IP monitoring solution troubleshooting
The switches in the complete monitoring network are divided into access switches, aggregation switches, and core switches.
The access layer switch is used for the access of the terminal network device. The monitoring network generally refers to the access of the front-end equipment such as the camera, and needs to provide the uplink to the aggregation layer. The network interface is dense and the cost is low.
The aggregation layer switch is the data aggregation point of multiple access layer switches. It must be able to process all communication data from the access layer devices and provide uplinks to the core layer. Therefore, the aggregation layer switch has higher performance and higher switching rate than the access layer switch.
The core layer switch is the ultimate bearer and aggregator of all traffic in the entire network. It provides high-speed forwarding communication to provide data transmission and optimization of each backbone network. It requires higher reliability, high-speed transmission and throughput, and is the entire monitoring network. The device with the highest performance requirements.
For the network failures that are often encountered in IP monitoring, such as failure, packet loss, crashes, etc., our company has summarized some troubleshooting methods.
The IP monitoring network troubleshooting is very different from the previous analog monitoring troubleshooting. The IP network is a whole system. The network devices that do not pass the location are likely to interact with each other, causing regional failures and even network failures.
Common network failures:
Cause: The network cable connection is incorrect, and the switch spanning tree protocol is not valid.
Cause: The network device is damaged, the network port is flooded, and the network cable is damaged.
Scope of influence: The entire subnet exists in the loop, causing the network to be severely staggered or even embarrassing.
Precautions: Enable port loop detection, enable automatic spanning tree (STP, RSTP, MSTP), and divide multiple subnets.
Troubleshooting: Understand the network topology, reduce the fault range by disconnecting the network link one by one, and gradually determine the fault location.
2. Data blocking
1. The transmitted data traffic exceeds the range (40%) that the network device bandwidth can withstand, resulting in data transmission being blocked.
Data traffic in IP monitoring = actual code stream of the camera * number of cameras * number of back-end streaming devices
Impact range: links with network bandwidth bottlenecks
Precautions: Understand the network topology, calculate data traffic, and choose a reasonable switch transmission bandwidth.
Troubleshooting: Understand the network topology, use the switch port traffic monitoring function to check the actual traffic of the port, adjust the link according to the actual traffic, and upgrade the device.
2. The device uses UDP to transmit, which requires high network requirements.
Impact range: Install links with UDP devices
Precautions: Use a better quality twisted pair cable, and the directly connected 100M switch will enable flow control.
Troubleshooting: Determine the transmission type of the device and upgrade the network device.
3. Other faults
Fiber attenuation is too large
Cause: poor fiber fusion, poor pigtail, ash or unplugged
Impact range: The link device loses packets or fails, and the network device crashes.
Precautions: Measure the fiber attenuation during installation to ensure that the link attenuation is within the normal range. Troubleshooting: Measure link light fading, adjust fiber attenuation to normal value or upgrade device
Network equipment failure
Cause: The device is not working properly due to chip soldering or poor device, packet loss or crash
Scope of influence: Data passed on this device
Troubleshooting: Is it a device issue or another issue by replacing statutoryity?
Network cable failure
Cause: The quality of the network cable is poor, and the impedance is too high, which affects the data transmission. If the network cable is not pressed well, the contact is poor, sometimes there is no; the network cable breaks the wire and causes the loop.
Scope of influence: the port device or the subnet device
Troubleshooting: Determine the fault location through other troubleshooting and test with the network cable tester.
4. Example analysis
one. Network connection diagram of a traffic police project
Each central computer room aggregation device MNB26G belongs to different VLANs
two. Symptoms Image 13 in the image, the ping packet has a delay
three. On-site troubleshooting steps:
1 Connect the test notebook to the spare network port of the MNB26G switch, and select a faulty camera ping65000 package (camera IP172.21.73.90) in the garden road in the cool new village section, with a delay of more than 200ms.
2 Unplug the fiber on VLAN 13 one by one, test the computer to keep the ping packet status, keep the fiber connection of the 172.21.23.90 camera back. After the other fibers are removed, the ping packet delay is still about 150ms, which proves that the network delay of 172.2.373.90 is caused. The main point of failure is at this point.
3Recover the back-end fiber to the front-end troubleshooting, connect the test computer to the front-end EII08-8E (8-port switch), ping172.21.73.90 delay 200ms or more, plug the EII08-8E network cable one by one, when unplugging 172.21.73.93 When the network cable of the camera is turned off, the delay phenomenon disappears immediately. Plugging in the network cable and unplugging any other network cable, the delay phenomenon is not improved.
4 Connect 172.21.73.93 to the MIB10-4E-4F-2G air supply port, the delay phenomenon remains, remove, delay phenomenon recovery, let the center cooperate with the test, after removing 172.21.73.93, the delay disappears and the delay is inserted. Generated, consistent with the front-end test results, confirmed that this camera has a delay of 172.21.73.93 caused by this camera, it is recommended to replace.
four. Analysis of the cause of the failure:
Because the monitoring network has a Layer 2 architecture in VLAN 13, the network packet is transparently transmitted in VLAN 13. If the front-end device is faulty and generates a large number of broadcast packets, it will directly cause delays in transmission of other devices on the transmission line. If not, the network packet can be excluded. The fault can even affect all network devices in VLAN 13. The impact range is large and it is not easy to find the fault point from the switch. Only the elimination method can be used to narrow the fault range and slowly determine the fault point.
Fives. Troubleshoot such failure scenarios:
1 As described in the on-site troubleshooting procedure, reduce the fault range by inserting and removing the optical fiber and the network cable to determine the fault point.
2 Because the switch used in this solution is a managed switch, you can configure the NMS address. If a similar fault occurs, you can access the switch through the NMS address. By narrowing the port, you can narrow the fault range and determine the fault point.
1. Access the NMB26G switch network management, ping the fault point camera, keep the ping window status, and close the other Gigabit optical ports on the MNB26G one by one. When the port is closed, there is no delay in the ping packet of the fault point, then the faulty device is closed. On this port, if all other ports are closed and the fault persists, the faulty device is at the front end of the link.
2. After determining the faulty equipment intersection, log in to the front-end MIB10-4E-4F-2G (4 electric 4 light 2 Gigabit) network management system, and close the network port one by one to further narrow the range. Finally, determine which port is closed and the fault is eliminated. There is a problem with the device on the port.
six. Reduce maintenance difficulty through a 3-layer network architecture
This network architecture replaces the Layer 2 switch MNB26G with MNB5728GF (three-layer all-optical switch).
The network planning adopts a three-layer architecture. Each intersection is a VLAN that is subordinate to a subnet and is uploaded to the core Cisco switch using the Layer 3 mode.
Advantages: When there is a device failure at the intersection, it only affects the intersection device, does not affect other intersection devices, and only needs to connect to the front-end switch network management, and determine the faulty device by closing the port and eliminate it.
A conference venue monitoring project
Field device connection method 1
Device name: one light four electric, one hundred mega transceiver, MIB112G-8E-4G, gigabit transceiver.
Solution overview: The IPC of each layer is transmitted from the network cable to the layer switch, and then transmitted to the total weak box through the optical fiber. The data is received by the corresponding 100 Mbps transceiver, and then transmitted to the MIB12G-8E-4G. The Gigabit transceiver of the equipment room, passed to the central switch.
Program features: The front-end equipment to the first floor of the total weak box has a 100-megabit bandwidth, and the back-to-machine room has a gigabit bandwidth.
The user reflects the problem: the live preview image will be stuck, and the Ping packet will be seriously lost.
Analyze the reasons:
The video server and storage device used by the customer on the site belong to two manufacturers respectively, and the protocol cannot be docked, resulting in all camera stream streams.
The four IPCs are connected to one 100M electrical port. The actual traffic is 6Mbps*4*2=48Mbps. It has exceeded the transmission limit of the 100M electrical port. When the data traffic is larger during the day, packet loss will occur.
5. Network management software
The network management software can monitor the content: port connection status, port enable status, port real-time traffic alarm