What questions do you ask when interviewing a network maintenance engineer?

Physical layer defines some electronic electrification standards on equipment interfaces, such as RJ45 interface and optical fiber interface. Transmission medium twisted pair, wireless, optical and electricity. wait for

-Data link layer

The second layer defines an important representation, MAC address, which must be unique in the LAN. It is also composed of 48 hexadecimal digits, with the first 24 digits representing the manufacturer and the last 24 digits being the serial number defined by the manufacturer. Sometimes, the MAC address just indicates the location of the device.

-Network layer

The network layer is used for logical routing addressing, and the most important protocol at this layer is IP protocol. Ip is divided into ARP, RARP, ICMP, IGMP, etc.

-Transport layer

This layer defines two important protocols, TCP and UDP. There is also the concept of port number. This layer is associated with a program or service on the host. For example, the web service of tcp 80 and the QQ program of udp 4000.

-Dialogue layer

The main function is to establish sessions and manage sessions. My understanding of dialogue is generally like this, such as telnet, a host, is a link to dialogue. When you open Baidu's webpage, you establish a session with Baidu's server.

-Presentation layer

Because the bottom layer transmits binary, the application layer can't directly recognize it. So according to the name of this layer, it can be directly understood that he is a translator. For example, a long string of data is "translated" into rmvb format and handed over to the high-level fast broadcast program, and another string of data is "translated" into MP3 format and handed over to the music player. In fact, there is a lot of work on this floor.

Compression, decompression, encryption, decryption, etc.

-Application layer

It provides an interface for users to operate, such as windows desktop or UNIX character interface.

Each of the seven layers of OSI works independently, but there is a relationship of "cooperation" and "compatibility" between them.

1.2[ Difference between Layer 3 Switch and Router]

Although both Layer 3 switches and routers can work at Layer 3, they are essentially different.

One is that the design function is different.

There are many multifunctional routers that can realize three-layer routing functions, including NAT address translation. Some provide layer 2 ports, and some are equipped with wireless functions. Then there is the function of firewall. But you can't divide it into switches or firewalls. Can only say that it is a multi-function router. The second layer exchange of firewall is only his additional function. The same is true of three-layer switching. Its main function is to solve the frequent communication of data in LAN. There are three layers of functions, but they are not necessarily much worse than routers.

Second, the application environment is different.

The routing function of three-layer switching is relatively simple, because it is more used in communication within the local area network, and its main function is data exchange.

The main function of router is routing addressing, which is suitable for different networks, such as local area network and wide area network, or different protocols.

Third, the implementation methods are different.

Routers can implement three-layer routing (or forwarding) based on software. When receiving a packet to be forwarded, it takes a series of complicated processes, such as looking up the routing table and the longest matching principle, to finally realize packet forwarding, which is slightly less efficient than three-layer switching. Three-layer switching is a way to realize three-layer functions based on hardware. After he successfully forwards a packet, he will record the correspondence between ip and MAC. When the data is forwarded again, it will be forwarded directly according to the previously recorded entries. This process is called "routing once, switching many times".

In a word, the biggest difference between a Layer 3 switch and a router is that a router can do port-based NAT, while a Layer 3 switch cannot. Routers can access the Internet directly by accessing optical fiber, but Layer 3 switches can't. Mainly, each interface of the three-layer switch has a proprietary MAC address and a specific ASIC integrated circuit.

.

1.3[ Difference between Static Routing and Dynamic Routing]

Static routing characteristics

Static routing is manually configured by the administrator, which is very accurate. But it is not flexible enough, and it is one-way. Considering the stability of static routing, resources (memory, cpu, link bandwidth) are saved. Often used in environments where the top layer of the network is not very large.

Characteristics of dynamic routing

The advantage of dynamic routing is that routers learn from each other by running dynamic routing protocols, which reduces the workload of engineers to some extent in large-scale network environment. There are many dynamic routing protocols, IGP and EGP. According to the working principle, IGP can be divided into link state type and distance vector type. But no matter what kind of dynamic protocol, he has to go through the following process.

1. Send routing information to other routers.

2. "Listen" to receive routing information from other routers.

3. Different dynamic routing protocols have different algorithms. Each routing protocol calculates the received routing information through its own unique algorithm, obtains the best routing entry and loads it into the routing table.

4. "Maintain" the routing table. When the TOP changes, update your routing table in time and send a change message.

In the production environment, we should have different network sizes and choose different routing protocols.

1.4[ Describe ACL and NAT]

ACL:acl access control list is a mechanism for making rules. It is used to tell the router which packets are allowed to access which resources and which are denied. Can be divided into two ways. One is the standard access control list, which can only be restricted according to the source address. It is an extended access control list, which can not only be filtered based on source address and destination address, but also be restricted based on port. ACL works by reading Layer 3 and Layer 4 of a packet and matching it with the entries in the access control list. If there is a match, it will be matched directly according to the policy (allow or deny) and will not be matched further. If there are no matching entries, the default rules will be followed.

NAT:nat network address translation is to solve the problem of insufficient ipv4 address space.

The principle of Nat is to replace the source ip+ port or the target ip+ port in the data packet, so as to achieve the situation that the private address can no longer be propagated on the public network or the release of the server in the local area network. Nat improves the security of LAN users to some extent.

Implementation can be roughly divided into static NAT and dynamic NAT PAT (port reuse).

1.5[ Describe VLAN]

VLAN is a means to avoid the influence of broadcast storm on large-scale users on the second floor.

Benefits brought by virtual local area network

Reduce broadcast storms

Improve certain security.

Simplify network management and make troubleshooting easy.

Vlan is a logical division of LAN, which can be realized in two ways.

1. Static vlan Port-based vlan (Generic)

2. vlan based on MAC address (suitable for mobile users)

Communication between Vlan needs to be configured with TRUNK. There are two packaging modes.

Isl？ Cisco's proprietary technology adds 30-byte identifiers to the head and tail of data frames.

Dot 1q？ Identify vlan ***4 bytes by marking after mac address? Public agreement

1.6 [RIP difference between RIP and OSPF]

Both of them belong to IGP protocol, and rip is a typical distance vector dynamic routing protocol. Ospf is a link-state protocol.

Rip is a broadcast update of the entire routing table (v2 is multicast), learning unknown route entries, and sometimes loops appear.

And there is no neighbor list, so the network convergence speed is slow. Because of the defect of loop, it is not suitable for larger networks.

Ospf multicast updates, and only when the TOP changes, the updated route will be placed in LSDB, and the route will be generated. Ospf itself has no loop, it has a hierarchical structure, while rip is flat and has no hierarchy. So ospf converges faster than rip. There are also concepts of DR and BDR in NBMA network, which promotes the convergence of ospf.

Rip management distance 120? The Ospf management distance is 1 10.

1.7[ Explain the meaning of the following terms]

Lan? Local LAN

Wan? Wan road

VLAN virtual local area network

WLAN wireless LAN

Virtual private network

AD？ Management distance is used to measure the trust value of different routing protocols to the same destination.

Metrics are used to judge the priority of the same routing protocol to the same destination.

1.8[ Briefly describe what stp is.

Stp？ Spanning tree protocol.

A good network should consider the redundancy of links, such as the redundancy of switches on the second floor to prevent problems caused by a single point of failure. But if the second floor is redundant, it will bring some problems:

1. broadcast storm, because the processing of unknown data frames in the second layer is broadcast, and the encapsulation structure of the second layer is not protected by TTL mechanism like the third layer. Therefore, once the broadcast storm occurs, other switches will follow the broadcast, causing the link to be blocked and paralyzed.

2. Copy 2. MAC address. Because of the working principle of the second layer, it will cause the switch to learn a MAC repeatedly, causing unnecessary waste of resources until the equipment is paralyzed.

3. number three. The MAC address table is unstable because you have to learn some addresses over and over again. Resulting in low forwarding efficiency.

The consequences of two-tier circulation are serious. Stp protocol is a logical excuse to break the loop and realize redundancy in a redundant environment. When the environment changes, it will automatically jump to the down interface.

1.9 [STP calculation flow]

1. Select the root bridge.

2. Select the root port

3. Select the specified port.

4. Specify a blocked port

2.0[ Describe HSRP]

Hot standby routing protocol is Cisco's proprietary gateway redundancy protocol. It is a hot standby group composed of a group of routers (at least 2 routers). Its function is to provide users with an uninterrupted gateway ip through which users can access the Internet. Even if the gateway equipment is down, it will not affect the normal work of customers.

Principle: The hot standby group includes the role of router in 4.

Active router: it is responsible for carrying the traffic sent to the virtual ip address and is the router that really forwards the user data traffic.

At the same time, send a hello packet to UDP 1985 to indicate your status, and there is only one in a group.

Standby router: Monitor the status of the entire HSRP group, that is, the status before becoming the next active router. Only one person in a group sends to the group at the same time.

Other routers: listen to the status of the entire HSRP group and are candidates for backup routers.

Virtual router: provide a virtual ip and MAC for the client? Active router forwarding is possible.

When the primary router is turned off, if the backup router does not receive the hello packet, it will become the primary router. This conversion process is imperceptible to users.

Please add a detailed explanation.