1.1 Identifying Fault Information
1.1.1 Determining the Fault Type
Faults are classified into forwarding, startup, hardware, and other faults.
|
Fault Type |
Description |
|
Forwarding faults |
Forwarding faults cause service interruption or slow network response. When a fault occurs, check forwarding problems first. Possible causes of forwarding faults: l Network link problems l Incorrect route configuration l Hardware problems l Attack defense problems |
|
Startup faults |
Startup faults occur during device startups or restarts. |
|
Hardware faults |
Hardware faults mainly comprise system, board, and interface faults. |
|
Other faults |
Other faults may be confronted by users, including: l License faults l Forgotten passwords |
1.1.2 Identifying Fault Information
1 Forwarding faults
Determine whether the fault is caused by slow connections or service interruption. If the fault is caused by service interruption, determine whether it is partial or complete interruption, whether the interruption affects TCP or UDP, whether the affected nodes have common symptoms, and whether the upstream and downstream devices are normal.
2 Startup faults
Check whether the board restart time is the same as the fault time and whether any anomaly occurs during the board operating. If there is an anomaly, ***yze it. For example, if the voltage is too high, check whether there are logs recording abnormal information, such as board registration failures and resetting.
3 Hardware faults
Determine whether a board or interface is faulty. Check whether the board or interface fault time is the same as the fault time. Check whether anomalies occurred during board operating. If there is an anomaly, identify it.
4 Other faults
Check whether the product capabilities specified in the license the customer purchased have reached the upper limit.
1.1.3 Determining the Faulty Path
Generally, a customer network consists of multiple small networks. If a service or transient interruption occurs, you need to first ***yze the network where the fault occurs to narrow the fault range and then determine the fast recovery solution based on the fault range.
