Maintaining a fault-tolerance threshold of a clusterstore during maintenance activities

Abstract

A method for maintaining fault tolerance in a storage cluster is provided. Embodiments include receiving, by a management component associated with a distributed data store on a cluster of host machines, a request to place a first host machine of the cluster of host machines in a maintenance mode, wherein the first host machine stores given data of the distributed data store. Embodiments include determining whether a second host machine that does not currently store any data of the distributed data store exists in the cluster of host machines. Embodiments include determining, based on whether the second host machine exists in the cluster of host machines, whether to transfer the given data of the distributed data store from the first host machine to the second host machine. Embodiments include initiating the maintenance mode on the first host machine.

We claim: 
     
         1 . A method for maintaining fault tolerance in a storage cluster, comprising:
 receiving, by a management component associated with a distributed data store on a cluster of host machines, a request to place a first host machine of the cluster of host machines in a maintenance mode, wherein the first host machine stores given data of the distributed data store;   determining, by the management component, whether to transfer the given data of the distributed data store from the first host machine to a second host machine of the cluster of host machines;   determining, by the management component, a number of failures to tolerate (FTT) of the distributed data store;   performing, based on the number of FTT of the distributed data store, at least one of:
 decrementing, by the management component, the number of FTT of the distributed data store by one, 
 deactivating, by the management component, the distributed data store; or 
 recreating, by the management component, a state associated with the given data of the distributed data store on the second host machine; and 
   after determining whether to transfer the given data of the distributed data store from the first host machine to the second host machine, initiating, by the management component, the maintenance mode on the first host machine.   
     
     
         2 . The method of  claim 1 , wherein the given data of the distributed data store is transferred from the first host machine to the second host machine. 
     
     
         3 . The method of  claim 1 , wherein the determining of whether to transfer the given data of the distributed data store from the first host machine to the second host machine comprises determining that the second host machine does not exist in the cluster of host machines, wherein the given data of the distributed data store is not transferred from the first host machine to the second host machine, and wherein the method further comprises:
 determining, by the management component, that the number of FTT of the distributed data store is greater than one; and   decrementing, by the management component, the number of FTT of the distributed data store by one.   
     
     
         4 . The method of  claim 1 , wherein the determining of whether to transfer the given data of the distributed data store from the first host machine to the second host machine comprises determining that the second host machine does not exist in the cluster of host machines, wherein the given data of the distributed data store is not transferred from the first host machine to the second host machine, and wherein the method further comprises:
 determining, by the management component, that the number of FTT of the distributed data store is zero; and   deactivating, by the management component, the distributed data store.   
     
     
         5 . The method of  claim 1 , wherein the determining of whether to transfer the given data of the distributed data store from the first host machine to the second host machine comprises determining that the second host machine exists in the cluster of host machines, and wherein the method further comprises:
 determining, by the management component, that the number of FTT of the distributed data store is zero; and   recreating, by the management component, the state associated with the given data of the distributed data store on the second host machine.   
     
     
         6 . The method of  claim 1 , further comprising removing, by the management component, the given data from the first host machine before initiating the maintenance mode on the first host machine. 
     
     
         7 . The method of  claim 1 , wherein the distributed data store comprises a clusterstore, and wherein the given data of the distributed data store comprises a replica related to the clusterstore. 
     
     
         8 . A system for maintaining fault tolerance in a storage cluster, comprising:
 at least one memory; and   at least one processor coupled to the at least one memory, the at least one processor and the at least one memory configured to:
 receive, by a management component associated with a distributed data store on a cluster of host machines, a request to place a first host machine of the cluster of host machines in a maintenance mode, wherein the first host machine stores given data of the distributed data store; 
 determine, by the management component, whether to transfer the given data of the distributed data store from the first host machine to a second host machine of the cluster of host machines; 
 determine, by the management component, a number of failures to tolerate (FTT) of the distributed data store; 
 perform, based on the number of FTT of the distributed data store, at least one of: 
 decrement, by the management component, the number of FTT of the distributed data store by one, 
 deactivate, by the management component, the distributed data store; or 
 recreate, by the management component, a state associated with the given data of the distributed data store on the second host machine; and 
 after determining whether to transfer the given data of the distributed data store from the first host machine to the second host machine, initiate, by the management component, the maintenance mode on the first host machine. 
   
     
     
         9 . The system of  claim 8 , wherein the given data of the distributed data store is transferred from the first host machine to the second host machine. 
     
     
         10 . The system of  claim 8 , wherein the determining of whether to transfer the given data of the distributed data store from the first host machine to the second host machine comprises determining that the second host machine does not exist in the cluster of host machines, wherein the given data of the distributed data store is not transferred from the first host machine to the second host machine, and wherein the at least one processor and the at least one memory are further configured to:
 determine, by the management component, that the number of FTT of the distributed data store is greater than one; and   decrement, by the management component, the number of FTT of the distributed data store by one.   
     
     
         11 . The system of  claim 8 , wherein the determining of whether to transfer the given data of the distributed data store from the first host machine to the second host machine comprises determining that the second host machine does not exist in the cluster of host machines, wherein the given data of the distributed data store is not transferred from the first host machine to the second host machine, and wherein the at least one processor and the at least one memory are further configured to:
 determine, by the management component, that the number of FTT of the distributed data store is zero; and   deactivate, by the management component, the distributed data store.   
     
     
         12 . The system of  claim 8 , wherein the determining of whether to transfer the given data of the distributed data store from the first host machine to the second host machine comprises determining that the second host machine exists in the cluster of host machines, and wherein the at least one processor and the at least one memory are further configured to:
 determine, by the management component, that the number of FTT of the distributed data store is zero; and   recreate, by the management component, a state associated with the given data of the distributed data store on the second host machine.   
     
     
         13 . The system of  claim 8 , wherein the at least one processor and the at least one memory are further configured to remove, by the management component, the given data from the first host machine before initiating the maintenance mode on the first host machine. 
     
     
         14 . The system of  claim 8 , wherein the distributed data store comprises a clusterstore, and wherein the given data of the distributed data store comprises a replica related to the clusterstore. 
     
     
         15 . A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to:
 receive, by a management component associated with a distributed data store on a cluster of host machines, a request to place a first host machine of the cluster of host machines in a maintenance mode, wherein the first host machine stores given data of the distributed data store;   determine, by the management component, whether to transfer the given data of the distributed data store from the first host machine to a second host machine of the cluster of host machines;   determine, by the management component, a number of failures to tolerate (FTT) of the distributed data store;   perform, based on the number of FTT of the distributed data store, at least one of:   decrement, by the management component, the number of FTT of the distributed data store by one,   deactivate, by the management component, the distributed data store: or recreate, by the management component, a state associated with the given data of the distributed data store on the second host machine; and   after determining whether to transfer the given data of the distributed data store from the first host machine to the second host machine, initiate, by the management component, the maintenance mode on the first host machine.   
     
     
         16 . The non-transitory computer-readable medium of  claim 15 , wherein determining whether the second host machine exists in the cluster of host machines comprises determining that the second host machine exists in the cluster of host machines, and wherein the given data of the distributed data store is transferred from the first host machine to the second host machine. 
     
     
         17 . The non-transitory computer-readable medium of  claim 15 , wherein the determining of whether to transfer the given data of the distributed data store from the first host machine to the second host machine comprises determining that the second host machine does not exist in the cluster of host machines, wherein the given data of the distributed data store is not transferred from the first host machine to the second host machine, and wherein the instructions, when executed by the one or more processors, further cause the one or more processors to:
 determine, by the management component, the number of FTT of the distributed data store is greater than one; and   decrement, by the management component, the number of FTT of the distributed data store by one.   
     
     
         18 . The non-transitory computer-readable medium of  claim 15 , wherein the determining of whether to transfer the given data of the distributed data store from the first host machine to the second host machine comprises determining that the second host machine does not exist in the cluster of host machines, wherein the given data of the distributed data store is not transferred from the first host machine to the second host machine, and wherein the instructions, when executed by the one or more processors, further cause the one or more processors to:
 determine, by the management component, that the number of FTT of the distributed data store is zero; and   deactivate, by the management component, the distributed data store.   
     
     
         19 . The non-transitory computer-readable medium of  claim 15 , wherein the determining of whether to transfer the given data of the distributed data store from the first host machine to the second host machine comprises determining that the second host machine exists in the cluster of host machines, and wherein the instructions, when executed by the one or more processors, further cause the one or more processors to:
 determine, by the management component, the number of FTT of the distributed data store is zero; and   recreate, by the management component, the state associated with the given data of the distributed data store on the second host machine.   
     
     
         20 . The non-transitory computer-readable medium of  claim 15 , wherein the instructions, when executed by the one or more processors, further cause the one or more processors to remove, by the management component, the given data from the first host machine before initiating the maintenance mode on the first host machine.