Namespace mirroring in an expandable storage volume

• US 9,684,571 B2
• Filed: 05/01/2013
• Issued: 06/20/2017
• Est. Priority Date: 05/01/2013
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Modukuri, Kiran Kumar, Chen, Su
• Primary Examiner(s): MORRISON, JAY A
• Application Number:
  US13/875,236
• Publication Number:
  US 20140330787A1
• Time in Patent Office: 1,511 Days
• Field of Search: None
• CPC Class Codes: G06F 11/1435 using file system or storage system metadata
• CPC Class Codes: G06F 11/20 using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
• CPC Class Codes: G06F 11/2056 by mirroring
• CPC Class Codes: G06F 2201/84 Using snapshots, i.e. a logical point-in-time copy of the data

Abstract

Technology for maintaining a backup of namespace metadata of an expandable storage volume is disclosed. In various embodiments, the expandable storage volume backs up metadata of a namespace constituent volume of the expandable storage volume into a namespace mirror volume. The namespace constituent volume is responsible for storing the metadata for data objects stored in multiple data constituent volumes of the expandable storage volume. In response to a signal indicating that the namespace constituent volume is unavailable, the namespace mirror volume replaces the role of the namespace constituent volume. The new namespace constituent volume continues to provide metadata for a data object of the data objects in response to an operation request for the data object.

Claims

1. A method comprising:

backing up metadata of a namespace constituent volume into a first namespace mirror volume, wherein the namespace constituent volume stores the metadata for data objects stored in data constituent volumes that are separate from the namespace constituent volume, wherein the first namespace mirror volume is separate from the namespace constituent volume and separate from the data constituent volumes;

determining to provide metadata indicated in a first metadata request from the namespace constituent volume for a first data object of the data objects based, at least in part, on a comparison of a frequency of access operations on the namespace constituent volume with a frequency of access operations on the first namespace mirror volume;

based on an indication that the namespace constituent volume is unavailable, redirecting the first metadata requests to the first namespace mirror volume instead of the namespace constituent volume; and

providing metadata for the first data object from the first namespace mirror volume.

2. The method of claim 1, wherein backing up the metadata of the namespace constituent volume into the first namespace mirror volume comprises:

copying a redirector object for a second data object of the data objects from the namespace constituent volume to the first namespace mirror volume, wherein the redirector object includes an object locator indicating a location of the second data object within a data constituent volume of the data constituent volumes.

3. The method of claim 1, wherein backing up the metadata of the namespace constituent volume into the first namespace mirror volume comprises:

backing up the metadata of the namespace constituent volume into the first namespace mirror volume according to a predetermined schedule.

4. The method of claim 1, wherein an expandable storage volume comprises the namespace constituent volume, the first namespace mirror volume, and the data constituent volumes.

5. The method of claim 1, wherein the metadata are organized as inode files stored in the namespace constituent volume, wherein each of the inode files includes metadata for a data object of the data objects.

6. The method of claim 1, wherein the first namespace mirror volume is located in a high-availability pair separate from another high-availability pair that contains the namespace constituent volume.

7. The method of claim 4, further comprising:

based on an operation to set the expandable storage volume offline, stopping the backing up the metadata of the namespace constituent volume to the first namespace mirror volume.

8. The method of claim 1, further comprising:

identifying a stub file for the first data object based, at least in part, on the metadata for the first data object; and

locating the first data object based, at least in part, on an object locator stored in the stub file.

9. The method of claim 1, further comprising:

adjusting a backup schedule based, at least in part, on a workload of the namespace constituent volume, wherein the metadata of the namespace constituent volume is backed up into the first namespace mirror volume according to the backup schedule.

10. The method of claim 1, wherein backing up the metadata of the namespace constituent volume into the first namespace mirror volume comprises:

backing up a portion of metadata of the namespace constituent volume into the first namespace mirror volume, wherein the portion of the metadata corresponds to an active file system including multiple data constituent volumes.

11. The method of claim 1, wherein backing up the metadata of the namespace constituent volume into the first namespace mirror volume comprises, based on receipt of a request to write first metadata to the namespace constituent volume, writing the first metadata to the namespace constituent volume and the first namespace mirror volume.

12. The method of claim 1, further comprising:

monitoring available storage space for a first aggregate, wherein the namespace constituent volume and the first namespace mirror volume are located on the first aggregate; and

based on determining that the available storage space for the first aggregate is insufficient to store the namespace constituent volume and the first namespace mirror volume, migrating the first namespace mirror volume to a second aggregate.

13. A non-transitory computer-readable storage medium having stored thereon instructions comprising machine executable code which, when executed by at least one machine, causes the at least one machine to:

in response to an instruction for generating a snapshot for an expandable storage volume having a namespace constituent volume, multiple data constituent volumes, and a first namespace mirror volume, generate snapshots for a namespace constituent volume and multiple data constituent volumes of the expandable storage volume without generation of a snapshot for the first namespace mirror volume, wherein the namespace constituent volume stores metadata for data objects stored in the multiple data constituent volumes;

in response to an instruction for restoring the expandable storage volume, copy metadata in the snapshot of the namespace constituent volume to the namespace constituent volume of the expandable storage volume, and restore the data constituent volumes to the snapshots for the data constituent volumes; and

after copying metadata to the namespace constituent volume,

create a second namespace mirror volume as a backup for the namespace constituent volume;

back up the metadata of the namespace constituent volume into the second namespace mirror volume; and

determine whether to direct requests for metadata to the second namespace mirror volume based, at least in part, on a comparison of a frequency of access operations on the namespace constituent volume with a frequency of access operations on the second namespace mirror volume.

14. The non-transitory computer-readable storage medium of claim 13, wherein the namespace constituent volume includes inode files for data objects stored in the data constituent volumes, wherein the inode files include pointers to stub files, wherein each of the stub files maintain an object locator for identifying aggregates and storage nodes that store a data object of the data objects.

15. A device comprising:

a processor; and

a memory containing machine readable medium having stored thereon instructions which, when executed by the processor, cause the device to,

back up metadata of a namespace constituent volume into a first namespace mirror volume, wherein the namespace constituent volume stores the metadata for data objects stored in multiple data constituent volumes that are separate from the namespace constituent volume, wherein the first namespace mirror volume is separate from the namespace constituent volume and separate from the data constituent volumes;

determine to provide metadata indicated in a first metadata request from the namespace constituent volume for a first data object of the data objects based, at least in part, on a comparison of a frequency of access operations on the namespace constituent volume with a frequency of access operations on the first namespace mirror volume;

based on an indication that the namespace constituent volume is unavailable, redirect the first metadata requests to the first namespace mirror volume instead of the namespace constituent volume; and

provide metadata for the first data object from the first namespace mirror volume.

16. The device of claim 15, wherein an expandable storage volume includes the namespace constituent volume, the first namespace mirror volume, and the data constituent volumes, and each of the data constituent volumes is stored on a node of a plurality of nodes of a network storage cluster.

17. The device of claim 15, wherein the metadata for the data objects include pointers to stub files, wherein the stub files include object locators pointing to locations storing content of the data objects.

18. The device of claim 15, further comprising instructions which, when executed by the processor, cause the device to:

adjust a backup schedule based, at least in part, on a workload of the namespace constituent volume, wherein the metadata of the namespace constituent volume is backed up into the first namespace mirror volume according to the backup schedule.

19. The device of claim 15, wherein the instructions which, when executed by the processor, cause the device to back up the metadata of the namespace constituent volume into the first namespace mirror volume comprises instructions which, when executed by the processor, cause the device to, based on receipt of a request to write first metadata to the namespace constituent volume, writing the first metadata to the namespace constituent volume and the first namespace mirror volume.

20. The device of claim 15, further comprising instructions which, when executed by the processor, cause the device to:

monitor available storage space for a first aggregate, wherein the namespace constituent volume and the first namespace mirror volume are located on the first aggregate; and

based on a determination that the available storage space for the first aggregate is insufficient to store the namespace constituent volume and the first namespace mirror volume, migrate the first namespace mirror volume to a second aggregate.