CN107147528A - One kind stores gateway intelligently anti-fissure system and method - Google Patents
One kind stores gateway intelligently anti-fissure system and method Download PDFInfo
- Publication number
- CN107147528A CN107147528A CN201710368277.3A CN201710368277A CN107147528A CN 107147528 A CN107147528 A CN 107147528A CN 201710368277 A CN201710368277 A CN 201710368277A CN 107147528 A CN107147528 A CN 107147528A
- Authority
- CN
- China
- Prior art keywords
- node
- network
- storage
- arbitrator
- fissure
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0654—Management of faults, events, alarms or notifications using network fault recovery
- H04L41/0663—Performing the actions predefined by failover planning, e.g. switching to standby network elements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/66—Arrangements for connecting between networks having differing types of switching systems, e.g. gateways
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/10—Active monitoring, e.g. heartbeat, ping or trace-route
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Cardiology (AREA)
- General Health & Medical Sciences (AREA)
- Computer And Data Communications (AREA)
Abstract
Gateway intelligently anti-fissure system and method is stored the invention discloses one kind, including application service layer, network link layer, storage management layer and storage device layer, wherein application-server layer is Client application system server, equipped with various application systems and database software；Network link layer, includes IP network and FC fiber optic networks；Storage management layer, including two storage gateways, as node, when there is a node failure, another node can be with normal work；Arbitrator is configured between two storage gateways, when node breaks down, the arbitrator arbitrates the state for judging each node, judge the node broken down；Storage device layer, includes at least two storage arrays.Intelligently anti-fissure system and method is compared with prior art for a kind of storage gateway of the present invention; propose three-level fissure judgment mechanism; Active and Inactive states that can be in data protection high availability scheme between effective decision node; so as to judge Primary nodes, the generation of fissure phenomenon is prevented.
Description
Technical field
It is specifically a kind of to store gateway intelligently anti-fissure system and method the present invention relates to field of computer technology.
Background technology
Big data epoch, data have become the vital core asset of every profession and trade.Exist in traditional disaster recovery solution
Downtime length when low resource utilization, poor availability, breaking down, the problems such as data recovery is slow, risk is high.Whether data
Whether safety, business are continuously run turns into the key that user weighs a disaster recovery solution without interruption.Conventional data centers store calamity
Standby typically to use active-standby mode, only after creation data central store failure, Disaster Preparation Center's storage just can adapter data access
Business, and this process needed to manually perform, and Disaster Preparation Center's corresponding business Lun manual actuations read-write is serviced；In addition, active and standby
The pattern of data center, under regular traffic working order, only main center is played a role, and the resource at standby center is constantly in
" awaiting orders " pattern, it is impossible at utmost play the efficiency of all resources.
Dual-active data center will be the trend of Future Data center development, and store the weight that dual-active is data center's dual-active
Will basis.Traditional calamity is the production center, a Disaster Preparation Center for deployment way, and Disaster Preparation Center's platform is in the shape that do not work
State, only when a disaster occurs, cardiaplegia in creation data, Disaster Preparation Center just start.This disaster recovery and backup systems face following challenge：
1. when the production center meets with power supply trouble, fire, big flood, the disaster such as earthquake, it is necessary to manually by service switching to Disaster Preparation Center,
Business interruption time is long, it is impossible to ensure that business continuity is run.2. Disaster Preparation Center can not externally provide service, throughout the year in idle shape
State, resource utilization is low.For traditional disaster recovery and backup systems build facing challenges, the dual-active solution based on storage meet the tendency of and
It is raw.In current storage dual-active technology, such as NodeA and NodeB can be by heartbeat detections with true between two storage gateways
Recognize other side's presence, in the presence of confirming less than other side by heartbeat detection, take over corresponding（It is shared）resource；If prominent
Between so, the heartbeat between NodeA and NodeB is not present（As network disconnects）, and NodeA and NodeB in fact all in
Active states, now NodeA will adapter NodeB resource, while NodeB will adapter NodeA resource,
At this moment it is exactly fissure（split-brain）
In " two-node cluster hot backup " High Availabitity（HA）In system, when " heartbeat " that contacts 2 nodes disconnects, constituted originally same
Individual entirety, the HA systems of coordination, just division turn into 2 independent individuals.Due to mutually losing contact, all think
That other side is out of order, the HA softwares on 2 nodes as " split ", " instinct " fight for " shared resource ", emulatively
Start " application service ", then will necessarily occur serious consequence：Or shared resource is carved up, " service " on both sides all rises not
Come；Or both sides " service " all gets up, but read while write " shared storage ", cause corrupted data（Common such as database wheel
The online daily record error ask）.As can be seen here, a kind of reliable and intelligent arbitration mechanism is created to prevent fissure from occurring heavy to closing
Will.Based on this, the present invention provides a kind of storage gateway intelligently anti-fissure system and method.
The content of the invention
The technical assignment of the present invention is to be directed to above weak point to store gateway intelligently anti-fissure system and side there is provided one kind
Method.
One kind storage gateway intelligently anti-fissure system, including application service layer, network link layer, storage management layer and storage
Mechanical floor, wherein,
Application-server layer is Client application system server, equipped with various application systems and database software；
Network link layer, is interconnected comprising IP network and FC fiber optic networks, and by IP interchangers and FC interchangers；
Storage management layer, including two storage gateways, as node, when there is a node failure, another node can be just
Often work, by the storage of upper-layer client end data write-in lower floor；Arbitrator is configured between two storage gateways, when node breaks down
When, the arbitrator arbitrates the state for judging each node, judges the node broken down；
Storage device layer, includes at least two storage arrays.
The arbitrator is a lock service or a share dish, can pass through ping orders between the arbitrator and node
Communicated.
During arbitrator's arbitration decision node state, its IP network heartbeat between node is first determined whether, works as heartbeat
Then the node breaks down during stopping；Do not stop, continue FC heartbeats judgement, i.e. optical fiber heartbeat and judge, then should when careful stop jumping is stopped
Node breaks down；Do not stop, continue IMPI judgements, the node for being possible to receive I/O Request continues as fault-free gateway
It is written and read service.
The network Heartbeat is the communication bag transmitted in the cluster by IP links between node, for monitor clustered node,
Whether network and network interface working condition are normal, prevent from the generation of isolated node to ensure the normal work of cluster.
The optical fiber heartbeat refers to the communication bag transmitted between clustered node by optical fiber link, for monitor clustered node,
Whether network and network interface working condition are normal, prevent from the generation of isolated node to ensure the normal work of cluster.
One kind stores gateway intelligently anti-fissure method, applied to dual-active data center, is configured with the dual-active data center
Two storage gateways are considered as two nodes, implementation step is, when difference occur in two nodes by two storage gateways, if
A third party arbitrator is put, is arbitrated respectively by arbitrator by two nodes for difference occur, judges the state of each node, really
Whether fixed each node breaks down.
The arbitrator is a lock service or a share dish.
Arbitrator's arbitration includes three kinds of arbitration mechanisms, is specially：
Network Heartbeat judges that two nodes for difference occur are distinguished into ping leads to arbitrators, then shows that breakpoint just exists when obstructed
Local terminal, general rule continues a kind of lower judgement；
Optical fiber heartbeat judges that two nodes for difference occur are distinguished into ping leads to arbitrators, then shows that breakpoint just exists when obstructed
Local terminal, general rule continues a kind of lower judgement；
IMPI judges, sends write IO request by arbitrator, when the node of Standby states receives write IO request, then passes through
Then oneself switched to Primary by IPMI by an other node forced shutdown, and the Primary nodes continue to provide read-write clothes
Business.
Heartbeat during the network Heartbeat judges is the communication bag transmitted in the cluster by IP links between node, for supervising
Whether normal survey clustered node, network and network interface working condition, prevent from the generation of isolated node to ensure the normal of cluster
Work；When judging, internetwork communication first determines whether all respective ping of node that difference occur in IP heartbeat passages, i.e., two
IP is once referred to, shows that breakpoint just goes out in local terminal if the obstructed arbitrators of ping, the node for allowing energy ping to lead to arbitrator is gone
Service.
The optical fiber heartbeat refers to the communication bag transmitted between clustered node by optical fiber link, for monitor clustered node,
Whether network and network interface working condition are normal, prevent from the generation of isolated node to ensure the normal work of cluster.
A kind of storage gateway of the present invention intelligently anti-fissure system and method compared to the prior art, with following beneficial effect
Really：
The a kind of of the present invention stores gateway intelligently anti-fissure system and method, it is proposed that three-level fissure judgment mechanism, can be in number
According to the Active and Inactive states between effective decision node in protection high availability scheme, so that judge Primary nodes,
The generation of fissure phenomenon is prevented, the safety of effectively protection data has extensive practical application；The system is designed in Internet,
Storage management layer and storage device layer full redundancy design, any one node breaks down, all without influence client's upper layer application
Continuation access；Design arbitration, can be a lock service or a share dish, design simple and convenient build；Three kinds of arbitrations
Judgment mechanism, preferentially walks network Heartbeat passage, if network Heartbeat line is obstructed, and system will attempt FC heartbeats communication, if both of the above
General rule does not start third level IPMI mechanism；Three-level judges the occurrence of preventing mechanism effectively can prevent fissure, is applicable model
Enclose extensively, with good application value.
Brief description of the drawings
Accompanying drawing 1 realizes schematic diagram for the present invention's.
Embodiment
Below in conjunction with the accompanying drawings and specific embodiment the invention will be further described.
As shown in Figure 1, a kind of storage gateway intelligently anti-fissure system, including application service layer, network link layer, storage
Management level and storage device layer, wherein,
Application-server layer is Client application system server, equipped with various application systems and database software；
Network link layer, is interconnected comprising IP network and FC fiber optic networks, and by IP interchangers and FC interchangers；
Storage management layer, including storage gateway 1 and storage gateway 2 are as node NodeA and NodeB, when there is a node failure
When, another node can be with normal work, by the storage of upper-layer client end data write-in lower floor；Configured between two storage gateways
Arbitrator, when node breaks down, the arbitrator arbitrates the state for judging each node, judges the node broken down；
Storage device layer, includes storage array 1 and storage array 2.
The arbitrator is a lock service or a share dish, can pass through ping orders between the arbitrator and node
Communicated.
During arbitrator's arbitration decision node state, its IP network heartbeat between node is first determined whether, works as heartbeat
Then the node breaks down during stopping；Do not stop, continue FC heartbeats judgement, i.e. optical fiber heartbeat and judge, then should when careful stop jumping is stopped
Node breaks down；Do not stop, continue IMPI judgements, the node for being possible to receive I/O Request continues as fault-free gateway
It is written and read service.
The network Heartbeat is the communication bag transmitted in the cluster by IP links between node, for monitor clustered node,
Whether network and network interface working condition are normal, prevent from the generation of isolated node to ensure the normal work of cluster.
The optical fiber heartbeat refers to the communication bag transmitted between clustered node by optical fiber link, for monitor clustered node,
Whether network and network interface working condition are normal, prevent from the generation of isolated node to ensure the normal work of cluster.
One kind stores gateway intelligently anti-fissure method, applied to dual-active data center, is configured with the dual-active data center
Two storage gateways are considered as two nodes, implementation step is, when difference occur in two nodes by two storage gateways, if
A third party arbitrator is put, is arbitrated respectively by arbitrator by two nodes for difference occur, judges the state of each node, really
Whether fixed each node breaks down.
The arbitrator is that the data transfer between a lock service or a share dish, storage gateway is handed over by optical fiber
Change planes progress.
Arbitrator's arbitration includes three kinds of arbitration mechanisms, is specially：
The first order：IP heartbeats are protected.
Heartbeat（Heartbeat）It is the communication bag transmitted in the cluster by IP links between Node nodes, for monitoring collection
Whether group node, network and network interface working condition are normal, prevent from the generation of isolated node to ensure the normal work of cluster.
To strengthen protection in design, the heartbeat of redundancy can be also added.Such as double " heartbeats "." splitting brain " chance of occurrence is reduced as far as possible.
Internetwork communication will preferentially walk IP heartbeat passages, if IP is obstructed, and system then attempts second level FC heartbeats communication.Illustrate：
When heartbeat is fully disconnected, the respective ping of 2 nodes once refers to IP, goes out not general rule does not show breakpoint at this
End, not only " heartbeat " has broken, the local terminal network link of externally offer " service " also break by road, even if starting（Or continue）Using
Service is not also used, and that abandons competition with regard to active, allows the one end for being capable of the logical reference IP of ping to go service.More insure one
A bit, the obstructed reference IP of a ping side is simply restarted with regard to self, those the shared moneys for being possible to also take up with thorough release
Source.
The second level：FC heartbeats are protected.
FC heartbeats refer to transmit communication bag by optical fiber link between cluster Node nodes, for monitoring clustered node, net
Whether network and network interface working condition are normal, prevent from the generation of isolated node to ensure the normal work of cluster.Principle and IP
Heartbeat is identical.Communication between gateway will preferentially walk network Heartbeat passage, if network Heartbeat line is obstructed, and system is led to FC heartbeats are attempted
Letter, if general rule does not start third level IPMI mechanism to both of the above.
The third level：IPMI mechanism.
If who there occurs failure between IP heartbeats and FC heartbeats are unable to decision node, system will start IPMI mechanism, when
When the gateway of Standby states receives write IO request, then oneself can be turned by IPMI an other gateway forced shutdown
For Primary, cause data inconsistent while writing data to prevent in the case of fissure client to same Lun and cause data
The generation of damage.
IP heartbeats, FC heartbeats and IPMI mechanism simultaneous faults, the probability that such case occurs are minimum, will not typically occur；
Present invention provide that if the situation of the whole failures of triple anti-fissure mechanism occurs, Primary gateways continue to provide read-write service,
Standby gateways will stop any read-write service, to ensure the uniformity of data.
The present invention uses multiple judgment mechanism, and intelligence finds each node storage gateway status, effectively reduces " splitting brain " chance
Generation.Three kinds of arbitration judgment mechanisms are specifically referred to, network Heartbeat passage are preferentially walked, if network Heartbeat line is obstructed, system will taste
FC heartbeats communication is tried, if general rule does not start third level IPMI mechanism to both of the above；Three-level judges effectively prevent fissure feelings
The generation of condition.
By embodiment above, the those skilled in the art can readily realize the present invention.But should
Work as understanding, the present invention is not limited to above-mentioned embodiment.On the basis of disclosed embodiment, the technical field
Technical staff can be combined different technical characteristics, so as to realize different technical schemes.
It is the known technology of those skilled in the art in addition to the technical characteristic described in specification.
Claims (10)
1. one kind stores gateway intelligently anti-fissure system, it is characterised in that including application service layer, network link layer, storage tube
Layer and storage device layer are managed, wherein,
Application-server layer is Client application system server, equipped with various application systems and database software；
Network link layer, is interconnected comprising IP network and FC fiber optic networks, and by IP interchangers and FC interchangers；
Storage management layer, including two storage gateways, as node, when there is a node failure, another node can be just
Often work, by the storage of upper-layer client end data write-in lower floor；Arbitrator is configured between two storage gateways, when node breaks down
When, the arbitrator arbitrates the state for judging each node, judges the node broken down；
Storage device layer, includes at least two storage arrays.
2. one kind according to claim 1 stores gateway intelligently anti-fissure system, it is characterised in that the arbitrator is one
Individual lock service or a share dish, can be communicated between the arbitrator and node by ping orders.
3. one kind according to claim 1 or 2 stores gateway intelligently anti-fissure system, it is characterised in that the arbitrator
When arbitrating decision node state, its IP network heartbeat between node is first determined whether, then the node occurs when careful stop jumping is stopped
Failure；Do not stop, continue FC heartbeats judgement, i.e. optical fiber heartbeat and judge that then the node breaks down when careful stop jumping is stopped；Do not stop
Only then continue IMPI to judge, be possible to receive the node of I/O Request as fault-free gateway and proceed read-write service.
4. a kind of intelligent anti-fissure system of storage gateway according to claim 3, it is characterised in that the network Heartbeat is
The communication bag transmitted in the cluster between node by IP links, for monitoring clustered node, network and network interface working condition
It is whether normal, the generation of isolated node is prevented to ensure the normal work of cluster.
5. one kind according to claim 3 stores gateway intelligently anti-fissure system, it is characterised in that the optical fiber heartbeat refers to
The communication bag transmitted between clustered node by optical fiber link, for monitoring clustered node, network and network interface working condition
It is whether normal, the generation of isolated node is prevented to ensure the normal work of cluster.
6. one kind stores gateway intelligently anti-fissure method, it is characterised in that applied to dual-active data center, the dual-active data center
In be configured with two storage gateways, by two storage gateways be considered as two nodes, implementation step is, when two nodes divide
During discrimination, a third party arbitrator is set, is arbitrated respectively by arbitrator by two nodes for difference occur, judges each node
State, it is determined that whether each node breaks down.
7. one kind according to claim 6 stores gateway intelligently anti-fissure method, it is characterised in that the arbitrator is one
Individual lock service or a share dish.
8. one kind according to claim 6 or 7 stores gateway intelligently anti-fissure method, it is characterised in that the arbitrator
Arbitration includes three kinds of arbitration mechanisms, is specially：
Network Heartbeat judges that two nodes for difference occur are distinguished into ping leads to arbitrators, then shows that breakpoint just exists when obstructed
Local terminal, general rule continues a kind of lower judgement；
Optical fiber heartbeat judges that two nodes for difference occur are distinguished into ping leads to arbitrators, then shows that breakpoint just exists when obstructed
Local terminal, general rule continues a kind of lower judgement；
IMPI judges, sends write IO request by arbitrator, when the node of Standby states receives write IO request, then passes through
Then oneself switched to Primary by IPMI by an other node forced shutdown, and the Primary nodes continue to provide read-write clothes
Business.
9. one kind according to claim 8 stores gateway intelligently anti-fissure method, it is characterised in that the network Heartbeat is sentenced
Heartbeat in disconnected is the communication bag transmitted in the cluster by IP links between node, for monitoring clustered node, network and network
Whether interfaces state is normal, prevents from the generation of isolated node to ensure the normal work of cluster；It is internetwork when judging
Communication first determines whether that the respective ping of node that difference occur in IP heartbeat passages, i.e., two once refers to IP, if ping is not
Logical arbitrator then shows that breakpoint just goes out in local terminal, and the node for allowing energy ping to lead to arbitrator goes service.
10. one kind according to claim 8 stores gateway intelligently anti-fissure method, it is characterised in that the optical fiber heartbeat
Refer to the communication bag transmitted between clustered node by optical fiber link, for monitoring clustered node, network and network interface work shape
Whether state is normal, prevents from the generation of isolated node to ensure the normal work of cluster.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710368277.3A CN107147528A (en) | 2017-05-23 | 2017-05-23 | One kind stores gateway intelligently anti-fissure system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710368277.3A CN107147528A (en) | 2017-05-23 | 2017-05-23 | One kind stores gateway intelligently anti-fissure system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107147528A true CN107147528A (en) | 2017-09-08 |
Family
ID=59779089
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710368277.3A Pending CN107147528A (en) | 2017-05-23 | 2017-05-23 | One kind stores gateway intelligently anti-fissure system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107147528A (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107846306A (en) * | 2017-09-30 | 2018-03-27 | 郑州云海信息技术有限公司 | A kind of full redundancy framework storage system |
| CN107918570A (en) * | 2017-10-20 | 2018-04-17 | 杭州沃趣科技股份有限公司 | A kind of method that dual-active system shares arbitration disk |
| CN109120522A (en) * | 2018-08-21 | 2019-01-01 | 郑州云海信息技术有限公司 | A kind of multipath state monitoring method and device |
| CN109144415A (en) * | 2018-08-01 | 2019-01-04 | 新华三云计算技术有限公司 | Nodal test method and sharing and storing device |
| CN110442094A (en) * | 2019-07-26 | 2019-11-12 | 新华三大数据技术有限公司 | A kind of distributed system referee method and device |
| CN110661599A (en) * | 2018-06-28 | 2020-01-07 | 中兴通讯股份有限公司 | HA implementation method, device and storage medium between main node and standby node |
| CN110912780A (en) * | 2019-12-13 | 2020-03-24 | 无锡华云数据技术服务有限公司 | High-availability cluster detection method, system and controlled terminal |
| CN111404980A (en) * | 2019-09-29 | 2020-07-10 | 杭州海康威视系统技术有限公司 | Data storage method and object storage system |
| CN111651291A (en) * | 2020-04-23 | 2020-09-11 | 国网河南省电力公司电力科学研究院 | A method, system, and computer storage medium for preventing split-brain in a shared storage cluster |
| 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 |
| CN114461428A (en) * | 2021-12-29 | 2022-05-10 | 国能朗新明南京环保科技有限公司 | An arbitration method for split-brain in a two-machine cluster |
| CN115811461A (en) * | 2023-02-08 | 2023-03-17 | 湖南国科亿存信息科技有限公司 | SAN shared storage cluster split brain prevention processing method and device and electronic equipment |
| CN116192617A (en) * | 2023-02-16 | 2023-05-30 | 广州广哈通信股份有限公司 | Dual-machine high-availability system |
| CN116248484A (en) * | 2023-03-09 | 2023-06-09 | 安超云软件有限公司 | Management method and device of cloud primary integrated machine, electronic equipment and storage medium |
| WO2023147737A1 (en) * | 2022-02-07 | 2023-08-10 | 苏州浪潮智能科技有限公司 | Processing method for preventing drbd split brain and related component |
| CN116743551A (en) * | 2022-09-30 | 2023-09-12 | 腾讯云计算（北京）有限责任公司 | Main-standby switching method and device of equipment gateway and computer readable storage medium |
| CN117614805A (en) * | 2023-11-21 | 2024-02-27 | 杭州沃趣科技股份有限公司 | Data processing system for monitoring state of data center |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2428251A1 (en) * | 2000-12-21 | 2002-06-27 | Daniel M. Price | Method of "split-brain" prevention in computer cluster systems |
| CN102402395A (en) * | 2010-09-16 | 2012-04-04 | 上海中标软件有限公司 | Quorum disk-based non-interrupted operation method for high availability system |
| CN103593266A (en) * | 2013-11-12 | 2014-02-19 | 浪潮(北京)电子信息产业有限公司 | ot standby method based on arbitration disk mechanism |
| CN103607310A (en) * | 2013-11-29 | 2014-02-26 | 华为技术有限公司 | Method for arbitration of remote disaster recovery |
-
2017
- 2017-05-23 CN CN201710368277.3A patent/CN107147528A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2428251A1 (en) * | 2000-12-21 | 2002-06-27 | Daniel M. Price | Method of "split-brain" prevention in computer cluster systems |
| CN102402395A (en) * | 2010-09-16 | 2012-04-04 | 上海中标软件有限公司 | Quorum disk-based non-interrupted operation method for high availability system |
| CN103593266A (en) * | 2013-11-12 | 2014-02-19 | 浪潮(北京)电子信息产业有限公司 | ot standby method based on arbitration disk mechanism |
| CN103607310A (en) * | 2013-11-29 | 2014-02-26 | 华为技术有限公司 | Method for arbitration of remote disaster recovery |
Non-Patent Citations (1)
| Title |
|---|
| 孙羚霞: ""信核本地高可用技术白皮书"", 《百度文库》 * |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107846306A (en) * | 2017-09-30 | 2018-03-27 | 郑州云海信息技术有限公司 | A kind of full redundancy framework storage system |
| CN107918570A (en) * | 2017-10-20 | 2018-04-17 | 杭州沃趣科技股份有限公司 | A kind of method that dual-active system shares arbitration disk |
| CN110661599A (en) * | 2018-06-28 | 2020-01-07 | 中兴通讯股份有限公司 | HA implementation method, device and storage medium between main node and standby node |
| CN109144415A (en) * | 2018-08-01 | 2019-01-04 | 新华三云计算技术有限公司 | Nodal test method and sharing and storing device |
| CN109144415B (en) * | 2018-08-01 | 2022-05-06 | 新华三云计算技术有限公司 | Node detection method and shared storage device |
| CN109120522B (en) * | 2018-08-21 | 2021-07-27 | 郑州云海信息技术有限公司 | Multipath state monitoring method and device |
| CN109120522A (en) * | 2018-08-21 | 2019-01-01 | 郑州云海信息技术有限公司 | A kind of multipath state monitoring method and device |
| CN110442094A (en) * | 2019-07-26 | 2019-11-12 | 新华三大数据技术有限公司 | A kind of distributed system referee method and device |
| CN111404980A (en) * | 2019-09-29 | 2020-07-10 | 杭州海康威视系统技术有限公司 | Data storage method and object storage system |
| CN110912780A (en) * | 2019-12-13 | 2020-03-24 | 无锡华云数据技术服务有限公司 | High-availability cluster detection method, system and controlled terminal |
| CN111651291A (en) * | 2020-04-23 | 2020-09-11 | 国网河南省电力公司电力科学研究院 | A method, system, and computer storage medium for preventing split-brain in a shared storage cluster |
| CN111651291B (en) * | 2020-04-23 | 2023-02-03 | 国网河南省电力公司电力科学研究院 | Method, system, and computer storage medium for preventing split-brain in a shared storage cluster |
| 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 |
| 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 |
| CN114461428A (en) * | 2021-12-29 | 2022-05-10 | 国能朗新明南京环保科技有限公司 | An arbitration method for split-brain in a two-machine cluster |
| WO2023147737A1 (en) * | 2022-02-07 | 2023-08-10 | 苏州浪潮智能科技有限公司 | Processing method for preventing drbd split brain and related component |
| CN116743551A (en) * | 2022-09-30 | 2023-09-12 | 腾讯云计算（北京）有限责任公司 | Main-standby switching method and device of equipment gateway and computer readable storage medium |
| CN116743551B (en) * | 2022-09-30 | 2024-07-23 | 腾讯云计算（北京）有限责任公司 | Main-standby switching method and device of equipment gateway and computer readable storage medium |
| CN115811461B (en) * | 2023-02-08 | 2023-04-28 | 湖南国科亿存信息科技有限公司 | SAN shared storage cluster brain crack prevention processing method and device and electronic equipment |
| CN115811461A (en) * | 2023-02-08 | 2023-03-17 | 湖南国科亿存信息科技有限公司 | SAN shared storage cluster split brain prevention processing method and device and electronic equipment |
| CN116192617A (en) * | 2023-02-16 | 2023-05-30 | 广州广哈通信股份有限公司 | Dual-machine high-availability system |
| CN116248484A (en) * | 2023-03-09 | 2023-06-09 | 安超云软件有限公司 | Management method and device of cloud primary integrated machine, electronic equipment and storage medium |
| CN116248484B (en) * | 2023-03-09 | 2024-03-22 | 安超云软件有限公司 | Management method and device of cloud primary integrated machine, electronic equipment and storage medium |
| CN117614805A (en) * | 2023-11-21 | 2024-02-27 | 杭州沃趣科技股份有限公司 | Data processing system for monitoring state of data center |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107147528A (en) | One kind stores gateway intelligently anti-fissure system and method | |
| US7536586B2 (en) | System and method for the management of failure recovery in multiple-node shared-storage environments | |
| US6785678B2 (en) | Method of improving the availability of a computer clustering system through the use of a network medium link state function | |
| EP2281240B1 (en) | Maintaining data integrity in data servers across data centers | |
| US6609213B1 (en) | Cluster-based system and method of recovery from server failures | |
| CN103209095B (en) | Method and device for preventing split brain on basis of disk service lock | |
| JP2005209201A (en) | Node management in high-availability cluster | |
| US20150339200A1 (en) | Intelligent disaster recovery | |
| CN101179432A (en) | Method of implementing high availability of system in multi-machine surroundings | |
| CN103019889A (en) | Distributed file system and failure processing method thereof | |
| JP2004532442A (en) | Failover processing in a storage system | |
| TWI701916B (en) | Method and device for self-recovering management ability in distributed system | |
| US20140095925A1 (en) | Client for controlling automatic failover from a primary to a standby server | |
| CN112181660A (en) | High-availability method based on server cluster | |
| CN103152414A (en) | High available system based on cloud calculation and implementation method thereof | |
| CN103905247B (en) | Two-unit standby method and system based on multi-client judgment | |
| CN101207512A (en) | A Redundant Disaster Backup and Switching Method for Database System | |
| CN107918570B (en) | Method for sharing arbitration logic disk by double-active system | |
| CN104469699B (en) | Cluster quorum method and more cluster coupled systems | |
| CN103257908A (en) | Software and hardware cooperative multi-controller disk array designing method | |
| CN103490914A (en) | Switching system and switching method for multi-machine hot standby of network application equipment | |
| CN114461428A (en) | An arbitration method for split-brain in a two-machine cluster | |
| US10664369B2 (en) | Determine failed components in fault-tolerant memory | |
| CN106254161A (en) | The quickly detection of node failure based on HDFS and restoration methods and system | |
| CN107071189A (en) | A kind of connection method of communication apparatus physical interface |
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 |
Application publication date: 20170908 |
|
| RJ01 | Rejection of invention patent application after publication |