CN103684648B

CN103684648B - A kind of ring network node clock effectiveness judgment method and device

Info

Publication number: CN103684648B
Application number: CN201210353572.9A
Authority: CN
Inventors: 万艳玲
Original assignee: ZTE Corp
Current assignee: ZTE Corp
Priority date: 2012-09-21
Filing date: 2012-09-21
Publication date: 2018-02-16
Anticipated expiration: 2032-09-21
Also published as: CN103684648A

Abstract

The invention discloses a kind of ring network node clock effectiveness judgment method and device, in methods described, when network link and configuration are normal, and each node is when possessing clock/time synchronized function, a detection with normal time synchronization PTP ports is chosen in looped network node from node, extra at least one detection is configured from port from node for the detection, and according to the detection from the BMC that port and time synchronized PTP ports include from port respectively with interacting for upstream equipment and at least two groups of reporting for calculating choose whether report and alarm, if the difference between at least two groups of reporting exceeds default alarm threshold, then report and alarm.The present invention is by the interaction of PTP messages, the calculating of time deviation the selection detecting port, port and compares, it is possible to achieve looped network node clock validity is judged, and report and alarm.

Description

A kind of ring network node clock effectiveness judgment method and device

Technical field

The present invention relates to the network clocking field of synchronization of technical field of network communication, in particular to a kind of looped network section Dot Clock availability deciding method and device.

Background technology

PTN（Packet Transport Network, Packet Transport Network）Refer to such a optical transport network framework and Particular technique：An aspect is provided between IP operation and bottom light-transmitting medium, it is directed to the burst of Packet Service flow Property and statistic multiplexing transmission requirement and design, as core and support that multi-service provides using Packet Service, there is lower totality Use cost（Total cost of ownership, abbreviation TCO）, while adhere to the advantage of optical transport, including high availability and Reliability, efficient bandwidth management mechanism and traffic engineering, easily OAM（Operation Administration and Maintenance, operation, management, safeguard）With webmaster, expansible, higher security etc..

At present, PTN realizes the transmission to frequency and time synchronizing signal, there is two class possible techniques：One is it is based on physics The simultaneous techniques of layer, such as the method for salary distribution G.8261 defined synchronous ethernet technology of reference clock signal；Second, as IEEE1588 V2（Also referred to as accurate time synchronization protocol, Precision Time Protocol, abbreviation PTP）Base as defined in agreement In the simultaneous techniques of packet bag.

Synchronous ethernet is a kind of based on traditional physical layer clocks simultaneous techniques, and the technology is from physical layer data code stream The high precision clock of network delivery is extracted, then is tracked and handles, system clock is formed, is entered in sending side using system clock Row data are sent, and so as to realize the Frequency Synchronization between different nodes, not by business load flow effect, are provided for system based on frequency The clock synchronizing function of rate, wherein, the credit rating information of clock can pass through special SSM（Synchronization Status Message, Synchronization Status Message）Frame is transmitted.

And IEEE1588v2 agreements, it is applied to such a master-slave synchronisation system, the synchronization system by ordinary clock and Boundary clock forms, and ordinary clock only possesses the ability for receiving the time, and boundary clock has the ability of passing time, when common The port of clock and boundary clock is generally in one kind under following three kinds of states in normal work：

Master state（master）, it is the source in sometime path to show the port；

Slave mode（slave）, show that the port is synchronized with a master（master）Clock；

Passive state（passive）, show that the port does not participate in synchronization message interaction to prevent that the time cyclic.

In system synchronization procedure, PTP time synchronization protocols and temporal information are issued, from clock master clock cycle Port receives the timestamp information that master clock port is sent, when system calculates time delay and principal and subordinate on principal and subordinate's circuit accordingly Between it is poor, and using time difference adjustment local zone time, the slave unit time is kept and the frequency and phase of main equipment time consistency.

It is general to use under ring network and the scene of network support synchronous ethernet and IEEE1588V2 time synchronizeds The scheme of full BC node hop-by-hop Time Transmissions, however, the mode of this hop-by-hop BC passing times has a risk：It is once middle There is clock failure in node, may result in IEEE1588V2 time synchronizeds and exception occurs, while involves downstream node, and this event Barrier downstream node can not be found out.

Referring to Fig. 1, node A is GM（Grandmaster Clock, best master clock）Clock, node B-F are BC（It is general Logical clock）Node.According to BMC（Best master clock algorithm）Algorithm, B lock A, C lock B, D lock C, once node B clock appearance is different Often, now, in the case of link is normal, node C and D time can be caused also to break down.

For the above-mentioned failure being likely to occur, it is necessary to propose a kind of detection means, to judge looped network node interior joint clock Whether source there is exception.

The content of the invention

In consideration of it, it is an object of the invention to provide a kind of ring network node clock effectiveness judgment method and device.

In order to realize the above-mentioned purpose of the present invention, it uses following technical scheme：

A kind of ring network node clock effectiveness judgment method, wherein, when network link and configure normally, and each node possesses During clock/time synchronized function, one is chosen in looped network node has the detection of normal time synchronization PTP ports from node, is The detection is detected from port and time synchronized PTP from the extra at least one detection of node configuration from port according to described At least two groups of reporting that the BMC that port includes calculates with interacting for upstream equipment respectively from port choose whether Report police, if the difference between at least two groups of reporting exceeds default alarm threshold, report and alarm.

Under preferred embodiment, the ring network node clock effectiveness judgment method comprises the following steps：

It is selected to refer to source node and detection from node, and looped network protection function is opened, wherein, it is described to detect from node energy It is enough according to and the time synchronized PTP ports of upstream equipment carry out PTP message interactions and to calculate the time between upstream equipment inclined Difference △ T, the reporting △ T are for correcting the node system time；

To specifying detection master port with reference to source node, detection is specified from port from node to detection, and configure detection main side Mouth carries out interacting for PTP messages with detecting from port, wherein, the detection master port and detection are not involved in BMC from port and transported Calculate；

Detect and calculate detection time deviation from port according to the message time stamp of the PTP messages received from detection master port △Tref；

△ T and △ Tref difference is calculated, the absolute value of the difference and default alarm threshold Threshold are carried out Compare, if the absolute value of the difference exceedes alarm threshold, report time anomaly to alert；If the absolute value of the difference does not surpass Alarm threshold is crossed, then does not report time anomaly to alert.

Under preferred embodiment, the detection is carrying out PTP messages from node with the time synchronized PTP ports of upstream equipment During interaction, the PTP messages encapsulation format can be L2 multicasts, L3 multicasts or L3 unicasts.

Under preferred embodiment, master port is being detected to being specified with reference to source node, and detection is specified from node to detection The step of port, under the pattern of boundary clock BC and transmission clock TC functions, with reference to source node and detection from node it Between node network element and upper hop node carry out normal time it is synchronous while, must possess the PTP reports that recognition detection port is sent Text and the function downstream transmitted, different type of messages, can use middle each TC nodes configuration and reference source node with And detect from node identical vlan domains, or configure with reference source node and detection from the respective IP address of node, with this The PTP messages for sending its recognition detection port.

Under preferred embodiment, master port is being detected to being specified with reference to source node, and detection is specified from node to detection From port, and configure detection master port with detection from port carry out detection messages the step interacted in, refer to source node and Detection sets the clock ID of opposite end respectively from node, and does legitimacy to the PTP messages that each detection port receives accordingly Check.

Under preferred embodiment, when the absolute value of the △ T and △ Tref difference is not less than alarm threshold, then choose The detection from the previous upstream node of node as new detection from node, and re-execute the looped network node clock and have Imitate sex determination method.

A kind of looped network node clock availability deciding device, including：

Judging unit, for when network link and configuration it is normal, and when each node possesses clock/time synchronized function, One is chosen in looped network node has the detection of normal time synchronization PTP ports from node, is configured additionally from node for the detection At least one detection is distinguished from port, and according to the detection from the BMC that port and time synchronized PTP ports include from port At least two groups of reporting calculated with interacting for upstream equipment choose whether report and alarm；

Alarm Unit, for when the difference between at least two groups of reporting exceeds default alarm threshold, Report and alarm.

Under preferred embodiment, the method that the judging unit carries out looped network node clock availability deciding includes following step Suddenly：

△ T and △ Tref difference is calculated, the absolute value of the difference and default alarm threshold Threshold are carried out Compare, if the absolute value of the difference exceedes alarm threshold, generate time anomaly alert command；If the absolute value of the difference Not less than alarm threshold, then time anomaly alert command is not generated.

Under preferred embodiment, when the absolute value of the △ T and △ Tref difference is not less than alarm threshold, list is judged It is first then choose the detection from the previous upstream node of node as new detection from node, and re-execute the looped network section Dot Clock availability deciding method.

The present invention can be seen that compared with prior art by the technical scheme of the invention described above, have below beneficial to effect Fruit：

1）, can be with by the interaction of PTP messages, the calculating of time deviation detecting the selection of port, port and comparing Realize and preliminary judgement is carried out to looped network node clock validity, in the case of being more highly preferred to, by repeatedly comparing, so as to enter One step determines abnormal nodes.

2）In the case where exception occurs in time deviation, notify manually to be intervened in a manner of alarm, it is complete so as to prevent Net time anomaly, strengthen the maintainability of synchronizing network.

Brief description of the drawings

Fig. 1 is IEEE1588V2 time synchronized loop network schematic diagrames；

Fig. 2 is to utilize BC+TC clock module detection time exception schematic diagrames；

Fig. 3 is looped network node clock availability deciding device Cleaning Principle schematic diagram provided in an embodiment of the present invention.

The realization, functional characteristics and excellent effect of the object of the invention, below in conjunction with specific embodiment and accompanying drawing do into The explanation of one step.

Embodiment

Technical scheme of the present invention is described in further detail with specific embodiment below in conjunction with the accompanying drawings, so that this The technical staff in field can be better understood from the present invention and can be practiced, but illustrated embodiment is not as the limit to the present invention It is fixed.

The purpose of the embodiment of the present invention is to provide a kind of ring network node clock effectiveness judgment method, and it is used to preventing complete It is abnormal to net time appearance, and in the case where exception occurs in the time, report and alarm, notice is manually intervened.

In consideration of it, a kind of ring network node clock effectiveness judgment method provided in an embodiment of the present invention comprises the following steps：

Referring to Fig. 3, it is assumed that exception is not present in network link and configuration, and each node possesses clock/time synchronized function, and energy Synchronize.

On node C in addition to having normal time synchronized PTP ports（BMC master ports（MasterPort）, BMC is from port （Slave Port））, at least one can be configured（Such as in the present embodiment can be 2 with option and installment）Detect from port （Reference Slave Port1、2）.In implementation process, this 3 groups from（slave）When port can calculate three groups respectively Between deviation（offset）.The time of source device selected by acquiescence is stable, reliable, so during node C calculate three groups Between deviation（offset）Between, it should difference very little, if three groups of reporting（offset）Between widely different, example Such as beyond default alarm threshold, now illustrate that exception occurs in the clock of some nodes in link, during so as to have influence on Between net synchronization capability, then report and alarm, notify human intervention.

In actual implementation process, ring network node clock effectiveness judgment method specific implementation step provided in an embodiment of the present invention It is rapid as follows：

Step 1, select with reference to source node and detection from node, open looped network protection function.

For that with reference to for source node, other trusted nodes outside non-detection node in looped network can be selected, be typically chosen Best master clock（Grandmaster clock）.

For detecting from node, it should calculate time synchronized PTP from port, and upstream equipment time Synchronous PTP master ports message interaction is normal（Detection messages encapsulation format can be L2 multicasts, L3 multicasts, L3 unicasts）, while energy The reporting between upstream equipment is calculated, is designated as △ T, the reporting is used for correcting the node system time.

Step 2, detection master port is being specified with reference to source node, detection is specified from port, the detection from node in detection Port is to be manually specified, and is not involved in BMC calculating.

Master and slave physical port as detection may be used for normal time synchronized simultaneously, in this case, it is necessary to preferential Ensure that nodal clock type is provided simultaneously with boundary clock and transmission clock（BC+TC patterns）Function, i.e. BC patterns are used for IEEE1588V2 time synchronizeds, TC patterns are used as detection IEEE1588V2（Referring to Fig. 2）.Under BC+TC patterns, host node is detected And detection from the node network element between node and upper hop node progress normal time it is synchronous while, it is necessary to possess recognition detection PTP messages that port is sent and the function of downstream transmitting, different type of messages, it can use and be configured in middle each TC nodes With detecting master and slave node identical vlan domains, or configuration detects the IP address of master and slave node, is sent out with this recognition detection port The PTP messages gone out.

In addition, detect master port and detection from port can also be newly-built PTP ports in equipment, such case Under, new physical connection need to be established, is adapted to the distance between node nearer mininet.

In addition, detection master port and detection need to correspond from the interaction between port.It is configured to detect master port hair Send sync（follow_up）, deslay_resp messages, receive dealy_req messages；Detection is configured to send from port Delay_req, and receive the message that detection master port is sent.

In order to ensure the correctness of timestamp, opposite end can be set on node with reference to source node and detection respectively Clock ID, the PTP messages thus received to detection port do validity checking.

Step 3, detect and stab calculating detection time deviation from port according to message time, be designated as △ Tref, the deviation is not For node system time complexity curve.

, generally can be with order to avoid detection link packet interaction is abnormal so as to cause △ Tref exception to be in itself present Two detection links of selection, that is, select 2 and refer to source node, and it respectively configures 1 detection master port, while in detection from node 2 detections of configuration carry out PTP message interactions, while calculate detection time deviation respectively from port.So, can be very When avoiding that wherein a link packet can not interact normally in big degree, the detection time deviation for causing to calculate occurs extremely.

Step 4, △ T and △ Tref difference is calculated, the absolute value of the difference and default alarm threshold are done into ratio Compared with the alarm threshold is designated as Threshold.

In preferred embodiment, for type identical node, possess identical acquiescence alarm threshold, while user can be with The alarm threshold is modified by configuring, preferably to safeguard.

If the absolute value of the difference exceedes alarm threshold, report time anomaly to alert, thereby it is assumed that from reference Source node on detection node this link with the presence of node clock failure, to facilitate user to determine whether；

If the absolute value of the difference does not report time anomaly to alert not less than alarm threshold.

In the case of not less than alarm threshold, it can not illustrate that nodal clock failure is not present in the link completely.It is false If there is clock exception in the tested node, it is possible to exist | T- △ Tref |<Threshold situation.So at this In preferred embodiment of the invention, in order to ensure the objectivity of testing result, it repeatedly can detect several nodes more, choose the inspection Survey from the previous upstream node of node as new detection from node, and re-execute the looped network node clock validity and sentence Method is determined, so as to progressively reduce the scope for determining malfunctioning node.

The present invention is described in further detail below in conjunction with the accompanying drawings：

Embodiment 1：The abnormal node of clock in network be present, but selected detection node clock is without exception.

With reference to figure 1, it is assumed that node A, i.e. best master clock（Grandmaster clock）Clock time be credible , there is exception in the B node clock in the network, but is not discovered, herein by taking D nodes as an example, progressively determines that B node is present and asks Topic.

D nodes have the 2 time synchronized PTP ports in left and right, participate in BMC and calculate, according to BMC result of calculations, when D nodes select Between synchronous C or E nodes（Assuming that D node times are synchronized with C nodes）.

The first step, confirm that nodes clock type is BC+TC, select with reference to source node and detection from node, open Open looped network protection function.

Selected A nodes are that D nodes are detection node with reference to source node, while looped network protection work(is opened on the node of A, D two Energy.By node D time synchronized from（Slave）Calculated on port and be designated as △ for the time deviation of system time amendment TD。

Second step, detection master port is being specified with reference to source node, detection is specified from port, the detection from node in detection Port is to be manually specified.

Detection messages encapsulation format is defaulted as L3 unicasts, a detection master port PortM is configured on node a, in node D Upper one detection of configuration is from port PortS, and configuration detects main section under all node TC patterns between the two detection nodes Point and detect from the IP of node, can also be in order to ensure normal IEEE1588V2 time synchronized message interactions between two-port The clock ID of other side is respectively configured on two nodes.

3rd step, detect and stab calculating detection time deviation from port according to message time.

The detection of D nodes calculates detection time deviation from port PortS ports, is designated as △ Tdref, the deviation, which is not used in, is System time complexity curve, only it is used for characterizing the time with the presence or absence of abnormal.

4th step, alarm threshold size is configured, is designated as ThresholdD, will | △ TD- △ TDref | with alarm threshold ThresholdD is compared.

Because B node clock is abnormal, occur | △ TD- △ TDref |>ThresholdD situation, above gives the correct time Between abnormality alarming notice manual intervention, link it is not disconnected, configure it is normal in the case of, now can tentatively judge, certain middle section There is exception in the clock of point.In order to determine specifically which node goes wrong, continue to handle in next step.

5th step, select C nodes be detection node, A nodes are remained as with reference to source node, continue executing with above-mentioned steps one to Step 4, if still had | △ TC- △ TCref |>ThresholdC, so far then there can be clock with initial guess, node B It is abnormal.

Embodiment 2：The abnormal node of clock in network be present, and selected detection node is exactly that clock abnormal section occurs Point.

With continued reference to Fig. 1, it is assumed that node A, i.e. best master clock（Grandmaster clock）Clock time be can Letter, there is exception in the B node clock in the network, but is not discovered, herein by taking B node as an example, determines that B node is present and asks Topic.

Selected A nodes are that B node is detection node with reference to source node, while looped network protection work(is opened on the node of A, B two Energy.By the time synchronized of B node from（Slave）Calculated on port and be designated as △ for the time deviation of system time amendment TB。

Detection messages encapsulation format is defaulted as L3 unicasts, a detection master port PortM is configured on A nodes, in B node Upper one detection of configuration is from port PortS, and configuration detects main section under all node TC patterns between the two detection nodes Point and detection, can be two in order to ensure normal IEEE1588V2 time synchronized message interactions between two-port from the IP of node The clock ID of other side is respectively configured on individual node.

The PortS ports of B node calculate detection time deviation, are designated as △ TBref, the deviation is not used in system time and repaiied Just, only it is used for characterizing the time with the presence or absence of abnormal.

4th step, alarm threshold size is configured, is designated as ThresholdB, will | △ TB- △ TBref | with alarm threshold ThresholdB is compared.

B node clock itself is abnormal, so occurring | △ TB- △ TBref |<ThresholdC situation, so as to not Report and alarm.In order to avoid missing to the abnormal judgement of B node, it is further processed.

5th step, one node adjacent with B node of increase continues to judge, such as C nodes, still selects A nodes as ginseng Source node is examined, copies step 1 to be had to the process of step 4 | △ TC- △ TCref |>ThresholdC, so far, Ke Yichu Step speculates that B node has clock exception, now, report and alarm.

The preferred embodiments of the present invention are the foregoing is only, are not intended to limit the scope of the invention, every utilization The equivalent structure or equivalent flow conversion that description of the invention and accompanying drawing content are made, or directly or indirectly it is used in other correlations Technical field, be included within the scope of the present invention.

Claims

A kind of 1. ring network node clock effectiveness judgment method, it is characterised in that when network link and configure normally, and each node When possessing clock/time synchronized function, a detection with normal time synchronization PTP ports is chosen in looped network node from section Point, it is that the detection is detected from port and time synchronized PTP from two extra detections of node configuration from port, and according to described At least two groups of reporting that the BMC that port includes calculates with interacting for upstream equipment respectively from port choose whether Report police, if the difference between at least two groups of reporting exceeds default alarm threshold, report and alarm.
2. ring network node clock effectiveness judgment method as claimed in claim 1, it is characterised in that comprise the following steps：

It is selected with reference to source node and detection from node, and open looped network protection function, wherein, the detection can be according to from node PTP message interactions are carried out according to the time synchronized PTP ports of upstream equipment and calculate the reporting between upstream equipment △ T, the reporting △ T are for correcting the node system time；

To with reference to source node specify detection master port, to detection from node specify detection from port, and configure detection master port with The interaction that PTP messages are carried out from port is detected, wherein, the detection master port and detection are not involved in BMC computings from port；

Detect and calculate detection time deviation △ from port according to the message time stamp of the PTP messages received from detection master port Tref；

△ T and △ Tref difference is calculated, the absolute value of the difference and default alarm threshold Threshold are compared Compared with if the absolute value of the difference exceedes alarm threshold, reporting time anomaly to alert；If the absolute value of the difference not less than Alarm threshold, then time anomaly is not reported to alert.
3. ring network node clock effectiveness judgment method as claimed in claim 2, it is characterised in that the detection exists from node When carrying out PTP message interactions with the time synchronized PTP ports of upstream equipment, the PTP messages encapsulation format can be L2 multicasts, L3 multicasts or L3 unicasts.
4. ring network node clock effectiveness judgment method as claimed in claim 2, it is characterised in that to referring to reference to source node Regular inspection surveys master port, and specifies detection the step of port from node to detection, in boundary clock BC and transmission clock TC It is synchronous with upper hop node progress normal time from the node network element between node with reference to source node and detection under the pattern of function While, must possess the recognition detection port PTP messages sent and the function of downstream transmitting, different type of messages can be with Using middle each TC nodes configuration with reference to source node and detection from node identical vlan domains, or configuration and reference source Node and detection make the PTP messages that its recognition detection port sends from the respective IP address of node, with this.
5. ring network node clock effectiveness judgment method as claimed in claim 4, it is characterised in that to referring to reference to source node Regular inspection surveys master port, and specifies detection from port from node to detection, and configures detection master port and carried out with detection from port In the step of the interaction of detection messages, the clock ID of opposite end is being set respectively with reference to source node and detection from node, and accordingly Validity checking is done to the PTP messages that each detection port receives.
6. ring network node clock effectiveness judgment method as claimed in claim 2, it is characterised in that as the △ T and △ When the absolute value of Tref difference is not less than alarm threshold, then the detection is chosen from the previous upstream node of node as new Detection from node, and re-execute the ring network node clock effectiveness judgment method.
A kind of 7. looped network node clock availability deciding device, it is characterised in that including：

Judging unit, for when network link and configuration it is normal, and when each node possesses clock/time synchronized function, in looped network One is chosen in node has the detection of normal time synchronization PTP ports from node, for two of the detection from node configuration additionally Detection is set with upstream respectively from port, and according to the BMC that the detection includes from port and time synchronized PTP ports from port The standby at least two groups of reporting for interacting and calculating choose whether report and alarm；

Alarm Unit, for when the difference between at least two groups of reporting exceeds default alarm threshold, reporting Alarm.
8. looped network node clock availability deciding device as claimed in claim 7, it is characterised in that the judging unit is carried out The method of looped network node clock availability deciding comprises the following steps：

It is selected with reference to source node and detection from node, and open looped network protection function, wherein, the detection can be according to from node PTP message interactions are carried out according to the time synchronized PTP ports of upstream equipment and calculate the reporting between upstream equipment △ T, the reporting △ T are for correcting the node system time；

To with reference to source node specify detection master port, to detection from node specify detection from port, and configure detection master port with The interaction that PTP messages are carried out from port is detected, wherein, the detection master port and detection are not involved in BMC computings from port；

Detect and calculate detection time deviation △ from port according to the message time stamp of the PTP messages received from detection master port Tref；

△ T and △ Tref difference is calculated, the absolute value of the difference and default alarm threshold Threshold are compared Compared with if the absolute value of the difference exceedes alarm threshold, generation time anomaly alert command；If the absolute value of the difference is not More than alarm threshold, then time anomaly alert command is not generated.
9. looped network node clock availability deciding device as claimed in claim 8, it is characterised in that the detection exists from node When carrying out PTP message interactions with the time synchronized PTP ports of upstream equipment, the PTP messages encapsulation format can be L2 multicasts, L3 multicasts or L3 unicasts.
10. looped network node clock availability deciding device as claimed in claim 8, it is characterised in that to reference to source node Detection master port is specified, and detection is specified the step of port from node to detection, in boundary clock BC and transmission clock It is same from the node network element between node and upper hop node progress normal time with reference to source node and detection under the pattern of TC functions While step, must possess the recognition detection port PTP messages sent and the function of downstream transmitting, different type of messages can To be configured and with reference to source node and detection from node identical vlan domains using in middle each TC nodes, or configuration is with referring to Source node and detection make the PTP messages that its recognition detection port sends from the respective IP address of node, with this.
11. looped network node clock availability deciding device as claimed in claim 8, it is characterised in that to reference to source node Detection master port is specified, and detection is specified from port from node to detection, and configures detection master port with detecting from port In the step of the interaction of row detection messages, the clock ID of opposite end is being set from node respectively with reference to source node and detecting, and according to This does validity checking to the PTP messages that each detection port receives.
12. looped network node clock availability deciding device as claimed in claim 8, it is characterised in that as the △ T and △ When the absolute value of Tref difference is not less than alarm threshold, judging unit is then chosen the detection and saved from the previous upstream of node Put as new detection from node, and re-execute the ring network node clock effectiveness judgment method.