Cluster Shared Volumes

• US 20090327798A1
• Filed: 06/27/2008
• Published: 12/31/2009
• Est. Priority Date: 06/27/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Das, Rajsekhar, D'Amato, Andrea, Nagar, Rajeev Y., Nishanov, Gor, Maeso, Gregorio
• Granted Patent: US 7,840,730 B2
• Time in Patent Office: Days
• US Class Current: 714/4.1
• CPC Class Codes: G06F 11/1482 by means of middleware or OS functionality
• CPC Class Codes: G06F 11/2035 without idle spare hardware
• CPC Class Codes: G06F 11/2046 where the redundant components share persistent storage G06F11/2043 takes precedence
• CPC Class Codes: G06F 16/13 File access structures, e.g. distributed indices arrangements of input from, or output to, record carriers G06F3/06
• CPC Class Codes: G06F 21/6218 to a system of files or objects, e.g. local or distributed file system or database
• CPC Class Codes: G06F 3/062 Securing storage systems
• CPC Class Codes: G06F 3/0635 by changing the path, e.g. traffic rerouting, path reconfiguration
• CPC Class Codes: G06F 3/0637 Permissions
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

Described is a technology by which a storage volume is shared by cluster nodes of a server cluster. In one implementation, each node includes a redirector that provides shared access to the volume from that node. The redirector routes file system metadata requests from applications and the like through a first (e.g., SMB) communications path to the owning node, and routes file system read and write data to the storage device through a second, high-speed communications path such as direct direct block level I/O. An owning node maintains ownership of the storage device through a persistent reservation mechanism that writes a key to a registration table associated with the storage device. Non-owning nodes write a shared key. The owning node validates the shared keys against cluster membership data, and preempts (e.g., removes) any key deemed not valid. Security mechanisms for controlling access are also described.

Claims

1. In a server cluster environment, a system comprising, a plurality of nodes, and a storage device coupled to the plurality of nodes, the storage device including a volume mounted on one of the nodes, and each other node including a redirector that provides shared access to the volume from that node.

2. The system of claim 1 wherein the node on which the volume is mounted comprises an owning node of the storage device, and each other node comprises a non-owning node, and wherein the owning node includes a mechanism for maintaining ownership of the storage device.

3. The system of claim 2 wherein the mechanism for maintaining ownership of the storage device comprises a persistent reservation mechanism that writes a key indicative of ownership to a registration table associated with the storage device, and each non-owning node writes a shared key to the registration table.

4. The system of claim 3 wherein the owning node validates shared keys against cluster membership data, and preempts any key deemed not valid based upon the cluster membership data.

5. The system of claim 1 wherein the redirector routes file system metadata requests through a first communications path to the owning node, and routes at least some file system read and write data to the volume through a second communications path that comprises a high-speed communications path relative to the first communications path.

6. The system of claim 1 wherein the redirector comprises a root instance that intercepts I/O requests, and a forwarding instance coupled to the root instance that forwards I/O data into a driver.

7. The system of claim 6 wherein the redirector selectively determines whether to route file system read and write data through the first or second communications path based on at least one current condition.

8. The system of claim 6 wherein the redirector maintains a handle table for converting an application'"'"'s virtual handle to an actual file system handle used by the volume, and changes the actual file system handle if the volume moves to a storage device of another node without changing the virtual handle.

9. The system of claim 1 further comprising a security component that uses a token to contain information corresponding to group or privilege data, or both group and privilege data, used to access a file.

10. The system of claim 1 wherein the volume is accessed via a filename corresponding to a single namespace independent of which node provides the filename.

11. In a computing environment, a method comprising,

receiving an I/O request at a first node of a cluster directed towards a volume that is mounted on a second node of the cluster;

and

processing the I/O request, including determining whether the I/O request may be handled by direct I/O to the volume, and if so, by redirecting the I/O request via direct I/O to the volume, and if not, by redirecting the I/O request over a network protocol to the second node for handling the I/O request.

12. The method of claim 11 wherein the second node owns the storage device corresponding to the volume, and further comprising, maintaining ownership of the storage device at the second node via a persistent reservation mechanism.

13. The method of claim 11 wherein determining whether the I/O request may be handled by direct I/O comprises determining whether the I/O request corresponds to metadata or read/write data, or determining whether the I/O request corresponds to buffered or unbuffered I/O, or both determining whether the I/O request corresponds to metadata or read/write data, and determining whether the I/O request corresponds to buffered or unbuffered I/O.

14. The method of claim 11 wherein processing the I/O request comprises detecting a host bus adapter failure to determine that I/O request may not be handled by direct I/O, pausing the I/O request, switching to an alternative path, and resuming the I/O request.

15. The method of claim 11 wherein determining whether the I/O request may be handled by direct I/O comprises determining a condition of a direct I/O communications path, or determining whether to communicate I/O data over the network protocol based on a throttling condition.

16. The method of claim 11 wherein the request is handled by direct I/O, and further comprising, determining whether the I/O request is of a type that may modify block allocation of a file corresponding to the I/O request, and if so, obtaining a block allocation of the file.

17. The method of claim 11 further comprising maintaining a handle table for converting an application'"'"'s virtual handle to an actual file handle, detecting failure of a node hosting a storage device corresponding to the volume or a host bust adapter on a node owning the storage device corresponding to the volume, and changing the actual file system handle without changing the virtual handle if the volume moves to a storage device of another node.

18. The method of claim 11 wherein the I/O request corresponds to metadata used for creating or opening a file, and further comprising, processing security information to grant access to the file.

19. One or more computer-readable media having computer-executable instructions, which when executed perform steps, comprising, persisting ownership of a storage device in a cluster having a cluster shared volume corresponding to that storage device, including writing a owner key into a data structure associated with the shared storage device, evaluated shared keys in the data structure against cluster membership data, and preempting any key that corresponds to a cluster node that is not identified as a cluster member via the cluster membership data.

20. The one or more computer-readable media of claim 19 having further computer-executable instructions comprising, using the shared keys to determine whether to grant access to the cluster shared volume.