CN107733702A - The method and apparatus that operational state of mainframe is managed in group system - Google Patents

Abstract

The invention discloses the method and apparatus that operational state of mainframe is managed in a kind of group system.Methods described includes：To the segregate management duration of the main frame after main frame can not communicate with other main frames in acquisition cluster file system CFS；The timing detection main frame can not be starting point with the time after other main-machine communications, broken down since the main frame；Before the duration of record is not up to the management duration, warning information is sent to cloud platform.

Description

Method and device for managing host running state in cluster system

technical field

The invention relates to the field of information processing, in particular to a method and device for managing the running state of a host computer in a cluster system.

Background technique

CFS (CFS Cluster File System, cluster file system) is a system in which multiple host nodes mount the same file system at the same time. The cluster file system enables all nodes running in the cluster to access storage devices concurrently through standard file system interfaces. This makes it easy to manage application systems that span the entire cluster.

If a host node cannot read and write storage within a certain period of time, the host will be isolated by the file system, that is, fence. In cloud computing products, if the host is fenced without the user's knowledge, it will inevitably lead to the interruption of the user's business processing and affect the normal processing of the business.

Contents of the invention

In order to solve the above technical problems, the present invention provides a method and device for managing the running state of a host in a cluster system, which can reduce the possibility that the host will be isolated by a file system.

In order to achieve the purpose of the present invention, the present invention provides a method for managing the running state of a host in a cluster system, including:

Obtain the management time from when the host in the CFS cluster file system cannot communicate with other hosts to when the host is isolated;

After detecting that the host cannot communicate with other hosts, start timing from the time when the host fails;

Before the recorded duration does not reach the management duration, an alarm message is sent to the cloud platform.

Wherein, the method also has the following characteristics: whether the host can communicate with other hosts is obtained through the following methods, including:

Detect the heartbeat response data sent by other hosts from the preset monitoring port, and get the detection result;

Comparing the detection result with the preset management strategy of the heartbeat response data to obtain the comparison result;

Based on the comparison result, it is determined whether the host can communicate with other hosts.

Wherein, the method also has the following characteristics: the heartbeat response data sent by other hosts is detected from the preset monitoring port, and the detection result is obtained, including:

Receive storage heartbeat packets sent by other hosts;

Reading data from a preset field in the preset stored heartbeat data packet;

Parse the content of the data to determine whether the data is heartbeat response data to the host.

Wherein, the method also has the following characteristics: the comparison of the detection result with the preset management strategy of the heartbeat response data to obtain the comparison result includes:

Judging whether the heartbeat response data sent by other hosts to the host is received;

If no heartbeat response data sent by other hosts to the host is received, then it is determined that the host cannot communicate with other hosts;

If the heartbeat response data sent by other hosts to the host is received, count the total number of hosts that send heartbeat response data to the host; judge whether the total is greater than the preset total threshold; if the total is less than the total threshold, it is determined that the host cannot communicate with other hosts.

Wherein, the method also has the following characteristics: before the recorded duration does not reach the management duration, after sending an alarm message to the cloud platform, the method also includes:

Detecting whether the host receives an operation request for fault handling;

If the host does not receive the fault processing operation request, it sends an alarm message to the cloud platform again.

A device for managing the running state of a host in a cluster system, comprising:

The obtaining module is used to obtain the management time from when the host in the CFS cluster file system cannot communicate with other hosts to when the host is isolated;

Timing module, used to start counting from the time when the host fails when it detects that the host cannot communicate with other hosts;

The alarm module is configured to send an alarm message to the cloud platform before the recorded time length reaches the management time length.

Wherein, the device also has the following characteristics: the device also includes:

The first detection module is used to detect heartbeat response data sent by other hosts from a preset monitoring port to obtain a detection result;

A comparison module, configured to compare the detection result with a preset management strategy for heartbeat response data to obtain a comparison result;

A determining module, configured to determine whether the host can communicate with other hosts according to the comparison result.

Wherein, the device also has the following characteristics: the first detection module includes:

A receiving unit, configured to receive storage heartbeat packets sent by other hosts;

A reading unit, configured to read data from a preset field in the preset stored heartbeat data packet;

The determining unit is configured to analyze the content of the data, and determine whether the data is heartbeat response data to the host.

Wherein, the device also has the following characteristics: the comparison module includes:

A first judging unit, configured to judge whether heartbeat response data sent by other hosts to the host is received;

The first processing unit is configured to determine that the host cannot communicate with other hosts if no heartbeat response data sent by the other host to the host is received;

A statistical unit, configured to count the total number of hosts sending heartbeat response data to the host if heartbeat response data sent by other hosts to the host is received;

A second judging unit, configured to judge whether the total number is greater than a preset total threshold;

A second processing unit, configured to determine that the host cannot communicate with other hosts if the total number is less than the total number threshold.

Wherein, the device also has the following characteristics: the device also includes:

The second detection module is used to detect whether the host receives an operation request for fault handling after sending an alarm message to the cloud platform before the recorded time length reaches the management time length;

The alarm module is further configured to send an alarm message to the cloud platform again if the host does not receive the fault handling operation request.

In the embodiment provided by the present invention, after obtaining the management duration for which the host is isolated, timing is performed on the failed host, and before the management duration is reached, an alarm message is sent to the cloud platform, so that when the host is about to be isolated by the file system, The purpose of timely alarming is to reduce the possibility of the host being isolated. Even if the host is finally isolated by the file system, the user can still understand the reason for the host fence through the notification information and improve user experience.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention may be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Description of drawings

The accompanying drawings are used to provide a further understanding of the technical solution of the present invention, and constitute a part of the description, and are used together with the embodiments of the application to explain the technical solution of the present invention, and do not constitute a limitation to the technical solution of the present invention.

Fig. 1 is a flow chart of the method for managing the running state of the host computer in the cluster system provided by the present invention;

FIG. 2 is a structural diagram of a device for managing the running state of a host in the cluster system provided by the present invention.

detailed description

In order to make the object, technical solution and advantages of the present invention more clear, the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined arbitrarily with each other.

The steps shown in the flowcharts of the figures may be performed in a computer system, such as a set of computer-executable instructions. Also, although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in an order different from that shown or described herein.

FIG. 1 is a flowchart of a method for managing the running state of a host in a cluster system provided by the present invention. The methods shown in Figure 1 include:

Step 101, obtaining the management time from when the host in the OCFS2 cluster system cannot communicate with other hosts to when the host is isolated;

Specifically, in the cluster file system, after a host cannot communicate with other hosts, it takes a period of time before it is isolated. Therefore, it is necessary to know the time in advance so as to perform management before the time is reached.

Step 102, after detecting that the host cannot communicate with other hosts, start timing from the time when the host fails;

Specifically, after knowing the management duration, once the host cannot communicate with other hosts, time the failure time of the failed host;

Step 103, before the recorded duration does not reach the management duration, send an alarm message to the cloud platform.

Specifically, when the host is about to be isolated by the file system, an alarm message is sent in time, so that it is convenient for the user to repair and reduce the possibility of being isolated.

In the embodiment of the method provided by the present invention, after obtaining the management duration for which the host is isolated, timing is performed on the failed host, and before the management duration is reached, an alarm message is sent to the cloud platform, so that when the host is about to be isolated by the file system , the purpose of timely alarming, thereby reducing the possibility of the host being isolated, even if the host is finally isolated by the file system, the user can also understand the reason for the host fence through the notification information, and improve user experience.

The method provided by the present invention is further described below:

In the OCFS2 system, the hosts use the storage heartbeat process to detect whether the connections between themselves and other nodes and storage devices are normal. The storage heartbeat process reads the storage heartbeat of other nodes every 2s, and writes the storage heartbeat of its own node at the same time. If a host node cannot read and write storage within a certain period of time, it will be isolated by the file system. Therefore, the detection of whether the host can communicate with other hosts provided by the present invention includes the following methods:

Detect the heartbeat response data sent by other hosts from the preset monitoring port, and get the detection result;

Comparing the detection result with the preset management strategy of the heartbeat response data to obtain the comparison result;

Based on the comparison result, it is determined whether the host can communicate with other hosts.

With the help of the existing transmission mechanism of the cluster system, the communication detection is completed, the existing protocol is used, and the detection purpose is completed without modification of the protocol, which is simple and convenient.

In the above detection method, the detection of heartbeat response data sent by other hosts from the preset monitoring port to obtain the detection result can also include:

Receive storage heartbeat packets sent by other hosts;

Reading data from a preset field in the preset stored heartbeat data packet;

Parse the content of the data to determine whether the data is heartbeat response data to the host.

Specifically, usually the hosts in the cluster system will negotiate to store specific data in a fixed field of the heartbeat data packet to identify the heartbeat response data. If the content of the data is the preset content in the data analysis, then It means that the stored heartbeat data packet has heartbeat response data to the host; if the content of the data is empty or not preset, it means that the stored heartbeat data packet has no heartbeat response data to the host.

In the above detection method, the comparison of the detection result with the preset management strategy of the heartbeat response data to obtain the comparison result includes:

Judging whether the heartbeat response data sent by other hosts to the host is received;

If no heartbeat response data sent by other hosts to the host is received, then it is determined that the host cannot communicate with other hosts;

If the heartbeat response data sent by other hosts to the host is received, count the total number of hosts that send heartbeat response data to the host; judge whether the total is greater than the preset total threshold; if the total is less than the total threshold, it is determined that the host cannot communicate with other hosts.

It should be noted that the threshold of the total number is generally greater than or equal to half of the total number in the cluster.

Of course, before the recorded duration does not reach the management duration, after sending an alarm message to the cloud platform, it is detected whether the host receives an operation request for troubleshooting; if the host does not receive the operation request for troubleshooting, Then send an alarm message to the cloud platform again.

Specifically, after the alarm is issued, if no response operation is received after a period of time, the alarm is issued again before the management time limit is reached, so as to remind the user to operate as soon as possible.

Further description is given below with the application examples provided by the present invention:

In order to reduce the possibility of the host being isolated, the present invention proposes a method for sharing storage heartbeat monitoring between hosts facing the OCFS2 cluster file system. The host obtains the heartbeat packet of the OCFS2 cluster file system by capturing packets, and then matches the fixed location information of the heartbeat packet , that is, the heartbeat response signal between the hosts in the OCFS2 cluster, to determine whether the own node is running normally. When the host can only receive the response signals from a few other hosts in the cluster or cannot receive the response signals from other host nodes in the cluster within a certain period of time, the host will determine that it is in an abnormal operating state, and before its own node fence, go up The layer platform sends notifications and sends warning messages to users through the user interface.

The method provided by this application example mainly includes the following steps:

Step 1. The host monitors its own fixed port 7777, and obtains the heartbeat packet of the host in the cluster by capturing packets;

Step 2. The host judges whether the heartbeat packet is the response of other hosts to itself by matching the fixed position of the captured heartbeat packet. If so, continue to capture packets; otherwise, continue to monitor;

Step 3. After a certain period of time, if the host has not received the heartbeat response information from other hosts in the cluster, or the number of hosts that have received the response is less than half of the hosts in the cluster, the host will send an alarm message to the platform;

Step 4. The platform monitors the information sent by the bottom layer. If there is no information sent from the bottom layer, it will continue to monitor; if there is a message sent, the platform will feed back to the UI after processing, and remind the user of the status of the bottom layer storage through the UI.

The method provided by the application example of the present invention can judge whether a single host in the OCFS2 cluster file system is working normally in a relatively short period of time, and if the host is not working normally, it will notify the user in advance by sending a warning message. Even if the host is finally fenced, the user will know the reason for the host fence through the alarm information.

FIG. 2 is a structural diagram of a device for managing the running state of a host in the cluster system provided by the present invention. The device shown in Figure 2 includes:

The obtaining module 201 is used to obtain the management time from when the host in the CFS cluster file system cannot communicate with other hosts to when the host is isolated;

The timing module 202 is used to start counting from the time when the host fails when it detects that the host cannot communicate with other hosts;

The alarm module 203 is configured to send an alarm message to the cloud platform before the recorded duration does not reach the management duration.

In an embodiment of the device provided by the present invention, the device further includes:

The first detection module is used to detect heartbeat response data sent by other hosts from a preset monitoring port to obtain a detection result;

A comparison module, configured to compare the detection result with a preset management strategy for heartbeat response data to obtain a comparison result;

A determining module, configured to determine whether the host can communicate with other hosts according to the comparison result.

In a device embodiment provided by the present invention, the first detection module includes:

A receiving unit, configured to receive storage heartbeat packets sent by other hosts;

A reading unit, configured to read data from a preset field in the preset stored heartbeat data packet;

The determining unit is configured to analyze the content of the data, and determine whether the data is heartbeat response data to the host.

In an apparatus embodiment provided by the present invention, the comparison module includes:

A first judging unit, configured to judge whether heartbeat response data sent by other hosts to the host is received;

The first processing unit is connected to the first disconnection unit, and is used to determine that the host cannot communicate with other hosts if no heartbeat response data sent by other hosts to the host is received;

A statistics unit, connected to the first judging unit, used to count the total number of hosts sending heartbeat response data to the host if heartbeat response data sent by other hosts to the host is received;

A second judging unit, connected to the statistical unit, for judging whether the total number is greater than a preset total threshold;

A second processing unit, connected to the second breaking unit, configured to determine that the host cannot communicate with other hosts if the total is less than the threshold of the total.

In an embodiment of the device provided by the present invention, the device further includes:

The second detection module is used to detect whether the host receives an operation request for fault handling after sending an alarm message to the cloud platform before the recorded time length reaches the management time length;

The alarm module is further configured to send an alarm message to the cloud platform again if the host does not receive the fault handling operation request.

Those of ordinary skill in the art can understand that all or part of the steps of the above-mentioned embodiments can be implemented using a computer program flow, the computer program can be stored in a computer-readable storage medium, and the computer program can be run on a corresponding hardware platform (such as system, device, device, device, etc.), and when executed, includes one or a combination of the steps of the method embodiment.

Optionally, all or part of the steps in the above embodiments can also be implemented using integrated circuits, and these steps can be fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module accomplish. As such, the present invention is not limited to any specific combination of hardware and software.

The devices/functional modules/functional units in the above embodiments can be realized by general-purpose computing devices, and they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices.

When each device/functional module/functional unit in the above-mentioned embodiments is realized in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, and the like.

The above is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Anyone skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present invention. Should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope described in the claims.

Claims (10)

1. A method for managing the running state of a host computer in a cluster system, characterized in that, comprising: Obtain the management time from when the host in the CFS cluster file system cannot communicate with other hosts to when the host is isolated; After detecting that the host cannot communicate with other hosts, start timing from the time when the host fails; Before the recorded duration does not reach the management duration, an alarm message is sent to the cloud platform. 2. The method according to claim 1, wherein whether the host can communicate with other hosts is obtained through the following methods, including: Detect the heartbeat response data sent by other hosts from the preset monitoring port, and get the detection result; Comparing the detection result with the preset management strategy of the heartbeat response data to obtain the comparison result; Based on the comparison result, it is determined whether the host can communicate with other hosts. 3. The method according to claim 2, wherein the detecting the heartbeat response data sent by other hosts from the preset monitoring port to obtain the detection result includes: Receive storage heartbeat packets sent by other hosts; Reading data from a preset field in the preset stored heartbeat data packet; Parse the content of the data to determine whether the data is heartbeat response data to the host. 4. The method according to claim 2 or 3, wherein the comparison of the detection result with the preset management strategy of the heartbeat response data to obtain the comparison result includes: Judging whether the heartbeat response data sent by other hosts to the host is received; If no heartbeat response data sent by other hosts to the host is received, then it is determined that the host cannot communicate with other hosts; If the heartbeat response data sent by other hosts to the host is received, count the total number of hosts that send heartbeat response data to the host; judge whether the total is greater than the preset total threshold; if the total is less than the total threshold, it is determined that the host cannot communicate with other hosts. 5. The method according to claim 1, characterized in that, before the recorded duration does not reach the management duration, after sending an alarm message to the cloud platform, the method further comprises: Detecting whether the host receives an operation request for fault handling; If the host does not receive the fault processing operation request, it sends an alarm message to the cloud platform again. 6. A device for managing the running state of a host computer in a cluster system, characterized in that it comprises: The obtaining module is used to obtain the management time from when the host in the CFS cluster file system cannot communicate with other hosts to when the host is isolated; Timing module, used to start counting from the time when the host fails when it detects that the host cannot communicate with other hosts; The alarm module is configured to send an alarm message to the cloud platform before the recorded time length reaches the management time length. 7. The device according to claim 6, further comprising: The first detection module is used to detect heartbeat response data sent by other hosts from a preset monitoring port to obtain a detection result; A comparison module, configured to compare the detection result with a preset management strategy for heartbeat response data to obtain a comparison result; A determining module, configured to determine whether the host can communicate with other hosts according to the comparison result. 8. The device according to claim 7, wherein the first detection module comprises: A receiving unit, configured to receive storage heartbeat packets sent by other hosts; A reading unit, configured to read data from a preset field in the preset stored heartbeat data packet; The determining unit is configured to analyze the content of the data, and determine whether the data is heartbeat response data to the host. 9. The device according to claim 7 or 8, wherein the comparison module comprises: A first judging unit, configured to judge whether heartbeat response data sent by other hosts to the host is received; The first processing unit is configured to determine that the host cannot communicate with other hosts if no heartbeat response data sent by the other host to the host is received; A statistical unit, configured to count the total number of hosts sending heartbeat response data to the host if heartbeat response data sent by other hosts to the host is received; A second judging unit, configured to judge whether the total number is greater than a preset total threshold; A second processing unit, configured to determine that the host cannot communicate with other hosts if the total number is less than the total number threshold. 10. The device according to claim 6, further comprising: The second detection module is used to detect whether the host receives an operation request for fault handling after sending an alarm message to the cloud platform before the recorded time length reaches the management time length; The alarm module is further configured to send an alarm message to the cloud platform again if the host does not receive the fault handling operation request.