Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Method for preventing split brain phenomenon from occurring on distributed replicated block device (DRBD) resource

A technology of disk mirroring and brain splitting, which is applied to the generation of response errors, error detection of redundant data in calculations, digital transmission systems, etc. It can solve problems such as user data loss and failure to guarantee user data security, and achieve prevention Effect of loss of written user data

Active Publication Date: 2012-05-16
CHINA STANDARD SOFTWARE
View PDF5 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] It can be seen that this split-brain automatic recovery strategy cannot guarantee the security of user data, that is, no matter which node is selected as the "sacrifice" node, there is always a possibility that data will be written to both nodes during a network failure. In this way, more or less, the user's data will always be lost

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method for preventing split brain phenomenon from occurring on distributed replicated block device (DRBD) resource
  • Method for preventing split brain phenomenon from occurring on distributed replicated block device (DRBD) resource
  • Method for preventing split brain phenomenon from occurring on distributed replicated block device (DRBD) resource

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0039] Embodiment 1: Assume that the network cable connected to DRBD on the DRBD master node is pulled out at a certain moment.

[0040] 1) If the method described in the present invention is not used, at this time, the DRBD nodes lose contact because the network cable is pulled out, and the nodes do not know each other's status, and the DRBD slave node will think that the master node is down and will Become the master node of DRBD by yourself, so that two DRBD master nodes will appear in the cluster at the same time, and user data will be written to two points at the same time without synchronization. Data written to the DRBD device during the failure is at risk of being lost when the network returns to normal.

[0041] 2) In the method described in the present invention (that is, using the modified OCF script of DRBD), when the cluster calls the drbd_monitor function again, the state on the DRBD master node is WFConnetion / Primary ( figure 2 Branch 5), and then call the fun...

Embodiment 2

[0044] Embodiment 2: If the switch connected to the DRBD network is down, that is, the network cables equivalent to the DRBD master and slave nodes are pulled out at the same time, at this time the DRBD master and slave nodes go to figure 2 Branches 5 and 6, call the check_network_status function to find out that there is a problem with the local network. Next, because the master node is to make the slave node standby, and the slave node is to make itself standby, the phenomenon and result of this situation are equivalent to the above Two cases of unplugging the network cable.

[0045] figure 2 The branch 1 and branch 2 are used to deal with the network failure caused by running the command "ifdown DRBD corresponding network card device" on the DRBD master and slave nodes respectively. The user fixes the failure and then manually enables the standby node. After the standby node is enabled, it will go to figure 2 Go up branch 3 or branch 4 as shown.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a method for preventing a split brain phenomenon from occurring on a distributed replicated block device (DRBD) resource. According to the method, through modifying a OCF(Open Cluster Framework) script of a DRBD, the connection status and the network status of the DRBD resource are monitored; a node of a cluster is reversed before the split brain phenomenon occurs; data cannot be written into the reversed node during the stoppage period; and all data are written into another node, and thereby the occurrence of the split brain phenomenon of the DRBD is completely eradicated and the integrity of user data is protected.

Description

technical field [0001] The invention relates to a method for preventing split-brain of disk image resources, in particular to a method for dealing with the split-brain problem of disk image (DRBD) resources in a high-availability cluster. Background technique [0002] Disk mirroring (DRBD, Distributed Replicated Block Device) is a distributed storage system on the Linux platform. It contains core modules, several user management tools and scripts, and is usually used in high availability (HA, high availability) clusters. DRBD is similar to RAID 1 (mirroring) of a disk array, except that RAID 1 is in the same computer, while DRBD uses the network to make partitions in different computers mirror. [0003] DRBD regards the device (partition) it manages as a resource, and each resource has two roles: primary (Primary) and secondary (Secondary). For the DRBD device of the Primary role, DRBD allows operations such as reading, writing, and mounting; the DRBD device of the Seconda...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): H04L12/24G06F11/14
Inventor 兰雨晴梁昕刘涛张永军夏颖
Owner CHINA STANDARD SOFTWARE
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com