CN102521099A

CN102521099A - Process monitoring method and process monitoring system

Info

Publication number: CN102521099A
Application number: CN2011103781220A
Authority: CN
Inventors: 刘列波
Original assignee: SHENZHEN TONGZHOU VIDEO MEDIA CO Ltd
Current assignee: Shenzhen Coship Electronics Co Ltd
Priority date: 2011-11-24
Filing date: 2011-11-24
Publication date: 2012-06-27

Abstract

The invention discloses a process monitoring method and a process monitoring system. The method comprises the following steps of: receiving a request for querying the list of monitored hosts from a client, acquiring the list of managed Windows and Linux hosts, and returning the list of the monitored hosts to the client; receiving information for polling a monitored host and the concerned process list of the host from the client, setting login information for logging in the host, and returning the concerned process list of the host to the client; and receiving information for querying the currently-running real-time process of the monitored host from the client, securely logging in the host based on SSH (Secure Shell) service, acquiring the list of all currently-running real-time processes of the host based on SNMP (Simple Network Management Protocol) service, filtering invalid process information according to the concerned process list of the host and a preset strategy, and outputting the acquired real-time process parameters to the client for exhibition. With the adoption of the process monitoring method and the process monitoring system, the cross-platform process monitoring can be achieved, and the cost for maintenance and development can be lowered.

Description

Process monitoring method and process monitoring system

Technical Field

The present invention relates to computer monitoring technologies, and in particular, to a process monitoring method and a process monitoring system.

Background

The process monitoring means monitoring various running processes in the Windows and Linux operating systems of the client, acquiring and analyzing various indexes including the memory utilization rate, the CPU occupation time, the current state of a program and the like during the process running, and reporting the running safety condition of the current system to a user according to various index information during the process running, so that the user can know the state of the process, and can timely maintain the process which has a fault when the process has a fault.

The existing process monitoring scheme is generally implemented based on a Windows Application Programming Interface (API) for a Windows operating system, for example, based on a Linux API for the Windows operating system, and a C-language development process monitoring program is implemented for the Linux operating system, for example, a C-language development process monitoring program is pre-implemented in the Windows operating system, and after the client is started, the process monitoring program is loaded to monitor the process started by the client, and a user may call a task manager in the Windows operating system through the Windows API to view related information of the current process operation, for example, a name of the current process operation, an operation path, CPU occupation time, memory usage, and the like, and may end a non-system process through a process termination control provided by a task manager interface.

As can be seen from the above, in the conventional process monitoring method, process monitoring between different operating systems (platforms) needs to be implemented through respective related application programming interfaces, for example, process monitoring in a Windows operating system can only be implemented through a Windows API provided by the Windows operating system, while process monitoring in a Linux operating system can only be implemented through a Linux API provided by the Linux operating system, and cross-platform process monitoring cannot be implemented, that is, process monitoring between different operating systems cannot be compatible, and even a process of another client cannot be monitored through one client; further, the process monitoring list is set by a program developer in advance, and cannot meet the configuration of the personalized requirements of the user; moreover, the process monitoring lists based on different operating systems (platforms) have higher development cost and larger maintenance workload.

Disclosure of Invention

In view of the above, the main objective of the present invention is to provide a process monitoring method, which achieves cross-platform process monitoring and reduces maintenance and development costs.

Another objective of the present invention is to provide a process monitoring system, which implements cross-platform process monitoring and reduces maintenance and development costs.

In order to achieve the above object, the present invention provides a process monitoring method, which comprises:

receiving a monitored host list query request of a client, acquiring a managed windows and linux host list, and then returning the monitored host list to the client;

receiving polling information of a monitored host selected by setting of a client and an attention process list of the host, setting login information for logging in the host, and returning the attention process list of the host to the client;

the method comprises the steps that a receiving user inquires real-time process information of a monitored host machine through a client side, the monitored host machine is safely logged in the host machine based on SSH service, all real-time process lists of the host machine in current operation are obtained based on SNMP service, invalid process information is filtered out according to a concerned process list of the host machine and a preset strategy, and then the obtained real-time process parameters are output to the client side for display.

The monitored host list includes: the IP address information of the network element, the name information of the monitoring process, whether alarm information is generated or not, the description information of the monitoring process, the default startup script information, the default shutdown script information and the default restart script information.

The monitored host list is configured in one or a combination of a database format and an extensible markup language format.

The process is composed of a plurality of process parameters, the process parameters including: the method comprises the following steps of process identification, process name, process manufacturer identifier, process running path, process starting parameter, process type, process running state, CPU using time and memory using size.

The process types include: unknown type, operating system type, device driver type, and application type.

The running state of the process comprises: operational, non-operational, and illegal.

The process monitoring comprises: process inquiry, process status polling, process start, process stop, and process restart.

The process starting comprises the following steps:

acquiring the name of the un-started process from the concerned process configuration list;

constructing a starting command character string to be executed according to the type of the monitored host, the process name and a starting mode selected by a user;

the process monitoring server executes the command on the monitored host computer through the SSH protocol to start the process.

Further comprising:

after the monitored host computer successfully executes the command, the SSH service returns the successful command execution information to the process monitoring server in a byte stream form;

the process monitoring server analyzes the returned result and acquires the process parameter data operated by the monitored host through SNMP;

and the process monitoring server outputs the process parameter data and the successful information of the starting process to the client requesting the process monitoring for displaying.

Further comprising:

and updating the set attention process list.

The updating comprises the following steps: adding a monitoring process, modifying the monitoring process and reconfiguring network element polling information.

The outputting the acquired real-time process parameters to the client for display comprises:

and if the preset warning strategy is determined to be a warning notification for notifying the user when the process fails or is suspended accidentally, after the process failure is determined, outputting warning information to notify the user and outputting the process parameter information of the failure to the client for display.

Further comprising:

and determining that the state of the concerned process which is failed or accidentally suspended is recovered to be normal, and sending alarm clearing information to a user.

The filtering out invalid process information according to the concerned process list of the host and a preset strategy, and then outputting the obtained real-time process parameters to the client for display comprises the following steps:

when the preset data acquisition interval time in the concerned process configuration list is up, determining that a host computer monitored by the requested process can be communicated by adopting a configured user name and a configured password through an SSH protocol;

acquiring a real-time process parameter data set of each monitored host according to the SNMP;

filtering the same process and the process with the process running state of being inoperable or illegal in the process parameter data set to obtain a filtered process parameter data set;

traversing the concerned processes in the concerned process configuration list, inquiring the processes with the same process names as the concerned processes in the filtered process parameter data set, and placing the inquired concerned processes in the real-time process set;

inquiring whether a process monitoring server generates offline warning information for an attention process in a real-time process set, if so, sending online warning clearing information of the attention process, and otherwise, sending offline warning information of the attention process;

and updating the stored attention process configuration list according to the attention process in the real-time process set.

The login information comprises: username, password, polling time, and whether to poll the information.

A process monitoring system, the process monitoring system based on a C/S architecture, comprising: a client and a process monitoring server, wherein,

the client receives the network element list updating notification of the process monitoring server, triggers a network element list acquiring request of the process monitoring server, and reloads a monitored network element list returned by the FE end of the process monitoring server to the interface; receiving a successful notification of real-time process polling of the process monitoring server, triggering to acquire a data request of the real-time process of the host, and reloading a real-time process list of returned goods of an FE (field programmable) end of the process monitoring server to an interface; triggering a polling information setting request of a monitored host, triggering an adding and deleting modification request of a concerned process, triggering a manual updating request of a real-time process, and triggering a modification request of the real-time process;

the process monitoring server receives a setting request of polling information of the monitored host, receives a request of an attention process list of the monitored host, receives a real-time process modification request of the monitored host, successfully modifies the database data through SQL statements, and then returns a processing result to the client through an FE end of the process monitoring server; after the polling information and the concerned process list of the monitored host are set, the process monitoring server realizes the safety authentication connected to the host based on the SSH service, acquires the current running real-time process list of the monitored host based on the SNMP service, filters out invalid processes according to the concerned process list and the filtering condition set by the monitored host, and finally returns the real-time process data to the client through the FE terminal of the process monitoring server.

The client comprises: a service forwarding module, a real-time process configuration module, an attention process management module and a display module,

the service forwarding module is used for forwarding various received RMI requests and SOCKET requests;

the attention process management module is used for forwarding an attention process configuration request of the client to the process monitoring server; receiving an attention process list returned by the process monitoring server, and adding, modifying and deleting the attention process result and forwarding the attention process list to the display module;

the real-time process configuration module is used for triggering the real-time process modification of the client, manually updating a request of the real-time process, and outputting a real-time process list of the monitored host inquired by the process monitoring server to the display module;

and the display module receives the information of the concerned process list, the real-time process list and the monitored host list and displays the information.

The client further comprises:

and the concerned process updating module is used for triggering the concerned process adding, deleting, modifying and checking requests of the monitored host, generating a concerned process list according to the data in the process monitoring server database after the XML file, and displaying the concerned process list at the client.

The process monitoring server comprises a front end and a back end, wherein,

the front end receives a process-concerned updating request of the client, a polling information setting request of the monitored host, a real-time process updating request of the monitored host, a process-concerned process of the monitored host, a real-time process query request and a request for acquiring all monitored host lists of the server and then forwards the requests to the rear end;

the back end receives the setting request of the polling information of the monitored host, receives the request of the concerned process list of the monitored host, receives the real-time process modification request of the monitored host, and returns the processing result to the client after the database data is successfully modified through SQL statements; after the polling information and the concerned process list of the monitored host are set, the BE terminal of the process monitoring server realizes the safety certification connected to the host based on the SSH service, acquires the current running real-time process list of the monitored host based on the SNMP service, filters out invalid processes according to the concerned process list and the filtering condition set by the monitored host, and finally returns the real-time process data to the FE.

The back end is further used for obtaining all monitored windows and Linux host lists of the server database, and a concerned process list and a real-time process list of a specific host.

The back end is further used for polling the process of the monitored host to change from online to offline, updating the stored real-time process list and outputting the information to the front end to inform the client requesting the process monitoring.

It can be seen from the foregoing technical solutions that, in the process monitoring method and the process monitoring system provided in the embodiments of the present invention, a setting request of polling information of a monitored host is received, a request of a process list of interest of the monitored host is received, a real-time process modification request of the monitored host is received, and after a database data is successfully modified through an SQL statement, a processing result is returned to a client through an FE terminal of the monitored host. After the polling information and the concerned process list of the monitored host are set, the process monitoring server realizes the safety authentication connected to the host based on the SSH service, acquires the current running real-time process list of the monitored host based on the SNMP service, filters out invalid processes according to the concerned process list and the filtering condition set by the monitored host, and finally returns the real-time process data to the client through the FE terminal of the process monitoring server.

Therefore, the difference of different operation platforms is shielded through the SNMP, the maintenance and development cost is reduced, the concerned process configuration list is set and stored, the concerned process of the user is obtained according to the SNMP, and then a script or a command is remotely executed through an SSH protocol, so that the process monitoring can be configured and the cross-platform process monitoring is realized.

Drawings

Fig. 1 is a flowchart illustrating a process monitoring method according to an embodiment of the present invention.

Fig. 2 is a schematic view of a process monitoring configuration list interface for user login entry according to an embodiment of the present invention.

Fig. 3 is a schematic flow chart illustrating a process parameter acquisition process by process status polling in a conventional process monitoring method.

Fig. 4 is a schematic structural diagram of a process monitoring system according to an embodiment of the present invention.

Fig. 5 is a schematic flow chart illustrating a process of acquiring process parameter data through SNMP according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.

The existing process monitoring method realizes process monitoring under a Windows operating system by calling a Windows API, or realizes process monitoring under a Linux operating system by calling the Windows API, so that the process monitoring cannot really cross platforms, and a process monitoring list is set by a program developer in advance, so that personalized requirement configuration of a user cannot be met.

The Simple Network Management Protocol (SNMP) is a standard application layer Protocol used for managing Network nodes, such as clients, servers, workstations, routers, switches, etc., in an IP Network, and adopts a separation structure of Protocol and Management information, and receives event report messages from the Network nodes (Network elements) through the SNMP, so that a Network Management System (NMS) can know problems occurring in each Network element in the Network, and the purpose of monitoring and managing different Network nodes (Network elements) in the NMS can be achieved by using the SNMP.

The Secure Shell (SSH) protocol is a protocol established on the basis of an application layer and a transport layer and providing security for a telnet session and other network services, has functions of safe and reliable user identity authentication, online information encryption transmission and the like, and can be applied to different operating platforms, so that the SSH protocol can be used for effectively preventing the problem of information leakage in the process of remote management.

According to the recommendations of the Internet Activity Board (IAB), all network transmission Control Protocol/Internet Protocol (TCP/IP) devices must be manageable. Therefore, in the embodiment of the invention, in consideration of the fact that the process monitoring is realized on the basis of process query, the SNMP can shield the difference between the Windows operating platform and the Linux operating platform, the process configuration file of the monitored client is set and stored to meet the personalized requirement of a user, the process currently running by the monitored client is queried in real time through the SNMP, the state of the monitoring process is reflected in real time through SNMP state polling, the process is started, stopped and restarted through an SSH protocol, and the user is informed through an alarm if the process is determined to have a fault once. The method comprises the steps of acquiring a process concerned by a user through an SNMP, generating corresponding SSH commands according to different monitored clients and different process operations, and then remotely executing scripts or commands through an SSH protocol, so that starting, stopping and restarting of the process are realized, the starting, stopping and restarting of the process can be compatible with a Windows operating platform and a Linux operating platform, and cross-platform (system) process monitoring is realized.

The general flow of the embodiment of the invention is as follows: receiving a request of a user for increasing, deleting, modifying, checking and paying attention to the process, and returning a processed result to the client; receiving a request of a user for setting the polling parameters of the host, and returning a result after setting to the client; receiving a request for updating the network element list, and outputting the network element list set to an online client for display; receiving a request of a user for inquiring a real-time process, realizing authority authentication based on a secure shell SSH protocol service, acquiring host side real-time process data based on a Simple Network Management Protocol (SNMP), filtering out repeated processes and invalid processes based on a specific filtering rule, matching with a concerned process configured by the host, sending a corresponding alarm according to an alarm rule after matching is completed, and finally outputting the processed real-time process to a client for display. By applying the invention, cross-platform process monitoring can be realized, and maintenance and development costs can be reduced.

Fig. 1 is a flowchart illustrating a process monitoring method according to an embodiment of the present invention. Referring to fig. 1, the process monitoring flow includes:

step 101, receiving a request of a client for inquiring a monitored host list, acquiring a managed windows and linux host list, and then returning the monitored host list to the client;

in this step, a process monitoring configuration request of a client is received, information of a monitored host based on an SNMP service and an SSH protocol service is acquired, and a process monitoring configuration list carrying the monitored host is returned to the client.

In this step, if the monitored host does not have the SNMP service and the SSH protocol service installed, it cannot be used as the monitored object, and in practical application, if the monitored host needs to install the service conforming to the SNMP and the SSH protocol service, reference may be made to the related technical documents, and details are not described herein.

In default, the monitored host is not configured with the attention process in advance, the attention process can be added in two ways, and detailed description is subsequently performed. After the process monitoring client is started, if a user needs to set a process concerned by the user and perform process monitoring, an concerned process updating request is sent to the process monitoring server to request for process configuration.

The process monitoring server monitors Windows or Linux monitored hosts found by topology (Topo), provides an interface according to TOPO to distinguish whether the monitored hosts are Windows operating systems or Linux operating systems, and loads the monitored hosts into corresponding default attention process lists according to the types of the operating systems.

Xml section content is as follows:

xml contains a < poll > node, which contains a < process > node, i.e. a process monitoring configuration list, and a < process > node contains a plurality of < search > nodes, one < search > node represents a monitoring process configuration information, and the meaning of each attribute of the < search > node is shown in table 1.

TABLE 1

ID	Network element name
		matchCriteria	Match mode (only support equality)
matchValue	Match value
		generateEvent	Whether to send an alert
severity	Severity level
		de scription	Process description information

In table 1, the network element refers to a monitored host.

The process monitoring configuration list is stored in a database table HostkeyProcess, and the meaning of each field in the table HostkeyProcess is shown in table 2.

TABLE 2

NeID	Ip address of network element
		ProcessName	Name of monitoring process
IsSendNotification	Whether or not to generate an alarm
		Description	Monitoring process description
startScript	Default startup script
		stopScript	Default offClosed script
restartScript	Default restart script

In table 2, an attention process is identified by a combination of a network element name and a process name, and an attention process configuration is performed by using an ip address of the network element and the attention process name as an identifier, for example, whether an alarm is generated, an attention process description, a default start script, a default close script, and a default restart script.

The concerned process list can be set in an Extensible markup language (XML) file format, and the concerned process can be configured by dynamically changing the XML file and storing the concerned process information of different monitored hosts in a database.

One process is composed of a plurality of process parameters, and the process parameters mainly comprise: process Identification (PID), process name, process manufacturer identifier, process execution path, process start parameter, process type, process execution status, CPU usage time, and memory usage size, wherein,

the process types include: unknown type, operating system type, device driver type, and application type;

Table 3 shows the main process parameters and their meaning contained by the process.

TABLE 3

Parameter name	Meaning of parameters
		Process Identification (PID)	The process unique identifier is an integer
Process name	The name of the process is a string of bytes between 0 and 64
		Process producer identifier	Object identifier, developer information uniquely identifying the process
Process execution path	The running path of the process is a character string with the length of 0 to 328 bytes
		Process start parameter	The parameter when the process is started is a character string with the length of 0 to 328 bytes
Type of process	Unknown, operating system, device driver, application program
		Process running state	Running, runnable, non-runnable, illegal
CPU usage time	Time that the process has used the CPU
		Memory usage size	The allocated memory size of the process

When the process monitoring system is initially started, a windows attention process list is loaded according to a monitored host selected by a user, for example, the selected host is a windows. Specifically, the following list information is returned to the client:

acquiring a monitored host list of a process monitoring server;

after the authority authentication of each monitored host is realized based on the SSH protocol service, the current running process of each monitored host is obtained through the SNMP, a real-time process configuration list of the monitored host is generated according to the obtained current running process of each monitored host, a preset process monitoring configuration list and monitoring process configuration information, and the real-time process configuration list is output to the process monitoring client.

Acquiring a focus process list of a monitored host;

the process of acquiring the real-time process of the monitored host through the SNMP is described in detail later.

102, receiving polling information of a monitored host selected by the setting of a client and an attention process list of the host, setting a user name, a password, polling time and whether polling information is required to log in the host, and returning the attention process list of the host to the client;

in this step, a real-time process query request of a user for the selected monitored host is received, and process monitoring is performed according to the concerned process configuration list.

In this step, the process monitoring includes: process inquiry, process polling, process starting, process stopping, process restarting and the like.

The client user requesting process monitoring selects the login user name of the monitored host through input, password verification is carried out, the request for acquiring the real-time process list triggered by the concerned process is set, and the real-time process list is reloaded to the client after the real-time process list sent by the FE end of the process monitoring server is received.

Fig. 2 is a schematic view of a real-time process monitoring configuration list interface for user login entry according to an embodiment of the present invention. Referring to fig. 2, a user logs in through a host user name and a host password, a process monitoring configuration list interface comprises a network element name of each monitored host, a concerned process list of the currently selected monitored host, information on whether to automatically poll a checkbox and data acquisition interval information, in fig. 2, the process monitoring configuration list interface is divided into two columns, the left column is a monitored host list and comprises four monitored hosts for the user to check, of course, as the monitored hosts may be numerous, in the left column, the monitored host can be searched according to the network element name, the right column is a process setting column of the selected monitored host concerned, and above is an automatically polled checkbox and a data acquisition interval information frame, the user can set automatic polling by checking the automatically polled checkbox, and corresponding data acquisition interval information is input in the data acquisition interval information frame, the following process list concerned by the selected monitored network element is set according to the set process information acquisition time, and the process list concerned by the selected monitored network element comprises the following steps: whether to monitor the checkbox, monitor the process name, monitor the process description, start the script configuration, close the script configuration, and restart the script configuration. The user can store the monitoring process into the database for actual monitoring by checking the monitoring check box. And when the user clicks different network element names, correspondingly displaying the network element attention process list of the monitored host.

In the embodiment of the invention, a user selects a process with monitoring process names http, java, ping and vsftpd in a network element with a network element name of 170.20.27.32 as a concerned process to monitor, starts an automatic polling function and sets a data acquisition interval to be 15 seconds. Of course, in practical application, other monitored hosts can be queried.

After the concerned process is set, the client sends a real-time process query request and forwards the request to the process monitoring server, the process monitoring server receives the request and realizes safety authentication through SSH service, real-time process data of the monitored host is obtained through SNMP, a real-time process list matched with the concerned process is matched from the obtained real-time process of the monitored host, and then the real-time process list is sent to the client through FE.

Certainly, in practical application, the user may also update the process of interest of the monitored host and set polling information of the monitored host, where the updating includes: adding a monitoring process, modifying the monitoring process, deleting the monitoring process, reconfiguring network element polling information and the like, wherein,

when a monitoring process needs to be added, a user enters a process monitoring host configuration interface, a monitoring process list is loaded from processmonitor.

The network element monitoring process existing in the database table HostKeyProcess is displayed as a check state on the interface, and the network element monitoring process not existing in the database table HostKeyProcess but existing in ProcessMonitor. xml is displayed as a check canceling state on the interface.

In this way, a user can add a monitoring process to a database table in the following two ways.

(1) Checking an optional monitoring process list;

(2) the monitoring process is manually added to the database table HostKeyProcess.

Modification of the monitoring process:

when a user enters a process monitoring host configuration interface, loading a monitoring process list from a process monitor.

(1) The Description modifies the Description information of the monitoring process;

(2) startScript, modifying the default starting script position of the monitoring process;

(3) stopScript, modifying the default closing script of the monitoring process;

(4) restartScript, modifies the default restart script of the monitoring process.

Reconfiguring the polling information of the network element:

the polling information of the network element can be configured on the host configuration panel, and after the configuration is successful, the polling information is stored in a database table NePoller, and the configurable network element polling information fields are as follows:

(1) the host user name is the user name connected with the host;

(2) the host password is the password connected with the host;

(3) whether to automatically poll or not, namely whether to poll the process parameter data of the host regularly or not, wherein the field parameters are valid or not only when the user name and the password are valid and the host is provided with SSH service;

(4) data acquisition interval, i.e. the period of polling, the data acquisition interval parameter is meaningful only if there is automatic polling tick.

After the setting is successful, NePoller is stored in a database table, and parameters in the NePoller table are shown in the table 4.

TABLE 4

NeID	Ip address of host
		userName	Host user name
password	Host password
		Type	Network element type

IsPoller	Whether to perform process status polling
		Period	Polling interval

The process monitoring is mainly carried out by checking the parameter data of the host process.

If the following two conditions are met, the jump to the host process parameter data panel can be determined through network element switching or clicking:

(1) the user has configured the network element polling information and the user name and password are correct, and the SSH protocol service is installed and started;

(2) the host installs and starts the SNMP service, and if the SNMP service is not installed, the host process parameter data cannot be acquired through the SNMP.

The real-time process query of the host is based on SNMP, the difference of network elements of different types can be shielded through the SNMP, and the process information currently running by the host is obtained by querying the table hrSWRunTable and the table hrSWRunPerfTable in the database. Specifically, the monitoring process of the network element is firstly queried from the database table HostKeyProcess, and then matched with the table hrSWRunTable and the table hrSWRunPerfTable acquired through SNMP, so as to acquire the process information currently running by the host.

The above-mentioned adding the monitoring process, modifying the monitoring process and configuring the polling information of the network element are that the user initiates a request at the process monitoring client, in the practical application, the process monitoring server can also initiate a request, mainly aiming at the newly found network element monitored by the process monitoring server, so as to query the network element list, automatically update the network element list and manually update the network element list in the pre-stored process monitoring configuration list, wherein,

and (3) inquiring the network element list:

and the user enters a process monitoring host configuration interface to load a new network element according to the network element discovered by the Topo background (process monitoring server) and the network element type list in the Processmonitor.

Automatic updating of the network element list:

when the Topo backstage newly finds the Windows or Linux host, all connected clients are informed to update the network element list through Socket communication.

Manual updating of the network element list:

the manual updating of the network element list is mainly applied to ensuring that the network element list at the process monitoring part is consistent with the database when the automatic updating fails. Specifically, the network element data to be updated is inquired from the network element type list in the processmonitor.

103, receiving real-time process information which is inquired by a user through a client and currently operated by a monitored host, safely logging in the host based on an SSH service, acquiring all real-time process lists currently operated by the host based on an SNMP service, filtering invalid process information according to an concerned process list of the host and a preset strategy, and outputting the acquired real-time process parameters to the client for display.

In this step, the acquired real-time process parameters matched with the concerned process are output to the client for display.

In this step, the real-time process is filtered according to a preset attention process: if the process is preset to be notified when the process fails or is suspended accidentally, after the process failure is determined, outputting alarm information to notify the user and outputting the process parameter information of the failure to a client for display; and if the process parameter concerned is preset, outputting the acquired process parameter information concerned to the client for display.

Further, for the case of outputting the alarm information, the method further includes:

and determining that the state of the concerned process which is failed or accidentally suspended is recovered to be normal, and sending alarm clearing information to a user. Therefore, real-time state management of the process is realized through process inquiry and the running state of the polling process, if the key process concerned by the user is stopped due to accidents, alarm information is generated to inform the user, and if the key process concerned by the user is recovered to be normal, alarm clearing information is sent.

Fig. 3 is a schematic flow chart illustrating a process parameter acquisition process by process status polling in a conventional process monitoring method. Referring to fig. 3, a process state polling method is adopted to query a process state, and the process polling is executed at a process monitoring server, and the process includes:

step 301, inquiring whether a polling switch is turned on, if so, executing step 302, otherwise, ending the process;

in this step, when the preset data collection interval time is up, if the Network Management System (NMS) queries that the automatic polling switch is turned on in the stored configuration list of the concerned process set in the client (Network element) requesting process monitoring, the process proceeds to step 302, otherwise, this polling is ended.

In the embodiment of the invention, the network element is a monitored windows or linux host, namely a monitored host, and the NMS comprises a process monitoring server and a process monitoring client.

Step 302, inquiring whether the selected monitored host can be communicated by adopting the configured user name and password through the SSH protocol, if so, executing step 303, otherwise, ending the process;

in this step, the monitored host authenticates the process monitoring server.

Step 303, acquiring a process parameter data set of the monitored network element according to the SNMP;

in this step, a process parameter acquisition request is sent to the monitored network element through the SNMP, and a real-time process parameter data set a of each monitored network element side is acquired and created.

Step 304, filtering the same process in the process parameter data set A to obtain a process parameter data set A1;

in this step, it is determined whether the two processes are the same, and the process name and the process operation path are used for matching. And if the process names and the process running paths of the two process parameters are the same, determining that the two processes are the same or repeated, and reserving one process.

305, filtering a process with a non-operational or illegal process operation state in the process parameter data set A1 to obtain a process parameter data set A2;

in this step, the processes with the running states of non-running and illegal (invalid) are filtered from the process parameter data set a1, and the running states of the processes are obtained as running and executable processes.

Step 306, obtaining an attention process configuration list set by a user from an NMS database to obtain an attention process set B;

in this step, the attention process of each monitored network element is set in an attention process configuration list set by the user, the attention process configuration list is configured by the user on the network element side requesting the monitoring process, the attention process of each network element can be added, deleted and modified, and the attention process is stored in the NMS database after the configuration is successful.

Step 307, if the number of elements in the attention process set B is 0, executing step 316, otherwise, executing step 308;

in this step, a process set C to be updated and a real-time process set D are created. And traversing the attention process set B, if the configured attention process does not exist in the attention process set B, namely the number of elements in the attention process set B is 0, entering a step 316, and otherwise, executing a step 308.

Step 308, for any concerned process in the concerned process set B, querying a process with the same process name as the concerned process in the process parameter data set a2, and placing the queried concerned process in the real-time process set D;

in this step, the real-time process corresponding to the process name of the process of interest in the process-of-interest set B is searched from the process parameter data set a2, and is placed in the real-time process set D.

Step 309, judging whether the number of elements in the real-time process set D is larger than 0, if so, executing step 310, otherwise, executing step 332;

in this step, if the number of elements in the real-time process set D is greater than 0, it indicates that there is an attention process matching the attention process set B in the process parameter data set a2 acquired through SNMP.

Step 310, inquiring whether the NMS generates the offline alarm information of the network element attention process, if so, executing step 311, otherwise, executing step 315;

in this step, the process is not online, including process non-runnability and process illegal.

Step 311, the NMS sends the online alarm clearing information of the network element attention process;

in this step, if the NMS has generated the information that the network element attention process does not generate the online alarm, that is, the network element attention process is abnormal and notified to the user before the current process status polling, the online alarm clearing information of the network element attention process is sent, and whether the alarm attribute generated by the network element attention process is false is modified. The alarm attribute is identified by a flag bit to indicate whether the network element attention process has generated an alarm, when the flag bit is true, the network element attention process is not on-line, and when the flag bit is false, the network element attention process is on-line.

Step 332, inquiring whether the NMS has generated an offline alarm of the network element attention process, if not, executing step 315, otherwise, executing step 313;

in this step, if the number of elements in the real-time process set D is 0, it indicates that there is no network element attention process matching the attention process set B in the process parameter data set a2 acquired through the SNMP, and the network element attention process is offline.

Step 313, the NMS sends the online alarm clearing information of the network element attention process;

step 314, adding the network element attention process into the attention process set C to be updated;

in this step, when the state of the process concerned by the network element changes from offline to online or from online to offline, it is necessary to modify whether the alarm field has been sent in the database, and thus, the NMS database needs to be updated.

In steps 313 to 314, if the NMS has not generated the no-online alarm of the network element attention process before, the NMS sends the no-online alarm of the network element attention process, and modifies whether the alarm attribute of the network element attention process is generated as True.

Step 315, sending the attention process off-line warning information, adding all processes in the attention process set C to be updated into the real-time process set D, and returning to the execution step 307;

in the step, if the number of elements in the real-time process set D is 0, adding a network element attention process to an attention process set C to be updated, and adding the attention process set C to be updated to the real-time process set D; otherwise, directly adding the network element attention process into the real-time process set D.

Step 316, updating the elements in the attention process set C to be updated into an NMS database;

step 317, the elements in the real-time process set D are updated into the NMS database.

At this point, the process polling process is finished, the query of the process state is equivalent to a simple process polling, and the process is similar to the process polling process and is not described again.

Fig. 4 is a schematic structural diagram of a process monitoring system according to an embodiment of the present invention. Referring to fig. 4, the system includes: a Client (Client) and a process monitoring Server (Server), the process monitoring system is based on a C/S structure, wherein,

the client receives the network element list updating notification of the process monitoring server, triggers a network element list acquiring request of the process monitoring server, and reloads a monitored network element list returned by the FE end of the process monitoring server to the interface; receiving a successful notification of real-time process polling of the process monitoring server, triggering to acquire a data request of the real-time process of the host, and reloading a real-time process list of returned goods of an FE (field programmable) end of the process monitoring server to an interface; in addition, the system is also responsible for triggering a polling information setting request of a monitored host, triggering an adding and deleting modification request of a concerned process, triggering a manual updating request of a real-time process and triggering a modification request of the real-time process.

And the process monitoring server receives a setting request of polling information of the monitored host, receives a request of a process list concerned by the monitored host, receives a real-time process modification request of the monitored host, successfully modifies the database data through SQL statements, and then returns a processing result to the client through an FE end of the process monitoring server. After the polling information and the concerned process list of the monitored host are set, the process monitoring server realizes the safety authentication connected to the host based on the SSH service, acquires the current running real-time process list of the monitored host based on the SNMP service, filters out invalid processes according to the concerned process list and the filtering condition set by the monitored host, and finally returns the real-time process data to the client through the FE terminal of the process monitoring server.

Wherein,

the client comprises: a service forwarding module, a real-time process configuration module, an attention process management module and a presentation module (not shown in the figure),

Preferably, the client further comprises:

In this embodiment, the attention process configuration list is generated by performing check in a check frame provided in the process monitoring configuration list, relative to the process monitoring configuration list.

The updating comprises the following steps: adding a monitoring process, deleting the monitoring process, modifying the monitoring process and reconfiguring the polling information of the network element.

The processes not selected in the attention process configuration list are selected, so that the increase of the monitoring processes can be realized, and the monitoring processes can be increased in the attention process configuration list through manual release. If the process selected in the concerned process configuration list is cancelled, the monitoring process can be deleted.

And selecting the monitoring process selected from the concerned process configuration list, and modifying the parameter data of the monitoring process, so that the modification of the monitoring process can be completed. Similarly, the reconfiguration of the network element polling information can be realized by modifying the host user name, the host password, whether to automatically poll and the data acquisition interval information in the concerned process configuration list.

As described above, when the process monitoring server monitors that a new network element is found or an old network element is deleted, the process monitoring configuration list needs to be updated, and the client requesting process monitoring is notified of the information, so that the process-concerned updating module of the client receives the process monitoring configuration update list output by the process monitoring server and updates the process-concerned configuration list, for example, for a newly added monitored host, the corresponding process information of the monitored host is added to the process-concerned configuration list, so that a user provides a selection when updating the process-concerned configuration list next time.

The process monitoring server comprises a Front End (FE) and a Back End (BE) (not shown in the figure), wherein,

the FE is used for receiving a process-concerned updating request of the client, a polling information setting request of the monitored host, a real-time process updating request of the monitored host, a process-concerned process of the monitored host, a real-time process query request and a request for acquiring all monitored host lists of the server and forwarding the requests to the back end;

BE, receive the setting request of the polling information of the monitored host computer, receive the request of the concerned process list of the monitored host computer, receive the real-time process modification request of the monitored host computer, after modifying the database data successfully through SQL statement, then return the processing result to the customer end; after the polling information and the concerned process list of the monitored host are set, the BE terminal of the process monitoring server realizes the safety certification connected to the host based on the SSH service, acquires the current running real-time process list of the monitored host based on the SNMP service, filters out invalid processes according to the concerned process list and the filtering condition set by the monitored host, and finally returns the real-time process data to the FE.

Further, the BE is used for acquiring all monitored windows and Linux host lists of the server database, and an attention process list and a real-time process list of a specific host.

Preferably, the BE is also operative to poll processes to the monitored host for changes from online to offline, update the stored real-time process list, and output this information to the front-end to notify the client requesting process monitoring.

The operations primarily responsible for FE and BE are described below.

The FE is mainly responsible for monitoring updating (e.g., adding, deleting, modifying, etc. of a network element), monitoring of polling information change of the network element, Remote Method Invocation (RMI) request of the client, and sending a Socket message (Socket) to the client after automatic polling is completed, that is, mainly responsible for receiving the request of the client, forwarding the request to the BE, receiving information output by the BE, and forwarding the information to the client.

The BE is responsible for main operations including querying a network management Database (DB, Database), i.e., a process monitoring configuration list, executing an SSH command through an SSH protocol, and acquiring process parameter data of a network element side through an SNMP.

The Client is mainly responsible for interaction with the user and display of process parameter data.

The relationship among Client, FE and BE is as follows:

the Client and the FE register the same Socket to perform Socket communication.

The Client communicates with the FE through RMI, and the FE communicates with the BE through RMI.

The Client is responsible for receiving a request of a user, calling the RMI interface of the FE to modify the configuration of the network element polling information (such as whether polling is carried out, a host user name, a host password, a polling interval and the like), paying attention to a process, a real-time process starting script, a restarting script and a closing script, and also calling the RMI interface of the FE, and then forwarding the RMI requests to the BE terminal by the FE, wherein the FE terminal does not carry out actual processing, and realizes the starting, stopping and restarting of the process and the manual updating of the process state by the BE terminal.

In addition, the Client and the FE also maintain Socket communication, and the Client mainly receives the following events from the FE:

I. when NMS finds a new network element, deletes a network element or modifies network element attributes, the Client receives a message for updating the network element list sent by the FE through the Socket, and then the Client reloads the network element list from the NMS database.

II. When the process of the network element in the NMS is automatically polled, the Client receives an update real-time process interface message sent by the FE through the Socket, and then the Client reloads the real-time process list from the NMS database.

And III, after the client successfully updates the network element polling information, sending an update polling task message to the FE, and then loading a latest network element polling task list from the NMS database by the FE of the server.

Fig. 5 is a schematic flow chart illustrating a process of acquiring process parameter data through SNMP according to an embodiment of the present invention. The SNMP obtains process parameter data, that is, the NMS obtains data in a Host-resource.mib table hrSWRunTable and a table hrSWRunPerfTable on the monitored client through the SNMP, where the table hrSWRunTable and the table hrSWRunPerfTable record process information of the currently monitored client in detail, referring to fig. 5, and the process includes:

step 501, a client selects an un-started process record and clicks to start an un-started process;

in this step, the client selects the un-started process through the attention process configuration list.

Step 502, obtaining the name of the un-started process;

step 503, constructing a starting command character string to be executed according to the network element type, the process name and the starting mode selected by the user;

in this step, the client can only perform a start operation on the process that is not started, and a start command can be constructed in three ways:

(1) the script starting mode is as follows: namely, directly executing the starting by using a starting script configured by a user;

(2) service starting mode: the method comprises the steps of starting a net start service name of a windows operating system and starting a service name start of a Linux operating system;

(3) user-defined starting mode: the user himself inputs any command to start.

Similarly, for the stop process, only the stop operation can be performed on the started process, and the stop process can construct a command in four ways:

(1) script stop mode: namely, directly executing the stop by using a stop script configured by a user;

(2) service stop mode: the method comprises the steps that a net stop service name of a windows operating system stops and a service name stop of a Linux operating system stops;

(3) forced stopping mode: the method comprises the steps of stopping the ID of a taskill/F/PID process of a windows operating system and stopping the kill-9PID of a Linux operating system;

(4) user-defined stopping mode: the user himself inputs an arbitrary command to stop.

For the restart process, only the start operation can be performed on the un-started process, and the command can be constructed in three ways:

(1) the script restart mode comprises the following steps: namely, directly executing the restart by using a restart script configured by a user;

(2) service restart mode: the method comprises the steps of including a net stop service name of a windows operating system, restarting a net start service name and resetting a service name of a Linux operating system;

(3) user-defined restarting mode: the user himself inputs any command to restart.

Step 504, an interface of the front end is called to forward the request for executing the starting process command to the front end;

505, the front end receives a request for starting a process of the monitored host and forwards the request to the back end;

step 506, the back end calls an SSH protocol to execute an SSH command on the monitored host;

step 507, the monitored host executes the SSH command and returns the execution result (successful start-up process) to the back end in the form of a byte stream;

step 508, the back end analyzes the returned result, and obtains the latest data of the monitored network element side through SNMP;

step 509, the back end stores the processed real-time process into the SNM database;

step 510, the back end returns the processed real-time process result to the front end;

step 511, the front end returns the processed real-time process result to the client;

in step 512, the client displays the result of the process start and the latest implementation process list data.

As can be seen from the above description, the process monitoring method, the process monitoring device and the process monitoring system according to the embodiments of the present invention shield the difference between the Windows operating platform and the Linux operating platform through the SNMP, and meet the requirement of the user for personalization by setting the process configuration file of the monitored client. Therefore, the SNMP can be used for inquiring the currently running process of the monitored client in real time, and the SSH protocol is used for starting, stopping and restarting the process, so that when the process fails, a user can be informed through an alarm. Has the following beneficial effects:

firstly, reflecting the state of a process in real time, so that the change of the process state can be reflected on a client of a user in real time;

secondly, the process monitoring can be configured by modifying the configuration file;

thirdly, the difference of the operating systems in the process monitoring is shielded, and the real cross-platform is realized;

and fourthly, the development cost can be reduced, the maintenance workload can be reduced, and the working efficiency of process monitoring is improved.

The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A process monitoring method, comprising:

2. The method of claim 1, wherein the list of monitored hosts comprises: the IP address information of the network element, the name information of the monitoring process, whether alarm information is generated or not, the description information of the monitoring process, the default startup script information, the default shutdown script information and the default restart script information.

3. The method of claim 2, wherein the monitored host list is arranged in one or a combination of a database format and an extensible markup language format.

4. The method of claim 1, wherein the process is comprised of a plurality of process parameters, the process parameters comprising: the method comprises the following steps of process identification, process name, process manufacturer identifier, process running path, process starting parameter, process type, process running state, CPU using time and memory using size.

5. The method of claim 4, wherein the process type comprises: unknown type, operating system type, device driver type, and application type.

6. The method of claim 4, wherein the running state of the process comprises: operational, non-operational, and illegal.

7. The method of claim 1, wherein the process monitoring comprises: process inquiry, process status polling, process start, process stop, and process restart.

8. The method of claim 7, wherein the process launch comprises:

9. The method of claim 8, further comprising:

10. The method of claim 1, further comprising:

and updating the set attention process list.

11. The method of claim 10, wherein the updating comprises: adding a monitoring process, modifying the monitoring process and reconfiguring network element polling information.

12. The method of claim 1, wherein outputting the obtained real-time process parameters to a client presentation comprises:

13. The method of claim 12, further comprising:

14. The method as claimed in claim 1, wherein said filtering out invalid process information according to the concerned process list of the host and a preset policy, and then outputting the obtained real-time process parameters to the client for presentation comprises:

15. The method of claim 1, wherein the login information comprises: username, password, polling time, and whether to poll the information.

16. A process monitoring system, based on a C/S architecture, comprising: a client and a process monitoring server, wherein,

the client receives the network element list updating notification of the process monitoring server, triggers a network element list acquiring request of the process monitoring server, and reloads a monitored network element list returned by the FE end of the process monitoring server to the interface; receiving a successful notification of real-time process polling of the process monitoring server, triggering to acquire a data request of the real-time process of the host, and reloading a real-time process list returned by an FE (field effect) end of the process monitoring server to an interface; triggering a polling information setting request of a monitored host, triggering an adding and deleting modification request of a concerned process, triggering a manual updating request of a real-time process, and triggering a modification request of the real-time process;

17. The system of claim 16, wherein the client comprises: a service forwarding module, a real-time process configuration module, an attention process management module and a display module,

18. The system of claim 17, wherein the client further comprises:

19. The system of claim 16, wherein the process monitoring service includes a front end and a back end, wherein,

20. The system of claim 19, wherein the backend is further configured to obtain a list of all monitored windows and Linux hosts of the server database, and a list of processes of interest and a list of real-time processes for a particular host.

21. The system of claim 20, wherein the back-end is further configured to poll processes to the monitored host for changes from online to offline, update the stored real-time process list, and output this information to the front-end to notify the client requesting process monitoring.