的主要功能、各種傳送Frame的運作模式以及如何學習來源端MAC位址等等,並且說明交換器在虛擬區域網路內的運作原理。


交換器如何學習來源端MAC位址 

剛剛提到交換器會把Frame內來源端的MAC位址與埠的對應表記錄到MAC資料庫中,而這資料庫所能記錄的筆數和設備的型號有關,例如Cisco Catalyst Switch 2950系列就能記錄8,192筆。 

假設現在有一台Cisco交換器,其中四個埠分別接到不同的電腦,其各個電腦的MAC位址如下圖所示: 




當交換器第一次開機後,MAC位址資料庫是空的。也因此,當交換器收到Frame之後,一定會從所有其他的埠轉發出去,除了來源埠之外。這種轉發到除來源埠以外的其他埠的動作,稱為「Flooding」。用這種Flooding的動作來轉送Frame很沒有效率,因為會浪費很多網路頻寬。 

假設現在A發送一個Frame要送給D,如下圖所示,則這個Frame會經由E0介面傳送給中間這個Cisco交換器,假設這台Cisco交換器是使用Store and Forward模式,則這個Frame到了Cisco交換器之後會先存放在暫時緩衝區中。 




接著,因為這台交換器還沒有學到這個MAC位址應該要送往哪個埠,所以預設上就會採用Flooding的作法,透過其他的埠把這個Frame轉送出去,而在收到這個Frame的同時,這台交換器同時也學到一件事情:從E0這個埠出去可以到達A這台機器(因為交換器能夠經由E0介面收到由A這台機器的MAC位址所發送的Frame)。 

因此,這個新學習到的資訊就成為MAC資料庫的第一筆資料。此時MAC資料庫中可能就有以下這樣的資料: 




在MAC資料庫中的每一筆資料並非永遠存在,一旦某筆資訊在MAC位址資料庫中經過300秒沒有更新,就會被強制移除,除非這筆資料的MAC位址所對應的電腦在300秒內繼續發送Frame到交換器上,這筆資訊的計時器才會重新開始計算。 

假設接著C想發送Frame給A,如下圖所示。這個Frame會經由E4這個埠傳送到中間這個交換器。緊接著,Cisco交換器會去查看MAC位址資料庫中是否可以找到A這台機器的MAC位址資訊。 




因為剛才這台交換器已經學到可以透過E0這個埠找到A,所以這個Frame就只會經由E0這個埠傳送出去,這是因為C送給A而學習到C的MAC位址與E2的關係。 

此時,MAC資料庫中的資料就會存放以下的對應關係: 




這樣的學習模式會一直進行,等到所連接之網段上的所有MAC位址都被學習過,這台交換器就能夠完完全全做到智慧型的轉發動作(Intelligent Forwarding)。 

不僅如此,這樣的學習也有助於Frame的篩選動作,因為Frame只會發送到真正需要送往的埠而已,這樣就可以節省整體的網路頻寬,這種篩選動作稱為「Frame Filtering」。 

經由集線器的Frame篩選過程 

假設目前在Cisco交換器和一般PC端之間架設集線器(Hub)設備,而這台Cisco交換器已經學到A和B兩台電腦的MAC位址和埠的對應關係,如下圖所示。 




若A想送Frame給B,一開始,A會把Frame送給集線器,而集線器因為沒有辦法像交換器一樣可以學習MAC位址,所以為了能夠送達目的地,集線器就會把這個Frame做Flooding,送往所有的埠,用亂槍打鳥的心態傳送Frame。 

這樣一來,B在不需要Cisco交換器轉送Frame的情況下就可以收到由A發送過來的Frame,此時,Cisco交換器收到這個Frame後,將不會再經由原來的埠送回給集線器。 

廣播與群播的轉送過程 

即使交換器已經學習完整個網路中各個電腦的MAC位址與交換器的埠的對應關係,對於廣播與群播的封包而言,也都會用Flooding的方式來轉送,算是比較特別的案例。 

不過也是因為廣播與群播所代表的MAC位址,根本不可能存在於真正的網路中,例如廣播的MAC位址是FFFF.FFFF.FFFF,所以在找不到這個MAC位址的情況下,MAC資料庫內根本不知道廣播與群播的MAC位址要送往哪裡,因為無從學習起,所以就會用Flooding的方式來解決。
Related Posts Plugin for WordPress, Blogger...