WO2020135445A1 - 对路由震荡的定位 - Google Patents

对路由震荡的定位 Download PDF

Info

Publication number
WO2020135445A1
WO2020135445A1 PCT/CN2019/128055 CN2019128055W WO2020135445A1 WO 2020135445 A1 WO2020135445 A1 WO 2020135445A1 CN 2019128055 W CN2019128055 W CN 2019128055W WO 2020135445 A1 WO2020135445 A1 WO 2020135445A1
Authority
WO
WIPO (PCT)
Prior art keywords
route
routing
shock
detected
table entry
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2019/128055
Other languages
English (en)
French (fr)
Inventor
杨帜
石翔
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
New H3C Technologies Co Ltd
Original Assignee
New H3C Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by New H3C Technologies Co Ltd filed Critical New H3C Technologies Co Ltd
Publication of WO2020135445A1 publication Critical patent/WO2020135445A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/16—Threshold monitoring
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00—Routing or path finding of packets in data switching networks
    • H04L45/54—Organization of routing tables

Definitions

  • the present disclosure relates to the field of network communication technology, and in particular, to a route oscillation positioning method, a server, and a machine-readable storage medium.
  • the core of the Internet to realize the interconnection of all things is that the network equipment realizes the forwarding of messages.
  • the normal forwarding of the network equipment lies in the stability and integrity of the routing table. If the entries in the routing table oscillate, it may cause problems such as packet loss and low rate in the forwarding process from the source device to the destination device; more importantly, the service will be in a stagnation state, and the loss caused is difficult to measure.
  • FIG. 1 is a schematic flowchart of a routing shock positioning method provided by an embodiment of the present disclosure
  • FIG. 2A is a schematic flowchart of another route oscillation positioning method according to an embodiment of the present disclosure.
  • FIG. 2B is a schematic flowchart of another routing shock positioning method according to an embodiment of the present disclosure.
  • FIG. 2C is a schematic flowchart of another route oscillation positioning method according to an embodiment of the present disclosure.
  • 2D is a schematic flowchart of a method for locating an oscillation source according to an embodiment of the present disclosure
  • FIG. 3 is a schematic diagram of a routing information collection time interval and a route oscillation period provided by an embodiment of the present disclosure
  • FIG. 4 is a schematic diagram of a correspondence relationship between a shock phenomenon and a cause of shock provided by an embodiment of the present disclosure
  • FIG. 5 is a schematic flow chart of a method for locating route oscillations according to an embodiment of the present disclosure
  • FIG. 6 is a schematic flowchart of a process for polling a device to be detected by a network management platform provided by an embodiment of the present disclosure
  • FIG. 7 is a schematic diagram of a waiting detection queue provided by an embodiment of the present disclosure.
  • FIG. 8 is a schematic flowchart of information collection of a shock route provided by an embodiment of the present disclosure.
  • FIG. 9 is a schematic diagram of a hardware structure of a server provided by an embodiment of the present disclosure.
  • FIG. 10 is a functional structural diagram of a routing oscillation positioning control logic provided by an embodiment of the present disclosure.
  • FIG. 1 is a schematic flow chart of a method for routing shock positioning provided by an embodiment of the present disclosure, wherein the routing shock positioning method can be applied to a network management platform, for example, a server running specific network management software, as shown in FIG. 1,
  • the route shock localization method may include steps 101 to 103.
  • Step 101 Perform pre-detection on the device to be detected to determine the target device that meets the condition for detecting the flapping route.
  • route flapping usually involves more network devices, and the routing information on each network device also usually has more. If route flapping occurs, all routing information of all network devices is directly analyzed by To locate the concussive route, the workload will be very large and the efficiency will be very low. Therefore, in order to improve the efficiency of route concussion and reduce the workload of route concussion, you need to pre-treat the detection equipment when you need to perform route concussion. Detection to screen out the devices to be detected that meet the condition of the oscillating route detection. The devices to be detected that meet the condition of the oscillating route detection are referred to as target devices in this document.
  • the device to be detected is pre-detected to determine the target device that meets the condition for detecting the oscillating route.
  • steps 1011 to 1013 may be included.
  • Step 1011 Polling the detection device according to the polling time interval of the device to be detected.
  • Step 1012 For any device to be detected, when it is detected that the number of routing table entries of the device to be detected changes, the polling time interval of the device to be detected is subtracted by the first preset time to update the polling time interval.
  • Step 1013 When the updated polling time interval is equal to the preset trigger detection time, perform a step of detecting the flapping route of the target device.
  • the network management platform may perform pre-detection on the device to be detected by polling, that is, every certain polling time interval.
  • the initial value of the polling time interval of each device to be detected can be set in advance, and the network management platform can pre-detect the corresponding device to be detected according to the initial value of the polling time interval of each device to be detected.
  • the polling time interval of different devices to be detected may be different.
  • the polling time interval of each device to be detected may also be the same.
  • the network management platform may pre-check each device to be tested every 30 seconds, that is, the polling time interval of each device to be tested is 30 seconds.
  • the pre-detection of the device to be detected by the network management platform may include obtaining the number of routing entries of the device to be detected, and determining whether the number of routing entries of the device to be detected has changed, ie this time (not the first time) Whether the number of the route entries of the device to be detected obtained is the same as the number of the route entries to be detected obtained last time is the same.
  • the network management platform determines that the number of routing entries of the device to be detected changes, it subtracts the first preset time from the polling time interval of the device to be detected.
  • the first preset time may be set according to actual scenes, such as 3 seconds, 5 seconds, and so on.
  • the network management platform may only record the number of routing entries, and does not need to judge whether the number of routing entries of the device to be detected is Changes.
  • the preset trigger detection time may be set according to the actual scenario, such as 15 seconds, 20 seconds.
  • the network management platform can determine that the device to be detected satisfies the condition of the oscillating route detection. Furthermore, the network management platform can determine that the device to be detected is the target device, and perform step 102, that is, perform the step of performing oscillating route detection on the target device.
  • the above determination of whether the device to be detected satisfies the oscillating route detection condition according to the polling time interval of the device to be detected is only a specific example of determining whether the device to be conditioned satisfies the oscillating route detection condition, and It is not a limitation on the protection scope of the present disclosure, that is, in the embodiment of the present disclosure, whether the device to be detected meets the condition of the oscillating route detection can also be determined in other ways, for example, the device to be detected whose number of route entries have changed can be directly determined as The target device that satisfies the detection condition of the shock route is not limited in the embodiment of the present disclosure.
  • the network management platform may determine whether each device to be tested needs to be pre-detected every second preset time .
  • the second preset time may be set according to actual scenes, such as 3 seconds, 5 seconds, and so on.
  • step 1011 polling the detection device according to the polling time interval of the device to be detected, may include step 10111 and step 10112.
  • Step 10111 Each time a second preset time passes, the remaining detection time of the device to be detected is subtracted from the second preset time to update the remaining detection time; wherein, for a single device to be detected, the remaining detection of the device to be detected
  • the initial value of time can be the same as the initial value of the polling interval
  • Step 10112 When the updated remaining detection time is less than or equal to 0, perform pre-detection on the device to be detected, and set the remaining detection time of the device to be detected as the polling time interval of the device to be detected.
  • the network management platform may subtract 5 seconds from the remaining detection time of the device to be detected every 5 seconds.
  • the remaining detection time of the device to be detected when the remaining detection time of the device to be detected is less than or equal to 0, pre-detect the device to be detected, and set the remaining detection time of the device to be detected as the polling time interval of the device to be detected .
  • the network management platform can subtract the remaining detection time of the device A to be detected every 5 seconds by 5 seconds, so that the remaining detection time changes to 13 seconds and 8 seconds in sequence , 3 seconds, -2 seconds.
  • the network management platform can continue to pre-detect the device A to be detected on the one hand, and set the remaining detection time of the device A to be detected to The polling time interval of the device A to be tested (in this case, 30 seconds).
  • the pre-detection of the device A to be detected by the network management platform may include: obtaining the number of routing table entries of the device A to be detected, and comparing the number of routing table entries with the device A to be detected last obtained If the number of routing table entries is the same, keep the polling interval of device A to be detected unchanged; otherwise, subtract the polling interval of device A to be detected by 5 seconds, and the polling interval update will become 25 second.
  • the remaining detection time of the device A to be detected can be reduced every 5 seconds After 5 seconds, the remaining detection time becomes 25 seconds, 15 seconds, 10 seconds, 5 seconds, and 0 seconds in sequence; when the remaining detection time of the device A to be detected is 0 (that is, less than or equal to 0) seconds, the network management platform On the one hand, the remaining detection time of the device A to be detected can be set as a polling time interval (that is, 25 seconds), and on the other hand, the device A to be detected continues to perform pre-detection.
  • the network management platform may determine that the device A to be detected is the target device, and perform the step of performing shock route detection on the target device.
  • the network management platform in order to make more reasonable use of the device resources of the network management platform, to avoid the network management platform pre-detection of each device to be detected is too concentrated, when setting the initial value of the remaining detection time, different to be detected The remaining detection time of the device can be set to not exactly the same.
  • the initial value of the remaining detection time of different devices to be detected can be obtained by the following formula:
  • Ti is the initial value of the remaining detection time of the device i to be detected, 1 ⁇ i ⁇ N1, where N1 is a natural number greater than or equal to 1, N1 is the number of devices to be detected; DevIDi is the device identification of the device i to be detected; T0 It is the maximum polling time interval.
  • the maximum polling time interval can be set according to the actual scenario, such as 30 seconds; the symbol "%" is the remainder operation.
  • the network management platform always determines whether to perform rounds on each device to be detected according to the initial value of the remaining detection time of each device to be detected Inquiry, the frequency of pre-detection of each device to be detected will be different, resulting in more times of pre-detection of some devices to be detected in the same time, and less pre-detection of other devices to be detected.
  • the network management platform checks each of the devices to be detected according to the initial value of the remaining detection time of each device to be detected After the device performs pre-detection, the remaining detection time of each device to be detected can be set as the polling interval.
  • the initial value of the polling time interval of each device to be detected may be the same.
  • the polling time interval of each detection device is adjusted according to the change in the number of routing table entries of each device to be detected.
  • the change in the number of routing table entries of the network device is not necessarily caused by route flapping, but may also be caused by user configuration reasons, such as the user adding or deleting routes, so
  • the network management platform detects the number of routing table entries for the device to be detected N2 consecutive times (N2 is a natural number greater than 1)
  • N2 is a natural number greater than 1
  • the polling time of the device to be detected is added to the third preset time until the polling time interval of the device to be detected reaches the preset maximum polling time interval.
  • the third preset time can be set according to the actual scenario, such as 3 seconds, 5 seconds, etc.; the maximum polling interval time can be configured by the administrator according to demand.
  • the maximum polling time interval may be the same as or different from the initial value of the foregoing polling time interval, which is not limited in the embodiment of the present disclosure.
  • Step 102 Perform shock route detection on the target device to determine whether there is a shock route.
  • the network management platform determines the target device that meets the condition of the oscillating route, and then can perform the oscillating route detection on the target device to determine whether there is the oscillating route.
  • performing oscillating route detection on the target device includes: when it is determined that there are multiple target devices, performing oscillating route detection on the target device in sequence.
  • the above-mentioned step 102, performing vibration route detection on the target device may include step 1021 and step 1022.
  • Step 1021 For any device to be detected, when it is determined that the device to be detected is a target device that satisfies the condition for detection of the shock route, the target device is added to the tail of the waiting detection queue;
  • Step 1022 Perform the shock route detection on the target devices in the waiting detection queue in order from the head to the tail of the waiting detection queue.
  • the network management platform can perform shock route detection on one network device at a time.
  • the network management platform may add the target device to the waiting detection queue.
  • the network management platform can take the target device from the queue for waiting detection in order from the head to the tail of the queue for checking the flapping route.
  • the network management platform may determine whether there is an oscillating route on the target device according to the routing information of the target device; wherein, the routing information may include a destination address and a routing table The time when the entry appears.
  • the destination address is used to identify the routing table entry, and different destination addresses correspond to different routing table entries.
  • the above determination of whether there is a flapping route on the target device according to the routing information of the target device may include:
  • M is a natural number greater than or equal to 1;
  • the step of determining the oscillation source related to the oscillation route in step 103 is performed; or, when the oscillation of the routing table entry When the number of times is less than the first preset threshold of the number of oscillations, and the total time of occurrence of the routing table entry is less than the first preset time threshold, the step of determining the oscillation source related to the oscillation route in step 103 is performed.
  • the appearance state of a routing table entry changes, the number of oscillations of the routing table entry increases by 1.
  • the network management platform can poll several times according to the routing information collection interval (which can be set according to the actual scenario) (it can be set according to the actual scenario, and M times are taken as an example in this article)
  • the routing table of the target device can poll several times according to the routing information collection interval (which can be set according to the actual scenario) (it can be set according to the actual scenario, and M times are taken as an example in this article)
  • the routing table of the target device can poll several times according to the routing information collection interval (which can be set according to the actual scenario) (it can be set according to the actual scenario, and M times are taken as an example in this article)
  • the routing table of the target device to collect routing information.
  • the network management platform may count the number of oscillations of each routing table entry on the target device according to the routing information of the target device.
  • the network management platform may use the route Add 1 to the number of oscillations of the table entry.
  • the network management platform collects the routing information of the target device five times. If the routing information of the target device collected in the five times, the routing table entry a appears in the first to third times, but does not appear in the fourth time, The fifth occurrence, the number of oscillations of routing table entry a is 2.
  • the network management platform may determine the route corresponding to the routing table entry whose oscillation number is greater than or equal to the first preset oscillation number threshold as the oscillation route, and execute step 103 To determine the source of the shock associated with the shock route.
  • the first preset threshold of the number of oscillations can be set according to the actual scene, such as 2 times, 3 times, and so on.
  • the route flapping time will generally be less than a specific time (usually 15 seconds), therefore, it can be further determined that the number of flapping times is less than the first preset number of flapping thresholds
  • the appearance time of the routing table entry determines whether the corresponding route is a flapping route.
  • the network management platform may further determine the total time that the routing table entry appears, that is, M The sum of the appearance time of the routing table entry in the routing information of the target device collected at the second time.
  • the routing table entry b appears in all five collections, and the appearance time of the routing table entry b acquired in five collections is t1 ⁇ t5, respectively. , Then the total time that the routing table entry b appears is t1+t2+t3+t4+t5.
  • the network management platform may also determine that the route corresponding to the routing table entry is a shock route and execute step 103 Steps to determine the source of the shock related to the shock route.
  • the first preset time threshold may be set according to actual scenarios.
  • the time length between two square nodes is the routing information collection interval (taking 30 seconds as an example), and the time length between triangle nodes is the routing oscillation period (assuming it is also 30 seconds) .
  • the routing information collection interval is always 30 seconds, because the routing oscillation period is the same as the routing information collection interval, Every time the result collected at the node that collects routing information is that the route does not exist, the result of the detection is that the route does not exist, and it is not that the route is flapping.
  • the route collection interval can be changed dynamically instead of staying the same all the time.
  • the routing information collection time interval increases or decreases as the number of completed routing information collection increases.
  • routing information collection interval can be calculated by the following formula:
  • T3 T2– ⁇ T1 ⁇ N3
  • T3 is the currently used routing information collection interval
  • T2 is the initial routing information collection interval
  • ⁇ T1 is the first preset decay time
  • N3 is the number of completed routing information collection.
  • a first preset decay time (that is, ⁇ T1 above) may be preset, and when the network management platform collects routing information, the first preset decay time interval and the first preset decay may be used according to the preset initial routing information collection time Time ⁇ T1 determines the time interval for each routing information collection.
  • the time interval for collecting routing information multiple times may also be implemented in other ways, for example, according to the preset increase time and completed.
  • the number of routing information collection times updates the routing information collection interval that is, the routing information collection interval increases with the number of routing information collections, or you can directly set multiple different routing information collection intervals and follow the multiple
  • the routing information collection is performed multiple times at different routing information collection intervals, and the specific implementation thereof will not be repeated here.
  • Step 103 For any target device, when there is a shock route on the target device, determine the shock source related to the shock route and determine the cause of the shock.
  • the source device that causes route flapping is referred to as a flapping source in this disclosure, that is, the flapping of multiple network devices in the network is caused by a problem with the flapping source.
  • the shock source related to the shock route may be determined, and the cause of the route shock may be analyzed from the source.
  • the above-mentioned determining the source of the oscillation route may include steps 201 to 203.
  • Step 201 For any concussive route, determine whether the concussive route is a direct route of the target device; if yes, go to step 202; otherwise, go to step 203.
  • step 202 it is determined that the target device is the source of oscillation of the oscillation route.
  • Step 203 Determine the shock source related to the shock route according to the routing protocol information corresponding to the shock route.
  • the routing protocol table is a table that records the routes propagated through a routing protocol, such as OSPF (Open Shortest Path First First Shortest Path First) protocol, BGP (Border Gateway Protocol), and so on.
  • OSPF Open Shortest Path First First Shortest Path First
  • BGP Border Gateway Protocol
  • the routing protocol information can be collected by collecting the link state database table ospfLsdbTable in the MIB (Management Information) Base of the device to be detected, the externally imported link state database table ospfExtLsdbTable, and the BGP receiving path attribute table bgp4PathAttrEntry, etc., and then according to the shock
  • the routing protocol information corresponding to the route determines the source of the shock.
  • the network management platform when the network management platform determines that a shock route exists on a target device, for any shock route on the target device, the network management platform can determine whether the shock route is a direct route of the target device. If the oscillating route is a direct connection route of the target device, the network management platform can determine that the target device is the oscillating source of the oscillating route; if the oscillating route is not a direct connection route of the target device, the network management platform can The routing protocol information corresponding to the shock route determines the shock source of the shock route.
  • the network management platform can determine whether the concussive route is a direct route of the target device according to the ipRouteType (route type) field in the routing table entry corresponding to the concussive route. For example, when the value of ipRouteType is 3, it indicates that the corresponding route is a direct route; otherwise, it is a non-direct route.
  • ipRouteType route type
  • the determining the source of the shock of the shock route according to the routing protocol information corresponding to the shock route includes:
  • the routing protocol table entry corresponding to the oscillating route is in an oscillating state, after the current target device completes the detection, the corresponding publisher device in the routing protocol table entry in the oscillating state is tested for the oscillating route until the oscillation route is determined Source of shock
  • step 103 If the routing protocol table entry corresponding to the oscillating route is not in an oscillating state, the step of determining the cause of the oscillating in step 103 is performed.
  • the network management platform when the network management platform determines that the shock route is not a direct route on the target device, the network management platform can query the routing protocol table entry corresponding to the shock route and determine whether the routing protocol table entry corresponding to the shock route is in Concussive state.
  • the network management platform may add the corresponding publisher device in the oscillating routing protocol table entry corresponding to the oscillating route to Wait to detect the head of the queue.
  • the detection order of the devices in the waiting detection queue by the network management platform is the order from the head to the tail.
  • the network management platform After the network management platform completes the detection of the current target device, it can perform the vibration route detection on the issuer device of the shock route until the source of the shock route is determined.
  • the concussive route is a direct route of the publisher device of the concussive route, or the routing protocol table entry corresponding to the concussive route on the publisher device of the concussive route is not in concussive state, then the concussive route can be determined
  • the publisher device is the source of the concussion route; if the concussion route is not a direct route of the concussion route's publisher device, and the routing protocol table entry corresponding to the concussion route on the concussion route's publisher device is in concussion state, Then, the publisher device of the oscillating route may be further determined according to the routing protocol table entry information on the publisher device of the oscillating route, and the source tracing detection may be continued as described in steps 201 to 203 above until the Source of shock.
  • the network management platform may use the routing protocol table entry a Determine the publisher device of the oscillating route (assuming network device B), and the network management platform can further determine the corresponding route of the oscillating route on network device B (that is, the route on network device B that has the same destination address as the oscillating route) Whether it is a direct route on network device B.
  • the network management platform can determine the publisher device of the concussive route (assumed to be network device C) according to the routing protocol table entry b.
  • the concussive route is on the corresponding route on network device C (that is, network device C and The route with the same destination address as the shock route is the direct route of the network device C, or the route corresponding to the shock route on the network device C is not the direct route of the network device C and the route (that is, the shock route is on the network device) Corresponding route on C)
  • the corresponding routing protocol table entry c is not in a state of shock on the network device C, and the network management platform may determine that the source of the shock of the shock route is the network device C.
  • the network management platform may determine that the target device is the shock source of the shock route.
  • routing protocol table entries there may be multiple routing protocol table entries corresponding to one oscillating route, for example, a routing protocol table entry of BGP (Border Gateway Protocol) and OSPF (Open Shorttest Path First Open The shortest path first) routing protocol table entry of the protocol, when any of the multiple routing protocol table entries is in a concussive state, it can be determined that the routing protocol table entry corresponding to the concussive route is in a concussive state, and according to the concussive state The routing protocol table entry determines the publisher of the concussive route, and its specific implementation will not be repeated here.
  • BGP Border Gateway Protocol
  • OSPF Open Shorttest Path First Open
  • determining whether a routing protocol table entry exists in a shock state on the target device may include:
  • For any target device poll the routing protocol table of the target device P times according to the routing protocol information collection interval to collect the routing protocol information; where the routing protocol information includes the destination address and the appearance time of the routing protocol table entry;
  • any routing protocol table entry when the number of oscillations of the routing protocol table entry is greater than or equal to the second preset threshold of oscillation times, after the detection of the current target device is completed, the corresponding Steps for the publisher device to perform flapping route detection; or, when the number of flapping of the routing protocol table entry is less than the second preset flapping threshold and the total time of the routing protocol table entry is less than the second preset time threshold, execute After the current target device completes the detection, perform the oscillating route detection step on the corresponding publisher device in the oscillating routing protocol table entry; where, when the appearance state of a routing protocol table entry changes, the routing protocol table The item's concussion count is increased by 1.
  • the routing protocol information collection interval increases or decreases as the number of completed routing protocol information collections increases.
  • routing protocol information collection interval can be obtained by the following formula:
  • T5 T4– ⁇ T2 ⁇ N4,
  • T5 is the currently used routing protocol information collection interval
  • T4 is the initial routing protocol information collection interval
  • ⁇ T2 is the second preset decay time
  • N4 is the number of completed routing protocol information collection.
  • the specific implementation of the routing protocol table entry that the network management platform determines whether there is a concussive state on the target device is similar to the specific implementation of the network management platform that determines whether there is a concussive route on the target device, and the embodiments of the present disclosure will not repeat them here.
  • the network management platform can collect routing information in parallel (to determine whether there is a concussive route) and routing protocol information (to determine whether there is a routing protocol table entry in a concussive state) ), the specific implementation of which will not be repeated here.
  • the initial routing information collection interval and the initial routing protocol information collection interval can be the same, and the routing information collection interval and the routing protocol information collection interval can be increased or decreased in the same manner, so that the network management platform can complete the routing synchronously
  • the collection of information and the collection of routing protocol information further improve the efficiency of routing shock localization, and its specific implementation will not be repeated here.
  • the network management platform determines the source of the oscillation of the oscillation route, it can also determine the cause of the oscillation that caused the oscillation route.
  • the above-mentioned determining the cause of oscillation may include: collecting the oscillation phenomenon of the oscillation source; and determining the oscillation cause according to the oscillation phenomenon of the oscillation source.
  • the network management platform may collect the vibration phenomenon of the vibration source and determine the cause of the vibration according to the vibration phenomenon of the vibration source.
  • the shock phenomenon can be determined based on the collected routing information and routing protocol information.
  • the shock phenomenon may include, but is not limited to, the routing table entries appear intermittently or are constantly refreshed (the routing table entries are constantly refreshed, which means that the routing table entry exists every time the routing information is collected, and the route appearance time is always a less than or equal to 15s ), whether the publisher device is continuously updated, and whether the routing protocol is continuously switched (for example, the same routing table entry, its corresponding route appears in the BGP protocol table in one collection, next time in the OPSF protocol, and next time It also appears in the BGP protocol table, it is considered that the routing protocol is constantly switching), whether the next hop of the route continues to change, and whether the overall route of the publisher is oscillating (that is, the OSPF routes published by the publisher corresponding to the same Router ID are oscillating) Status, which is usually caused by duplicate router ID).
  • Causes of oscillation can include, but are not limited to, physical or protocol-level connectivity, unstable physical links, inconsistent OSPF timers (OSPF timers), mismatched MTU (Maximum Transmission Unit) at both ends, and routes in different routes
  • OSPF timers inconsistent OSPF timers
  • MTU Maximum Transmission Unit
  • the above-mentioned determination of the cause of the oscillation according to the oscillation phenomenon of the oscillation source may include: querying the pre-configured correspondence between the oscillation phenomenon and the oscillation cause according to the oscillation phenomenon of the oscillation source to determine the oscillation that matches the oscillation phenomenon of the oscillation source the reason.
  • the correspondence between the oscillation phenomenon and the cause of oscillation may be pre-configured.
  • the corresponding relationship between the oscillating phenomenon and the oscillating cause can be shown in Figure 4, where the large box is the oscillating cause and the small box is the oscillating phenomenon; the solid small box is the occurring oscillating phenomenon, and the dashed small box is not occurring The oscillating phenomenon; the left side of the vertical dashed line is for a oscillating route, and the right side of the vertical dashed line is for the entire publisher device.
  • a certain number of bits can be used to record the correspondence between the oscillating phenomenon and the cause of the oscillating, where the number of bits can match the number of oscillating phenomena.
  • the corresponding oscillating phenomenon introduced into each other by routing can be expressed as "011100"
  • the bit value corresponding to the occurrence of the oscillating phenomenon is 1, and the oscillating phenomenon has not occurred
  • the corresponding bit value is 0.
  • the oscillation phenomenon corresponding to the repeated route advertisement shown in the third row can be expressed by "110100".
  • the network management platform After the network management platform determines the vibration source of the vibration route, it can collect the vibration phenomenon of the vibration source and query the corresponding relationship between the pre-configured vibration phenomenon and the vibration source according to the vibration phenomenon of the vibration source to determine the vibration phenomenon with the vibration source Match the cause of the shock.
  • the shock phenomenon can be recorded in the manner of the above bits.
  • the oscillating phenomenon of collecting the oscillating source refers to the collection of the oscillating phenomenon corresponding to the oscillating route; in the process of collecting the oscillating phenomenon corresponding to the oscillating route, for the oscillating phenomenon associated with the publisher, it is necessary to determine whether the same router ID is the same Whether the OSPF routes published by the corresponding publisher are in a state of shock.
  • the vibration phenomenon of the vibration source when no matching vibration reason is found according to the vibration phenomenon of the vibration source, the vibration phenomenon of the vibration source can be recorded, and the user, such as a network administrator, analyzes the corresponding vibration reason in the subsequent process and increases Correspondence between the new shock phenomenon and the cause of shock.
  • one route on one network device in the network when it is oscillated, it may cause multiple routes of multiple network devices to oscillate, and accordingly, when the oscillation source of one oscillating route After determining the cause of the oscillation and taking corresponding measures, the oscillation of multiple routes of multiple network devices in the network may all return to normal.
  • the network management platform may only need to Detect a target device with a concussion route, and perform traceback detection on one of the concussion routes on the target device to determine the concussion source and the cause of the concussion, and record the determined concussion source and the cause of the concussion, as recorded In the alarm log, users such as network administrators take corresponding measures.
  • the network management platform can relocate the route flapping according to the pre-detection, detection, and traceability analysis results described above. The specific implementation will not be repeated here. .
  • the route shock localization method may include the following steps.
  • the network management platform reads the device ID (abbreviated as DevID) of the device to be detected from the local database, initializes it, and joins the polling queue.
  • DevID is the unique identifier of the device to be detected on the network management platform.
  • the identifier of the device to be detected is DevIDi
  • DevID is a 32-bit unsigned integer value.
  • the network management platform may maintain the attribute fields shown in Table 1.
  • the initial polling time interval InitCyclingTime and the trigger detection time InitTriggerTime can be read from the configuration file, and then the polling time of each device to be detected
  • the interval DevCyclingtime is assigned to the initial polling interval InitCyclingTime (in this article, 30s is used as an example).
  • the initial value of the remaining detection time DevLeftTime of each device to be detected is DevID%T0.
  • the maximum polling interval may be equal to the initial value of the polling interval, or it may not be equal to the initial value of the polling interval. This method of setting the initial value effectively disperses polling of the devices to be detected, avoiding polling too many devices at one time.
  • the initial value of the remaining detection time is DevIDi%30.
  • the network management platform subtracts 5s from the remaining detection time of the device to be detected every 5s, then the pre-detection of the device to be detected is divided into 6 batches, where, The initial value of the remaining inspection time is a batch of 0 to 5 seconds, the initial value of the remaining inspection time is a batch of 6 to 10 seconds... The initial value of the remaining inspection time is a batch of 26 to 29 seconds.
  • the remaining detection time of each device to be detected may be The value of DevLeftTime is set to the polling interval of the device to be detected.
  • the initial value of the polling interval of each device to be detected is 30s, and the polling interval will be updated as the number of routing table entries changes during the pre-detection process.
  • the system polling timer time is set to 5s, that is, every 5s, the remaining detection time DevLeftTime of all the devices to be tested is polled; if a device to be tested with DevLeftTime less than or equal to 0 is polled, the device to be tested is polled Pre-test.
  • the pre-detection process obtains the number of routing table entries of the device to be detected through SNMP (Simple Network Management Protocol, Simple Network Management Protocol) GET (Get).
  • SNMP Simple Network Management Protocol, Simple Network Management Protocol
  • the polling interval of the device to be tested is reduced by 5s, that is, to shorten the time to be tested
  • the pre-detection period of the device and set the value of the number of remaining stable DevStableLeft to 5.
  • DevStableLeft will be reduced by 1.
  • the polling interval DevCyclingtime will be restored by 5s (that is, plus 5s) until it returns to the original The polling interval is 30s.
  • the polling time interval of the device to be detected is reduced to the trigger detection time InitTriggerTime, for example, 15s, it indicates that the device to be detected has at least 3 routing table entries If the number changes, at this time, it can trigger the detection of the flapping route of the device to be detected.
  • the implementation process of the network management platform polling the device to be detected may be as shown in FIG. 6.
  • the device to be detected (that is, the target device) that needs to trigger the shock route detection during the pre-detection process
  • it can be added to the waiting detection queue (Wait2CheckList), where the target devices in the waiting detection queue are in order from head to tail Perform the detection, that is, each time the oscillating route test is performed, a target device is taken from the head of the waiting detection queue for detection, and the target device newly added to the waiting detection queue (the pre-detection process determines that the oscillating route detection needs to be triggered to be detected The device) is added to the end of the waiting detection queue.
  • the target device F when the target device F is determined to need to trigger the shock route detection in the pre-detection process, the target device F may be added to the waiting detection At the end of the queue, after waiting for the detection of the target devices A to E in the detection queue, the target device F can be detected.
  • the information collection of concussive routing may include two repeated information collection processes (b, c) in parallel, which are repeatedly collecting routing information and routing protocol information of the target device, and according to the collected routing information and routing The protocol information determines the shock route and locates the shock source of the shock route.
  • the information collection of shock routing can include the following steps:
  • the network management platform may maintain the attribute fields shown in Table 2.
  • the initial information collection time interval InitIntervalTime, decay time ReduceTime, and collection times CollectNum can be read directly from the configuration file by the network management platform, and InitIntervalTime is the initial value of the information collection time interval.
  • CollectNum is the number of collections.
  • the default value of CollectNum is 10, that is, the default is to collect the routing information and routing protocol information of the device 10 times.
  • ReduceTime is the decay time.
  • the initial value of the routing information collection interval and the initial value of the routing protocol information collection interval are the same (both InitIntervalTime), the number of routing information collection and the routing protocol information collection frequency are the same (both CollectNum) 2.
  • the first preset decay time and the second preset decay time are the same (both ReduceTime) as an example for description.
  • InitIntervalTime-decay time ReduceTime ⁇ number of completed routing information collections DevRouteTblCollectTimes, to avoid the situation that the route flapping period is the same as the routing information collection interval and the route flapping cannot be detected.
  • the routing information collected once can be shown in Table 3.
  • the network management platform After the network management platform collects the routing information 10 times, it can merge the routing information collected 10 times and obtain the information shown in Table 4.
  • the network management platform implements the collection of routing protocol information similar to that of routing information collection.
  • the network management platform combines the routing protocol information collected 10 times and obtains the information shown in Table 5.
  • the way to determine route flapping may mainly include:
  • oTurnTimes View the number of oscillations oTurnTimes; where, for any routing table entry, if oTurnTimes ⁇ 2, it is determined that the routing table entry corresponds to the routing oscillation; and when the routing table entry oTurnTimes ⁇ 1, then view the total time of routing occurrence RouteAgeSum, if RouterAgeSum ⁇ GetTimes ⁇ 15s, it is determined that the route corresponding to the routing table entry has not been shaken. On the contrary, when the oTurnTimes ⁇ 1 of the routing table entry, and RouterAgeSum ⁇ GetTimes ⁇ 15s, it is determined that the route corresponding to the routing table entry is flapping.
  • the shock route it can be placed in the FlapRouteInfoList (shock route information list).
  • the publisher device of the route can be determined according to the corresponding routing protocol information, and it is determined whether the routing protocol table entry corresponding to the oscillating route is in an oscillating state, and if so, the oscillating route The publisher's device is added to the head of the waiting queue.
  • the location of the oscillating routes may be achieved through traceability detection.
  • the network management platform can add the oscillating route's publisher device to the head of the waiting detection queue, that is, the next one to perform the oscillating route detection is Publisher device for concussive routing.
  • the publisher device of the concussive route is theoretically the concussion source that causes the concussion of the route
  • the next target device that performs concussion route detection points to the concussion source which can more quickly identify the cause of the concussion and improve the efficiency of concussion route location.
  • the cause of the shock can be further determined, and its specific implementation is described below; when the network management platform detects that there is no shock route, the current process ends, and Re-add the target device that performs the flapping route detection to the polling queue, and perform pre-detection in the manner described above.
  • the oscillating route detection on the target device may be stopped.
  • the network management platform can generate corresponding error logs so that users such as network administrators can take corresponding measures.
  • the correspondence between the cause of oscillation and the phenomenon of oscillation as shown in FIG. 3 may be pre-configured.
  • the network management platform After the network management platform locates the oscillating source in the above manner, it can query the corresponding relationship between the oscillating cause and the oscillating phenomenon shown in FIG. 3 according to the oscillating phenomenon of the oscillating source to determine the oscillating reason that matches the oscillating phenomenon of the oscillating source.
  • the library can be used to locate the cause of the shock. The efficiency of the cause of the shock is higher and the workload is less.
  • a target device that meets the condition for detecting the flapping route is determined, and the flapping route detection is performed on the target device to determine whether there is a flapping route
  • the concussion route on the target device determine the concussion source of the concussion route, and determine the cause of the concussion, thus realizing route concussion positioning.
  • the server may include a processor 901 and a machine-readable storage medium 902 that stores machine-executable instructions.
  • the processor 901 and the machine-readable storage medium 902 can communicate via a system bus 903. And, by reading and executing the machine-executable instructions corresponding to the routing shock localization control logic in the machine-readable storage medium 902, the processor 901 can execute the routing shock localization method described above.
  • the machine-readable storage medium 902 mentioned herein may be any electronic, magnetic, optical, or other physical storage device, and may contain or store information, such as executable instructions, data, and so on.
  • the machine-readable storage medium may be: RAM (Radom Access Memory), volatile memory, non-volatile memory, flash memory, storage drive (such as a hard disk drive), solid-state drive, any type of storage disk (Such as optical discs, DVDs, etc.), or similar storage media, or a combination thereof.
  • the above routing oscillation positioning control logic may include: a pre-detection unit 1001, a detection unit 1002, and a positioning unit 1003.
  • the pre-detection unit 1001 is used for pre-detection of the device to be detected to determine the target device that meets the condition for detection of the shock route;
  • the detection unit 1002 is configured to perform a shock route detection on the target device to determine whether a shock route exists;
  • the positioning unit 1003 is configured to, for any of the target devices, when there is a oscillating route on the target device, determine the oscillating source of the oscillating route and determine the cause of the oscillating.
  • the pre-detection unit 1001 is specifically configured to poll the device to be detected according to the polling time interval of the device to be detected; for any device to be detected, when the device to be detected is detected When the number of routing table entries changes, subtract the first preset time from the polling interval of the device to be detected; when the polling interval of the device to be detected is equal to the preset trigger detection time, execute the Steps for the device to perform flapping route detection.
  • the pre-detection unit 1001 is specifically configured to subtract the second preset time from the remaining detection time of the device to be detected every second preset time; When the remaining detection time is less than or equal to 0, the device to be detected is pre-detected, and the remaining detection time of the device to be detected is set as the polling time interval of the device to be detected.
  • the detection unit 1002 is specifically configured to determine, for any target device that performs oscillating route detection, whether there is an oscillating route on the target device according to the routing information of the target device; wherein, the routing information includes the purpose Address and routing table entry time.
  • the detection unit 1002 is specifically configured to poll the routing table of the target device M times according to the routing information collection interval to collect routing information, where M is a natural number greater than or equal to 1; for any route Table entry, when the number of oscillations of the routing table entry is greater than or equal to the first preset oscillation number threshold, the route corresponding to the routing table entry is determined to be an oscillation route; when the number of oscillations of the routing table entry is less than the first predetermined oscillation A threshold of times, and the total time that the routing table entry appears is less than the first preset time threshold, the route corresponding to the routing table entry is determined to be a shock route; wherein, when the appearance state of a routing table entry changes, the The number of oscillations of routing table entries increases by 1.
  • the routing information collection interval increases or decreases as the number of completed routing information collections increases.
  • the positioning unit 1003 is specifically configured to determine whether the oscillating route is a direct route of the target device for any oscillating route; if so, determine that the target device is the oscillating source of the oscillating route; otherwise , According to the routing protocol information corresponding to the oscillating route, determine the oscillating source of the oscillating route.
  • the positioning unit 1003 is specifically configured to determine whether the routing protocol table entry corresponding to the oscillating route is in an oscillating state; if so, after the detection of the current target device is completed, the routing protocol table entry in the oscillating state The corresponding publisher device in the oscilloscope performs oscillation route detection until the oscillation source of the oscillation route is determined; if not, the step of determining the cause of oscillation is performed.
  • the positioning unit 1003 is further configured to poll the routing protocol table of the target device P times according to the routing protocol information collection interval for any target device to collect routing protocol information, where P is greater than Natural number equal to 1; wherein, the routing protocol information includes the destination address and the appearance time of the routing protocol table entry; for any routing protocol table entry, when the number of oscillations of the routing protocol table entry is greater than or equal to the second preset oscillation number threshold , After performing the detection of the current target device, performing the oscillating route detection step on the corresponding publisher device in the oscillating routing protocol table entry; or, when the oscillating number of the routing protocol table entry is less than the second preset oscillating Times threshold, and when the total time of the routing protocol table entry is less than the second preset time threshold, after performing detection of the current target device, perform oscillating routing on the corresponding publisher device in the oscillating routing protocol table entry The detection step; wherein, when the appearance state of a routing protocol table entry changes, the number of oscillations of the routing protocol table entry
  • the routing protocol information collection interval increases or decreases as the number of completed routing protocol information collections increases.
  • the positioning unit 1003 is specifically configured to collect the vibration phenomenon of the vibration source; determine the cause of the vibration according to the vibration phenomenon of the vibration source.
  • the positioning unit 1003 is specifically configured to query the pre-configured correspondence between the oscillating phenomenon and the oscillating cause according to the oscillating phenomenon of the oscillating source to determine the oscillating cause matching the oscillating phenomenon of the oscillating source .
  • Embodiments of the present disclosure also provide a machine-readable storage medium that stores computer-executable instructions, such as the memory 902 in FIG. 9, and the computer-executable instructions can be executed by the processor 901 in the electronic device shown in FIG. 9 to Achieve the above-mentioned routing shock location method.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Environmental & Geological Engineering (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

提供一种路由震荡定位方法、服务器及机器可读存储介质。作为一个示例，该方法包括：对待检测设备进行预检测，以确定满足震荡路由检测条件的目标设备；对所述目标设备进行震荡路由检测，以确定是否存在震荡路由；对于任一所述目标设备，当该目标设备上存在震荡路由时，确定与所述震荡路由相关的震荡源，并确定震荡原因。

Description

对路由震荡的定位 技术领域
本公开涉及网络通信技术领域，尤其涉及一种路由震荡定位方法、服务器及机器可读存储介质。
背景技术
互联网实现万物互联的核心在于网络设备实现报文转发，网络设备的正常转发在于路由表的稳定完整。如果路由表中条目出现震荡，轻则会导致源设备到目的设备的转发过程出现丢包，速率低等问题；重则导致业务处于停滞状态，带来的损失难以估量。
因此，当发生路由震荡时，如何实现路由震荡定位成为一个亟待解决的问题。
附图说明
图1是本公开实施例提供的一种路由震荡定位方法的流程示意图；
图2A是本公开实施例提供的另一种路由震荡定位方法的流程示意图；
图2B是本公开实施例提供的另一种路由震荡定位方法的流程示意图；
图2C是本公开实施例提供的另一种路由震荡定位方法的流程示意图；
图2D是本公开实施例提供的一种定位震荡源的方法的流程示意图；
图3是本公开实施例提供的一种路由信息收集时间间隔和路由震荡周期的示意图；
图4是本公开实施例提供的一种震荡现象与震荡原因的对应关系的示意图；
图5是本公开实施例提供的一种路由震荡定位方法的流程示意图；
图6是本公开实施例提供的一种网络管理平台对待检测设备进行轮询的流程示意图；
图7是本公开实施例提供的一种等待检测队列的示意图；
图8是本公开实施例提供的一种震荡路由的信息收集的流程示意图；
图9是本公开实施例提供的一种服务器的硬件结构示意图；
图10是本公开实施例提供的一种路由震荡定位控制逻辑的功能结构图。
具体实施方式
为了使本技术领域的人员更好地理解本公开实施例中的技术方案，并使本公开实施例的上述目的、特征和优点能够更加明显易懂，下面结合附图对本公开实施例中技术方案作进一步详细的说明。
图1为本公开实施例提供的一种路由震荡定位方法的流程示意图，其中，该路由震荡定位方法 可以应用于网络管理平台，例如，运行有特定网络管理软件的服务器，如图1所示，该路由震荡定位方法可以包括步骤101～步骤103。
步骤101、对待检测设备进行预检测，以确定满足震荡路由检测条件的目标设备。
本公开实施例中，考虑到路由震荡通常会涉及到较多的网络设备，而各网络设备上的路由信息也通常会较多，若发生路由震荡时，直接通过分析所有网络设备的全部路由信息来定位震荡路由，则工作量会很大，效率会很低，因此，为了提高路由震荡定位的效率，并减少路由震荡的工作量，当需要进行路由震荡定位时，可以先对待检测设备进行预检测，以筛选出满足震荡路由检测条件的待检测设备，满足震荡路由检测条件的待检测设备在本文中称为目标设备。
在本公开一个实施方式中，上述步骤101，对待检测设备进行预检测，以确定满足震荡路由检测条件的目标设备，如图2A所示，可以包括步骤1011～步骤1013。
步骤1011，按照待检测设备的轮询时间间隔对待检测设备进行轮询。
步骤1012，对于任一待检测设备，当检测到该待检测设备的路由表条目数量发生变化时，通过将该待检测设备的轮询时间间隔减去第一预设时间，更新所述轮询时间间隔。
步骤1013，当该更新的轮询时间间隔等于预设触发检测时间时，执行对目标设备进行震荡路由检测的步骤。
具体的，在该实施例中，网络管理平台可以通过轮询的方式，即每隔一定的轮询时间间隔，对待检测设备进行预检测。
相应地，可以预先设置各待检测设备的轮询时间间隔的初始值，网络管理平台可以根据各待检测设备的轮询时间间隔的初始值对对应的待检测设备进行预检测。
其中，不同的待检测设备的轮询时间间隔可以不同。
当然，各待检测设备的轮询时间间隔也可以相同。例如，网络管理平台可以每隔30秒，即各待检测设备的轮询时间间隔均为30秒，对各待检测设备进行预检测。
在该实施例中，网络管理平台对待检测设备进行预检测可以包括获取该待检测设备的路由条目的数量，并判断该待检测设备的路由条目的数量是否发生变化，即此次(非首次)获取到的该待检测设备的路由条目的数量与上一次获取到的该待检测的路由条目的数量是否相同。当网络管理平台确定该待检测设备的路由条目的数量发生变化时，将该待检测设备的轮询时间间隔减去第一预设时间。该第一预设时间可以根据实际场景设定，如3秒、5秒等。
其中，对于网络管理平台第一次获取到某个待检测设备的路由条目的数量时，网络管理平台可以仅记录该路由条目的数量，而不需要去判断该待检测设备的路由条目的数量是否发生变化。
在该实施例中，对于任一待检测设备，当该待检测设备的轮询时间间隔等于预设触发检测时间时，该预设触发检测时间可以根据实际场景设定，如15秒、20秒等，网络管理平台可以确定该待检测设备满足震荡路由检测条件，进而，网络管理平台可以确定该待检测设备为目标设备，并执行 步骤102，即执行对目标设备进行震荡路由检测的步骤。
在本公开实施例中，上述根据待检测设备的轮询时间间隔确定待检测设备是否满足震荡路由检测条件仅仅是本公开实施例中确定待检测设备是否满足震荡路由检测条件的一个具体示例，而并不是对本公开保护范围的限定，即在本公开实施例中，也可以通过其他方式确定待检测设备是否满足震荡路由检测条件，例如，可以将路由条目的数量发生变化的待检测设备直接确定为满足震荡路由检测条件的目标设备，本公开实施例对此不做限定。
进一步地，在本公开实施例中，为了避免网络管理平台一直实时监测各待检测设备是否需要进行预检测，网络管理平台可以每隔第二预设时间判断一次各待检测设备是否需要进行预检测。该第二预设时间可以根据实际场景设定，如3秒、5秒等。
相应地，在一个示例中，如图2B所示，上述步骤1011，按照待检测设备的轮询时间间隔对待检测设备进行轮询，可以包括步骤10111和步骤10112。
步骤10111，每经过第二预设时间，将待检测设备的剩余检测时间减去第二预设时间，更新所述剩余检测时间；其中，对于单个待检测设备而言，待检测设备的剩余检测时间的初始值可以与轮询时间间隔的初始值设置的相同；
步骤10112，当更新的剩余检测时间小于等于0时，对该待检测设备进行预检测，并将该待检测设备的剩余检测时间置为该待检测设备的轮询时间间隔。
举例来说，假设第二预设时间为5秒，网络管理平台可以每经过5秒，将待检测设备的剩余检测时间减去5秒。
对于任一待检测设备，当该待检测设备的剩余检测时间小于等于0时，对该待检测设备进行预检测，并将该待检测设备的剩余检测时间置为待检测设备的轮询时间间隔。
举例来说，以待检测设备A为例，假设待检测设备A的轮询时间间隔的初始值为30秒，剩余检测时间为18秒，第一预设时间为5秒，第二预设时间为5秒，预设触发检测时间为15秒，则网络管理平台可以每隔5秒，将待检测设备A的剩余检测时间减去5秒，由此剩余检测时间依次变化为13秒、8秒、3秒、-2秒。当待检测设备A的剩余检测时间为-2秒(即小于等于0)时，网络管理平台一方面可以继续对待检测设备A进行预检测，另一方面将待检测设备A的剩余检测时间置为待检测设备A的轮询时间间隔(此时为30秒)。
其中，如前面步骤1012所述，网络管理平台对待检测设备A进行预检测可以包括：获取待检测设备A的路由表条目数量，并比较该路由表条目数量和上一次获取到的待检测设备A的路由表条目数量，若相同，则保持待检测设备A的轮询时间间隔不变；否则，将待检测设备A的轮询时间间隔减去5秒，此时轮询时间间隔更新变为25秒。
以待检测设备A的轮询时间间隔变为25秒为例，网络管理平台将待检测设备A的剩余检测时间置为30秒之后，可以每隔5秒将待检测设备A的剩余检测时间减去5秒，由此剩余检测时间依次变为25秒、15秒、10秒、5秒、0秒；当待检测设备A的剩余检测时间为0(即小于等于0)秒 时，网络管理平台一方面可以将待检测设备A的剩余检测时间置为轮询时间间隔(即25秒)，另一方面，对待检测设备A继续进行预检测。
按照上述方式，当待检测设备A的轮询时间间隔为15秒时，网络管理平台可以确定待检测设备A为目标设备，并执行对目标设备进行震荡路由检测的步骤。
进一步地，在本公开实施例中，为了更合理的利用网络管理平台的设备资源，避免网络管理平台对于各待检测设备的预检测过于集中，在设置剩余检测时间的初始值时，不同待检测设备的剩余检测时间可以设置成不完全相同。
举例来说，不同的待检测设备的剩余检测时间的初始值可以通过以下公式：
Ti＝DevIDi％T0
其中，Ti为待检测设备i的剩余检测时间的初始值，1≤i≤N1，其中N1为大于等于1的自然数，N1为待检测设备的数量；DevIDi为待检测设备i的设备标识；T0为最大轮询时间间隔，该最大轮询时间间隔，可以根据实际场景设定，如30秒；符号“％”为取余运算。
在本公开实施例中，当各待检测设备的剩余检测时间的初始值不完全相同时，若网络管理平台始终根据各待检测设备的剩余检测时间的初始值确定是否对各待检测设备进行轮询，则各待检测设备进行预检测的频率会不相同，导致相同时间内部分待检测设备被预检测的次数较多，而另一部分待检测设备被预检测的次数较少，因此，为了在避免各待检测设备进行预检测的时间过于集中的情况下，保证各待检测设备进行预检测的时间间隔尽量保持一致，网络管理平台根据各待检测设备的剩余检测时间的初始值对各待检测设备进行预检测之后，可以将各待检测设备的剩余检测时间置为轮询时间间隔。其中，各待检测设备的轮询时间间隔的初始值可以相同，在后续流程中，根据各待检测设备的路由表条目数量的变化情况分别对各检测设备的轮询时间间隔进行调整，其具体实现将在下文中结合具体实例进行说明，本公开实施例在此不做赘述。
进一步地，在本公开实施例中，考虑到网络设备的路由表条目数量的变化并不一定是由于路由震荡引起的，也有可能是由于用户配置的原因导致的，如用户添加或删除路由，因此，为了提高所确定的满足震荡路由检测的目标设备的准确性，对于任一待检测设备，当网络管理平台连续N2次(N2为大于1的自然数)检测到该待检测设备的路由表条目数量未发生变化时，将该待检测设备的轮询时间加上第三预设时间，直至该待检测设备的轮询时间间隔达到预设最大轮询时间间隔。其中，第三预设时间可以根据实际场景设定，如3秒、5秒等；最大轮询间隔时间可以由管理员根据需求进行配置。
在本公开实施例中，对于任一待检测设备，最大轮询时间间隔与前述的轮询时间间隔的初始值可以相同，也可以不同，本公开实施例对此不做限定。
步骤102、对目标设备进行震荡路由检测，以确定是否存在震荡路由。
本公开实施例中，网络管理平台通过对待检测设备进行预检测，确定了满足震荡路由条件的目标设备之后，可以对目标设备进行震荡路由检测，以确定是否存在震荡路由。在一个示例中，对所 述目标设备进行震荡路由检测，包括：当确定出存在多个目标设备时，依次对所述目标设备进行震荡路由检测。
在本公开一个实施例中，如图2C所示，上述步骤102，对目标设备进行震荡路由检测，可以包括步骤1021和步骤1022。
步骤1021，对于任一待检测设备，当确定该待检测设备为满足震荡路由检测条件的目标设备时，将该目标设备加入到等待检测队列的尾部；
步骤1022，按照从等待检测队列的头部到尾部的顺序，依次对等待检测队列中的目标设备进行震荡路由检测。
在该实施例中，考虑到一台网络设备上的路由表条目的数量可能会很多，若网络管理平台同时对多台网络设备进行震荡路由检测，会导致网络管理平台资源压力过载，进而可能会导致处理卡顿，甚至后台崩溃，因此，为了降低网络管理平台的负载，提高网络管理平台的可靠性，网络管理平台可以每次对一台网络设备进行震荡路由检测。
相应地，在该实施例中，对于任一待检测设备，当网络管理平台确定该待检测设备为满足震荡路由检测条件的目标设备时，网络管理平台可以将该目标设备加入等待检测队列。
当需要对目标设备进行震荡路由检测时，网络管理平台可以按照从等待检测队列的头部到尾部的顺序，依次从等待检测队列中取出目标设备进行震荡路由检测。
本公开实施例中，对于任一进行震荡路由检测的目标设备，网络管理平台可以根据该目标设备的路由信息确定该目标设备上是否存在震荡路由；其中，该路由信息可以包括目的地址以及路由表条目出现时间。
其中，目的地址用于标识路由表条目，不同的目的地址对应不同的路由表条目。
在本公开一个实施例中，上述根据该目标设备的路由信息确定该目标设备上是否存在震荡路由，可以包括：
根据路由信息收集时间间隔轮询M次该目标设备的路由表，以收集得到路由信息，M为大于等于1的自然数；
对于任一路由表条目，当该路由表条目的震荡次数大于等于第一预设震荡次数阈值时，执行步骤103中确定与震荡路由相关的震荡源的步骤；或，当该路由表条目的震荡次数小于第一预设震荡次数阈值，且该路由表条目出现的总时间小于第一预设时间阈值时，执行步骤103中确定与震荡路由相关的震荡源的步骤。其中，当一个路由表条目的出现状态发生变化时，该路由表条目的震荡次数加1。
在该实施例中，对于任一目标设备，网络管理平台可以按照路由信息收集时间间隔(可以根据实际场景设定)轮询若干次(可以根据实际场景设定，本文中以M次为例)该目标设备的路由表，以收集路由信息。
在该实施例中，网络管理平台可以根据目标设备的路由信息，统计目标设备上的各路由表条目的震荡次数。
其中，对于任一路由表条目，当该路由表条目在相邻两次收集的目标设备的路由信息中的出现状态发生变化例如从有到无或从无到有时，网络管理平台可以将该路由表条目的震荡次数加1。
举例来说，假设网络管理平台收集了5次目标设备的路由信息，若在该5次收集的目标设备的路由信息中，路由表条目a在第1～3次出现，第4次未出现，第5次出现，则路由表条目a的震荡次数为2。
在该实施例中，网络管理平台确定了各路由表条目的震荡次数之后，可以将震荡次数大于等于第一预设震荡次数阈值的路由表条目对应的路由确定为震荡路由，并执行步骤103中的确定与震荡路由相关的震荡源的步骤。第一预设震荡次数阈值可以根据实际场景设定，如2次、3次等。
此外，在该实施例中，考虑到当发生路由震荡时，路由震荡出现的时间一般会小于特定时间(通常为15秒)，因此，还可以进一步根据震荡次数小于上述第一预设震荡次数阈值的路由表条目出现的时间确定对应的路由是否为震荡路由。
相应地，对于任一路由表条目，当网络管理平台确定该路由表条目的震荡次数小于第一预设震荡次数阈值时，网络管理平台还可以进一步确定该路由表条目出现的总时间，即M次收集的该目标设备的路由信息中该路由表条目的出现时间的和。
举例来说，假设共收集了5次目标设备的路由信息，路由表条目b在5次收集中均出现，且5次收集中所获取到的该路由表条目b的出现时间分别为t1～t5，则该路由表条目b出现的总时间为t1+t2+t3+t4+t5。
在该实施例中，当网络管理平台确定该路由表条目的出现的总时间小于第一预设时间阈值时，网络管理平台也可以确定该路由表条目对应的路由为震荡路由，并执行步骤103中确定与震荡路由相关的震荡源的步骤。第一预设时间阈值可以根据实际场景设定。
进一步地，在本公开实施例中，考虑到当路由震荡周期与路由信息收集时间间隔相同时，可能会检测不出存在的震荡路由。
举例来说，请参见图3，假设两个方形节点之间的时间长度为路由信息收集时间间隔(以30秒为例)，三角节点之间的时间长度为路由震荡周期(假设也是30秒)。在圆形节点处，某路由由于震荡原因消失，在三角节点处，该路由由于震荡原因重新出现，若路由信息收集时间间隔始终保持为30秒，由于路由震荡周期与路由信息收集时间间隔相同，每次在收集路由信息的节点处收集的结果都是此路由不存在，检测得到的结果是此路由不存在，并不是此路由震荡。
为了避免上述问题的发生，路由收集时间间隔可以动态变化，而不是一直保持不变。
相应地，在本公开一个实施例中，路由信息收集时间间隔随着已完成的路由信息收集次数的增多而增大或减小。
举例来说，路由信息收集时间间隔可以通过以下公式：
T3＝T2–△T1×N3
其中，T3为当前使用的路由信息收集时间间隔，T2为初始路由信息收集时间间隔，△T1为第一预设衰减时间，N3为已完成的路由信息收集次数。
在该实施例中，可以预先设置第一预设衰减时间(即上述△T1)，当网络管理平台进行路由信息收集时，可以根据预先设置的初始路由信息收集时间间隔以及该第一预设衰减时间△T1确定每一次进行路由信息收集的时间间隔。
上述根据第一预设衰减时间以及已完成的路由信息收集次数更新路由信息收集时间间隔的实现方式仅仅是本公开实施例中使多次进行路由信息收集的时间间隔不完全相同的一种具体实现，而并不是对本公开保护范围的限定，即在本公开实施例中，也可以通过其他方式实现使多次进行路由信息收集的时间间隔不完全相同，例如，可以按照预设增长时间以及已完成的路由信息收集次数更新路由信息收集时间间隔，即路由信息收集时间间隔随着路由信息收集次数的增多而增大，或者，可以直接设置多个不同的路由信息收集时间间隔，并按照该多个不同的路由信息收集时间间隔进行多次路由信息收集，其具体实现在此不做赘述。
步骤103、对于任一目标设备，当该目标设备上存在震荡路由时，确定与该震荡路由相关的震荡源，并确定震荡原因。
本公开实施例中，考虑到在发生路由震荡的情况下，往往有多台网络设备同时处于震荡，且大多数网络设备是处于被动引起震荡的状态。如果不分优先级的将所有网络设备一起检测，网络管理设备在性能上就会遇到瓶颈，且无法快速确定引起路由震荡的原因，因此，为了降低网络管理平台的负荷，并提高确定震荡原因的效率，可以采用寻找造成路由震荡的源头设备的方式实现路由震荡定位。造成路由震荡的源设备，在本公开中称为震荡源，即网络中多台网络设备的震荡均是由于震荡源产生了问题而导致的。
相应地，在本公开实施例中，当网络管理平台确定存在震荡路由时，可以确定该震荡路由相关的震荡源，并从源头分析引起路由震荡的原因。
在本公开一个实施例中，如图2D所示，上述确定该震荡路由的震荡源，可以包括步骤201～步骤203。
步骤201，对于任一震荡路由，判断该震荡路由是否为该目标设备的直连路由；若是，则转至步骤202；否则，转至步骤203。
步骤202，确定该目标设备为该震荡路由的震荡源。
步骤203，根据该震荡路由对应的路由协议信息确定与该震荡路由相关的震荡源。
其中，路由协议表为记录经由某一路由协议，如OSPF(Open Shortest Path First开放式最短路径优先)协议、BGP(Border Gateway Protocol,边界网关协议)协议等，传播的路由的表。具体的， 可以通过收集待检测设备的MIB(Management Information Base)中的链路状态数据库表ospfLsdbTable、外部引入链路状态数据库表ospfExtLsdbTable以及BGP接收路径属性表bgp4PathAttrEntry等收集得到路由协议信息，进而根据震荡路由对应的路由协议信息确定震荡源。
在该实施例中，当网络管理平台确定某一目标设备上存在震荡路由时，对于该目标设备上的任一震荡路由，网络管理平台可以判断该震荡路由是否为该目标设备的直连路由。若该震荡路由为该目标设备的直连路由，则网络管理平台可以确定该目标设备为该震荡路由的震荡源；若该震荡路由不为该目标设备的直连路由，网络管理平台可以根据该震荡路由对应的路由协议信息确定该震荡路由的震荡源。
其中，对于任一震荡路由，网络管理平台可以根据该震荡路由对应的路由表条目中的ipRouteType(路由类型)字段确定该震荡路由是否为目标设备的直连路由。例如，当ipRouteType的值为3时，表明对应的路由为直连路由；否则，为非直连路由。
在一个示例中，上述根据该震荡路由对应的路由协议信息确定该震荡路由的震荡源，包括：
判断该震荡路由对应的路由协议表条目是否处于震荡状态；
若该震荡路由对应的路由协议表条目处于震荡状态，则在当前的目标设备完成检测之后，对处于震荡状态的路由协议表条目中对应的发布者设备进行震荡路由检测，直至确定该震荡路由的震荡源；
若该震荡路由对应的路由协议表条目未处于震荡状态，则执行步骤103中的确定震荡原因的步骤。
在该示例中，当网络管理平台确定震荡路由不是目标设备上的直连路由时，网络管理平台可以查询该震荡路由对应的路由协议表条目，并判断该震荡路由对应的路由协议表条目是否处于震荡状态。
在该示例中，当网络管理平台确定该震荡路由对应的路由协议表条目处于震荡状态时，网络管理平台可以将该震荡路由对应的处于震荡状态的路由协议表条目中对应的发布者设备加入到等待检测队列的头部。
其中，网络管理平台对等待检测队列中的设备的检测顺序为从头部到尾部的顺序。
网络管理平台完成对当前的目标设备的检测之后，可以对该震荡路由的发布者设备进行震荡路由检测，直至确定该震荡路由的震荡源。
其中，若该震荡路由为该震荡路由的发布者设备的直连路由，或者，该震荡路由的发布者设备上对应该震荡路由的路由协议表条目未处于震荡状态，则可以确定该震荡路由的发布者设备为该震荡路由的震荡源；若该震荡路由不是该震荡路由的发布者设备的直连路由，且该震荡路由的发布者设备上对应该震荡路由的路由协议表条目处于震荡状态，则可以进一步根据该震荡路由的发布者设备上的路由协议表条目信息确定该震荡路由的发布者设备，并按照上述步骤201～步骤203中描述的 方式继续进行溯源检测，直至确定该震荡路由的震荡源。
举例来说，假设网络设备A上的路由表条目a对应的路由为震荡路由，且路由表条目a对应的路由协议表条目a也处于震荡状态，则网络管理平台可以根据该路由协议表条目a确定该震荡路由的发布者设备(假设为网络设备B)，网络管理平台可以进一步确定该震荡路由在网络设备B上对应的路由(即网络设备B上与该震荡路由的目的地址相同的路由)是否是网络设备B上的直连路由。若该震荡路由在网络设备B上对应的路由不是网络设备B的直连路由，且该路由(即震荡路由在网络设备B上对应的路由)对应的路由协议表条目b在网络设备B上处于震荡状态，则网络管理平台可以根据该路由协议表条目b确定该震荡路由的发布者设备(假设为网络设备C)，当该震荡路由在网络设备C上对应的路由(即网络设备C上与该震荡路由的目的地址相同的路由)为网络设备C的直连路由，或，该震荡路由在网络设备C上对应的路由不是网络设备C的直连路由且该路由(即震荡路由在网络设备C上对应的路由)对应的路由协议表条目c在网络设备C上未处于震荡状态，则网络管理平台可以确定该震荡路由的震荡源为网络设备C。
在该示例中，当网络管理平台确定震荡路由对应的路由协议表条目未处于震荡状态时，则网络管理平台可以确定目标设备为该震荡路由的震荡源。
在本公开实施例中，一条震荡路由对应的路由协议表条目可以有多条，如可以同时存在BGP(Border Gateway Protocol,边界网关协议)协议的路由协议表条目和OSPF(Open Shortest Path First开放式最短路径优先)协议的路由协议表条目，当该多条路由协议表条目中任一处于震荡状态时，即可确定该震荡路由对应的路由协议表条目处于震荡状态，并根据该处于震荡状态的路由协议表条目确定震荡路由的发布者，其具体实现在此不做赘述。
在本公开一个实施例中，确定目标设备上是否存在震荡状态的路由协议表条目，可以包括：
对于任一目标设备，根据路由协议信息收集时间间隔轮询P次该目标设备的路由协议表，以收集得到路由协议信息；其中，路由协议信息包括目的地址以及路由协议表条目出现时间；
对于任一路由协议表条目，当该路由协议表条目的震荡次数大于等于第二预设震荡次数阈值时，执行在当前的目标设备完成检测之后，对处于震荡状态的路由协议表条目中对应的发布者设备进行震荡路由检测的步骤；或，当该路由协议表条目的震荡次数小于第二预设震荡次数阈值，且该路由协议表条目出现的总时间小于第二预设时间阈值时，执行在当前的目标设备完成检测之后，对处于震荡状态的路由协议表条目中对应的发布者设备进行震荡路由检测的步骤；其中，当一个路由协议表条目的出现状态发生变化时，该路由协议表条目的震荡次数加1。
在一个示例中，路由协议信息收集时间间隔随着已完成的路由协议信息收集次数的增多而增大或减小。
举例来说，路由协议信息收集时间间隔可以通过以下公式：
T5＝T4–△T2×N4，
其中，T5为当前使用的路由协议信息收集时间间隔，T4为初始路由协议信息收集时间间隔，△T2 为第二预设衰减时间，N4为已完成的路由协议信息收集次数。
其中，网络管理平台确定目标设备上是否存在震荡状态的路由协议表条目的具体实现与网络管理平台确定目标设备上是否存在震荡路由的具体实现相类似，本公开实施例在此不做赘述。
在本公开实施例中，为了提高路由震荡定位的效率，网络管理平台可以并行地进行路由信息的收集(确定是否存在震荡路由)以及路由协议信息的收集(确定是否存在震荡状态的路由协议表条目)，其具体实现，在此不做赘述。
相应地，上述初始路由信息收集时间间隔以及初始路由协议信息收集间隔可以相同，且路由信息收集时间间隔以及路由协议信息收集间隔可以按照相同方式进行增大或减少，以便网络管理平台可以同步完成路由信息的收集和路由协议信息的收集，进一步提高路由震荡定位的效率，其具体实现在此不做赘述。
进一步地，在本公开实施例中，网络管理平台确定了震荡路由的震荡源之后，还可以确定引起震荡路由的震荡原因。
在本公开一个实施例中，上述确定震荡原因，可以包括：收集震荡源的震荡现象；根据震荡源的震荡现象确定震荡原因。
在该实施例中，网络管理平台确定了震荡路由的震荡源之后，可以收集震荡源的震荡现象，并根据震荡源的震荡现象确定震荡原因。
其中，震荡现象可以根据收集的路由信息和路由协议信息确定。具体的，震荡现象可以包括但不限于路由表条目间隔性出现或不断被刷新(路由表条目不断刷新特指每一次收集路由信息时该路由表条目都存在，路由出现时间始终是一个小于等于15s的值)、发布者设备是否不断更新、路由协议是否不断切换(例如，同一路由表条目，其对应的路由在一次收集中出现在BGP协议表中，下一次出现在OPSF协议中，再下一次又出现在BGP协议表中，则认为路由协议在不断切换)、路由下一跳是否不断变化、是否存在发布者整体路由均在震荡(即同一Router ID对应的发布者发布的OSPF路由均处于震荡状态，其通常由Router ID重复导致)等。
震荡原因可以包括但不限于物理上或者协议层面的联通性、物理链路不稳定、OSPFtimer(OSPF计时器)不一致、两端MTU(Maximum Transmission Unit，最大传输单元)不匹配、路由在不同的路由协议间相互引入情况下，引入路由的源头优先级设置不正确、路由重复发布、RouterID(路由器标识)重复或唯一标识符重复引起震荡、NQA(Network Quality Analyzer，网络质量分析)和静态路由联动引起震荡等。
在一个示例中，上述根据震荡源的震荡现象确定震荡原因，可以包括：根据震荡源的震荡现象，查询预先配置的震荡现象与震荡原因的对应关系，以确定与震荡源的震荡现象匹配的震荡原因。
在该示例中，可以预先配置震荡现象与震荡原因的对应关系。
例如，该震荡现象与震荡原因的对应关系可以如图4所示，其中，大框中为震荡原因，小框中为震荡现象；实线小框为发生的震荡现象，虚线小框为未发生的震荡现象；竖虚线左侧是针对一个震荡路由来说，竖虚线右侧是对于发布者设备整体来说的。
在该示例中，可以使用一定数量的比特位来记录震荡现象与震荡原因的对应关系，其中比特位的数量可与震荡现象的数量匹配。
以图4第二行所示震荡现象与震荡原因的对应关系为例，路由相互引入对应的震荡现象可以使用“011100”来表示，震荡现象发生了对应的bit位值为1，震荡现象未发生对应的bit位值为0。第三行所示的路由重复发布对应的震荡现象可以使用“110100”来表示。
当网络管理平台确定了震荡路由的震荡源之后，可以收集震荡源的震荡现象，并根据震荡源的震荡现象，查询预先配置的震荡现象与震荡源的对应关系，以确定与震荡源的震荡现象匹配的震荡原因。可选的，震荡现象可以按照上述比特位的方式记录。
其中，收集震荡源的震荡现象是指针对震荡路由对应的震荡现象的收集；其中，在针对震荡路由对应的震荡现象的收集过程中，对于与发布者关联的震荡现象，需要确定是否同一Router ID对应的发布者发布的OSPF路由是否均处于震荡状态。
在本公开实施例中，当根据震荡源的震荡现象未查询到匹配的震荡原因时，可以记录该震荡源的震荡现象，由用户如网络管理员在后续流程中分析对应的震荡原因，并增加新的震荡现象与震荡原因的对应关系。
此外，当存在两个不同的震荡原因对应的震荡现象完全相同时，可以拓展震荡现象集合，即增加新的震荡现象，并更新所配置的震荡现象和震荡原因的对应关系，其具体实现在此不做赘述。
再者，在本公开实施例中，考虑到当网络中一个网络设备上的一条路由发生震荡时，可能会引起多台网络设备的多条路由发生震荡，相应地，当一条震荡路由的震荡源和震荡原因确定并采取相应措施之后，网络中多台网络设备的多条路由的震荡可能均会恢复正常，因此，当满足震荡路由检测条件的目标设备存在多台时，网络管理平台可以仅需检测出一台存在震荡路由的目标设备，并针对该目标设备上的其中一条震荡路由进行溯源检测，以确定震荡源和震荡原因即可，并记录所确定的震荡源和震荡原因，如记录到告警日志中，由用户如网络管理员采取相应措施。
其中，若用户采取相应措施之后，网络中仍然存在路由震荡，则网络管理平台可以重新按照上文中描述的预检测、检测以及溯源分析结果的方式进行路由震荡定位，其具体实现在此不做赘述。
为了使本领域技术人员更好地理解本公开实施例提供的技术方案，下面结合具体实例对本公开实施例提供的技术方案进行说明。
请参见图5，在该实施例中，路由震荡定位方法可以包括以下步骤。
一、获取设备列表
网络管理平台从本地数据库中读取出待检测设备的设备ID(简称为DevID)，并初始化，加入轮询队列中。其中，DevID为待检测设备在网络管理平台上的唯一标识，例如待检测设备i的标识为DevIDi，且DevID是一个32位的unsigned int(无符号整形)类型值。
二、轮询待检测设备
在该实施例中，为了实现待检测设备的轮询，网络管理平台可以维护表1所示的属性字段。
表1
Figure PCTCN2019128055-appb-000001
Figure PCTCN2019128055-appb-000002
在该实施例中，网络管理平台初始化过程中，可以从配置文件中读取出初始的轮询时间间隔InitCyclingTime和触发检测时间InitTriggerTime这两个初始值，再将每个待检测设备的轮询时间间隔DevCyclingtime赋值为初始化的轮询时间间隔InitCyclingTime(本文中以30s为例)。
其中，每个待检测设备的剩余检测时间DevLeftTime的初始值为DevID％T0，实施例中以最大轮询时间间隔T0＝30秒为例进行说明。最大轮询时间间隔可等于轮询时间间隔的初始值，也可不等于轮询时间间隔的初始值。利用这一设置初始值的方式有效分散轮询待检测设备，避免一次性轮询过多的设备。
举例而言，对于待检测设备i，其剩余检测时间的初始值为DevIDi％30。
以系统轮询计时器时间设置为5s为例，即网络管理平台每隔5s，将待检测设备的剩余检测时间减去5s，则待检测设备的预检测被划分为6个批次，其中，剩余检测时间的初始值为0～5秒的一个批次，剩余检测时间的初始值6～10秒的一个批次…剩余检测时间的初始值为26～29秒的一个批次。
为了避免各待检测设备的进行预检测的时间间隔差异性过大，在按照待检测设备的剩余检测时间DevLeftTime的初始值触发对待检测设备进行预检测之后，可以将各待检测设备的剩余检测时间DevLeftTime的值置为待检测设备的轮询时间间隔。
其中，各待检测设备的轮询时间间隔的初始值均为30s，轮询时间间隔会随着预检测过程中路由表条目数的变化情况而更新。
在该实施例中，系统轮询计时器时间设置为5s，即每隔5s轮询所有待检测设备的剩余检测时间DevLeftTime；若轮询到DevLeftTime小于等于0的待检测设备，对该待检测设备进行预检测。
其中，预检测流程通过SNMP(Simple Network Management Protocol，简单网络管理协议) GET(获取)的方式获取待检测设备的路由表条目数。对于任一待检测设备，若发现该待检测设备的路由表条目数不同于上次的路由表条目数，则将该待检测设备的轮询时间间隔DevCyclingtime减少5s，即缩短对这一待检测设备的预检测周期，并将稳定剩余次数DevStableLeft这一数值置为5。后续每次发现路由表条目数与上一次比没有变化，即将DevStableLeft减1，如果连续5次路由表条目数没有变化，则将轮询时间间隔DevCyclingtime恢复5s(即加5s)，直到恢复到初始的轮询时间间隔30s。
在该实施例中，对于任一待检测设备，若该待检测设备的轮询时间间隔DevCyclingtime减小到触发检测时间InitTriggerTime，例如15s时，则表明该待检测设备至少出现了3次路由表条目数发生变化的情况，此时，可以触发对该待检测设备进行震荡路由检测。
其中，网络管理平台对待检测设备进行轮询的实现流程可以如图6所示。
三、震荡路由检测
对于在预检测过程确定需要触发震荡路由检测的待检测设备(即目标设备)，可以将其加入等待检测队列(Wait2CheckList)，其中，该等待检测队列中的目标设备按照从头部到尾部的顺序进行检测，即每次进行震荡路由检测时，从等待检测队列的头部取出一台目标设备进行检测，新加入该等待检测队列的目标设备(通过预检测流程确定需要触发震荡路由检测的待检测设备)加入到等待检测队列尾部。
举例来说，如图7所示，假设等待检测队列中包括目标设备A～E，当目标设备F在预检测流程中被确定为需要触发震荡路由检测时，可以将目标设备F加入到等待检测队列的尾部，当等待检测队列中的目标设备A～E均完成检测后，可以对目标设备F进行检测。
在该实施例中，通过使用队列的方式，保证同一时间只有一台设备处于检测状态，避免了同一时间有多台设备同时处于震荡状态时，多台设备同一时间检测导致的网络管理平台的设备资源压力过载，轻则导致处理卡顿，重则导致设备崩溃，提高了网络管理平台的可靠性。
1、震荡路由的信息收集
在该实施例中，震荡路由的信息收集可以包括两个重复收集信息过程(b，c)并行进行，分别是重复收集目标设备的路由信息和路由协议信息，并根据所收集的路由信息和路由协议信息确定震荡路由以及定位震荡路由的震荡源。
如图8所示，震荡路由的信息收集可以包括以下步骤：
a、初始化检测参数
在该实施例中，为了实现震荡路由的信息收集，网络管理平台可以维护表2所示的属性字段。
表2
Figure PCTCN2019128055-appb-000003
在该实施例中，初始信息收集时间间隔InitIntervalTime、衰减时间ReduceTime、收集次数CollectNum三项可以由网络管理平台直接从配置文件读取，InitIntervalTime是信息收集时间间隔的初始值。CollectNum是收集次数，为了达到收集信息的准确性，默认CollectNum取值为10次，即默认收集10次设备的路由信息和路由协议信息。ReduceTime为衰减时间。
在该实施例中，以路由信息收集时间间隔的初始值和路由协议信息收集时间间隔的初始值相同(均为InitIntervalTime)、路由信息的收集次数和路由协议信息的收集次数相同(均为CollectNum)、第一预设衰减时间和第二预设衰减时间相同(均为ReduceTime)为例进行说明。
以路由信息收集为例，网络管理平台对任一目标设备进行路由信息收集时，当前使用的路由信息收集时间间隔CurInterValTime＝初始间隔收集设备信息时间InitIntervalTime–衰减时间ReduceTime×已完成的路由信息收集次数DevRouteTblCollectTimes，以避免路由震荡周期与路由信息收集时间间隔相同的情况而导致的检测不出路由震荡。
b、路由信息收集
在该实施例中，单次收集的路由信息可以如表3所示。
表3
Figure PCTCN2019128055-appb-000004
网络管理平台收集了10次路由信息之后，可以合并10次收集的路由信息，并得到表4所示的信息。
表4
Figure PCTCN2019128055-appb-000005
c、路由协议信息收集
在该实施例中，网络管理平台对路由协议信息的收集的实现与路由信息收集的实现相类似，网络管理平台合并10次收集的路由协议信息，并得到表5所示的信息。
表5
属性名称 属性解释
strRouteDest 目的地址
EmergeTime 路由协议表条目出现次数统计
strRouteAdvID 发布者路由ID
LSDBAgeSum； 路由协议表条目出现时间总和
2、筛选震荡路由
在该实施例中，确定路由震荡的方式可以主要包括：
查看震荡次数oTurnTimes；其中，对于任一路由表条目，若oTurnTimes≥2，则确定该路由表条目对应的路由震荡；而当路由表条目的oTurnTimes≤1时，则查看路由出现时间总和RouteAgeSum，若RouterAgeSum≥GetTimes×15s，则确定该路由表条目对应的路由未发生震荡，反之，当路由表条目的oTurnTimes≤1，且RouterAgeSum<GetTimes×15s，则确定该路由表条目对应的路由震荡。
其中，对于震荡路由，可以放置在FlapRouteInfoList(震荡路由信息链表)中。
在该实施例中，对于任一震荡路由，可以根据对应的路由协议信息确定该路由的发布者设备，并确定该震荡路由对应的路由协议表条目是否处于震荡状态，若是，则将该震荡路由的发布者设备加入到等待检测队列的头部。
在该实施例中，考虑到在全网发生路由震荡的情况下，往往有多台网络设备同时发生路由震荡，且大多数网络设备是处于被动引起路由震荡的状态。如果不分优先级的将所有网络设备一起检测，后台检测程序在性能上就会遇到瓶颈，通过队列(等待检测队列)的方式有效地解决了该问题。
但是考虑到每次对一台网络设备进行震荡路由检测的方式在进行震荡路由定位的效率上可能会比较差，因此，该实施例可以通过溯源检测的方式实现震荡路由的定位。
在该实施例中，当震荡路由对应的路由协议表也处于震荡状态时，网络管理平台可以将该震荡路由的发布者设备加入到等待检测队列的头部，即下一个进行震荡路由检测的就是震荡路由的发布者设备。
由于震荡路由的发布者设备理论上就是引起路由震荡的震荡源，因此下一个进行震荡路由检测的目标设备指向震荡源，可以更快速找出震荡原因，提高震荡路由定位的效率。
在该实施例中，当网络管理平台通过上述方式定位到震荡路由时，可以进一步确定震荡原因，其具体实现在下文中描述；当网络管理平台检测到不存在震荡路由时，则结束当前流程，并将进行震荡路由检测的目标设备重新加入轮询队列，按照上文中描述的方式进行预检测。
在该实施例中，对于任一目标设备，若在对该目标设备进行震荡路由检测的过程中出现错误时，可以停止对该目标设备进行震荡路由检测。此时，网络管理平台可以生成对应的错误日志， 以便用户诸如网络管理员采取相应的措施。
四、确定震荡原因
在该实施例中，可以预先配置如图3所示的震荡原因与震荡现象的对应关系。
网络管理平台按照上述方式定位了震荡源之后，可以根据震荡源的震荡现象查询图3所示的震荡原因与震荡现象的对应关系，确定与震荡源的震荡现象匹配的震荡原因。
在该实施例中，通过配置震荡原因与震荡现象的对应关系，并根据震荡源的震荡现象在该对应关系查询匹配的震荡原因，仅需通过SNMP协议的标准MIB(Management Information Base，管理信息库)库即可实现震荡原因的定位，其震荡原因定位的效率较高，且工作量较少。
通过以上描述可以看出，在本公开实施例提供的技术方案中，通过对待检测设备进行预检测，确定满足震荡路由检测条件的目标设备，并对目标设备进行震荡路由检测，确定是否存在震荡路由，对于任一目标设备，当该目标设备上存在震荡路由时，确定该震荡路由的震荡源，并确定震荡原因，实现了路由震荡定位。
以上对本公开提供的方法进行了描述。下面对本公开提供的装置进行描述。
图9为本公开实施例提供的一种服务器的硬件结构示意图。该服务器可包括处理器901、存储有机器可执行指令的机器可读存储介质902。处理器901与机器可读存储介质902可经由系统总线903通信。并且，通过读取并执行机器可读存储介质902中与路由震荡定位控制逻辑对应的机器可执行指令，处理器901可执行上文描述的路由震荡定位方法。
本文中提到的机器可读存储介质902可以是任何电子、磁性、光学或其它物理存储装置，可以包含或存储信息，如可执行指令、数据，等等。例如，机器可读存储介质可以是：RAM(Radom Access Memory，随机存取存储器)、易失存储器、非易失性存储器、闪存、存储驱动器(如硬盘驱动器)、固态硬盘、任何类型的存储盘(如光盘、dvd等)，或者类似的存储介质，或者它们的组合。
如图10所示，从功能上划分，上述路由震荡定位控制逻辑可以包括：预检测单元1001、检测单元1002以及定位单元1003。
预检测单元1001，用于对待检测设备进行预检测，以确定满足震荡路由检测条件的目标设备；
检测单元1002，用于对所述目标设备进行震荡路由检测，以确定是否存在震荡路由；
定位单元1003，用于对于任一所述目标设备，当该目标设备上存在震荡路由时，确定所述震荡路由的震荡源，并确定震荡原因。
在可选实施例中，预检测单元1001，具体用于按照所述待检测设备的轮询时间间隔对所述待检测设备进行轮询；对于任一待检测设备，当检测到该待检测设备的路由表条目数量发生变化时，将该待检测设备的轮询时间间隔减去第一预设时间；当该待检测设备的轮询时间间隔等于预设触发 检测时间时，执行对所述目标设备进行震荡路由检测的步骤。
在可选实施例中，预检测单元1001，具体用于每经过第二预设时间，将所述待检测设备的剩余检测时间减去所述第二预设时间；当所述待检测设备的剩余检测时间小于等于0时，对所述待检测设备进行预检测，并将所述待检测设备的剩余检测时间置为所述待检测设备的轮询时间间隔。
在可选实施例中，检测单元1002，具体用于对于任一进行震荡路由检测的目标设备，根据该目标设备的路由信息确定该目标设备上是否存在震荡路由；其中，所述路由信息包括目的地址以及路由表条目出现时间。
在可选实施例中，检测单元1002，具体用于根据路由信息收集时间间隔轮询M次该目标设备的路由表，以收集得到路由信息，其中M为大于等于1的自然数；对于任一路由表条目，当该路由表条目的震荡次数大于等于第一预设震荡次数阈值时，所述路由表条目对应的路由被确定为震荡路由；当该路由表条目的震荡次数小于第一预设震荡次数阈值，且该路由表条目出现的总时间小于第一预设时间阈值时，所述路由表条目对应的路由被确定为震荡路由；其中，当一个路由表条目的出现状态发生变化时，该路由表条目的震荡次数加1。
在可选实施例中，所述路由信息收集时间间隔随着已完成的路由信息收集次数的增多而增大或减小。
在可选实施例中，定位单元1003，具体用于对于任一震荡路由，判断该震荡路由是否为该目标设备的直连路由；若是，则确定该目标设备为该震荡路由的震荡源；否则，根据该震荡路由对应的路由协议信息确定该震荡路由的震荡源。
在可选实施例中，定位单元1003，具体用于判断该震荡路由对应的路由协议表条目是否处于震荡状态；若是，则在当前的目标设备完成检测之后，对处于震荡状态的路由协议表条目中对应的发布者设备进行震荡路由检测，直至确定该震荡路由的震荡源；若否，则执行确定震荡原因的步骤。
在可选实施例中，定位单元1003，还用于对于任一目标设备，根据路由协议信息收集时间间隔轮询P次该目标设备的路由协议表，以收集得到路由协议信息，其中P为大于等于1的自然数；其中，所述路由协议信息包括目的地址以及路由协议表条目出现时间；对于任一路由协议表条目，当该路由协议表条目的震荡次数大于等于第二预设震荡次数阈值时，执行在当前的目标设备完成检测之后，对处于震荡状态的路由协议表条目中对应的发布者设备进行震荡路由检测的步骤；或，当该路由协议表条目的震荡次数小于第二预设震荡次数阈值，且该路由协议表条目出现的总时间小于第二预设时间阈值时，执行在当前的目标设备完成检测之后，对处于震荡状态的路由协议表条目中对应的发布者设备进行震荡路由检测的步骤；其中，当一个路由协议表条目的出现状态发生变化时，该路由协议表条目的震荡次数加1。
在可选实施例中，所述路由协议信息收集时间间隔随着已完成的路由协议信息收集次数的增多而增大或减小。
在可选实施例中，定位单元1003，具体用于收集所述震荡源的震荡现象；根据所述震荡源的震荡现象确定震荡原因。
在可选实施例中，定位单元1003，具体用于根据所述震荡源的震荡现象，查询预先配置的震荡现象与震荡原因的对应关系，以确定与所述震荡源的震荡现象匹配的震荡原因。
本公开实施例还提供了一种存储有计算机可执行指令的机器可读存储介质，例如图9中的存储器902，所述计算机可执行指令可由图9所示电子设备中的处理器901执行以实现上文描述的路由震荡定位方法。
需要说明的是，在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
以上所述仅为本公开的较佳实施例而已，并不用以限制本公开，凡在本公开的精神和原则之内，所做的任何修改、等同替换、改进等，均应包含在本公开保护的范围之内。

Claims (14)

  1. 一种路由震荡定位方法，包括：
    对待检测设备进行预检测，以确定满足震荡路由检测条件的目标设备；
    对所述目标设备进行震荡路由检测，以确定是否存在震荡路由；
    对于任一所述目标设备，当该目标设备上存在震荡路由时，确定与所述震荡路由相关的震荡源，并确定震荡原因。
  2. 根据权利要求1所述的方法，其特征在于，所述对待检测设备进行预检测，以确定满足震荡路由检测条件的目标设备包括：
    按照所述待检测设备的轮询时间间隔对所述待检测设备进行轮询；
    对于任一待检测设备，
    当检测到该待检测设备的路由表条目数量发生变化时，通过将该待检测设备的轮询时间间隔减去第一预设时间，更新所述轮询时间间隔；
    当该更新的轮询时间间隔等于预设触发检测时间时，执行对所述目标设备进行震荡路由检测的步骤。
  3. 根据权利要求2所述的方法，其特征在于，所述按照所述待检测设备的轮询时间间隔对所述待检测设备进行轮询，包括：
    每经过第二预设时间，通过将所述待检测设备的剩余检测时间减去所述第二预设时间，更新所述剩余检测时间；
    当所述更新的剩余检测时间小于等于0时，对所述待检测设备进行预检测，并将所述待检测设备的剩余检测时间置为所述待检测设备的轮询时间间隔。
  4. 根据权利要求1所述的方法，其特征在于，所述对所述目标设备进行震荡路由检测，包括：
    对于任一目标设备，根据该目标设备的路由信息确定该目标设备上是否存在震荡路由；
    其中，所述路由信息包括目的地址以及路由表条目出现时间。
  5. 根据权利要求4所述的方法，其特征在于，所述根据该目标设备的路由信息确定该目标设备上是否存在震荡路由，包括：
    根据路由信息收集时间间隔轮询M次该目标设备的路由表，以收集得到路由信息，其中M为大于等于1的自然数；
    对于任一路由表条目，
    当该路由表条目的震荡次数大于等于第一预设震荡次数阈值时，所述路由表条目对应的路由被确定为震荡路由；其中，当一个路由表条目的出现状态发生变化时，该路由表条目的震荡次数加1；
    或，当该路由表条目的震荡次数小于第一预设震荡次数阈值，且该路由表条目出现的总时间小于第一预设时间阈值时，所述路由表条目对应的路由被确定为震荡路由；其中，当一个路由表条目的出现状态发生变化时，该路由表条目的震荡次数加1。
  6. 根据权利要求5所述的方法，其特征在于，所述路由信息收集时间间隔随着已完成的路由信 息收集次数的增多而增大或减小。
  7. 根据权利要求1所述的方法，其特征在于，所述确定与所述震荡路由相关的震荡源，包括：
    对于任一震荡路由，
    判断该震荡路由是否为该目标设备的直连路由；
    若是，则确定该目标设备为与该震荡路由相关的震荡源；
    否则，根据该震荡路由对应的路由协议信息确定与该震荡路由相关的震荡源。
  8. 根据权利要求7所述的方法，其特征在于，所述根据该震荡路由对应的路由协议信息确定与该震荡路由相关的震荡源，包括：
    判断该震荡路由对应的路由协议表条目是否处于震荡状态；
    若是，则在当前的目标设备完成检测之后，对处于震荡状态的路由协议表条目中对应的发布者设备进行震荡路由检测，直至确定与该震荡路由相关的震荡源；
    若否，则执行确定震荡原因的步骤。
  9. 根据权利要求8所述的方法，其特征在于，所述方法还包括：
    对于任一目标设备，根据路由协议信息收集时间间隔轮询P次该目标设备的路由协议表，以收集得到路由协议信息，P为大于等于1的自然数；其中，所述路由协议信息包括目的地址以及路由协议表条目出现时间；
    对于任一路由协议表条目，
    当该路由协议表条目的震荡次数大于等于第二预设震荡次数阈值时，执行在当前的目标设备完成检测之后，对处于震荡状态的路由协议表条目中对应的发布者设备进行震荡路由检测的步骤；其中，当一个路由协议表条目的出现状态发生变化时，该路由协议表条目的震荡次数加1；
    或，当该路由协议表条目的震荡次数小于第二预设震荡次数阈值，且该路由协议表条目出现的总时间小于第二预设时间阈值时，执行在当前的目标设备完成检测之后，对处于震荡状态的路由协议表条目中对应的发布者设备进行震荡路由检测的步骤；其中，当一个路由协议表条目的出现状态发生变化时，该路由协议表条目的震荡次数加1。
  10. 根据权利要求9所述的方法，其特征在于，所述路由协议信息收集时间间隔随着已完成的路由协议信息收集次数的增多而增大或减小。
  11. 根据权利要求1所述的方法，其特征在于，所述确定震荡原因，包括：
    收集所述震荡源的震荡现象；
    根据所述震荡源的震荡现象确定震荡原因。
  12. 根据权利要求11所述的方法，其特征在于，所述根据所述震荡源的震荡现象确定震荡原因，包括：
    根据预先配置的震荡现象与震荡原因的对应关系，确定与所述震荡源的震荡现象匹配的震荡原因。
  13. 一种服务器，包括处理器和机器可读存储介质，所述机器可读存储介质存储有能够被所述 处理器执行的机器可读指令，所述处理器被所述机器可读指令促使执行如权利要求1-12任一项所述的路由震荡定位方法。
  14. 一种机器可读存储介质，所述机器可读存储介质存储有机器可执行指令，在被处理器调用和执行时，所述机器可执行指令促使所述处理器执行如权利要求1-12任一项所述的路由震荡定位方法。
PCT/CN2019/128055 2018-12-24 2019-12-24 对路由震荡的定位 Ceased WO2020135445A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811585124.5 2018-12-24
CN201811585124.5A CN110708207B (zh) 2018-12-24 2018-12-24 路由震荡定位方法、服务器及机器可读存储介质

Publications (1)

Publication Number Publication Date
WO2020135445A1 true WO2020135445A1 (zh) 2020-07-02

Family

ID=69192983

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/128055 Ceased WO2020135445A1 (zh) 2018-12-24 2019-12-24 对路由震荡的定位

Country Status (2)

Country Link
CN (1) CN110708207B (zh)
WO (1) WO2020135445A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111337784B (zh) * 2020-04-07 2023-01-20 中车株洲电力机车有限公司 一种电力机车辅机测试的控制方法及系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040017769A1 (en) * 2002-04-11 2004-01-29 International Business Machines Corporation Method of establishing a route redundancy in a data transmission system using static routes
CN1870538A (zh) * 2006-05-08 2006-11-29 国家数字交换系统工程技术研究中心 一种实现故障管理的方法及系统
US20120030523A1 (en) * 2010-07-28 2012-02-02 At&T Intellectual Property I, L.P. Alarm Threshold For BGP Flapping Detection
CN102868621A (zh) * 2012-08-31 2013-01-09 武汉烽火网络有限责任公司 一种利用异步方式实现大容量路由快速写硬件的方法
CN106992877A (zh) * 2017-03-08 2017-07-28 中国人民解放军国防科学技术大学 基于sdn架构的网络故障检测与修复方法

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7180864B2 (en) * 2002-02-27 2007-02-20 Lucent Technologies Inc. Method and apparatus for exchanging routing information within an autonomous system in a packet-based data network
CN101155175B (zh) * 2006-09-27 2011-06-15 华为技术有限公司 一种基于bgp协议的出路由过滤的方法和装置
CN101394302B (zh) * 2007-09-19 2012-04-04 华为技术有限公司 控制路由震荡的方法和设备
CN101227262B (zh) * 2008-02-20 2011-12-07 中兴通讯股份有限公司 路由抑制系统及方法
CN101404614B (zh) * 2008-11-05 2011-01-26 中国移动通信集团江苏有限公司 一种路由振荡探测方法
CN106470154A (zh) * 2015-08-17 2017-03-01 中兴通讯股份有限公司 边界网关协议bgp路由源处理方法及装置
CN111277494B (zh) * 2016-02-16 2021-08-13 华为技术有限公司 一种报文传输方法及装置
CN106059934B (zh) * 2016-06-27 2020-01-03 新华三技术有限公司 一种路由信息处理方法及装置
CN108123848B (zh) * 2017-12-19 2020-10-09 新华三技术有限公司 设备监控方法及装置
CN108696433B (zh) * 2018-07-24 2021-04-06 新华三技术有限公司 路由振荡抑制方法及bgp路由设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040017769A1 (en) * 2002-04-11 2004-01-29 International Business Machines Corporation Method of establishing a route redundancy in a data transmission system using static routes
CN1870538A (zh) * 2006-05-08 2006-11-29 国家数字交换系统工程技术研究中心 一种实现故障管理的方法及系统
US20120030523A1 (en) * 2010-07-28 2012-02-02 At&T Intellectual Property I, L.P. Alarm Threshold For BGP Flapping Detection
CN102868621A (zh) * 2012-08-31 2013-01-09 武汉烽火网络有限责任公司 一种利用异步方式实现大容量路由快速写硬件的方法
CN106992877A (zh) * 2017-03-08 2017-07-28 中国人民解放军国防科学技术大学 基于sdn架构的网络故障检测与修复方法

Also Published As

Publication number Publication date
CN110708207A (zh) 2020-01-17
CN110708207B (zh) 2021-10-29

Similar Documents

Publication Publication Date Title
US11646952B2 (en) Method and network device for tagging network traffic flows
CN107104824B (zh) 一种网络拓扑确定方法和装置
US10764214B1 (en) Error source identification in cut-through networks
US20150331777A1 (en) System and method of generating data center alarms for missing events
US20160050148A1 (en) Controlling the reactive caching of wildcard rules for packet processing, such as flow processing in software-defined networks
US12206572B2 (en) Performance measurement, telemetry, and OAM in MPLS networks using entropy labels
WO2019185051A1 (zh) 一种基于集成流表转发报文的方法及装置
CN113542128B (zh) 一种发送路由信息的方法和装置
US12132640B2 (en) Method for monitoring running state of peer, apparatus, and storage medium
WO2021093465A1 (zh) 发送报文、接收报文以进行oam的方法、装置及系统
Laffranchini et al. Measurements as first-class artifacts
WO2023193498A1 (zh) 网络的可靠性评估方法、装置、计算设备以及存储介质
CN113965492A (zh) 一种数据流统计方法及装置
CN109981768A (zh) 分布式网络存储系统中的io多路径规划方法及设备
WO2020135445A1 (zh) 对路由震荡的定位
CN116094957A (zh) 端口连接检测方法、装置及存储介质
WO2023009314A1 (en) Performance measurement, telemetry, and oam in mpls networks using entropy labels
US20180302305A1 (en) Data center automated network troubleshooting system
CN118612315A (zh) 一种基于dpu的网络报文的处理方法、装置、设备及介质
US11848837B2 (en) Network telemetry based on application-level information
WO2024055584A1 (zh) 时延测量方法、电子设备、计算机可读介质
US12500828B2 (en) Optimizing path tracing to enable network assurance in existing network hardware
KR20250060480A (ko) Ｓｄｎ 제어기와 이와 연결된 ｓｄｎ 스위치를 이용한 네트워크 토폴로지 구축방법
WO2024186628A1 (en) Hardware accelerated path tracing analytics
EP4677824A1 (en) Optimizing path tracing to enable network assurance in existing network hardware

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19903428

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19903428

Country of ref document: EP

Kind code of ref document: A1