Mediator assisted switchover between clusters

Patent metadata

Field	Value
Publication number	US-12032459-B2
Application number	US-202217748277-A
Country	US
Kind code	B2
Filing date	May 19, 2022
Priority date	Nov 20, 2020
Publication date	Jul 9, 2024
Grant date	Jul 9, 2024

Abstract

Techniques are provided for metadata management for enabling automated switchover in accordance with a configuration of storage solution that expresses a preference for either maintaining availability (e.g., a non-zero RPO mode) of the storage solution or avoiding data loss (e.g., a zero RPO mode). In one example, responsive to detecting a switchover trigger event, a node of a local cluster of a cross-site storage solution determines whether performance of an automated switchover from a failed cluster to a surviving cluster of the cross-site storage solution is enabled. Responsive to an affirmative determination, the node selectively proceeds with the automated switchover based on the configuration.

First claim

What is claimed is:

1. A method comprising:
after detecting, by a first node of a plurality of nodes of a first cluster of a cross-site storage solution including a mediator node, the first cluster operable at a first site, and a second cluster operable at a second site and having a plurality of nodes, a switchover trigger event, determining, by the first node, whether performance of an automated switchover from a failed cluster to a surviving cluster of the cross-site storage solution is enabled;
after said determining being affirmative, further determining, by the first node, a switchover mode in which the cross-site storage solution is configured to operate;
after the switchover mode is determined to be indicative of a preference to avoid data loss over maintaining availability of the cross-site storage solution, proceeding, by the first node, with the automated switchover if a local operation log maintained by the first node is in a state of synchronization with a remote operation log maintained by the second cluster; and
after the switchover mode is determined to be indicative of a preference to maintain availability of the cross-site storage solution over avoiding data loss, proceeding, by the first node, with the automated switchover regardless of the state of synchronization between the local operation log and the remote operation log.

2. The method of claim 1 , further comprising prior to the switchover event, enabling or disabling, by the first node, the performance of the automated switchover based on success or failure of execution of a write operation to store metadata, specifying operational state information of the first node and switchover information.

3. The method of claim 2 , further comprising:
obtaining, by the first node, results of an initial quorum vote, before execution of the write operation, wherein the initial quorum vote corresponds to at least one of a first local vote derived from whether a set of local mailbox storage devices are operational, a first remote vote derived from whether a set of remote mailbox storage devices are operational, or a first mediator vote derived from whether a mediator mailbox storage device is operational;
executing, by the first node, the write operation to store the metadata;
performing, by the first node, a final quorum vote, after execution of the write operation, corresponding to at least one of a second local vote derived from whether the set of local mailbox storage devices are operational, a second remote vote derived from whether the set of remote mailbox storage devices are operational, or a second mediator vote derived from whether the mediator mailbox storage device is operational; and
designating, by the first node, the write operation as either successful or failed based upon at least one of the initial quorum vote or the final quorum vote.

4. The method of claim 3 , wherein the metadata is written to the set of local mailbox storage devices owned by the first node and located at the first cluster, the set of remote mailbox storage devices owned by the first node and located at a second storage cluster, and the mediator storage device owned by the first node and located at the mediator node.

5. The method of claim 3 , further comprising determining the first local vote as a first value based upon at least one local mailbox storage device of the set of local mailbox storage devices being operational, otherwise, determining the first local vote as a second value.

6. The method of claim 3 , further comprising determining the first remote vote as a first value based upon at least one remote mailbox storage device of the set of remote mailbox storage devices being operational, otherwise, determining the second remote vote as a second value.

7. The method of claim 3 , further comprising determining the first mediator vote as a first value based upon the mediator mailbox storage device being operational, otherwise, determining the first mediator vote as a second value.

8. The method of claim 3 , further comprising comparing the initial quorum vote to the final quorum vote to determine whether to designate the write operation as successful or failed.

9. The method of claim 3 , further comprising responsive to the switchover trigger event, obtaining, by the first node, a reservation on the mediator mailbox storage device to block a node of the plurality of nodes of the second cluster from performing the switchover operation.

10. A non-transitory machine readable medium having stored thereon instructions, which when executed by a first node of a plurality of nodes of a first cluster of a cross-site storage solution including a mediator node, the first cluster operable at a first site, and a second cluster operable at a second site and having a plurality of nodes, cause the first node to:
after detecting a switchover trigger event, determine whether performance of an automated switchover from a failed cluster to a surviving cluster of the cross-site storage solution is enabled;
after an affirmative determination, further determine a switchover mode in which the cross-site storage solution is configured to operate;
after the switchover mode is determined to be indicative of a preference to avoid data loss over maintaining availability of the cross-site storage solution, proceed with the automated switchover if a local operation log maintained by the first node is in a state of synchronization with a remote operation log maintained by the second cluster; and
after the switchover mode is determined to be indicative of a preference to maintain availability of the cross-site storage solution over avoiding data loss, proceed with the automated switchover regardless of the state of synchronization between the local operation log and the remote operation log.

11. The non-transitory machine readable medium of claim 10 , wherein the instructions further cause the first node to prior to the switchover event, enable or disable the performance of the automated switchover based on success or failure of execution of a write operation to store metadata, specifying operational state information of the first node and switchover information.

12. The non-transitory machine readable medium of claim 11 , wherein the instructions further cause the first node to:
obtain results of an initial quorum vote before execution of the write operation, wherein the initial quorum vote corresponds to at least one of a first local vote derived from whether a set of local mailbox storage devices are operational, a first remote vote derived from whether a set of remote mailbox storage devices are operational, or a first mediator vote derived from whether a mediator mailbox storage device is operational;
execute the write operation to store the metadata;
perform a final quorum vote after execution of the write operation, corresponding to at least one of a second local vote derived from whether the set of local mailbox storage devices are operational, a second remote vote derived from whether the set of remote mailbox storage devices are operational, or a second mediator vote derived from whether the mediator mailbox storage device is operational; and
designate the write operation as either successful or failed based upon at least one of the initial quorum vote or the final quorum vote.

13. The non-transitory machine readable medium of claim 12 , wherein the metadata is written to the set of local mailbox storage devices owned by the first node and located at the first cluster, the set of remote mailbox storage devices owned by the first node and located at a second storage cluster, and the mediator storage device owned by the firs

Assignees

• Netapp Inc

Inventors

• Krishnan Sasidharan
• Arumugham Kalaivani
• Bansal Preksha
• Ekkaladevi Vijay Kumar Chakravarthy
• Bartlett Ryan Edward

Classifications

• G06F11/2028 Primary

  eliminating a faulty processor or activating a spare · CPC title

• G06F11/203

  using migration · CPC title

• G06F11/1658

  Data re-synchronization of a redundant component, or initial sync of replacement, additional or spare unit · CPC title

• G06F11/1435

  using file system or storage system metadata · CPC title

• G06F11/2048

  where the redundant components share neither address space nor persistent storage · CPC title