CN108984320A - A kind of anti-fissure method and device of message queue cluster - Google Patents
A kind of anti-fissure method and device of message queue cluster Download PDFInfo
- Publication number
- CN108984320A CN108984320A CN201810682010.6A CN201810682010A CN108984320A CN 108984320 A CN108984320 A CN 108984320A CN 201810682010 A CN201810682010 A CN 201810682010A CN 108984320 A CN108984320 A CN 108984320A
- Authority
- CN
- China
- Prior art keywords
- message queue
- alarm
- node
- fissure
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/54—Interprogram communication
- G06F9/546—Message passing systems or structures, e.g. queues
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/32—Monitoring with visual or acoustical indication of the functioning of the machine
- G06F11/324—Display of status information
- G06F11/327—Alarm or error message display
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3447—Performance evaluation by modeling
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Computer Hardware Design (AREA)
- Computer And Data Communications (AREA)
Abstract
The embodiment of the present application discloses a kind of anti-fissure method and device of message queue cluster, the method includes the status informations according to node when message queue exception, building alerts model, includes N number of element for characterizing the status information of the node in the alarm model;Time series analysis carried out to N number of element at the node current time respectively, obtains the corresponding element prediction value of N number of element described in subsequent time;If the corresponding element prediction value of the first element is greater than or equal to the corresponding alarm threshold of first element, alerted, first element is the either element in N number of element.In the embodiment of the present application, to modeling analysis the reason of causing message queue fissure, obtain the threshold value for causing each element of message queue, and using these threshold values as alarm conditions, in addition each alarm element is predicted, and alarm is provided according to predicted value, it effectively prevent the appearance of message queue fissure.
Description
Technical field
This application involves field of computer technology, more particularly to a kind of anti-fissure method and device of message queue cluster.
Background technique
OpenStack is a series of combination of Open-Source Tools (or open source projects), mainly carrys out structure using pond virtual resource
Build and manage private clound and public cloud.It can provide to store including calculating, network, storage, authentication, mirror image and object and take
Core cloud computing service including business, user can come together to create cloud framework that is unique, can disposing by being bundled.
HA (High Available, high availability cluster) is the effective solution for guaranteeing business continuance.Usual feelings
Under condition, the High Availabitity between message queue cluster is carried out by mirror queue (mirror image pattern), the number of mirror image pattern queue
According on all mirror images portion to all nodes.Any one node failure in this way, will not influence the use of entire cluster.In reality
On now, there is a set of election algorithm inside mirror queue, a master (host node) and several slaver can be selected
(from node).Master and slaver checks whether that connection disconnects by constantly sending heartbeat each other.In HA system, when
When contacting " heartbeat " disconnection of 2 nodes, it was integral originally, the HA system of coordination, just division becomes 2 independences
Individual.Due to mutually losing connection, which is all thought as that other side is out of order, and the HA software on two nodes is striven
It robs " shared resource ", striven " application service ", will lead to that shared resource is carved up, 2 sides " service " all cannot get up in this way;Or 2
Side " service " is all got up, but reads while write " shared storage ", leads to corrupted data, the online log of common such as database poll
Error.
For the fissure countermeasure between message queue cluster, following scheme is generallyd use in the prior art: first, it adds superfluous
Remaining heartbeat, for example, reducing " fissure " occurrence probability to the greatest extent using diplocardia wire jumper;Second, arbitration mechanism is set, such as sets
It sets with reference to IP (such as gateway IP), when heartbeat is fully disconnected, the respective ping of 2 nodes once refers to IP, and general rule does not show
Breakpoint just goes out in local terminal, and not only " heartbeat ", the local terminal network link of also simultaneous externally " service " are disconnected, even if starting (or continuation)
Application service is not also used, that just actively abandons competing, allow can the logical one end with reference to IP ping gone to service.More insurance one
A bit, the obstructed side with reference to IP of ping self is restarted, and those of is possible to also take up shared resource with thorough release.
But in above scheme, fissure can not be avoided by addition redundancy heartbeat or setting arbitration mechanism completely, only
It is to reduce odds.Therefore, a kind of more preferably anti-fissure scheme of message queue cluster urgently occurs.
Summary of the invention
A kind of anti-fissure method and device of message queue cluster is provided in the embodiment of the present application, in favor of solving existing skill
Message queue cluster cannot avoid the problem that fissure completely in art.
In a first aspect, the embodiment of the present application provides a kind of method of anti-fissure of message queue cluster, comprising:
According to the status information of node when message queue exception, building alerts model, includes being used in the alarm model
N number of element of the status information of the node is characterized, wherein one alarm threshold of each element arrangements, N >=1;
Time series analysis is carried out to N number of element at the node current time respectively, obtains N described in subsequent time
The corresponding element prediction value of a element;
If the corresponding element prediction value of the first element is greater than or equal to the corresponding alarm threshold of first element, carry out
Alarm, first element are the either element in N number of element.
Optionally, the method also includes:
If the alarm number of the node is more than preset alarm frequency threshold value, the message team on the node is closed
Column.
Optionally, the alarm model is a trinary data group, and the element of the trinary data group is cpu busy percentage, interior
Deposit utilization rate and network fluctuation coefficient.
Optionally, the alarm threshold of the cpu busy percentage is cpu busy percentage historical data average value;The memory utilizes
The alarm threshold of rate is memory usage historical data average value;The alarm threshold of the network fluctuation coefficient is network fluctuation system
Number historical data average value.
Optionally, N number of element to the node current time carries out time series analysis respectively, under acquisition
The corresponding element prediction value of N number of element described in one moment, comprising:
Based on Arima algorithm, time series analysis is carried out to N number of element at the node current time respectively, is obtained
The corresponding element prediction value of N number of element described in subsequent time.
Second aspect, the embodiment of the present application provide a kind of device of anti-fissure of message queue cluster, comprising:
Module is constructed, for the status information according to node when message queue exception, building alarm model, the alarm mould
It include N number of element for characterizing the status information of the node in type, wherein one alarm threshold of each element arrangements, N >=1;
Prediction module carries out time series analysis for N number of element to the node current time respectively, obtains
The corresponding element prediction value of N number of element described in subsequent time;
Alarm module, if being greater than or equal to the corresponding announcement of first element for the corresponding element prediction value of the first element
Alert threshold value, then alerted, and first element is the either element in N number of element.
Optionally, described device further include:
Closedown module closes the section if the alarm number for the node is more than preset alarm frequency threshold value
Message queue on point.
Optionally, the alarm model is a trinary data group, and the element of the trinary data group is cpu busy percentage, interior
Deposit utilization rate and network fluctuation coefficient.
Optionally, the alarm threshold of the cpu busy percentage is cpu busy percentage historical data average value;The memory utilizes
The alarm threshold of rate is memory usage historical data average value;The alarm threshold of the network fluctuation coefficient is network fluctuation system
Number historical data average value.
Optionally, the prediction module is specifically used for being based on Arima algorithm, to the N at the node current time
A element carries out time series analysis respectively, obtains the corresponding element prediction value of N number of element described in subsequent time.
In the embodiment of the present application, it to modeling analysis the reason of causing message queue fissure, obtains and causes message queue
The threshold value of each element, and using these threshold values as alarm conditions, in addition each alarm element is predicted, and according to prediction
Value provides alarm, operation maintenance personnel can be noted that before message queue is abnormal, and after alarm generates, automatic to kill
Process effectively prevent the appearance of message queue fissure.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, for those of ordinary skill in the art
Speech, without creative efforts, is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of anti-fissure method flow schematic diagram of message queue cluster provided by the embodiments of the present application;
Fig. 2 is a kind of apparatus structure schematic diagram of the anti-fissure of message queue cluster provided by the embodiments of the present application.
Specific embodiment
In order to make those skilled in the art better understand the technical solutions in the application, below in conjunction with the application reality
The attached drawing in example is applied, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described implementation
Example is merely a part but not all of the embodiments of the present application.Based on the embodiment in the application, this field is common
The application protection all should belong in technical staff's every other embodiment obtained without making creative work
Range.
First below to the invention relates to concept be illustrated.
The cloud computing of OpenStack: one open source manages platform project, is combined by several main components and completes tool
Body running.Project objective is to provide that implementation is simple, can extend, enrich on a large scale, the cloud computing platform that standard is unified.
OpenStack services the solution of (IaaS) by the service of the various complementations facility that provides the foundation, and each service provides
API is to be integrated.
Hypervisor:Hypervisor is a kind of middleware software operated between physical server and operating system
Layer, allows multiple operating systems and a set of underlying physical hardware of Application share, therefore also can be regarded as in virtual environment
" member " operating system, it can coordinate to access all physical equipments and virtual machine on server, also be virtual machine monitor
(Virtual Machine Monitor).Hypervisor is the core of all virtualization technologies.Multiplexing is supported to make non-interruptedly
The ability of load migration is the basic function of Hypervisor.When server is started and carried out Hypervisor, it can be to every
One virtual machine distributes suitable memory, CPU, network and disk, and loads the client operating system of all virtual machines.
Message queue (RabbitMQ): the framework of Openstack, which determines, to be needed to realize not using message queue mechanism
With the communication of intermodule, by information authentication, message conversion, message routing framework mode, bring benefit is exactly that can be mould
Utmostly decoupled between block, client do not need concern server-side position and whether there is, only need to by message queue into
The transmission of row information.RabbitMQ is suitble to be deployed in the flexible scale system environments of a topology, is effectively ensured
Disparate modules, different node, between different processes message communicating timeliness, can effectively support OpenStack cloud platform system
Scale deployment, resilient expansion, the demand of flexible architecture and information security.
Message queue network partition (fissure): in High Availabitity (HA) system, when " heartbeat " of 2 nodes of connection disconnects
When, it was integral originally, the HA system of coordination, just division becomes 2 independent individuals.Due to mutually losing connection,
All it is thought as that other side is out of order.HA software on two nodes fights for " shared resource ", has striven and " answered as " split "
With service ", it will lead to that shared resource is carved up, 2 sides " service " all cannot get up in this way;Or 2 side " service " all get up, but
" shared storage " is read while write, corrupted data is caused, the common online log such as database poll malfunctions.
The HA design of OpenStack cluster is divided into two aspects: stateless and stateful.Wherein, statelessly refer to the clothes
The data for needing persistence will not be locally stored in the example run of being engaged in, and multiple examples are for the knot of the same request response
Fruit is completely the same.This kind of service is after the creation of Netease's honeycomb cloud platform, by the load balancing inside k8s, when access should
After the request of service reaches service one end, load balancing can find an example at random come the response for completing the request and (be at present
Poll).The example of this kind of service may stop because of some reasons or re-create (when such as dilatation), and at this moment, these stop
All information (in addition to log and monitoring data) in example only will all lose (restarting container i.e. can lose).So if you
Container instance in need to retain important information, and it is desirable that can back up in order to can restore later at any time, then
It is recommended that you create stateful service.Stateful service (Stateful Service): refer to that the example of the service can be by one
Divided data is backed up at any time, and when creating a new stateful service, can by these data of Backup and Restore, with
Achieve the purpose that data persistence.Stateful service can only have an example, therefore not support " automation services capacity regulating ".One
As for, database service or need can in the application program of local file system storage configuration file or other permanent datas
Stateful service is used with creation.To create stateful service, it is necessary to meet several premises: service mirror image to be created
(image) storage volume (Volume) must be defined in Dockerfile, because only that the data in catalogue where storage volume
When can be backed up creation service, it is necessary to the disk space size of storage volume distribution is assigned to, if while creation service
Which need to restore data in a backup before, then also to indicate the storage volume with Backup and Restore.
In general stateless resume module be it is fairly simple, basic ideas are to run multiple nodes or clothes parallel
Be engaged in module and to they carry out load balancing.Typical example is the Web server cluster an of website, is often added using front end
The LoadBanlace server of LVS or Nginx etc solves the problems, such as HA, wherein the High Availabitity of LVS and Nginx is mainly benefit
It is solved with Keepalived, Heartbeat etc. based on route redundancy protocol VRRP or heartbeat arbitration mechanism.And for stateful
Module, realize HA there are mainly two types of mode, one is multinode be based on distributed consensus agreement (such as Paxos,
Raft agreement etc.) the identical state of maintenance, typically representing has Zookeeper, Rabbitmq;One is based on master slave mode
Either synchronously or asynchronously duplication is to safeguard identical state, such as Mysql, Redis.The former is more complex for both modes, in some fields
Performance can be very low under scape, and the latter is insufficient in terms of data consistency and retractility.
For the fissure countermeasure between message queue cluster, following scheme is generallyd use in the prior art: first, it adds superfluous
Remaining heartbeat, for example, reducing " fissure " occurrence probability to the greatest extent using diplocardia wire jumper;Second, arbitration mechanism is set, such as sets
It sets with reference to IP (such as gateway IP), when heartbeat is fully disconnected, the respective ping of 2 nodes once refers to IP, and general rule does not show
Breakpoint just goes out in local terminal, and not only " heartbeat ", the local terminal network link of also simultaneous externally " service " are disconnected, even if starting (or continuation)
Application service is not also used, that just actively abandons competing, allow can the logical one end with reference to IP ping gone to service.More insurance one
A bit, the obstructed side with reference to IP of ping self is restarted, and those of is possible to also take up shared resource with thorough release.
But fissure can not be avoided by addition redundancy heartbeat or setting arbitration mechanism completely, it is only reduction
Probability, and can only prevent heartbeat caused by network fluctuation disconnect phenomenon.For example, two heartbeats preferably at most avoid network wave
It is dynamic, but it is seemingly-dead unrelated with network caused by memory, cpu.When a message queue node is caused due to memory, cpu
Response is very slow, and when causing Message Queuing Services seemingly-dead, diplocardia wire jumper can not then be solved.
Based on this, the embodiment of the present application provides a kind of anti-fissure method of message queue cluster, and Fig. 1 is the embodiment of the present application
The anti-fissure method flow schematic diagram of a kind of message queue cluster provided, as shown in Figure 1, it is mainly included the following steps that.
Step S101: it according to the status information of node when message queue exception, constructs and alerts model, in the alarm model
N number of element including the status information for characterizing the node, wherein one alarm threshold of each element arrangements, N >=1.
In the embodiment of the present application, the status information may include cpu busy percentage, memory usage, network state etc.,
Specifically, the reason of message queue cluster exception is analyzed, enumerates all possible factors for causing message queue exception, example
Such as, the cpu busy percentage of node, memory usage, network state etc..Correspondingly, the alarm model can be a triple S
={ C, M, N }, wherein C is cpu busy percentage, and M is memory usage, and N is network fluctuation coefficient.
In addition, it is also necessary to which alarm threshold, which is arranged, for each element specifically can calculate separately ternary in the past period
The average value of each element in group, S (avg)={ C (avg), M (avg), N (avg) }.
Step S102: time series analysis is carried out to N number of element at the node current time respectively, is obtained next
The corresponding element prediction value of N number of element described in moment;
In the embodiment of the present application, time series analysis can be carried out to triple interior element based on Arima algorithm respectively,
Obtain the predicted value of each element in subsequent time triple.Specifically, to cpu busy percentage, memory usage, network fluctuation system
Number carries out time series analysis, obtains cpu busy percentage predicted value, memory usage predicted value and network fluctuation coefficient prediction respectively
Value.
Step S103: if the corresponding element prediction value of the first element is greater than or equal to the corresponding alarm threshold of first element
Value, then alerted, and first element is the either element in N number of element.
In the embodiment of the present application, which can be cpu busy percentage predicted value, memory usage predicted value or net
Network coefficient of variation predicted value.For example, then being carried out when cpu busy percentage predicted value is greater than or equal to the alarm threshold for its setting
Alarm.
Certainly, those skilled in the art can be set according to actual needs warning strategies, for example, can be set to when all
When element is more than alarm threshold, just alerted, it should all be within the protection scope of the application.
In addition, the generation of fissure phenomenon in order to prevent, also sets up alarm frequency threshold value, when a certain in the embodiment of the present application
When the alarm number of node is more than that the alarm vehicle of setting damages threshold value, no matter which kind of state message queue cluster is in, then preferential to close
The message queue on the node is closed, that is, kills the message queue on the node, prevents the appearance of fissure.
In the embodiment of the present application, it to modeling analysis the reason of causing message queue fissure, obtains and causes message queue
The threshold value of each element, and using these threshold values as alarm conditions, in addition each alarm element is predicted, and according to prediction
Value provides alarm, operation maintenance personnel can be noted that before message queue is abnormal, and after alarm generates, automatic to kill
Process effectively prevent the appearance of message queue fissure.
Corresponding with above method embodiment, present invention also provides a kind of device of anti-fissure of message queue cluster, Fig. 2
For a kind of apparatus structure schematic diagram of the anti-fissure of message queue cluster provided by the embodiments of the present application, as shown in Fig. 2, it is mainly wrapped
It includes with lower module.
Module 201 is constructed, for the status information according to node when message queue exception, building alarm model, the announcement
It include N number of element for characterizing the status information of the node in alert model, wherein one alarm threshold of each element arrangements, N
≥1;
Prediction module 202 carries out time series analysis for N number of element to the node current time respectively,
Obtain the corresponding element prediction value of N number of element described in subsequent time;
Alarm module 203, if corresponding more than or equal to first element for the corresponding element prediction value of the first element
Alarm threshold, then alerted, first element be N number of element in either element.
In an alternative embodiment, described device further includes closedown module 204, if the alarm number for the node
More than preset alarm frequency threshold value, then the message queue on the node is closed.
In an alternative embodiment, the alarm model is a trinary data group, and the element of the trinary data group is
CPU utilization rate, memory usage and network fluctuation coefficient.
In an alternative embodiment, the alarm threshold of the cpu busy percentage is cpu busy percentage historical data average value;
The alarm threshold of the memory usage is memory usage historical data average value;The alarm threshold of the network fluctuation coefficient
For network fluctuation coefficient historical data average value.
In an alternative embodiment, the prediction module is specifically used for being based on Arima algorithm, current to the node
N number of element at moment carries out time series analysis respectively, obtains the corresponding element prediction of N number of element described in subsequent time
Value.
In the embodiment of the present application, it to modeling analysis the reason of causing message queue fissure, obtains and causes message queue
The threshold value of each element, and using these threshold values as alarm conditions, in addition each alarm element is predicted, and according to prediction
Value provides alarm, operation maintenance personnel can be noted that before message queue is abnormal, and after alarm generates, automatic to kill
Process effectively prevent the appearance of message queue fissure.
In the specific implementation, the application also provides a kind of computer storage medium, wherein the computer storage medium can store
There is program, which may include step some or all of in each embodiment provided by the present application when executing.The storage is situated between
Matter can be magnetic disk, CD, read-only memory (English: read-only memory, abbreviation: ROM) or random storage memory
Body (English: random access memory, referred to as: RAM) etc..
It is required that those skilled in the art can be understood that the technology in the embodiment of the present application can add by software
The mode of general hardware platform realize.Based on this understanding, the technical solution in the embodiment of the present application substantially or
Say that the part that contributes to existing technology can be embodied in the form of software products, which can deposit
Storage is in storage medium, such as ROM/RAM, magnetic disk, CD, including some instructions are used so that computer equipment (can be with
It is personal computer, server or the network equipment etc.) execute certain part institutes of each embodiment of the application or embodiment
The method stated.
Same and similar part may refer to each other between each embodiment in this specification.Implement especially for terminal
For example, since it is substantially similar to the method embodiment, so being described relatively simple, related place is referring in embodiment of the method
Explanation.
Above-described the application embodiment does not constitute the restriction to the application protection scope.
Claims (10)
1. a kind of method of the anti-fissure of message queue cluster characterized by comprising
According to the status information of node when message queue exception, it includes for characterizing in the alarm model that building, which alerts model,
N number of element of the status information of the node, wherein one alarm threshold of each element arrangements, N >=1;
Time series analysis is carried out to N number of element at the node current time respectively, obtains N number of member described in subsequent time
The corresponding element prediction value of element;
If the corresponding element prediction value of the first element is greater than or equal to the corresponding alarm threshold of first element, accused
Alert, first element is the either element in N number of element.
2. the method for the anti-fissure of message queue cluster according to claim 1, which is characterized in that further include:
If the alarm number of the node is more than preset alarm frequency threshold value, the message queue on the node is closed.
3. the method for the anti-fissure of message queue cluster according to claim 1 or 2, which is characterized in that the alarm model
For a trinary data group, the element of the trinary data group is cpu busy percentage, memory usage and network fluctuation coefficient.
4. the method for the anti-fissure of message queue cluster according to claim 3, which is characterized in that the cpu busy percentage
Alarm threshold is cpu busy percentage historical data average value;The alarm threshold of the memory usage is memory usage history number
According to average value;The alarm threshold of the network fluctuation coefficient is network fluctuation coefficient historical data average value.
5. the method for the anti-fissure of message queue cluster according to claim 1, which is characterized in that described to work as to the node
N number of element at preceding moment carries out time series analysis respectively, obtains the corresponding element prediction of N number of element described in subsequent time
Value, comprising:
Based on Arima algorithm, time series analysis is carried out respectively to N number of element at the node current time, under acquisition
The corresponding element prediction value of N number of element described in one moment.
6. a kind of device of the anti-fissure of message queue cluster characterized by comprising
Module is constructed, for the status information according to node when message queue exception, constructs and alerts model, in the alarm model
N number of element including the status information for characterizing the node, wherein one alarm threshold of each element arrangements, N >=1;
Prediction module carries out time series analysis for N number of element to the node current time respectively, obtains next
The corresponding element prediction value of N number of element described in moment;
Alarm module, if being greater than or equal to the corresponding alarm threshold of first element for the corresponding element prediction value of the first element
Value, then alerted, and first element is the either element in N number of element.
7. the device of the anti-fissure of message queue cluster according to claim 6, which is characterized in that further include:
Closedown module is closed on the node if the alarm number for the node is more than preset alarm frequency threshold value
Message queue.
8. the device of the anti-fissure of message queue cluster according to claim 6 or 7, which is characterized in that the alarm model
For a trinary data group, the element of the trinary data group is cpu busy percentage, memory usage and network fluctuation coefficient.
9. the device of the anti-fissure of message queue cluster according to claim 8, which is characterized in that the cpu busy percentage
Alarm threshold is cpu busy percentage historical data average value;The alarm threshold of the memory usage is memory usage history number
According to average value;The alarm threshold of the network fluctuation coefficient is network fluctuation coefficient historical data average value.
10. the device of the anti-fissure of message queue cluster according to claim 6, which is characterized in that the prediction module, tool
Body is used to be based on Arima algorithm, carries out time series analysis respectively to N number of element at the node current time, obtains
The corresponding element prediction value of N number of element described in subsequent time.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810682010.6A CN108984320A (en) | 2018-06-27 | 2018-06-27 | A kind of anti-fissure method and device of message queue cluster |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810682010.6A CN108984320A (en) | 2018-06-27 | 2018-06-27 | A kind of anti-fissure method and device of message queue cluster |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108984320A true CN108984320A (en) | 2018-12-11 |
Family
ID=64538630
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810682010.6A Pending CN108984320A (en) | 2018-06-27 | 2018-06-27 | A kind of anti-fissure method and device of message queue cluster |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108984320A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109677465A (en) * | 2018-12-29 | 2019-04-26 | 卡斯柯信号有限公司 | Distributed real-time systems framework for track traffic synthetic monitoring system |
CN110225133A (en) * | 2019-06-20 | 2019-09-10 | 恒生电子股份有限公司 | Message method, node, device, system and relevant device |
CN113204437A (en) * | 2020-01-31 | 2021-08-03 | 慧与发展有限责任合伙企业 | Connection of application instances to client devices |
CN113377702A (en) * | 2021-07-06 | 2021-09-10 | 安超云软件有限公司 | Method and device for starting two-node cluster, electronic equipment and storage medium |
CN113438111A (en) * | 2021-06-23 | 2021-09-24 | 华云数据控股集团有限公司 | Method for restoring RabbitMQ network partition based on Raft distribution and application |
CN114528350A (en) * | 2022-02-18 | 2022-05-24 | 苏州浪潮智能科技有限公司 | Cluster split brain processing method, device and equipment and readable storage medium |
CN115174356A (en) * | 2022-07-27 | 2022-10-11 | 济南浪潮数据技术有限公司 | Cluster alarm reporting method, device, equipment and medium |
CN113296971B (en) * | 2020-07-14 | 2024-04-19 | 阿里巴巴集团控股有限公司 | Method, device and equipment for expanding and shrinking volume and processing message queue |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20060010101A1 (en) * | 2004-07-08 | 2006-01-12 | Yasuhiro Suzuki | System, method and program product for forecasting the demand on computer resources |
CN103178990A (en) * | 2011-12-20 | 2013-06-26 | 中国移动通信集团青海有限公司 | Network device performance monitoring method and network management system |
CN103513983A (en) * | 2012-06-25 | 2014-01-15 | 国际商业机器公司 | Method and system for predictive alert threshold determination tool |
CN106375115A (en) * | 2016-08-30 | 2017-02-01 | 东软集团股份有限公司 | Resource distribution method and device |
CN106909487A (en) * | 2017-01-18 | 2017-06-30 | 北京盛世全景科技股份有限公司 | It is applied to the method for early warning and device of information system |
-
2018
- 2018-06-27 CN CN201810682010.6A patent/CN108984320A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20060010101A1 (en) * | 2004-07-08 | 2006-01-12 | Yasuhiro Suzuki | System, method and program product for forecasting the demand on computer resources |
CN103178990A (en) * | 2011-12-20 | 2013-06-26 | 中国移动通信集团青海有限公司 | Network device performance monitoring method and network management system |
CN103513983A (en) * | 2012-06-25 | 2014-01-15 | 国际商业机器公司 | Method and system for predictive alert threshold determination tool |
CN106375115A (en) * | 2016-08-30 | 2017-02-01 | 东软集团股份有限公司 | Resource distribution method and device |
CN106909487A (en) * | 2017-01-18 | 2017-06-30 | 北京盛世全景科技股份有限公司 | It is applied to the method for early warning and device of information system |
Non-Patent Citations (2)
Title |
---|
ANOOP S. KUMAR等: ""Forecasting HPC Workload Using ARMA Models and SSA"", 《2016 INTERNATIONAL CONFERENCE ON INFORMATION TECHNOLOGY (ICIT)》 * |
仇沂: ""分布式环境中的性能异常预测监控"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109677465A (en) * | 2018-12-29 | 2019-04-26 | 卡斯柯信号有限公司 | Distributed real-time systems framework for track traffic synthetic monitoring system |
CN110225133A (en) * | 2019-06-20 | 2019-09-10 | 恒生电子股份有限公司 | Message method, node, device, system and relevant device |
CN110225133B (en) * | 2019-06-20 | 2022-01-21 | 恒生电子股份有限公司 | Message sending method, node, device, system and related equipment |
CN113204437A (en) * | 2020-01-31 | 2021-08-03 | 慧与发展有限责任合伙企业 | Connection of application instances to client devices |
US11824947B2 (en) | 2020-01-31 | 2023-11-21 | Hewlett Packard Enterprise Development Lp | Connecting application instances to client devices |
CN113296971B (en) * | 2020-07-14 | 2024-04-19 | 阿里巴巴集团控股有限公司 | Method, device and equipment for expanding and shrinking volume and processing message queue |
CN113438111A (en) * | 2021-06-23 | 2021-09-24 | 华云数据控股集团有限公司 | Method for restoring RabbitMQ network partition based on Raft distribution and application |
CN113377702A (en) * | 2021-07-06 | 2021-09-10 | 安超云软件有限公司 | Method and device for starting two-node cluster, electronic equipment and storage medium |
CN113377702B (en) * | 2021-07-06 | 2024-03-22 | 安超云软件有限公司 | Method and device for starting two-node cluster, electronic equipment and storage medium |
CN114528350A (en) * | 2022-02-18 | 2022-05-24 | 苏州浪潮智能科技有限公司 | Cluster split brain processing method, device and equipment and readable storage medium |
CN114528350B (en) * | 2022-02-18 | 2024-01-16 | 苏州浪潮智能科技有限公司 | Cluster brain fracture processing method, device, equipment and readable storage medium |
CN115174356A (en) * | 2022-07-27 | 2022-10-11 | 济南浪潮数据技术有限公司 | Cluster alarm reporting method, device, equipment and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108984320A (en) | A kind of anti-fissure method and device of message queue cluster | |
JP4505763B2 (en) | Managing node clusters | |
CN102402395B (en) | Quorum disk-based non-interrupted operation method for high availability system | |
CN109729129A (en) | Configuration modification method, storage cluster and the computer system of storage cluster | |
CN112069265B (en) | Synchronization method of configuration data, business data system, computer system and medium | |
US10728085B1 (en) | Model-based network management | |
CN109729111A (en) | Method, equipment and computer program product for managing distributing type system | |
CN110795503A (en) | Multi-cluster data synchronization method and related device of distributed storage system | |
JP2000137694A (en) | System and method for supplying continuous data base access by using common use redundant copy | |
US20210216417A1 (en) | Hot-standby redundancy control system, method, control apparatus, and computer readable storage medium | |
CN111935244B (en) | Service request processing system and super-integration all-in-one machine | |
CN112217847A (en) | Micro service platform, implementation method thereof, electronic device and storage medium | |
CN108600316A (en) | Data managing method, system and the equipment of cloud storage service | |
CN107135097A (en) | The disaster tolerance system and disaster recovery method filed based on bookkeeping | |
CN112527567A (en) | System disaster tolerance method, device, equipment and storage medium | |
CN113515408A (en) | Data disaster tolerance method, device, equipment and medium | |
US8489721B1 (en) | Method and apparatus for providing high availabilty to service groups within a datacenter | |
CN111092754B (en) | Real-time access service system and implementation method thereof | |
CN105471986B (en) | A kind of Constructing data center Scale Revenue Ratio method and device | |
CN112231399A (en) | Method and device applied to graph database | |
CN116886696A (en) | Cloud edge collaboration-based application deployment method and system | |
US9973569B2 (en) | System, method and computing apparatus to manage process in cloud infrastructure | |
CN116192885A (en) | High-availability cluster architecture artificial intelligent experiment cloud platform data processing method and system | |
CN105468446A (en) | Linux based method for realizing high availability of HPC job scheduling | |
JPH04311251A (en) | Multiprocessor system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181211 |